Evolution and the Social Mind: Evolutionary Psychology and Social Cognition

  • 46 552 9
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up

Evolution and the Social Mind: Evolutionary Psychology and Social Cognition

EVOLUTION AND THE SOCIAL MIND The Sydney Symposium of Social Psychology series This book is Volume 9 in the Sydney Sym

3,568 896 2MB

Pages 345 Page size 432 x 648 pts

Report DMCA / Copyright


Recommend Papers

File loading please wait...
Citation preview


The Sydney Symposium of Social Psychology series This book is Volume 9 in the Sydney Symposium of Social Psychology series. The aim of the Sydney Symposia of Social Psychology is to provide new, integrative insights into key areas of contemporary research. Held every year at the University of New South Wales, Sydney, each symposium deals with an important integrative theme in social psychology, and the invited participants are leading researchers in the field from around the world. Each contribution is extensively discussed during the symposium and is subsequently thoroughly revised into book chapters that are published in the volumes in this series. For further details see the website at www.sydneysymposium.unsw.edu.au Previous Sydney Symposium of Social Psychology volumes: SSSP 1. FEELING AND THINKING: THE ROLE OF AFFECT IN SOCIAL COGNITION ** ISBN 0-52164223-X (Edited by J. P. Forgas). Contributors: Robert Zajonc, Jim Blascovich, Wendy Berry Mendes, Craig Smith, Leslie Kirby, Eric Eich, Dawn Macaulay, Len Berkowitz, Sara Jaffee, EunKyung Jo, Bartholomeu Troccoli, Leonard Martin, Daniel Gilbert, Timothy Wilson, Herbert Bless, Klaus Fiedler, Joseph P. Forgas, Carolin Showers, Anthony Greenwald, Mahzarin Banaji, Laurie Rudman, Shelly Farnham, Brian Nosek, Marshall Rosier, Mark Leary, Paula Niedenthal, Jamin Halberstadt. SSSP 2. THE SOCIAL MIND: COGNITIVE AND MO TIVATIONAL ASPECTS OF INTERPERSONAL BEHAVIOR** ISBN 0-521-77092-0 (Edited b y J. P. Forgas, K. D. Williams, & L. Wheeler). Contributors: William & Claire McGuire , Susan Andersen, Ro y Baumeister, Joel Cooper, Bill Cr ano, Garth Fletcher, Joseph P. Forgas, Pascal Huguet, Mike Hogg, Mar tin Kaplan, Norb Kerr, John Nezlek, Fred Rhodewalt, Astrid Schuetz, Constantine Sedikides , Jeffry Simpson, Richard Sorrentino , Dianne Tice, Kip Williams, Ladd Wheeler. SSSP 3. SOCIAL INFLUENCE: DIRECT AND INDIRECT PR OCESSES* ISBN 1-84169-038-4 (Edited by J. P. Forgas & K. D. Williams). Contributors: Robert Cialdini, Eric Knowles, Shannon Butler, Jay Linn, Bibb Latané, Martin Bourgeois, Mark Schaller, Ap Dijksterhuis, James Tedeschi, Richard Petty, Joseph P. Forgas, Herbert Bless, Fritz Strack, Eva Walther, Sik Hung Ng, Thomas Muss weiler, Kipling Williams, Lara Dolnik, Charles Stangor, Gretchen Sechrist, John Jost, Deborah Terry, Michael Hogg, Stephen Har kins, Barbara David, John Turner, Robin Mar tin, Miles Hewstone, Russell Spears, Tom Postmes, Martin Lea, Susan Watt. SSSP 4. THE SOCIAL SELF: COGNITIVE, INTERPERSONAL, AND INTERGROUP PERSPECTIVES** ISBN 1-84169-062-7 (Edited b y J. P. Forgas & K. D. Williams). Contributors: Eliot R. Smith, Thomas Gilovich, Monica Biernat, Joseph P. Forgas, Stephanie J. Moylan, Edward R. Hirt, Sean M. McCrea, Frederick Rhodewalt, Michael Tragakis, Mark Leary, Roy F. Baumeister, Jean M. Twenge, Natalie Ciarocco , Dianne M. Tice, Jean M. Twenge, Brandon J. Schmeichel, Ber tram F. Malle, William Ic kes, Marianne LaFrance, Yoshihisa Kashima, Emiko Kashima, Anna Clark, Marilynn B. Brewer, Cynthia L. Pickett, Sabine Otten, Christian S. Crandall, Diane M. Mackie, Joel Cooper, Michael Hogg, Stephen C. Wright, Art Aron, Linda R. Tropp, Constantine Sedikides. SSSP 5. SOCIAL JUDGMENTS: IMPLICIT AND EXPLICIT PR OCESSES** ISBN 0-521-82248-3. (Edited by J. P. Forgas, K. D. Williams, & W. von Hippel). Contributors: Herbert Bless, Marilynn Brewer, David Buss, Tanya Chartrand, Klaus Fiedler, Joseph P. Forgas, David Funder, Adam Galinsky, Martie Haselton, Denis Hilton, Lucy Johnston, Ar ie Kruglanski, Matthew Lieberman, John McClure , Mario

Mikulincer, Norbert Schwarz, Philip Sha ver, Diederik Stapel, Jerr y Suls, William v on Hippel, Michaela Waenke, Ladd Wheeler, Kipling Williams, Michael Zarate. SSSP 6. SOCIAL MO TIVATION: CONSCIOUS AND UNCONSCIOUS PR OCESSES** ISBN 0-52183254-3 (Edited by J. P. Forgas, K. D. Williams, & S. M. Laham). Contributors: Henk Aar ts, Ran Hassin, Trish Devine, Joseph P. Forgas, Jens F orster, Nira Liberman, Judy Harackiewicz, Leanne Hing, Mar k Zanna, Michael Kernis, Paul Lewicki, Steve Neuberg, Doug Kenrick, Mark Schaller, Tom Pyszczynski, Fred Rhodewalt, Jonathan Schooler, Steve Spencer, Fritz Strack, Roland Deutsch, Howard Weiss, Neal Ashkanasy, Kip Williams , Trevor Case, Wayne Warburton, Wendy Wood, Jeffrey Quinn, Re x Wright, Guido Gendolla. SSSP 7. THE SOCIAL OUTCAST: OSTRACISM, SOCIAL EXCLUSION, REJECTION, AND BULLYING* ISBN 1-84169-424-X (Edited b y K. D. Williams, J. P. Forgas, & W. von Hippel). Contributors: Kipling D. Williams, Joseph P. Forgas, William von Hippel, Lisa Zadro , Mark R. Leary, Roy F. Baumeister, and C. Nathan DeWall, Geoff MacDonald, Rachell Kingsb ury, Stephanie Sha w, John T. Cacioppo, Louise C . Hawkley, Naomi I. Eisenberger, Matthew D. Lieberman, Rainer Romero-Canyas, Geraldine Downey, Jaana Juvonen, Elisheva F. Gross, Kristin L. Sommer, Yonata Rubin, Susan T. Fiske, Mariko Yamamoto, Jean M. Twenge, Cynthia L. Pickett, Wendi L. Gardner, Megan Knowles, Michael A. Hogg, Julie Fitness, Jessica L. Lakin, Tanya L. Chartrand, Kathleen R. Catanese and Dianne M. Tice, Lowell Gaertner, Jonathan Iuzzini, Jaap W. Ouwerkerk, Norbert L. Kerr, Marcello Gallucci, Paul A. M. Van Lange, Marilynn B. Brewer. SSSP 8. AFFECT IN SOCIAL THINKING AND BEHA VIOR* ISBN 1-84169-454-1 (Edited b y J. P. Forgas). Contributors: Joseph P. Forgas, Carrie L. Wyland, Simon M. Laham, Mar tie G. Haselton, Timothy Ketelaar, Piotr Winkielman, John T. Cacioppo, Herbert Bless, Klaus Fiedler, Craig A. Smith, Bieke David, Leslie D. Kirby, Eric Eich, Dawn Macaulay, Gerald L. Clore, Justin Storbeck, Roy F. Baumeister, Kathleen D. Vohs, Dianne M. Tice, Dacher Keltner, E. J. Horberg, Christopher Oveis, Elizabeth W. Dunn, Constantine Sedikides, Tim Wildschut, Jamie Arndt, Clay Routledge, Yaacov Trope, Eric I. Igou, Christopher T. Burke, Felicia A. Huppert, Ralph Erber, Susan Markunas, Joseph Ciarrochi, John T. Blackledge, Janice R. Kelly, Jennifer R. Spoor, John G. Holmes, Danu B. Anthony.

* Published by Psychology Press; ** Published by Cambridge University Press

EVOLUTION AND THE SOCIAL MIND Evolutionary Psychology and Social Cognition

Edited by

Joseph P. Forgas University of New South Wales

Martie G. Haselton University of California at Los Angeles

William von Hippel University of New South Wales

Published in 2007 by Psychology Press 270 Madison Avenue New York, NY 10016 www.psypress.com

Published in Great Britain by Psychology Press 27 Church Road Hove, East Sussex BN3 2FA www.psypress.com

Copyright © 2007 by Psychology Press

This edition published in the Taylor & Francis e-Library, 2011. To purchase your own copy of this or any of Taylor & Francis or Routledge’s collection of thousands of eBooks please go to ww w.eBookstore.tandf.co.uk. Psychology Press is an imprint of the Taylor & Francis Group, an informa business Cover design by Lisa Dynan All rights reserved. No part of this book may be reprinted or reproduced or utilized in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Library of Congress Cataloging-in-Publication Data A catalog record for this book is available from the Library of Congress.

ISBN 0-203-83778-9 Master e-book ISBN

ISBN: 978-1-84169-458-0 (hbk)

Contents About the Editors Contributors Preface 1

xiii xv xvii

Evolutionary Psychology and Social Thinking: History, Issues, and Prospects


William von Hippel, Martie G. Haselton, and Joseph P. Forgas Evolutionary Psychology and Social Cognition A Natural Affinity / 3 Different “Why” Questions / 5 Content-Specificity and Adaptive Design / 6 Refining the Metaphor for Social Cognition / Interface with Modern Evolutionary Biology / What can Social Cognition Bring to the Table? Overview of the Book / 14



7 8 / 9


The Social Brain Hypothesis and its Relevance to Social Psychology


R. I. M. Dunbar The Social Brain Hypothesis / 25 The Structure of Human Social Networks / 26 Two Specializations of Social Cognition / 28 3

The Evolution of Social Inference Processes: The Importance of Signaling Theory Steven W. Gangestad and Randy Thornhill Signaling Systems / 34 Processing Incidental Effects Deception / 43 Conclusion / 45





How the Mind Warps: 49 A Social Evolutionary Perspective on Cognitive Processing Disjunctions Douglas T. Kenrick, Andrew W. Delton, Theresa E. Robertson, D. Vaughn Becker, and Steven L. Neuberg Our Basic Model of How Fundamental Motives Influence Cognitive Processes / 51 An Unexpected Disjunction between Visual Attention, Frequency Estimation, and Memory / 51 An Opposite Disjunction for Outgroup Males / 57 Suppression and Amplification / 58 Disjunctions’ Functions / 60 A General Model of the Biases Underlying Disjunctions / 62 Some Empirical Implications of Considering Disjunctions in Evolutionary/Ecological Terms / 63 Conclusion / 66 II. THE EVOLUTIONARY PSYCHOLOGY OF AFFECT AND COGNITION


Appraisals, Emotions, and Adaptation


Phoebe C. Ellsworth Theories of Emotion / 73 Implications of Appraisal Theories / 81 Ambiguous Situations and Incomplete Emotions / 83 How Different are Modular and Appraisal Theories, Really? 6



The Evolutionary Bases of Social and Moral Emotions: Dominance, Submission, and True Love


Ross Buck Motivation, Emotion, and Communication: The Developmental-Interactionist View / 89 Altruism / 95 Attachment and Higher-Level Social and Moral Emotions / 99 Conclusions / 102 7

The Strange Cognitive Benefits of Mild Dysphoria: On the Evolutionary Advantages of Not Being Too Happy Joseph P. Forgas Introduction / 107 The Evolutionary Functions of Affect / 108 Contemporary Cognitive Approaches / 110 The Empirical Evidence / 111 The Interpersonal Benefits of Negative Affect / Conclusions / 121





Evolution, Social Cognition, and Depressed Mood: Exploring the Relationship Between Depression and Social Risk Taking


Paul B. T. Badcock and Nicholas B. Allen Introduction / 125 Darwinian Models of Depressed Mood / 126 Theories of Resource/Energy Conservation / 126 Social Theories of the Evolution of Depression / 127 The Social Risk Hypothesis: An Integrative View / 130 Depression and Cognition About Social Risk / 133 Depression and Reduced Social Risk Taking / 133 Recent Studies on Depression and Risk Propensity / 134 Discussion and Conclusions / 137


Coevolved Cognitive Mechanisms in Mate Search: Making Decisions in a Decision-shaped World


Peter M. Todd Searching for a Space / 145 The Big Picture: Ecological Rationality / 146 Sequential Decision Making in Mate Choice / 148 Strategies for Mutual Mate Search / 152 Summary and Connections / 156 10

An Evolutionary Account of Strategic Pluralism in Human Mating: Changes in Mate Preferences Across the Ovulatory Cycle


Jeffry A. Simpson and Jonathon LaPaglia Strategic Pluralism and Human Mating: Patterned Changes in Women’s Mate Preferences Across the Ovulatory Cycle / 161 Basic Evolutionary Concepts / 162 Mating Strategies in Humans / 164 Study 1 / 166 Study 2 / 169 Broader Theoretical Considerations / 173 11

Aligning Evolutionary Psychology and Social Cognition: Inbreeding Avoidance as an Example of Investigations into Categorization, Decision Rules, and Emotions


Debra Lieberman Introduction / 179 What is a Computational Theory of Mind and Why Should Social-Cognitive Scientists Care? / 181


x CONTENTS Selection Pressures Guiding the Evolution of Inbreeding Avoidance Mechanisms / 182 An Information-Processing View of Inbreeding Avoidance: What Would a Well-Engineered System for Inbreeding Avoidance Look Like? / 184 Empirical Investigation of Systems for Inbreeding Avoidance / 190 Conclusion / 191 12

The Self in Intimate Relationships: A Social Evolutionary Account


Garth J. O. Fletcher and Nickola C. Overall A Conceptual and Methodological Backdrop / The Role of the Self in Mate Selection / 197 The Self Never Sleeps / 205 Conclusions / 207



A Social Cognitive Evolutionary Approach to Jealousy: The Automatic Evaluation of One’ s Romantic Rivals


Abraham P. Buunk, Karlijn Massar, and Pieternel Dijkstra The Importance of Jealousy / 213 Inventory of Relevant Rival Characteristics / 214 Experimentally Manipulating Rival Characteristics / Body Build / 219 Sexual Versus Emotional Infidelity / 222 Conclusion / 225 14


Cognitive and Social Adaptations for Leadership and Followership: Evolutionary Game Theory and Group Dynamics


Mark van Vugt and Rob Kurzban Leadership and Followership in an Evolutionary Framework An Evolutionary Game Analysis of Leadership / 231 Non-human Evidence for Leadership / 235 Leadership in Humans / 236 Discussion / 238 15



Proximate and Ultimate Origins of a Bias for Prototypical Faces: An Evolutionary Social Cognitive Account Jamin Halberstadt The Prototypicality Bias as an Adaptation / 247 Domain Specificity of the Prototypicality Bias / 248 The Prototypicality Bias as a Side-Effect / 250 Conclusion / 258




The Social Prediction Dynamic: A Legacy of Cognition and Mixed Motives


Oscar Ybarra, Matthew C. Keller, Emily Chan, Andrew S. Baron, Jeffrey Hutsler, Stephen M. Garcia, Jeffrey Sanchez-Burks, and Kimberly Rios Morrison Social Pressures on Cognitive Evolution / 264 On Social Scientists Trying to Predict People / 265 The Averseness of Being the Target of Prediction / 268 The Social Prediction Dynamic: Delineating the Theoretical Framework Situational Factors Moderate the Need to be Unpredictable / 272 Conclusion / 275 17

The Evolution of an Ostracism Detection System




Jennifer R. Spoor and Kipling D. Williams Introduction / 279 The Evolutionary Importance of Inclusion in Groups / 280 Model of Ostracism Detection / 281 Experimental Tests of the Ostracism Detection System / 283 The Indiscriminate Ostracism Detection System / 286 Implications and Conclusions / 288 18

The Behavioral Immune System: Its Evolution and Social Psychological Implications


Mark Schaller and Lesley A. Duncan The Past / 295 The Present / 299 The Future / 305 Author Index Subject Index

308 320


About the Editors Joseph P. Forgas received his DPhil and subsequently a DSc from the University of Oxford. He is currently Scientia Professor Psychology at the University of New South Wales, Sydney, Australia. He has also spent various periods of time working at the Universities of Giessen, Heidelberg, Stanford, Mannheim, and Oxford. His enduring interest is in studying the role of cognitive and affective processes in interpersonal behavior. His current projects investigate how mood states can influence everyday social judgments and interaction strategies. He has published some 17 books and more than 130 articles and chapters in this area. He has been elected Fellow of the Academy of Social Sciences in Australia, the Hungarian Academy of Sciences, the American Psychological Society, and the Society for Personality and Social Psychology. He is recipient of the Alexander von Humboldt Research Prize (Germany) and a Special Investigator Award, and a Professorial Research Fellowship from the Australian Research Council. Martie G. Haselton received her PhD in Psychology in the year 2000 from the University of Texas, Austin. She is currently associate professor of Communication Studies and Psychology at the University of California, Los Angeles. She serves as co-editor in chief at Evolution and Human Behavior, board member of the UCLA Center for Behavior, Evolution, and Culture, associate director of the NSF-funded Interdisciplinary Relationship Science Program at UCLA, and elected member at large of the Human Behavior and Evolution Society. She has dozens of scientific publications on bias in social judgment, psychological changes in women across the ovulatory cycle, and the evolution of human mate choice and sexuality. William von Hippel received his BA from Yale University, and his PhD from the University of Michigan. He taught at Ohio State University for 11 years prior to coming to the University of New South Wales in 2001. He has over 50 publications and served as Associate Editor of Psychological Science and Personality and Social Psychology Bulletin. His main research areas include social cognitive aging, executive functioning, and evolutionary psychology.


Contributors Nicholas B. Allen, University of Melbourne, Australia

Joseph P. Forgas, University of New South Wales, Sydney, Australia

Paul B. T. Badcock, University of Melbourne, Australia

Steven W. Gangestad, University of New Mexico, Albuquerque, USA

Andrew S. Baron, Harvard University, Cambridge, USA

Stephen M. Garcia, University of Michigan, Ann Arbor, USA

D. Vaughn Becker, Arizona State University, Polytechnic Campus, Mesa, USA

Jamin Halberstadt, University of Otago, Dunedin, New Zealand

Ross Buck, University of Connecticut, Storrs, USA Abraham P. Buunk, University of Groningen, The Netherlands

Martie G. Haselton, University of California, Los Angeles, USA Jeffrey Hutsler, University of Nevada, Reno, USA

Emily Chan, Colorado College, Colorado Springs, USA

Matthew C. Keller, Virginia Commonwealth University, Richmond, USA

Andrew W. Delton, University of California, Santa Barbara, USA

Douglas T. Kenrick, Arizona State University, Tempe, USA

Pieternel Dijkstra, University of Groningen, The Netherlands

Rob Kurzban, University of Pennsylvania, Philadelphia, USA

R. I. M. Dunbar, University of Liverpool, Liverpool, UK

Jonathon LaPaglia, University of Minnesota, Minneapolis, USA

Lesley A. Duncan, University of British Columbia, Vancouver, Canada

Debra Lieberman, University of Hawaii, Honolulu, USA

Phoebe C. Ellsworth, University of Michigan, Ann Arbor, USA

Karlijn Massar, University of Groningen, The Netherlands

Garth J. O. Fletcher, University of Canterbury, New Zealand

Kimberly Rios Morrison, Stanford University, Stanford, USA xv


Steven L. Neuberg, Arizona State University, Tempe, USA

Randy Thornhill, University of New Mexico, Albuquerque, USA

Nickola C. Overall, University of Auckland, New Zealand

Peter M. Todd, Indiana University, Bloomington, USA, and Max Planck Institute, Berlin, Germany

Theresa E. Robertson, University of California, Santa Barbara, USA Jeffrey Sanchez-Burks, University of Michigan, Ann Arbor, USA Mark Schaller, University of British Columbia, Vancouver, Canada Jeffry A. Simpson, University of Minnesota, Minneapolis, USA Jennifer R. Spoor, Butler University, Indianapolis, USA

Mark van Vugt, University of Kent, Canterbury, UK William von Hippel, University of New South Wales, Sydney, Australia Kipling D. Williams, Purdue University, West Lafayette, USA Oscar Ybarra, University of Michigan, Ann Arbor, USA

Preface he collection of chapters in this book seeks to combine the study of human social cognition—the way we think, decide, plan and analyse social situations—with an evolutionary framework that considers these activities in light of evolutionary adaptations for solving problems of survival and reproduction faced by our ancestors over thousands of generations. The very suggestion that social thinking is enabled by evolved psychological adaptations is controversial. People often bridle at the suggestion that their most intimate thoughts and decisions—the partners they choose, the people they like, the courses of action they prefer—are tied in important ways to their biology and evolutionary heritage. This book seeks to provide an up-to-date integration of some of the most recent developments in social cognition with research on evolutionary influences on social thinking. Arguably, one of the most intriguing recent developments in social psychology has been the growing recognition of evolutionary theories as relevant to explaining many kinds of complex human social behaviours. The key objective of this book is to provide an informative, scholarly yet readable overview of recent advances in research on the application of evolutionary psychology to the study of social cognition, and to offer a closer integration between these two, hitherto largely separate research traditions. The chapters included here all argue that a complete understanding of social cognitive processes is inconceivable without paying full attention to the way human thinking and decision making have been shaped by the manifold demands of the ancestral environment our forebears faced. Indeed, as Robin Dunbar (chapter 2) argues, the very evolution of that marvellous computational organ, the human brain, is intimately tied to the social cognitive demands of co-ordinating and negotiating interactive behaviours in human groups of ever-increasing size and complexity. The chapters offer important new insights into the way everyday social thinking operates, and how evolutionary principles can illuminate such intriguing questions as: Why do people find prototypical faces more attractive than atypical ones? Why do people behave in ways that make it more difficult to make valid predictions about them? How do men and women go about assessing




potential mates, and making choices? What makes men and women jealous, and why? What are the adaptive functions of negative affect? Are emotional reactions to social situations shaped by evolutionary experiences? Why do we pay attention to some kinds of information, and ignore others? Do evolutionary influences play a role in such social cognitive phenomena as beliefs, attitudes, judgments, prejudice and group preferences? Can we identify evolutionary principles in the way people respond to, and cope with social exclusion and ostracism? We recognize, of course, that no single book could possibly include everything that is interesting and exciting in the current applications of evolutionary ideas to social cognitive phenomena. In selecting and inviting our contributors, we aimed to achieve coverage that is representative of the exciting new developments in social cognition and evolutionary psychology, while simultaneously selecting some of the best examples of clear theorizing and careful research in this area. This book is divided into four parts. Part I deals with some of the fundamental issues concerned with the application of evolution theories to social cognitive phenomena. In Part II we look at the way evolution illuminates our understanding of the links between affect and social cognition. Part III surveys recent research on the evolutionary psychology of the social cognitive processes involved in romantic relationships. Part IV surveys a variety of areas where evolutionary principles have been used to illuminate interpersonal and intergroup processes, such as jealousy, leadership and followership, preferences for different faces, and ostracism.

THE ORIGINS OF THIS BOOK: THE SYDNEY SYMPOSIUM OF SOCIAL PSYCHOLOGY SERIES This book is the ninth volume in the Sydney Symposium of Social Psychology series, held every year at the University of New South Wales, Sydney. Perhaps a few words are in order about the origins of this volume, and the Sydney Symposium of Social Psychology series in general. First, we should emphasize that this is not simply an edited book in the usual sense. The objective of the Sydney Symposia is to provide new, integrative understanding in important areas of social psychology by inviting leading researchers in a particular field to a threeday residential Symposium in Sydney. This Symposium has received generous financial support from the University of New South Wales, allowing the careful selection and funding of a small group of leading researchers as contributors. Draft papers by all contributors are prepared and circulated well in advance of the symposium and are placed on a dedicated website. Thus, participants had an opportunity to review and revise their papers in the light of everybody else’s draft contribution even before they arrived in Sydney. A vital part of the preparation of this book has been the intensive three-day face-to-face meeting between all invited contributors. Sydney Symposia are characterized by open, free-ranging and critical discussion between all participants,


with the objective being to explore points of integration and contrast between the proposed papers. A further revision of each chapter is prepared soon after the Symposium, incorporating many of the shared points that emerged in our discussions. Thanks to these collaborative procedures, the book does not simply consist of a set of chapters prepared in isolation. Rather, this Sydney Symposium volume represents a collaborative effort by a leading group of international researchers intent on producing a wide-ranging and up-to-date review of research on evolutionary influences on social thinking. We hope that the published papers will succeed in conveying some of the sense of fun and excitement we all shared during the Symposium. For more information on the Sydney Symposium series and details of our past and future projects please see our website (www.sydneysymposium.unsw.edu.au). Eight previous volumes of the Sydney Symposium series have been published. All Sydney Symposium books feature original contributions from leading international researchers on key issues in social psychology. Detailed information about our earlier volumes can be found on the series page in this book, and also on our website. Given its breadth of coverage, the present book should be useful both as a basic reference book, and as an informative textbook to be used in advanced courses dealing with social cognition and courses on evolutionary psychology. The main target audience for this book comprises researchers, students and professionals in all areas of the social and behavioural sciences, such as social, cognitive, clinical, counselling, personality, organizational and applied psychology, as well as sociology, communication studies and cognitive science. The book is written in a readable yet scholarly style, and students at the undergraduate and at the graduate level should find it an engaging overview of the field and thus useful as a textbook in courses dealing with evolutionary psychology and social cognition. The book should also be of particular interest to people working in applied areas where using and understanding processes of social thinking is important, such as clinical, counselling, educational, forensic, marketing, advertising and organizational psychology, and health psychology. We want to express our thanks to people and organizations who helped to make the Sydney Symposium of Social Psychology series, and this ninth volume in particular, a reality. Producing a complex multi-authored book such as this is a lengthy and sometimes challenging task. We have been very fortunate to work with such an excellent and co-operative group of contributors. Our first thanks must go to them. Because of their help and professionalism, we were able to finish this project on schedule. Past friendships have not been frayed, and we are all still on speaking terms; indeed, we hope that working together on this book has been as positive an experience for them as it has been for us. The idea of organizing the Sydney Symposia owes much to discussions with, and encouragement by, Kevin McConkey, Chris Fell, Peter Lovibond and numerous others at the University of New South Wales. Our colleagues at the School of Psychology at UNSW, Rebekah East, Norman Chan, Vera Thomson, Liz Goldenberg and others, have, with their advice, support and sheer hard work,



helped to share the burden of preparing and organizing the symposium and the ensuing book. We also wish to acknowledge financial support from the Australian Research Council and the University of New South Wales, support that was of course essential to get this project off the ground. Most of all, we are grateful for the love and support of our families who have put up with us during the many months of work that went into producing this book. Joseph Forgas, Martie G. Haselton, and William von Hippel September 2006


Evolutionary Psychology and Social Thinking


Evolutionary Psychology and Social Cognition A Natural Affinity Different “Why” Questions Content-Specificity and Adaptive Design Refining the Metaphor for Social Cognition Interface with Modern Evolutionary Biology What can Social Cognition Bring to the Table? Overview of the Book sychology has undergone a profound paradigmatic shift in the past few decades. For most of the second half of the 20th century, a kind of unquestioning belief in the power of environmental influences on social thinking and behavior has ruled supreme in the social sciences. This environmentalist ideology rested on some notably fallacious scientific claims, such as Margaret Mead’s now debunked arguments that even patterns of mating behavior are essentially culturally determined. In hindsight, it is puzzling why well-meaning psychologists and social scientists should have chosen to deny the obvious for so long—that biological, genetic, and evolutionary influences do play a fundamental role in understanding social behavior. In one way, we may regard this book as a celebration of the belated return of balance to theories about human cognition and behavior. Using twin studies, numerous converging lines of evidence now show that there is a significant genetic contribution even to high-level, elaborate social cognitive processes that shape




our attitudes, beliefs, and interpersonal strategies. Complementary research programs have repeatedly demonstrated the cross-cultural universality of a large number of sophisticated social behaviors including emotional communication, partner selection, gender differences in mating strategies, and the like. The message of these research programs was not universally welcomed. Many adherents of ideological environmentalism saw any evidence for genetic or evolutionary influences on behavior as a grievous threat to their belief that the potential for reengineering of social arrangements has no natural limits. It is partly for such political reasons that research into evolutionary influences on social cognition and behavior has been so controversial for so long. This book seeks to bring together some of the most recent research and theorizing in the field of social cognition and evolutionary psychology, in an attempt to show that there are significant benefits that can be derived by adopting evolutionary principles in the scientific study of social thinking.

EVOLUTIONARY PSYCHOLOGY AND SOCIAL COGNITION To commemorate the 100th anniversary of the publication of Darwin’s psychological treatise, The Expression of Emotions in Man and Other Animals (Darwin, 1872/1965), a young Berkeley Zoologist named Ghiselin gave an address at the American Psychological Association Conference. In the address, subsequently published in Science (Ghiselin, 1973), Ghiselin pointed out that Darwin’s radically new way of studying behavior—which he called “evolutionary psychology”—hadn’t fully caught on. The study of white rats and college sophomores missed the mark, and much of what purported to be evolutionary psychology was a “warmed over version of scala naturae which arranged beings . . . from highest to lowest” (God to man to brutes to plants, p. 179). Nonetheless, he argued, there clearly seemed to be promise—not only in understanding the emotions, as many had already acknowledged, but also the important role played by sexual selection in human behavior and in so-called “higher” attributes such as moral sentiments. In the three and half decades since this address, the field of evolutionary psychology has seen dramatic progress. In the early 1990s the publication of The Adapted Mind (Barkow, Cosmides, & Tooby, 1992; see also Buss, 1996; Buss & Kenrick, 1998) established the outlines of the paradigm of evolutionary psychology. In the coming years (as the contributions in the current volume aptly demonstrate), researchers began to take seriously the notion that the brain, like the body, is rich in evolved design. As Darwin promised, the evolutionary approach has begun to provide a new foundation for psychological theorizing— and along with it, new insights. By a coarse count, there are hundreds of new discoveries that probably would not have been found without explicit evolutionary psychological theorizing (perhaps thousands by a fine count; see Buss, 2005). The discoveries span the domains of affect and emotion (see chapters in this


volume by Badcock & Allen, chapter 8; Buck, chapter 6; Ellsworth, chapter 5; Forgas, chapter 7), cooperation and sociality (Dunbar, chapter 2; Lieberman, chapter 11; Spoor & Williams, chapter 17), leadership (van Vugt & Kurzban, chapter 14), social perception and inference (Gangestad & Thornhill, chapter 3; Halberstadt, chapter 15; Kenrick, Delton, Robertson, Becker, & Neuberg, chapter 4; Schaller & Duncan, chapter 18; Ybarra et al., chapter 16), kinship (Lieberman, chapter 11), morality (Buck, chapter 6; Lieberman, chapter 11), and, of course, romantic relationships, mating, and sexuality (Buunk, Massar, & Dijkstra, chapter 13; Fletcher & Overall, chapter 12; Gangestad & Thornhill, chapter 3; Simpson & LaPaglia, chapter 10; Todd, chapter 9). One of the key areas in which evolutionary theorizing has played an increasing role is social psychology. In a search of all papers published in the Journal of Personality and Social Psychology between 1985 and 2004, Webster (2006) found an increasing linear trend in the number of articles published on evolutionary psychology that was comparable in magnitude to other emerging areas in social psychology (including stereotyping and prejudice and emotion and motivation). Thus, consistent with the breadth of the contributions in this volume, evolutionary psychology has played an increasing role in social psychology over the last two decades.

A NATURAL AFFINITY Social and evolutionary psychology are both concerned with humans as highly social animals, and thus they have a natural affinity. For example, evolutionary theorists like Dunbar (chapter 2) note that humans’ long history of living in groups requires adaptations for social living, including sophisticated capacities for representing the mental states of others (and others’ mental representations of oneself; and others’ representations of one’s own representations of others’ representations; and so on). Along similar lines, biological anthropologists Boyd and Richerson (2006) argue that humans are unusually cooperative relative to other animals, and that human sociality requires and enables an extraordinary capacity for culture, teaching, and social learning. These are the very capacities that many social scientists regard as central to human nature. Interpersonal communication is another area where the interests of evolutionary and social psychologists have been converging for some time. Darwin’s (1872/1965) classic work on the communication of emotions in man and animals presaged the recent rapid development of empirical research on the mechanisms that influence nonverbal communication between individuals. There is now clear experimental and neuropsychological evidence documenting the close links between emotional expressions and emotional experiences, and there is good reason to believe that some of our most powerful interpersonal signals, such as eye gaze, touching, spacing, and gestures, are also produced by deep-seated evolved mechanisms, as Darwin (1872/1965) proposed.



Likewise, some of the foundational studies in social psychology conducted by Asch (1956) and Milgram (1963) documented the surprising extent to which humans are susceptible to social influence. William James (1890) remarked that “solitary confinement is by many regarded as a mode of torture too cruel and unnatural for civilised countries to adopt” and social psychologists followed James’ insight by documenting how devastating social rejection can be and how sensitive people are in detecting it. For example, Spoor and Williams (chapter 17) describe experiments using minimal cues of rejection such as being left out of a computerized ball toss. Participants in these studies quickly show signs of dejection, and neuroimaging reveals an increase in brain activation associated with the experience of pain: rejection literally hurts (Eisenberger, Lieberman, & Williams, 2003). In sum, there is considerable overlap in the topics of interest to social and evolutionary psychologists. Social and evolutionary psychology also share a preference for the cognitive level of description. Social psychologists never bought the argument that behavior should be the only unit of analysis, and evolutionary psychologists have similarly suggested that it is most appropriate to theorize about psychological adaptations (rather than to limit one’s analysis to overt behavior; Buss, 1995; Cosmides & Tooby, 1994). Thus, social and evolutionary psychologists are both inclined to theorize about the design of the cognitive adaptations underlying social behavior. Lieberman’s analysis (chapter 11) of inbreeding avoidance and familial cooperation beautifully illustrates the cognitive approach. She asked what cues were available to ancestral humans that would have allowed them to reliably estimate kinship and what kind of cognitive procedures could take them into account to produce both sexual aversion and cooperation. She hypothesized that coresidence duration during childhood could serve as an appropriate cue, and a well-designed system would vary the intensity of the motivations underlying sexual avoidance and cooperation depending on length of coresidence (and thus with the probability of relatedness). Similarly, Todd’s chapter (chapter 9) points out that the social environment can be uncertain and overwhelmingly complex; thus, evolution should favor decision rules that use a small amount of ecologically-valid information and limited processing time to come up with what are often very good choices, when they are employed in the proper environments. This led to the surprising discovery that assessing fewer possible mates rather than more resulted in optimal mate choice in computer simulations (searching too long can result in passing the best mate by or running out of time altogether; see Todd, chapter 9). (For other excellent examples see contributions by Dunbar, chapter 2; Kenrick et al., chapter 4; Halberstadt, chapter 15; Schaller & Duncan, chapter 18; and Spoor & Williams, chapter 17.)


DIFFERENT “WHY” QUESTIONS Along with a natural affinity based on mutual interests, there are also differences between the approaches of social cognition and evolutionary psychology. Chief among these concerns the nature of explanation. Social psychologists tend to focus on explanations involving proximate mechanisms—attention, cognitive load, schemas, stress, frustration, norms, etc. Evolutionary psychologists, on the other hand, tend to focus on functional explanations, often leaving the details of proximate mechanisms unexamined. Thus, social and evolutionary psychologists may approach the same topic very differently. For example, in understanding certain types of mating decisions, social psychologists my invoke norms, such as sex roles or cultural taboos. Such forces clearly exist. Parents attempt to influence the mating decisions of their children, including, for example, the sexual permissiveness of their daughters. Evolutionary psychologists, on the other hand, begin by asking what adaptive problems mechanisms of mate choice address. Thus, an evolutionary psychologist might locate key theoretical causes at the distal level. In mate choice, for example, evolutionary theorists might argue that females are less sexually permissive than males because of their necessarily larger investment in offspring and greater costs associated with poor mate choice (Trivers, 1972). Of course, evolutionary theorists propose that proximate evolved decision rules execute the functions they study—for example, a selective sociosexual psychology in women or adjustments of sexual aversion based on coresidence—but evolutionary psychologists are less likely to focus their empirical efforts on the examination of precisely how these rules are implemented. Few social psychologists reject the notion that the entire human body, including the brain, has been shaped by natural selection. Reticence instead stems from questions about the utility of distal explanations. As Conway and Schaller (2002) point out, whereas proximate mechanisms can usually be tested directly, distal evolutionary mechanisms usually cannot. So, social psychologists have often asked what can be gained by this sort of seemingly irrefutable conjecturing. Evolutionary psychologists have responded that well-formulated evolutionary hypotheses lead to specific predictions and are therefore falsifiable, evolutionary explanations are required for a complete explanation of behavior, and most importantly, they are useful as they guide researchers to new domains of inquiry, new ideas about behavior, and ultimately new findings (e.g., Buss, 1995; Buss, Haselton, Shackelford, Bleske, & Wakefield, 1998). As evolutionary psychology has progressed, this promise has been more fully realized. The predictions of evolutionary theories have become sophisticated, more specific, and genuinely difficult to reconcile with alternative accounts, thus making the utility of evolutionary theorizing harder to doubt. An interesting irony, as we note below, is that these advances may be due in part to a more complete integration of proximate and ultimate levels of explanations. For example, some social psychologists have found recent evolutionary work on menstrual cycle effects particularly compelling (see, e.g., Simpson & LaPaglia,



chapter 10). This work may be compelling precisely because researchers have linked women’s behaviors to specific proximate mechanisms (cycling hormones) that would seem to rule out the traditional domain-general, sociocultural explanations that are often touted as alternative explanations (e.g., mass media effects, social roles, etc.).

CONTENT-SPECIFICITY AND ADAPTIVE DESIGN Before we elaborate on why evolutionary psychology needs the theories and methodologies of social cognition researchers, we suggest that there are several other lessons that evolutionary psychology might teach social cognition researchers. Perhaps most importantly, when starting with first principles, evolutionary psychologists tend to arrive at predictions about specialized, contentrich psychological mechanisms designed to address specific adaptive problems. The chapters in this volume describe specific psychologies of kinship (Lieberman, chapter 10), leadership (van Vugt & Kurzban, chapter 14), jealousy (Buunk et al., chapter 13), mind-reading (Dunbar, chapter 2), and mate selection (Todd, chapter 9), to name a few. The information processing rules involved in each of these areas might overlap with those involved in others (e.g., mindreading likely plays a role across social domains), but they will also contain specialized features. In their joint research program, Kenrick, Neuberg, Maner, Schaller, and colleagues have collected some of the best evidence for content-specificity in social judgments (see e.g., Kenrick et al., chapter 4). For example, given the differing demands of mate choice and physical self-protection, Maner et al. (2005) hypothesized that participants induced to feel these motivational states would show qualitatively different biases in interpreting the facial expressions of others, and in specific cases these biases would differ for the sexes. After viewing a scary film clip, participants in the studies conducted by these authors rated neutral male faces as angrier especially when they were the faces of outgroup members (Maner et al., 2005). Thus, when cues indicate increased danger, men and women become vigilant about the potential threat of aggressive others. After viewing a romantic film clip, men but not women in the studies saw more romantic interest in the faces of neutral opposite-sex faces, especially attractive faces (Maner et al., 2005). Overperceiving sexual interest may function to help individuals avoid missing sexual opportunities, which would have benefited ancestral men more than ancestral women (Haselton & Buss, 2000), and thus would have selected for precisely the sort of bias Maner et al. documented. In describing a related research program, Schaller and Duncan (chapter 18) argued that one of the most daunting adaptive problems faced by ancestral humans was exposure to communicable pathogens. Thus, these authors argued that humans are endowed with a “psychological immune system” that tracks heuristic cues to disease and leads to behavioral avoidance of contamination.


Because judgments of disease threat are uncertain, the psychological immune system may be adaptively overinclusive (Kurzban & Leary, 2001) by responding to a variety of physical disfigurements, even when they are caused by accidents rather than illness. Similarly, humans are more susceptible to pathogens carried by people from foreign lands than those in their local environments (whose pathogens are familiar to their physical immune system). Therefore, the psychological immune system may lead to a tendency to stigmatize and avoid various classes of individuals, including disfigured people and foreign immigrants. Schaller and colleagues predicted that these prejudices should increase when disease threat is activated, and, as predicted, participants who score highly on a measure of vulnerability to disease showed stronger implicit associations between images of disabled individuals and disease concepts (Park, Faulkner, & Schaller, 2003). In a related study (Faulkner, Schaller, Park, & Duncan, 2004), participants were shown slide show images of accidents or diseases. Relative to the accident condition, those in the disease condition allocated less funding to an effort to recruit immigrants from subjectively foreign lands (allocation of funds to recruiting immigrants from familiar locations did not show this effect; Faulkner et al., 2004). In sum, theorizing about specific adaptive problems has led to successful predictions about the differing effects of mating, self-protection, and pathogen avoidance on how information is processed. In social cognition, key concepts and proposed mechanisms tend to be far more general and content-free. For example, theories of stereotyping, social memory, and attribution tend not to propose specific mechanisms for how these processes differ with different types of targets, and indeed theorizing is even rare concerning how social versus nonsocial cognition differ. Given the evidence that human brains are adapted in part from the influence of other humans (Dunbar, chapter 2), it seems likely that certain social cognitive routines are only initiated in certain social situations, and that these routines differ in important ways by the type of target. Thus, it might be useful to infuse theorizing about social cognitive processes with more content by starting with adaptive problems rather than assuming that cognitive procedures will be domain-general (see Alexander, Brewer, & Herrmann, 1999). In addition to leading to new predictions, this type of integration of content and process might also enable researchers to fill in some of the “conceptual holes” in social psychology (see Daly, Salmon, & Wilson, 1997; Kenrick et al., chapter 4).

REFINING THE METAPHOR FOR SOCIAL COGNITION Although social cognition is now much more concerned with motivation and emotion than it was 10 or 20 years ago, we suggest that the nature of that motivation is still underspecified and poorly understood. It is undoubtedly a step in the right direction to conceptualize humans as motivated tacticians rather than cognitive misers, but what motivates humans, and why are they being



tactical? Evolutionary theory provides a well-articulated perspective on these questions that can focus social cognition research on the fundamental goals chronically held by humans in different situations. For example, Trivers’ work on parental investment (1972) and reciprocal altruism (1971) provides hugely influential mid-level theories about important motivations and tactics, but evolutionary theorizing also provides much more specific predictions as well. A wonderful example of such a prediction can be found in the chapter by Ybarra et al. in the current volume (chapter 16), in which they propose that sometimes it is in our interests to be figured out by others and sometimes it is not. Consistent with this perspective, cues that suggest competition lead people to ensure that their behaviors and motivations are opaque, whereas cues that suggest cooperation lead people to be more transparent in their actions. More broadly, however, the evolutionary approach to the motivated tactician clearly reveals that people were never as dumb as they were made out to be when the field was focused on bias and error. Rather, when people must make decisions under uncertainty (which is probably most of the time), evolutionary processes appear to have shaped human decisions to be biased in favor of the least costly error (Haselton & Buss, 2000, 2003; Haselton & Nettle, 2006; Kenrick et al., chapter 4; Schaller & Duncan, chapter 18; Spoor & Williams, chapter 17). Furthermore, when problems are framed in a manner that is consistent with experience, or with adaptively recurrent problems such as cheater detection, previously unsolvable problems suddenly become simple for even uneducated individuals (Cosmides & Tooby, 1996, 2005). People thus appear much more like savvy bookies playing the odds than hopeless incompetents who somehow manage to survive despite themselves (see Todd, chapter 9). Evolutionary psychology thus suggests that it is time to refine our metaphor of the motivated tactician by considering what motivates him, what motivates her, and what tactics are likely to have emerged in the never-ending social cognitive arms race of manipulation, cooperation, and competition among individuals and groups as people navigate an increasingly complex social world.

INTERFACE WITH MODERN EVOLUTIONARY BIOLOGY Lastly, principles of modern evolutionary biology may be brought to bear on theorizing in social cognition. Gangestad and Thornhill (chapter 3) point out that social inference permeates the topics studied within social psychology and yet surprisingly few researchers have made use of the vast evolutionary literature on signaling theory. Signaling theory suggests two types of systems involved in social inference, signaling systems and nonsignaling systems. The first type of system involves the transmission of factual information from senders to receivers. The second type of system is one in which receivers make inferences on the basis of cues emitted by targets, but targets possess no adaptations designed to convey information to perceivers; instead, receivers base their inferences on incidental


effects. Signaling systems tend to collapse if information is not factual—over time, receivers will cease attending to deceptive signals. Given that the perfect alignment of fitness interests is rare in the animal kingdom, senders will rarely benefit from always signaling their true intentions. The consequence is that social inference is probably based on coevolving adaptations in senders and receivers in which senders are selected to often conceal cues indicating their true intentions and receivers are selected to detect them (also see Ybarra et al., chapter 16). This perspective leads to new predictions about the conditions under which communication is characterized by deception or honesty; it may explain many of the cases in which social inference fails; and it may drive theorists to ask themselves whether the social inference psychologies they propose obey principles derived from modern evolutionary biology (see Gangestad & Thornhill, chapter 3).

WHAT CAN SOCIAL COGNITION BRING TO THE TABLE? By now we hope that it’s clear what evolutionary psychology can bring to the study of social cognition, but the question remains concerning what social cognition can offer to the study of evolutionary psychology. In answering this final question, we find it helpful to consider the predominant criticisms of evolutionary psychology and how social cognition might address them. To enable a preliminary examination of these criticisms, we sent an e-mail to the list serve for social psychologists (SPSP) and evolutionary biologists (Evoldir) asking list members about their perception of evolutionary psychology and whether they have found it helpful to their understanding of human behavior. Despite the rather disparate perspectives of members of these two groups (and keeping in mind that respondents were undoubtedly not a representative sample of their lists), responses to this request showed a great deal of similarity across the two lists (see Table 1.1). In both lists evolutionary psychology evokes strong feelings, with approximately equal numbers voicing strong support for the enterprise, strong reservations (or even animosity) for evolutionary psychology, or a bit of both. Indeed, several respondents requested that all identifying information be removed from their responses prior to any dissemination. Examination of the responses of those who are critical of evolutionary psychology revealed a variety of concerns with the science among both social psychologists and evolutionary biologists (for a more thorough discussion of criticisms of evolutionary psychology, see Hagen, 2005). Furthermore, among both disciplines, dissatisfaction with the science most often centered around the concern that evolutionary psychologists tell “just so” stories about how we became the way we are, and that these stories cannot be adequately tested. This criticism is surprising, given the enormous amount of progress made by evolutionary psychologists in the last 20 years. Nevertheless, a substantial portion of the psychological and biological community appears to regard the enterprise


3.5 4 (15%)

No details/other

1.5 2 (12%)

SPSP Evoldir Totals


SPSP Evoldir Totals

.5 1.5 (8%)

Implications/ political uses

6 1 (14%)

Hypothesis Generation (72%) (42%)


7 5.5 (42%)

4 1.5 (19%)

Just So stories/not Not X-cultural/not tested socialization

28.5 6 (70%)


1 1.5 (8%)

Not X-species

1 3 (10%)

Not genetic/ biologically naive

(28%) (58%)


Responses could be placed in up to two categories if more than one issue was raised in the response (a very common situation), in which case each of the two categories received .5 of a response. Thus, the total number of respondents in both lists who made positive or negative responses in a particular category can be inferred from column totals. The total number of social psychologists and evolutionary biologists who made positive or negative comments can be inferred from the row totals (although this number is probably less informative than the nature of the reasons, as it seems more likely to be influenced by the likely nonrepresentative nature of the sample).

No details/other


Table 1.1 Frequency Counts of Social/Personality Psychologists (SPSP) and Biologists (Evoldir) Who Endorsed a Particular Category of Positive or Negative Reactions to Evolutionary Psychology


with doubt or disdain, in part because they perceive the product of this research as the telling of tales that are largely impossible to disconfirm. With these criticisms in mind, we turn now to the topic of how a social cognitive approach might help address these concerns, and thereby further the discipline of evolutionary psychology. Although evolutionary psychologists are already methodological pluralists, the most obvious contribution that social cognition can make to the study of evolutionary psychology is by providing new methodological tools that can supplement the frequent reliance on self-report as a means of testing evolutionary hypotheses. Because social cognitive theorists are concerned with the more microlevel workings of the social mind, from the beginning of the enterprise they have been dubious about people’s ability to self-report on these processes. Thus, researchers in social cognition have borrowed and developed a large set of procedures and tools that enable the study of social functioning without directly asking people what they are thinking. A variety of social cognitive methods now exist for the assessment of goals, attention, accessibility, and (more controversially) attitudes, even if people are unwilling or unable to directly report on the contents of their mind. Greater use of these procedures is critically important for the development of evolutionary psychology as a discipline. First of all, measures of social cognitive processing often provide clearer and more in-depth evidence for the phenomena of interest than is possible through self-report. Indeed, the methods of social cognition are uniquely well suited for the study of proximal mechanisms, as they allow researchers to get a handle on the processes by which the mind executes important evolved functions. Evidence of these advantages of a social cognitive approach to evolutionary psychology abounds in this book. For example, in their work on the behavioral immune system, Schaller and Duncan use reaction time to show that people have an automatic association between bodily disfigurement and disease, and that this association is stronger among people who perceive themselves as particularly vulnerable to disease. Indeed, Schaller and Duncan (chapter 18) show that people automatically activate thoughts of disease even when they are confronted by a bodily disfigurement that they know is not contagious (such as a birth mark or obesity), yet these automatically activated thoughts of disease are not evident when people encounter individuals who appear healthy but are known to carry a contagious and dangerous disease. Such findings speak to the primacy of appearances in activating automatic disease cognitions, and suggest a possible functional role for these cognitions in premedical societies. These data also provide an excellent example of the benefits of a social cognitive approach to evolutionary psychology, as not only would people be loath to report such negative responses to disabled, disfigured, or obese others, but it is also unlikely that people would be able to introspect about such associations in their mind between surface-level abnormalities and fear of disease. The chapter by Kenrick et al. (chapter 4) also provides a clear example of



the advantages of evolutionary social cognition in elucidating some of the details of proximal mechanisms for achieving distal functions. In their research, they find that both males and females focus greater visual attention on attractive members of the opposite sex, but only females also spend more time encoding attractive members of the same sex. Furthermore, although women initially attend more to the faces of attractive men, they do not show an advantage in memory for these attractive male faces. Thus, in line with evolutionary reasoning that physical attractiveness is valued in both sexes but is more important in women than men, eye gaze translates into memory for attractive women but not for attractive men. In addition, and as predicted by an evolutionary account, these differences in eye gaze and memory are themselves moderated by factors such as the relationship and ovulatory status of the individuals doing the perceiving. Second, social cognitive measures can help researchers avoid the criticism that evolutionary psychologists are simply tapping societal stereotypes of how people believe they are supposed to think and act, and that these stereotypes don’t reflect the real workings of the mind. For example, consider the evolutionary prediction that men should be particularly jealous of high-status rivals, whereas women should be more jealous of beautiful rivals. When men and women self-report such an effect, critics of the evolutionary approach respond that society suggests that men should be concerned with status and women concerned with beauty, and thus people dutifully self-report that these are their concerns, when in fact they may not really be at issue. Setting aside why society should shape the mind in such evolutionarily advantageous ways (as sociocultural critics of the evolutionary approach are typically unswayed by the argument that culture is a mechanism for achieving evolutionary goals), these criticisms are clearly addressed by experiments such as those reported by Buunk et al. (chapter 13). In their chapter, these authors show that nonconscious priming of status or attractiveness cues leads to gender differences in jealousy, whereby high mate value females are more jealous when primed with attractiveness cues and high mate value males are more jealous when primed with status cues. Low mate value individuals were uniformly jealous, regardless of priming. This sort of pattern of responses is very difficult to explain away as an artifact of social stereotypes or culturally induced values, particularly when moderated by individual differences (such as perceived mate value in these experiments, or relationship or ovulatory status in the experiments of Kenrick et al., chapter 4). Third, and for the very reasons outlined above, social cognitive approaches have the potential to address what appears to be the most widespread scientific criticism of evolutionary psychology by psychologists and biologists who are dubious of the enterprise. That is, a social cognitive approach to evolutionary psychology is very well suited to confront the criticism that evolutionary theorists are weaving a bunch of “just so” stories about the origins and development of human nature. By couching research in terms of information processing, with a clear emphasis on proximal mechanisms underlying distal evolutionary goals,


evolutionary social cognition has the potential to address these important criticisms. For a wonderful example of this marriage of evolutionary theory and social cognition, we need only consider Halberstadt’s chapter (chapter 15), in which he describes his research on preferences for prototypical faces and the possible evolutionary origins of these preferences. Beginning with the hypothesis that the preference for prototypical faces is an adaptation for enhancing mate quality, Halberstadt’s first experiments reveal the unexpected result that prototypical members of almost all categories are preferred, whether positive or negative and whether natural or artificial. This finding suggests that perhaps the preference for prototypical faces is not an adaptation for mating, but rather is simply an aspect of a more general preference for items that are easily categorized. To test this possibility, he then partials out the subjective familiarity of the different category members, and finds that the sense of familiarity mediates the effect of prototypicality for artificial categories but not for natural categories. This result narrows the range of possible causes for preferences for prototypicality as found in nature, but still leaves open a wide variety of reasons for why humans prefer prototypicality in animals as well as humans. Halberstadt (chapter 15) then tests the possibility that prototypes are preferred because prototypicality communicates safety, but finds that sometimes prototypical animals are actually the more dangerous members of their category. This result leads to a test of an alternative possibility that is closer to his original hypothesis, which is that natural prototypes are preferred because they represent an overgeneralization of a mechanism adapted for the perception of humans. Consistent with this possibility, the magnitude of the correlation between prototypicality and attractiveness in animals is found to be a direct function of the similarity of the animals to humans, with animals that are more similar to humans showing an increased association between prototypicality and attractiveness. Because similarity between these animal categories and humans was only measured rather than manipulated, and thus potentially confounded with other factors, Halberstadt then constructed drawings of faces and described them as sketches of either criminals or aliens. Consistent with the idea that preference for prototypicality is an adaptation for face perception, prototypical faces were preferred for both humans and aliens, but the latter and not the former effect was mediated by the subjective familiarity of the face. This sequence of experiments represents a superb combination of basic research in categorization and prototypicality—long areas of interest within social cognition—with evolutionary principles concerned with attraction and mate selection. This series of experiments demonstrates how both disciplines can inform each other, and how an evolutionary social cognitive approach leads to unique predictions that are readily tested. Had Halberstadt (chapter 15) not focused on underlying mechanisms such as perceived familiarity and perceptual fluency, this research might have started and ended with the speculation that preference for prototypical faces may be an adaptation for finding high quality



mates. Although more experimentation is clearly necessary, Halberstadt’s social cognitive approach to evolutionary psychology has moved beyond speculation about ultimate goals to specific tests of underlying hypotheses, with the end result that we can be much more confident about the evolutionary origins of preferences for prototypical faces.

OVERVIEW OF THE BOOK This book is organized into four main parts. Following this introductory chapter, the first part deals with some of the fundamental theoretical issues that form the foundation of an evolutionary framework for social cognition. Robin Dunbar (chapter 2) describes his “social brain” hypothesis that postulates a close interdependence between the evolution of human brain and human group size, suggesting that brain evolution was most likely driven by that primordial social cognitive task, the need to navigate and coordinate the activities of ever larger and ever more sophisticated and effective interacting social groups. In the next chapter, Steve Gangestad and Randy Thornhill (chapter 3) outline a carefully elaborated evolutionary framework for understanding the most basic social cognitive processes, the way social inferences are formed. Doug Kenrick and his colleagues (chapter 4) focus on a key aspect of evolutionary social cognition— the fact that certain kinds of information contents receive preferential treatment, as illustrated by the emergence of numerous domain-specific mechanisms that appear to be adapted to solve specific fitness problems. Evolutionary approaches thus offer a complementary, content-focused framework to the traditional information processing models that characterize mainstream social cognitive research seeking universal (and mostly domain-general) explanatory principles. The second part of the book discusses the evolutionary psychology of affect and cognition. Adaptive emotional reactions to social events require sophisticated cognitive appraisal strategies, and Phoebe Ellsworth (chapter 5) outlines how appraisal theories of emotion can inform evolutionary theorizing. Ross Buck (chapter 6) argues that the origins of moral emotions that regulate so much of our social thinking and behavior go back to the dawn of evolution, and he points to the ubiquity of such emotional reactions in everyday social communication. Joseph Forgas (chapter 7) argues that negative affective states have an important adaptive function, recruiting more focused and accommodative thinking strategies that produce identifiable benefits in many social cognitive tasks. In the final chapter in this part, Paul Badcock and Nicholas Allen (chapter 8) suggest that depressed moods may also serve another adaptive function, reducing risk taking and possibly competitive behaviors. The third part of the book looks at one of the social cognitive problems most intensively studied from an evolutionary perspective: the evolutionary psychology of mate selection. Peter Todd (chapter 9) offers a cognitiveevolutionary analysis of how the all-important adaptive task of mate selection is


performed. Jeffry Simpson and Jonathon LaPaglia (chapter 10) describe the intriguing phenomenon of patterned changes in mate preferences across the ovulatory cycle of women, changes that are consistent with evolutionary theories but would be difficult to explain in terms of alternative cognitive models of decision making. Debra Lieberman (chapter 11) looks at the evolutionary problem of inbreeding avoidance, and shows how humans use information readily available from their environment to solve the problem of whom to avoid as a mate. Garth Fletcher and Nicola Overall (chapter 12) offer an intriguing discussion of the role of the self in mate-selection decisions, and in particular, how assessment of one’s own and one’s partner’s value may impact mate choices and relationship functioning. The final, and largest part of the book features chapters that adopt an evolutionary perspective in analyzing a variety of interpersonal and intergroup processes involving social cognitive mechanisms. The phenomenon of jealousy is the topic of the chapter by Abraham Buunk and his colleagues (chapter 13), who argue that men and women are of a different mind when it comes to jealousy. Women are more likely to be jealous of a rival’s physical attractiveness, while men are more jealous of a rival’s status and dominance, judgmental differences that are readily understandable in terms of evolutionary mechanisms. Mark van Vugt and Rob Kurzban (chapter 14) offer an incisive analysis of the evolutionary psychology of leadership and followership, an approach that has the capacity to greatly enrich existing theorizing in these domains. Jamin Halberstadt (chapter 15) looks at the evolutionary basis of preferences for prototypical faces, and suggests that such preferences are indeed likely to endure because prototypicality indicates greater reproductive fitness in a potential mate. Oscar Ybarra and his colleagues (chapter 16) discuss the notion of human unpredictability as the basic problem of social perception and prediction, and suggest that humans may possess adaptive mechanisms that predispose us to avoid being too easily known and predicted. Jennifer Spoor and Kipling Williams (chapter 17) look at the evolutionary psychology of social rejection and ostracism, and discuss the importance of an ostracism detection system. In the final chapter, Mark Schaller and Lesley Duncan (chapter 18) propose the existence of an evolved behavioral immune system that plays an important role in influencing a variety of social cognitive phenomena, such as norms, values, attitudes, and social communication. In their entirety, these chapters offer a broad and integrated overview of the many specific domains and research areas where evolutionary theorizing has contributed to social cognitive research in recent years. As editors, we hope that readers will find these contributions as exciting and intriguing as we did, and we hope that collecting them in one volume will stimulate further interest in the rapidly expanding interface of evolutionary psychology and social cognition.


16 EVOLUTION AND THE SOCIAL MIND REFERENCES Alexander, M. G., Brewer, M. B., & Herrmann, R. K. (1999). Images and affect: A functional analysis of out-group stereotypes. Journal of Personality and Social Psychology, 77, 78–93. Asch, S. E. (1956). Studies of independence and conformity: I. A minority of one against a unanimous majority. Psychological Monographs, 70(9, Serial No. 416). Barkow, J. H., Cosmides, L., & Tooby, J. (Eds.). (1992). The adapted mind: Evolutionary psychology and the generation of culture. New York: Oxford University Press. Boyd, R., & Richerson, P. J. (2006). Culture and the evolution of the human social instincts. In S. Levinson & N. Enfield (Eds.), Roots of human sociality (pp. 453–477). Oxford, UK: Berg. Buss, D. M. (1995). Evolutionary psychology: A new paradigm for psychological science. Psychological Inquiry, 6(1), 1–30. Buss, D. M. (1996). The evolutionary psychology of human social strategies. In E. T. Higgins & A. W. Kruglanski (Eds.), Social psychology: Handbook of basic principles (pp. 3–38). New York: Guilford Press. Buss, D. M. (2005). The handbook of evolutionary psychology. Hoboken, NJ: Wiley. Buss, D. M., Haselton, M. G., Shackelford, T. K., Bleske, A. L., & Wakefield, J. (1998). Adaptations, exaptations, and spandrels. The American Psychologist, 53, 533–548. Buss, D. M., & Kenrick, D. T. (1998). Evolutionary social psychology. In D. T. Gilbert, S. T. Fiske, & L. Gardner (Eds.), The handbook of social psychology (Vol. 2, 4th ed., pp. 982–1026). New York: McGraw-Hill. Conway, L. G., III, & Schaller, M. (2002). On the verifiability of evolutionary psychological theories: An analysis of the psychology of scientific persuasion. Personality and Social Psychology Review, 6(2), 152–166. Cosmides, L., & Tooby, J. (1994). Beyond intuition and instinct blindness: The case for an evolutionarily rigorous cognitive science. Cognition, 50, 41–77. Cosmides, L., & Tooby, J. (1996). Are humans good intuitive statisticians after all? Rethinking some conclusions from the literature on judgment under uncertainty. Cognition, 58(1), 1–73.

Cosmides, L., & Tooby, J. (2005). Neurocognitive adaptations for social exchange. In D. M. Buss (Ed.), The handbook of evolutionary psychology (pp. 584–627). Hoboken, NJ: Wiley. Daly, M., Salmon, C., & Wilson, M. (1997). Kinship: The conceptual hole in psychological studies of social cognition and close relationships. In J. A. Simpson & D. T. Kenrick (Eds.), Evolutionary social psychology (pp. 265–296). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. Darwin, C. (1965). The expression of emotions in man and other animals. Chicago: University of Chicago Press. (Original work published 1872) Eisenberger, N. I., Lieberman, M. D., & Williams, K. D. (2003). Does rejection hurt? An fMRI study of social exclusion. Science, 302, 290–292. Faulkner, J., Schaller, M., Park, J. H., & Duncan, L. A. (2004). Evolved disease-avoidance mechanisms and contemporary xenophobic attitudes. Group Processes and Intergroup Relations, 7(4), 333–353. Ghiselin, M. T. (1973). Darwin and evolutionary psychology: Darwin initiated a radically new way of studying behavior. Science, 179(4077), 964–968. Hagan, E. H. (2005). Controversial issues in evolutionary psychology. In D. M. Buss (Ed.), The handbook of evolutionary psychology (pp. 145–173). Hoboken, NJ: Wiley. Haselton, M. G., & Buss, D. M. (2000). Error management theory: A new perspective on biases in cross-sex mind reading. Journal of Personality and Social Psychology, 78, 81–91. Haselton, M. G., & Buss, D. M. (2003). Biases in social judgment: Design flaws or design features? In J. Forgas, K. Williams, & B. von Hippel (Eds.), Responding to the social world: Implicit and explicit processes in social judgments and decisions (pp. 23–43). New York: Cambridge University Press. Haselton, M. G., & Nettle, D. (2006). The paranoid optimist: An integrative evolutionary model of cognitive biases. Personality and Social Psychology Review, 10, 47–66. James, W. (1890). Principles of psychology. New York: Henry Holt.


Kurzban, R., & Leary, M. R. (2001). Evolutionary origins of stigmatization: The functions of social exclusion. Psychological Bulletin, 127(2), 187–208. Maner, J. K., Kenrick, D. T., Becker, D. V., Robertson, T. E., Hofer, B., Neuberg, S. L., et al. (2005). Functional projection: How fundamental social motives can bias interpersonal perception. Journal of Personality and Social Psychology, 88(1), 63–78. Milgram, S. (1963). Behavioral study of obedience. Journal of Abnormal and Social Psychology, 67(4), 371–378. Park, J. H., Faulkner, J., & Schaller, M. (2003). Evolved disease-avoidance processes and contemporary anti-social behavior: Prejudicial

attitudes and avoidance of people with physical disabilities. Journal of Nonverbal Behavior, 27(2), 65–87. Trivers, R. L. (1971). The evolution of reciprocal altruism. Quarterly Review of Biology, 46, 35–57. Trivers, R. L. (1972). Parental investment and sexual selection. In B. Campbell (Ed.), Sexual selection and the descent of man, 1871–1971 (pp. 136–179). Chicago: Aldine. Webster, G. D. (2006, January). Evolutionary theory’s increasing role in personality and social psychology. Poster presented at the fourth annual Evolutionary Psychology Preconference at the Society for Personality and Social Psychology, Palm Springs, CA.






The Social Brain Hypothesis and its Relevance to Social Psychology


The Social Brain Hypothesis The Structure of Human Social Networks Two Specializations of Social Cognition umans, like most mammals, are intensely social. In many ways, primates’ success from an evolutionary perspective is a direct consequence of that sociality. Primate societies are implicit social contracts that allow some of the problems of survival and reproduction to be solved co-operatively. Social contracts of this kind work because they allow relevant problems to be solved more efficiently by individuals collaborating in their solution. However, social contracts require individuals to be willing to forgo some of their more immediate personal interests in order to benefit from greater returns later through group-level co-operation. If too many individuals act in their own selfish interests, the cohesion of the group will be threatened, simply because others will end up paying the costs of sociality. When these costs become a burden and begin to outweigh the benefits of co-operation, group stability is undermined, leading to the rapid collapse of the contract (and, of course, society). Gürerk, Irlenbusch, and Rockenbach (2006) provide some evidence to support this claim in the context of small group economic games: They showed that when punishment is prevented, individuals rapidly (within the time frame of repeatedencounters laboratory games) switch allegiance to another group where punishment is permitted. In effect, free-riding does lead to the rapid breakdown of group cohesion. The real issue here seems to be the cognitive demands of maintaining the stability of relationships through time. That process of negotiation is far from




simple. At its most basic, it requires co-ordination and compromise, and often subtle adjustments to ensure that group members do not drift apart during foraging. But it also requires the ability to manage conflict in such a way as to reduce its natural tendency to cause groups to dissipate, and to cope effectively with the ever-present threat generated by rivals. In the limit, and certainly in the human case, it requires individuals to be able to understand another’s perspective sufficiently well to appreciate what kinds of adjustments are necessary to create the levels of “bondingness” required to keep a group together. One element of that is knowing when to trust another individual. However, the very fact that these processes depend on relatively sophisticated cognitive skills leaves them open to abuse. Free-riders, who take the benefits of sociality without paying all the costs, are an inevitable by-product of a biological system that is built on cognitive flexibility rather than genetic hardwiring or endocrinological determinism (as is the case with many social insects, for example). Free-riders destabilize the social contract because they oblige others to pick up the bill. If that bill becomes too large, it will wipe out the benefits that these individuals expect to accrue from sociality and their willingness to co-operate in the contract will soon evaporate. It is important, at this point, to note that comparative and developmental psychologists have taken to referring to the cognitive processes involved by the term “social cognition”. By this, they mean the suite of explicitly cognitive processes (causal reasoning, analogical reasoning, memory etc.) that allow an individual to understand another’s mind states. The most important of these (at least in so far as the broad social cognitive developmental literature is concerned) is the phenomenon known as theory of mind (ToM), the capacity to appreciate that another individual has a mind like one’s own. The essence of the phenomenon lies in the more conventional cognitive mechanisms that make these kinds of inferences possible, but social cognition should properly be seen as the emergent property of the system when these more fundamental cognitive processes are applied explicitly to social contexts. There is an implicit sense that individuals of those species that have this capacity are especially sensitive to social contexts, and that this context brings into play a specialized suite of cognitive mechanisms not normally used in more mundane physical world contexts (e.g., conventional causal reasoning). This usage has perhaps been unfortunate, because the term “social cognition” already has a rather specific meaning within social psychology. Nonetheless, I will stick with the conventional comparative psychology usage here, and use the term to refer to the capacity to understand the workings of another individual’s mind. Although the case for social cognition is well established in principle, we do not have a very clear idea about what is involved. We know much about only one aspect of social cognition, and this is the phenomenon known as “theory of mind”. And even then, our knowledge is limited mainly to its natural history and we know almost nothing about the phenomenon itself as a cognitive process (Roth & Lesley, 1998). Indeed, there is a long-running debate in the cognitive


developmental literature as to whether theory of mind itself is a bona fide module (in the conventional evolutionary psychology sense advocated by Tooby & Cosmides, 1990) or just an emergent property of more fundamental executive functions (for an overview, see Barrett, Henzi & Dunbar, 2003; Mitchell, 1997). Be that as it may, the emergent properties of theory of mind seem to have some reality. As the ability to understand another’s mental state, to understand that another individual believes something to be the case, it forms a component of what philosophers refer to as “intentionality”, the state of mind associated with terms like believing, intending, supposing, desiring and so on. In this context, ToM is identified with second order intentionality (“I believe that you intend [. . . to do something]”). First order intentionality is equivalent to having knowledge of one’s own mind state (“I believe [. . . that something is the case]”), and is thought to be the condition for most—if not all—higher vertebrates, although it could well be more widespread in the animal kingdom. Second order intentionality (theory of mind itself ) is a major transition point in child development and occurs at around age 4 years as a very discrete phase change in a child’s understanding of the social world (Astington, 1993). The transition is important because it allows children to appreciate that another individual may hold a belief about the world that is different from that held by the child (a so-called “false belief”). At this point, the child is able to factor another individual’s belief state into its calculation of events. This has an important consequence for the child’s capacity to interact with the social world in which it is embedded because it facilitates two activities that it could not previously do: It can engage in fictional play (pretending that the doll can drink tea) and it can lie (deliberately make statements about the condition of the world that it knows to be untrue in order to mislead another individual and cause them to behave in a way that will benefit the child). It is important to appreciate that the latter claim does not mean to say that children younger than 4 years of age cannot lie; rather, it is that the way they lie is very different. Once they have theory of mind, they can use intuitive knowledge about how another individual sees the world to feed it information that will mislead it. The philosopher’s notion of intentionality is implicitly reflexive: In principle, intentionality forms an embedded hierarchy of mind states that has no end. “I intend that you suppose that I want you to understand that I believe that you wish . . .”—the sequence is limited only by the human mind’s ingenuity in constructing meaningful sentences reflecting the mind states of those involved. However, it should be self-evident that the human mind cannot manage to keep track of an unlimited number of these mind states, and in practice the limit seems to occur at fifth order (Figure 2.1) (Kinderman, Dunbar & Bentall, 1998; Stiller & Dunbar, in press). Even so, this is quite impressive, given that 4-yearolds can master only two orders and most animals (with the possible exception of great apes) only one. In an everyday sense, it is perhaps obvious that these kinds of social cognition play an important role in human social behaviour. Ybarra et al. (chapter 16,



FIGURE 2.1 Normal adults typically start to fail multi-level intentionality tasks at sixth order

(redrawn from Kinderman, Dunbar, & Bentall, 1998).

this volume), for example, provide us with evidence that humans strive to maintain a level of behavioural unpredictability on competitive—but not co-operative—tasks, suggesting that we may prefer to make things easier for our neighbours when we need to integrate our behaviour with theirs. That we can make that distinction in itself implies that we are using theory of mind to adjust the cognitive load we place on those with whom we interact. There is some evidence to suggest that this form of social cognition is relatively expensive in wetware terms (i.e., in terms of neural processing demands). With the very limited data we have available, there is some evidence that achievable levels of intentionality across species of primates are correlated with neocortex volume, and specifically with frontal lobe volume (Dunbar, 2003a). This in itself is interesting (the frontal lobe is the principal seat of those crucial cognitive executive functions thought to underpin ToM), but the main lesson for the moment is that the computational demands of intentionality (and, more generally, social cognition) seem to be very heavy. This may in turn provide us with a possible candidate for the selection pressures that have driven brain evolution in primates in particular, and perhaps mammals in general. This has given rise to what has become known as the “social brain hypothesis” and it is to a more detailed consideration of this that I now turn.


THE SOCIAL BRAIN HYPOTHESIS For a long time, the reason why primates have so much larger brains for body size than all other vertebrates (Jerison, 1973) remained somewhat unclear, although the presumption inevitably was that this must have to do with the kinds of ecological strategies that different species pursued. However, during the late 1980s, Byrne and Whiten (1988) proposed that the answer lay in the unusual complexity and sophistication of the primate social world. They couched this in terms of the use of tactical deception and coalition formation, and referred to their hypothesis as the “Machiavellian Intelligence Hypothesis”. However, because reference to Machiavelli raised implications of political machination that were unintended by the original authors, the hypothesis was later renamed the “Social Intelligence Hypothesis” or the “Social Brain Hypothesis” (Barton & Dunbar, 1997; Dunbar, 1998). The principal evidence adduced in support of the social brain hypothesis is that social group size correlates with relative neocortex volume in primates, whereas a selection of indices of ecological constraints does not correlate well with neocortex volume (Dunbar, 1992; Dunbar & Schultz, in press). The finding that social group size correlates with relative neocortex size in primates, and that humans seem to fit into the same quantitative pattern (Dunbar, 1993) has since been generalized in two different directions. One has been to show that a whole suite of behavioural (as opposed to strictly demographic) indices also correlate with neocortex size. These have included the size of grooming cliques (essentially, coalitions) (Kudo & Dunbar, 2001), male mating strategies (Pawłowski, Lowen, & Dunbar, 1997), the amount of social (as opposed to object or solitary) play (Lewis, 2001), and the amount of tactical deception (Byrne & Corp, 2004). This shift of emphasis to actual behavioural strategies was given further support by Joffe’s (1997) demonstration that, in primates, neocortex size correlates best with the length of the juvenile period, whereas the brain as a whole correlates best with the period of parental investment (gestation plus lactation). The significance of this is that the juvenile period is the period of socialization, the period during which the animals’ social skills are developed and honed (often, of course, by imitation and social learning). Thus, Joffe’s finding offers evidence of an important role for what amounts to software development: The size of the neocortex (i.e., its neural computational capacity) is important, but part of that importance lies in its capacity to absorb and use information and experience—and that itself is probably a function of the quantity of neural machinery that can be brought to bear on the problem. The finding that neocortex volume correlates with some behavioural phenomena that inevitably involve social skills suggests that there might be a correlation between a species’ capacities in terms of social cognition and its neocortex volume. Since social cognition seems to have its primary representation in terms of intentionality, this seems an obvious place to look. Though we have yet to



explore this aspect of social cognition in detail, there is some prima facie evidence to suggest that the levels of intentionality to which individual species can aspire may be a function of brain volume. The most direct evidence for this is the fact that the limiting achievable levels of intentionality for Old World monkeys, apes and adult humans (taken to be first, second and fifth order intentionality, respectively) are a linear function of frontal lobe volume. This finding is important in so far as it may provide an explanation why certain kinds of social phenomena are found only in humans. I will have more to say about this in the next two sections.

THE STRUCTURE OF HUMAN SOCIAL NETWORKS In the previous section, I pointed out that, among primates, there is a correlation between social group size and relative neocortex volume. This relationship yields a predicted group size for humans, based on the human neocortex volume, of about 150 individuals (give or take some error variance). Evidence culled from both the ethnographic and the sociological literatures indicates that groups of this size are particularly common in a wide range of human societies (Dunbar, 1993). Not only are the clans of hunter-gatherers the world over of about this size, but the same value characterizes a range of contemporary social groupings, including military units (it is the standard size of the company in modern armies the world over), business organization (it demarcates the point at which hierarchical management structures become necessary), church congregations (a detailed survey suggested that 200 individuals was the absolute upper limit for a coherent, well-integrated congregation), and the typical size of personal social networks (i.e., all those people whom one knows personally) (Barrett, Dunbar & Lycett, 2002; Dunbar, 1993). However, human communities, like those of all primates, are not homogenous social groupings. They are highly structured: Not everyone interacts with everyone else. One way to think of them is as a series of circles of acquaintanceship that surround an individual, rather as the ripples on a pond spread outwards from the point of a stone’s impact. In a recent study (Zhou, Sornette, Hill, & Dunbar, 2005), we were able to show that the sizes of these various grouping levels have a consistent and natural scaling ratio that is almost exactly three. The number of individuals included within each of the successive layers is typically 5, 15, 50, 150, 500 and 1500. These seem to correspond to the following wellestablished human groupings: the support clique of best friends (5), the sympathy group (12–15), the number of individuals contacted at least once a month (and the overnight camp in hunter-gatherers) (30–50), the social network (150), mega-bands in hunter-gatherer societies (500) and tribal groupings in traditional societies (1500). (Note that each layer includes all those individuals included within the inner layers: In other words, the 15 includes the 5 that form the innermost layer, and the 15 themselves are subsumed within the 50 of the next layer.)


What seems to be important about these grouping levels is that they seem to reflect degrees of familiarity or intimacy. In an analysis of Christmas card distribution lists, Hill and Dunbar (2003) showed that these groupings correspond to both levels of intimacy and frequencies of contact. All those individuals who fall into a given ring share similar degrees of intimacy to the person at the centre, and are contacted with roughly equivalent frequency (allowing for geographical distance). What this seems to suggest is that we have only a fixed number of slots at each level of intimacy. Once these are filled up, we cannot easily add new individuals to our social world. It is not entirely clear whether this reflects a purely cognitive constraint (we can only manage so many relationships at a given level of intimacy) or a constraint of time (we can only interact with so many individuals per day, and the frequency of contact limits the quality of the relationship). Either way, it seems that this relationship has important consequences for the pattern of social relationships in two particular respects. First, if someone moves away, the fact that we can contact them less frequently inevitably means that our relationship with them gradually decays: In time, they will naturally slide down through the levels of acquaintanceship until, eventually, they drop off the edge of the 150 primary personal relationships altogether. Obviously, in the modern world, we have a variety of mechanisms for maintaining contact (post, telephone, e-mail), but these merely slow down the rate of decay. We have to make the effort to maintain the relationship, and if we fail to do so the relationship will inexorably decay. Second, if we want to add someone new to our inner circles, then the likelihood is that an existing member will have to drop out to make room for them. This is not, of course, to say that the numerical boundaries are absolutely rigid (they clearly are not), but there are limits to which we can put pressure on them. If the issue really is one of time budgeting, then, inevitably, time spent with one individual must result in less time being spent with others, and hence a natural decay in the quality of the relationship that we have with these particular individuals. There are a number of important individual differences in respect of the size of our social networks. First, there is a slight but statistically very robust sex difference in the size of networks at any given level: On average, those of women are larger than those of men (albeit, of course, with considerable overlap) (Dunbar & Spoor, 1995). This accords well with the finding that there is a correlation between an individual’s achievable level of intentionality and the size of at least the innermost layers of his/her social network size (Stiller & Dunbar, in press). Since, on average, women perform better than men on false belief and other social cognition tasks, it is to be expected that they would naturally have correspondingly larger social circles. In addition, we have been able to show, in other studies, that at least some psychiatric conditions (including both bipolar disorder and schizophrenia) disrupt social cognition capacities, resulting in the loss of the higher orders of intentionality (Kerr, Dunbar, & Bentall, 2003; Swarbrick, 2000). It may not be surprising, therefore, that individuals who suffer from these conditions have rather limited social lives.



TWO SPECIALIZATIONS OF SOCIAL COGNITION I want, in this final section, to suggest that social cognition may have important consequences for two areas of human endeavour and experience that often seem to be overlooked in this context. They are story-telling and religion, and might in many ways be considered the centrepiece of what it is to be human. Story-telling and religion have two key features in common: They require us to be able to imagine virtual worlds that have no immediate experiential content and they require us to assume that those whom we ask to share these imaginings with us can genuinely follow us on this journey. I can perhaps show this most easily in the context of drama. Consider Shakespeare sitting down to compose his play Othello. His problem is this: To make a play interesting, you need to have at least three characters, and you have to ensure that your audience understands what is going on inside all these characters’ minds—anything less, and the story becomes a dull narrative. So, the audience must understand that Iago intends that Othello believes that his wife Desdemona was in love with Cassio—which would probably not be much more than idle fantasy by Desdemona were Iago not also able to convince Othello that Cassio himself also wanted the same outcome. So stated, it will be obvious that, if the audience also has to factor Cassio’s complicity into the equation to make the deception convincing for Othello, it has to be able to work at fifth order intentionality. But to do this, Shakespeare himself must operate at one level higher: He must intend that the audience understands . . . etc. Shakespeare was having to work comfortably at sixth order intentionality, and this is now one level beyond the normal limits for most adult humans. We can look at the play another way that tells much the same story. In practice, of course, the story is played out in a series of scenes in which a variety of characters come and go, and express their feelings and concerns. If we look at how Shakespeare handles the structure of his plays in more detail, then we find that a typical scene involves four speaking characters (Stiller, Nettle, & Dunbar, 2004). As it happens, this mirrors very closely indeed the limits on natural human conversations (Dunbar, Duncan, & Nettle, 1995). So when Shakespeare was constructing the play and putting the story and characters together scene by scene, he was actually working with four characters’ mind states, and he was thus expecting his audience to work at fifth order if they were going to be able to follow the convolutions of the plot as it developed. At this point, Shakespeare himself, of course, had to work at sixth order. One interpretation of this analysis is that the difference between a good story-teller (be they dramatist or novelist) is that they can take their audience to their natural cognitive limits and still make it work. In doing so, of course, it is important that they do not push the audience beyond their natural limit, otherwise the story will become incoherent for them. Indeed, those of Shakespeare’s plays that are conventionally regarded as difficult for audiences (Titus Andronicus is a notable example) are precisely those in which he tries to include too many


characters. So good story-tellers tread a fine line here: they must judge the audience’s capacities so that they do not under-challenge them (or the story will seem uninteresting), but at the same time they must not over-challenge them (else they will lose all but the small number of cognitively very skilled individuals). What probably makes story-telling cognitively demanding is that we (audience and composer alike) have to imagine a fictional world. Even though, in a conventional drama, the characters are on stage in front of us, nonetheless we have to make a leap of imagination to suppose that these characters (the actors) are someone other than whom they actually are. To be able even to begin to imagine that possibility, we need minimally second order intentionality (theory of mind). As I have already noted, Barrett et al. (2003) have argued that this ability to work in a virtual (or fictional) world may be especially demanding cognitively and may explain why great apes (and hence humans) have such large neocortices compared to other primates. I have suggested elsewhere (Dunbar, 2003b, 2004) that the same demands are true of religion. Religion in any meaningful sense requires us to imagine that there is another universe that we cannot directly see and touch: this spirit world exists along side the physical world we experience directly, and has its own reality as well as being able to interact with our world. We can conceive that such a world exists, but to do so minimally requires second order intentionality (theory of mind). I must believe that the world as I experience it is not all there is. Though not necessarily a statement about other minds, we can, I think, argue a case here that believing in two simultaneous but separate worlds requires us to run two belief states in parallel. In effect, I have to believe that something is both true (the world is as I see it) and not true (the world as I see it is not the only possible world) at the same time. So far so good, but to make this interesting as a social phenomenon, I have to believe that you also believe this to be the case. But even with what is now third order intentionality, we still do not have anything approximating religion: merely the possibility of two individuals believing in the existence of another parallel universe. To make it into something even approximating a real religion, we both have to believe that there are beings (spirits?) in this parallel universe who themselves have intentions [about how we should behave]. To do this, we require fourth order intentionality. Even so, this can only provide us with a rather impoverished form of religion—you can accept that I believe in this parallel spirit world, but you do not have to believe in it yourself. I refer to this as social religion, and distinguish it from religion in the communal sense that seems to me to be fundamental to the nature of religion—i.e., such that it can be used to enforce conformation to the communal will. To achieve this, we have to add one further layer of intentionality: You and I have to understand that we both believe that these denizens of the spirit world either intend that we should behave with appropriate intentions [otherwise they will exact retribution] or might be willing to intervene in the world on our behalf. In effect, I intend that you believe that we understand that the spirits intend that we act with righteous intent. At this point, we are at fifth order. Now we have what I



would identify as communal religion: There is something that we both have to agree on, and that something involves minds that can enforce their will on us, be willing to help us out by intervening in the future of the world etc. I raise these two examples here specifically because, in many ways, they represent the essence of what it is to be human: culture, and the way in which we humans build the complex world of the imagination in which we live and breathe. The fact that they both seem to depend on fifth order intentionality (and hence on our unusually large neocortices) may explain why they are unique to modern humans and have no counterpart of any kind in any other living lineage of animals. It is my contention that nothing that we do (or any of our ancestors have ever done) comes close to being as cognitively demanding as story-telling and religion. Indeed, I have explicitly argued (Dunbar, 2003b, 2004) that our fifth order capacities really evolved explicitly to make religion possible. This is not for reasons intrinsic to religion itself (which should properly be seen as just another form of story-telling), but rather because religion has provided us with a mechanism for bonding (and bonding in a particularly deep sense) social groups that, by primate standards, are relatively large, and hence especially prone to the free-rider problem.

REFERENCES Astington, J. W. (1993). The child’s discovery of the mind. Cambridge, MA: Cambridge University Press. Barrett, L., Henzi, S. P., & Dunbar, R. I. M. (2003). Primate cognition: From “what now?” to “what if?” Trends in Cognitive Science, 7, 494–497. Barrett, L., Dunbar, R. I., & Lycett, J. E. (2002). Human evolutionary psychology. Basingstoke, UK/Princeton, NJ: Palgrave Macmillan/Princeton University Press. Barton, R. A., & Dunbar, R. I. M. (1997). Evolution of the social brain. In A. Whiten & R. W. Byrne (Eds.), Machiavellian intelligence II (pp. 240–263). Cambridge, UK: Cambridge University Press. Byrne, R. W., & Corp, N. (2004). Neocortex size predicts deception rate in primates. Proceedings of the Royal Society of London, 271, 1693–1699. Byrne, R. W., & Whiten, A. (1988). Machiavellian intelligence. Oxford, UK: Oxford University Press. Dunbar, R., & Spoor, M. (1995). Social networks, support cliques and kinship. Human Nature, 6, 273–290.

Dunbar, R. I. M. (1992). Neocortex size as a constraint on group size in primates. Journal of Human Evolution, 22, 469–493. Dunbar, R. I. M. (1993). Coevolution of neocortex size, group size and language in humans. Behavioral and Brain Sciences, 16, 681–735. Dunbar, R. I. M. (1998). The social brain hypothesis. Evolutionary Anthropology, 6, 178–190. Dunbar, R. I. M. (2003a). Why are apes so smart? In P. Kappeler & M. Pereira (Eds.), Primate life histories and socioecology (pp. 285–298). Chicago: Chicago University Press. Dunbar, R. I. M. (2003b). The social brain: Mind, language and society in evolutionary perspective. Annual Review of Anthropology, 32, 163–181. Dunbar, R. I. M. (2004). The human story. London: Faber & Faber. Dunbar, R. I. M., Duncan, N., & Nettle, D. (1995). Size and structure of freely forming conversational groups. Human Nature, 6, 67–78. Dunbar, R. I. M., & Shultz, S. (in press). Understanding primate brain evolution.


Philosophical Transactions of the Royal Society, London. Gürerk, Ö., Irlenbusch, B., & Rockenbach, B. (2006). The competitive advantage of sanctioning institutions. Science, 312, 108–111. Hill, R. A., & Dunbar, R. (2003). Social network size in humans. Human Nature, 14, 53–72. Jerison, H. J. (1973). Evolution of the brain and intelligence. New York: Academic Press. Joffe, T. H. (1997). Social pressures have selected for an extended juvenile period in primates. Journal of Human Evolution, 32: 593–605. Kerr, N., Dunbar, R., & Bentall, R. (2003). Theory of mind deficits in bipolar affective disorder. Journal of Affective Disorders, 73, 253–259. Kinderman, P., Dunbar, R. I. M., & Bentall, R. P. (1998). Theory-of-mind deficits and causal attributions. British Journal of Psychology, 89, 191–204. Kudo, H., & Dunbar, R. I. M. (2001). Neocortex size and social network size in primates. Animal Behavior, 62, 711–722. Lewis, K. (2001). A comparative study of primate play behaviour: Implications for the study of cognition. Folia Primatologica, 71, 417–421.

Mitchell, P. (1997). Introduction to theory of mind. London: Arnold. Pawłowski, B. P., Lowen, C. B., & Dunbar, R. I. M. (1998). Neocortex size, social skills and mating success in primates. Behaviour, 135, 357–368. Roth, D., & Leslie, A. M. (1998). Solving belief problems: Toward a task analysis. Cognition, 66, 1–31. Stiller, J., & Dunbar, R. I. M. (in press). Perspective-taking and social network size in humans. Social Networks. Stiller, J., Nettle, D., & Dunbar, R. I. M. (2004). The small world of Shakespeare’s plays. Human Nature, 14, 397–408. Swarbrick, R. (2000). A social cognitive model of paranoid delusions. PhD thesis, University of Manchester, UK. Tooby, J., & Cosmides, L. (1990). The psychological foundations of culture. In J. H. Barkow, L. Cosmides, & J. Tooby (Eds.), The adapted mind (pp. 19–136). Oxford, UK: Oxford University Press. Zhou, W.-X., Sornette, D., Hill, R. A., & Dunbar, R. (2005). Discrete hierarchical organization of social group sizes. Proceedings of the Royal Society of London, Series B, 272, 439–444.



The Evolution of Social Inference Processes

The Importance of Signaling Theory STEVEN W. GANGESTAD RANDY THORNHILL

Signaling Systems Processing Incidental Effects Deception Conclusion harles Darwin delayed publication of Origin of Species (1859) because its profound implications for human origins might lead to its premature rejection. Indeed, his book said nothing explicit about humans until the final pages and, even then, merely tantalized readers: “In the distant future, . . . [p]sychology will be based on a new foundation, that of the necessary acquirement of each mental power and capacity by gradation. Light will be thrown on the origin of man and his history.” A dozen years passed before Darwin said more. In 1871 and 1872, he published The Descent of Man and Selection in Relation to Sex and The Expression of the Emotions in Man and Animals, which contain the first words addressing the topic of this Sydney symposium, Darwinian perspectives on human social cognition. Sexual selection, the major focus of the first volume, fundamentally implicated perception of mating prospects’ traits. The functions of emotional expression and its impact on a social community, the topic of the second, involve social inference at their core. Evolutionary psychologists have investigated many phenomena over the past two decades, including cheater detection, mate preferences, reciprocal altruism, kin recognition, cooperation, friendship and trust, and many more. Most implicate social inference and not surprisingly so: Strategic interaction typically




entails inferences about others. That is, the critical information processed is information about other individuals. In this chapter, we use evolutionary biology to flesh out a framework for talking about processes underlying social inference, defined very broadly here to be any inference an individual makes about a feature or state of another individual. The few specific illustrations we discuss are typically familiar to evolutionary psychologists (e.g., inferences involving sexual attractiveness, intentions to cheat, kinship, and female fertility. But the framework applies to social inference more generally (see also, e.g., Searcy & Nowicki, 2005). Briefly, we expand upon the following themes: 1.


3. 4. 5.

Social inference largely relies on two broad kinds of target–perceiver system. The first kind is a signaling system. Here, targets possess specialized adaptations to emit signals to perceivers. Perceivers possess adaptations to receive and respond to these signals. Target and perceiver adaptations have coevolved. The second kind of system is one in which receivers have adaptations to make inferences based on specific information emitted by targets, but targets possess no adaptations designed to convey information to perceivers. Instead, social inference is based on incidental effects of target adaptations that have noncommunicative functions. A signaling system cannot evolve and stably persist if communication does not benefit both targets and perceivers. As perceivers do not benefit by responding to deceptive information, signaling systems will, in general, be reliable or “honest.” In systems in which perceivers respond to incidental effects emitted by targets, perceiver adaptations may be benign or detrimental to targets. The criterion that differentiates the two sorts of systems is simply whether targets have adaptations for signaling. The conclusion that systems will not be deceptive must be qualified in a number of ways.

SIGNALING SYSTEMS The hallmark of a signaling system is coadaptation of targets and perceivers: Targets possess adaptations that function to convey information to perceivers. And perceivers have adaptations that function to process the information conveyed by targets. In evolutionary biology, a function is a beneficial effect of a trait that led it to be selected (Williams, 1966): The function of eyes is seeing and the function of wings is flight. To say that target adaptations function to convey information to perceivers is to say that the beneficial effects that led these traits to evolve occurred through their impact on perceivers’ adaptations to differentially respond as a function of perceiving the traits.


Signaling systems have received much attention from evolutionary biologists in the past two decades. We discuss two major kinds of signals: signals of quality and signals of intent. We do not address signals of need, though their evolution obeys principles similar to those for the other two (e.g., Searcy & Nowicki, 2005).

Signals of Condition Quality or condition refers to an individual’s ability to successfully interact with the environment to acquire and effectively expend energetic resources (e.g., Rowe & Houle, 1996). It is not merely “health” (e.g., Grammer, Fink, Møller, & Thornhill, 2003); in some circumstances, individuals of superior condition may be more prone to disease than others (see below; see also Thornhill & Gangestad, 1999a). For a signal of condition to evolve, individuals in better condition must benefit from the signal, and receivers must benefit from discriminating individuals’ condition. The benefits to signals of condition (or a specific component of condition) most discussed in the literature are mating benefits. Individuals in superior condition may make better mates for a variety of reasons: fitter genes to pass on to offspring; greater ability to provide material benefits such as protection or food; greater fertility (e.g., more viable sperm, greater ability to provision offspring through gestation); absence of disease. It pays receivers to discriminate mate value and those in better condition may receive mating benefits from signaling. Signals of condition may evolve due to other benefits. For instance, perceivers may size up competitors so that they avoid contests in which they would surely lose. Or signals may play a role in predator–prey interactions. Predators may identify weak prey (thereby reducing predation costs) and so strong prey may signal that they are not easy to catch. These examples illustrate an important point: Although both targets and perceivers must benefit from a signaling system for it to evolve, the interaction context in which the signal functions need not be cooperative in nature; it can be highly antagonistic. A signaling system is at equilibrium when neither the signaler nor the receiver benefits from a change (i.e., in signal sent or preference exercised) if the other retains its strategy. For a signal to be an “honest” indicator of one’s quality at equilibrium, a reliable relation between the signaler’s quality and the signal strength must persist. Zahavi (1975) introduced the idea that the costliness of a trait ensures its honesty. He proposed the intuitive notion that individuals who can afford a costly handicap are more viable than those who can’t. Costly signalers can afford to “waste” some of their viability and still have residual viability greater than that of less costly signalers, which renders the handicapping trait an “honest” signal of viability. Costs may be due to the signal’s size, complexity, or other feature requiring effort to produce. Costs can also be mediated socially (see Male Facial Masculinity).


In the past 15 years, honest signaling through handicapping has been quantitatively modeled. Zahavi’s (1975) insight that honest signals of condition are costly has withstood the test of time, but some of his intuitions about why this is so have not. For a signal to validly indicate condition at equilibrium, the size of signal that maximizes individuals’ own fitness must vary as a function of condition. Those of lower quality do not cheat and produce a bigger signal because they are actually worse off by doing so. Though individuals of lower quality gain benefits from the increased size of the signal (for instance, mating benefits), the costs they pay to produce that larger signal (generally, costs in the currency of survival ability) more than offset those gains. For condition to predict optimal signal size, higher quality individuals either get greater benefits out of the signal or pay lower costs for marginal gains in signal size. They might get greater benefits if they live longer to enjoy them. They may pay lower costs because they need not dig as deep into their overall “budget” to increase the size of the signal; what they give up to increase signal size affects their well-being less than what the individual in worse condition must sacrifice to increase signal size. According to Zahavi’s original verbal argument (1975), individuals in better condition can give up more and still be more viable than individuals in worse condition. Big signalers, then, should be healthier and more viable than those in worse condition. While intuitively appealing, thus conclusion does not follow from current signaling models. At equilibrium, individuals of highest quality may have the same, higher, or even lower viability (or health) compared to small signalers, depending on specific parameters of the system (e.g., Getty, 2002). Quality and mortality can actually positively covary in a population, with the highest quality individuals dying, on average, at younger ages than lower quality individuals (Kokko, Brooks, McNamara, & Houston, 2002). This outcome can occur when just a few winners in the signaling game win big. Individuals on the cusp of winning big may dig deep (in costs) to make a big push to be winners, whereas individuals not close to being a big winner may hold back, paying small costs for small signals. This situation might exist in lekking species in which males collectively gather and display to females and a few male winners garner most matings. Signals that reveal differences in quality may also evolve (Grafen, 1990). These amplify pre-existing indicators of quality between individuals (or transmit those differences in an easily perceived signal), so that perceivers can detect them more readily. When establishing territories, male bullfrogs attend to the depth of pitch of other bullfrogs’ croaks, which reflects bullfrog size and thereby quality (e.g., Emlen, 1976). Bullfrogs croak to reveal quality. Croaking is energetically costly. It need not have evolved because costs and hence levels of croaking vary with quality; rather, it may have evolved merely because croaking captures and transmits differences in quality.


How Signals “Get Off the Ground” That reliable signaling systems are stable when signals predict individuals’ quality is easy to understand: It pays mate choosers to choose on the basis of the signal, which reinforces the signaling sex to possess a big signal, thereby stabilizing the signal as an indicator of quality. How the signal becomes predictive of quality is harder to explain. For a trait to evolve as a signal of quality, it must somehow predict quality before it actually qualifies as a signal. In the context of mating, theorists propose two main scenarios by which traits become associated with quality. The first route is preferred-signal-through-sensory-bias. Here, a trait is initially preferred not because mate choosers benefit from the preference. Instead, the preference is a by-product of a sensory adaptation that has a function unrelated to mate choice. For instance, suppose mate choosers are drawn to “redness” because ripe fruits are red (i.e., “attend-to-red” reflects food preference adaptation). Potential mates who exhibit redness, though not edible, attract attention and thereby benefit on the mating market. Initially, red mates are no better than nonred mates. As redness (or size of a red trait) becomes exaggerated (and costly) as a result of selection for it, however, individuals in best condition become best able to display it. Over time, then, the trait becomes an indicator of quality (see Kokko, Brooks, Jennions, & Morley, 2003). This process presumably explains, for instance, how the peacock’s tail evolved to be a signal of quality. The sensory bias model is sometimes proposed as an alternative to an honest quality signaling model (e.g., Kirkpatrick & Ryan, 1991). In the sensory bias model, mate-choosers prefer, say, a big red tail merely because it gains attention, not because it is associated with condition. In the scenario above, initially this sensory bias model does apply. The problem with the sensory bias model as a complete explanation of signaling systems (and hence a viable, real-world alternative to the honest quality signaling model), however, is that it is not evolutionarily stable. As the preferred trait becomes exaggerated, individuals in better condition produce it at lower cost and the signal becomes an indicator of quality. The second route is that the preferred trait varies with quality prior to it evolving as a signal. Individuals in better condition generally have more energy to allocate to traits important to survival and reproduction. In many species, as individuals have greater amounts of energy available or are healthier, they allocate a larger proportion to traits that foster immediate reproduction and a lower proportion to survival traits. Individuals have but one life to live. When energy budgets are low or individuals are sick, it often makes adaptive sense for individuals to protect that one life by engaging in mortality reduction efforts (e.g., sequestering energy reserves to maintain survival, allocating effort to immune function). When individuals’ condition is more favorable to survival, they may reproductively benefit from allocating a greater amount of energy into reproductive traits. Hence, for example, women’s estrogen and fertility levels


increase with their energy stores and energy balance (e.g., Ellison, 2001). Certain traits—often reproductive ones—tend to particularly differentiate individuals varying in condition. Because these traits covary with condition, selection can lead to adaptations in mate choosers to prefer mates who exhibit these traits. When such adaptations evolve, the discriminating traits are sexually selected for their signal value as well for functions they had previously. Their added benefit as signals lead individuals to allocate more effort into developing them, resulting in exaggeration. This process presumably explains mate choice within many lekking species. Females choose males who can hold central territories. Ability to hold central territories probably covaried with male condition before it was a signal females prefer. Subsequently, females did prefer it, males allocated even more effort to it, and ritualized displays of the ability evolved. This process may explain how females in many species come to assess the outcomes of male–male competition.

Probable Examples of Condition Signals Women store two kinds of fat, android fat and gynoid fat. Gynoid fat, which is particularly rich in long-chain polyunsaturated fatty acids thought to be important for fetal and newborn brain growth, is stored in specialized depots in the breasts, hips, and buttocks and is reserved for pregnancy and lactation. Estrogen facilitates storage of gynoid fat. More generally, estrogen facilitates a shift of energetic resources in women toward reproduction (e.g., it increases fertility). Women in better condition can presumably afford to allocate more energy for reproduction. Hence, storage of gynoid fat probably predicts condition or components of it (fertility and reproductive value; Jasienska, Ziomkiewicz, Ellison, Lipson, & Thune, 2004). Display of gynoid fat depots may have become exaggerated to signal condition and reproductive value to men. Men’s preference for a relatively small waist-to-hip ratio (around. 7; e.g., Singh, 1993; Streeter & McBurney, 2003; Thornhill & Grammer, 1999) may be part of a gynoid fat signaling system.

The Waist-to-Hip Ratio.

Male Facial Masculinity. Though women do not find masculine male faces more attractive in general, they prefer more masculine faces when fertile in their menstrual cycles than when not (see Penton-Voak & Perrett, 2001). Men with more masculine faces may also be preferred by women as short-term mates (e.g., Little, Jones, Penton-Voak, Burt, & Perrett, 2002; Rhodes, Simmons, & Peters, 2005). Possibly, women do not prefer them as long-term mates because masculine men may be less faithful and responsible (e.g., Penton-Voak & Perrett, 2001). Development of male facial masculinity is facilitated by testosterone (e.g., Swaddle & Reierson, 2002), which is thought to increase allocation of effort into mating and mate seeking through male–male competition (see Ellison, 2001). Ancestrally, men in better condition may have been able to afford to put larger amounts of effort into mating. Possibly, then, male facial masculinity was shaped


by selection to signal condition and women evolved to respond to facial masculinity as a signal of condition. One question is what costs keep male facial masculinity honest. The amount of energetic effort required to increase facial masculinity is probably minimal and unlikely to maintain the honesty of the signal. More likely, the costs are socially mediated. In some species of birds, males with large color patches or badges signal to other males their intrasexual competitive abilities, which puts them in the fray of competition with other such males. Males who do not truly possess advertised competitive abilities suffer large costs in the process of being tested, which keep the signals honest. Male facial masculinity (and voice quality; e.g., Puts, Gaulin, & Verdolini, 2006) may function similarly and hence have also evolved as a signal that regulates male–male competition.

Signals of Intent Signals of intent have received less attention than signals of condition, but may be very important in some species. In the courtship process, for instance, individuals of one sex may benefit from knowing whether individuals of the other sex will help care for offspring after mating. The factors that create a stable system of signals of intent should be analogous to those that stabilize systems involving signals of condition (Andrews, 2001). Signals are honest when it doesn’t pay individuals to “cheat”—to dishonestly emit a signal. For signals of intent to be honest, it should not pay individuals lacking the intent to emit a signal of intent. Romantic love may have evolved as an honest signal of intent. A way to signal interest in and commitment to a particular person is to intently focus on them and conspicuously ignore other potential mates. The signal is honest if someone who has no commitment to another but rather intends to desert can’t actually benefit from ignoring other potential mates. Frank (1988) and Gonzaga, Haselton, Smurda, Davies, and Poore (2006) have discussed the function of love in similar terms, though with less explicit emphasis on the rationale behind honest signaling. If this signaling view is correct, some consequences follow. First, because the signal is costly, the equilibrium state should be that signalers emit a signal of size just sufficient for it to be honest. Because the function is to induce an inference of intent, it does not pay to produce a signal more costly than one sufficient to induce the inference. Second, the size of the signal sufficient to induce intent should depend on the extent to which the signaler is perceived to have options. Signalers perceived to have few options may have a difficult time convincing perceivers that they are giving up options by intently focusing on one individual. Hence, the size of the signal should vary by signalers’ condition or other valued mate characteristics (i.e., it should be greater in individuals lower in mate value). Third, a cost of oversignaling (in addition to needless expenditure of effort), then, is that perceivers may infer a signaler to be of lower quality than they would otherwise judge because, again, all else equal, a bigger signal is associated with


lower quality. Fourth, if one sex could benefit from deserting more than the other sex could benefit from deserting, the size of signal required to be honest should be larger for the former sex than the latter sex. If men can benefit from deserting more than can women, they should, on average, signal more strongly than do women. The idea that romantic love evolved as a signal has been minimally explored to date (see Galperin, Gonzaga, Laird, & Haselton, 2006).

Noncostly Signaling Systems We’ve discussed costly honest signals of condition and intent. Costliness ensures their honesty. Can honest signals ever be noncostly? They can be when there are no potential conflicts of interest between signalers and perceivers (Searcy & Nowicki, 2005). In cases of honest signaling of condition or intent, those in poor condition or without intent have potential interest in falsely signaling, which conflicts with the interests of perceivers. In other circumstances, no such conflicts exist. For instance, major histocompatibility (MHC) alleles code for cellsurface markers that the immune system uses to detect foreign pathogens. In some species, possibly including humans, individuals obtain genes for offspring compatible with their own by choosing mates with MHC alleles dissimilar to their own (e.g., Penn & Potts, 1999; Wedekind, Seebeck, Bettens, & Paepke, 1995). People (Pause et al., 2005) and mice (Yamazaki, Beauchamp, Curran, Baird, & Boyse, 2000) can detect chemical signatures of MHC alleles through scent. In these systems, individuals may have no interest in deceiving others about their own MHC genotype, as all individuals presumably benefit through mating with an MHC compatible mate. Perhaps detection of MHC genotypes, then, is an example of an honest, noncostly signaling system. Alternatively, MHC detection may not involve a signaling system. Instead, human perceivers may detect MHC signatures that are mere by-products of MHC (specifically, self-peptides presented by MHC molecules that are shed from skin cells; Leinders-Zufall et al., 2004; see Processing of Incidental Effects).

Constraints on Signaling Systems Principles that apply to signaling systems constrain which ones we should entertain as possibly true. For signaling systems to evolve, both signalers and perceivers should benefit. Hence, signaling systems should generally be honest. At equilibrium, signalers should not typically mislead perceivers. In addition, if signals are costly, signalers should enjoy benefits that pay for costs. There are some hypotheses that these constraints rule out, including those concerning the evolution of women’s permanent breasts and gynoid fat deposits. A number of researchers have proposed that these signals are deceptive: (1) Low, Alexander, and Noonan (1987) argued that breast, buttock, and thigh fat deceptively signal female quality (see critiques by Anderson, 1988; Caro & Sellen,


1990); (2) Miller (1996) proposed that women’s breasts deceptively signal pregnancy, leading men to provision nonpregnant large-breasted females (see also Smith, 1984); (3) a common view is that women’s bodily ornaments deceptively signal peak cycle-related fertility throughout the cycle (indeed, permanently; see Thornhill & Gangestad, 2006), which stems from the more general view that sexual swellings in nonhuman primates signal peak cycle-related fertility (see below). As these theories argue that men are duped by female signals, all propose signaling systems that are not evolutionarily stable and hence unlikely to exist in nature. Many species of primates develop sexual swellings (enlarged, often differentially colored anogenital areas) around the time of ovulation (though typically they are not restricted to ovulation, and in some species are even more exaggerated in subfertile adolescent females). Sexual swellings are widely thought to “advertise” ovulation. Pagel (1994) has argued that this view is probably wrong. Females pay costs for swellings. They must experience benefits to pay for the costs. The argument that swellings advertise fertility presumes that females benefit by “waking up” males to be interested in them. Yet males should be strongly selected to detect when females are fertile on their own using byproducts, should they be available (see below). Indeed, males in primate species that lack swellings detect female fertility using by-products (typically, scents). And in chimpanzees, female by-products associated with cycle-related fertility, not swellings per se, are particularly effective at motivating male sexual behavior (e.g., Deschner, Heistermann, Hodges, & Boesch, 2004). A priori, a more plausible theory of these signals, then, is that they signal female condition, to which male delivery of material benefits to females is sensitive. Females may signal condition around ovulation because males pay particular attention to females at that time. Again, in some species swellings are more exaggerated in subfertile adolescents than adult females, though males prefer adult females as sex partners. Possibly, females benefit by signaling to males their condition just prior to entering the reproductive period. Evidence that primate sexual swellings function to advertise female condition is mixed (e.g., Domb & Pagel, 2001; Zinner, Nunn, van Schaik, & Kappeler, 2004).

PROCESSING OF INCIDENTAL EFFECTS In signaling systems, both targets and perceivers have adaptations that play a role. In many cases in which perceivers make inferences about targets, targets do not possess adaptations that function to signal. Rather, perceivers detect incidental effects or by-products of adaptations that have other functions. As just discussed, males in many primate species (and, indeed, nonprimate species) detect female fertility using scent cues related to estrogen levels (e.g., breakdown products of estrogen or ovarian function; e.g., Deschner et al., 2004;


Engelhardt, Pfiefer, Heistermann, & Niemitz, 2004; for a review, see Thornhill & Gangestad, 2006). Females do not possess adaptations that function to produce these cues. Evidence for adaptation is to be found in design. In these instances, there is no evidence that females have specialized mechanisms for producing or disseminating the breakdown products that males detect. The breakdown products are mere by-products. As females have no adaptation for producing cues, they do not pay costs to “signal” males. Males have adaptations to detect cues. But the cues they detect have no signaling function or, indeed, any function; they are breakdown products (Thornhill & Gangestad, 2006). Several studies indicate that women smell more pleasant and sexier to men when fertile than when infertile in their menstrual cycles (e.g., Kuukasjarvi et al., 2004; Singh & Bronstad, 2001; Thornhill et al., 2003; cf. Thornhill & Gangestad, 1999b). Women probably do not possess adaptations to produce and disseminate a scent that lead males to be attracted to them when fertile. Rather, women probably excrete by-products associated with estrogen or ovarian function that men have evolved to find pleasant. Though men have adaptations to perceive female fertility status, women probably have no adaptations to “advertise” fertility. That is, women probably do not pay costs to produce signals to induce men to detect their fertility; instead, men have been selected to detect cues of female fertility in absence of a specialized female signal.

Do Targets Benefit from Perceiver Adaptation? When targets signal, both targets and perceivers should benefit. When perceivers detect incidental effects, targets may benefit—but detection of them could also be detrimental to targets. In many species in which males detect female fertility through by-products, females probably benefit. If males find fertile females, females pay fewer costs to wait or search for mates and suffer fewer harmful effects from ardent males sexually harassing them when they are infertile. To say that females benefit, however, is not to say that male detection of female scent performs a function for females. A function is a beneficial effect that led a trait to be selected and evolve. Again, females probably did not evolve traits designed to signal fertility. When perceiver detection of incidental effects is detrimental to targets, perceivers and targets have conflicting interests over the perceptual process; perceivers benefit from accurate perception, whereas signalers are harmed by it. Conflicting interests between individuals can lead to recurrent antagonistic coevolution, with no stable equilibrium (e.g., Rice & Holland, 1998). For instance, hosts evolve adaptations to defeat pathogens; pathogens evolve adaptations to undermine those adaptations, which lead to counteradaptations in hosts, and so on. When conflicting interests over perceptual processes exist, targets may evolve to suppress or confound cues picked up by perceivers; perceivers may evolve more sensitive detection of cues, which may lead to greater suppression of cues, and so on.


Conflicts of Interest in Detection of Cycle-Based Fertility in Humans In some species male perception of female fertility status is probably detrimental to females. When males and females pair-bond and cooperatively care for offspring, for instance, females may often be better off when males don’t know when they are fertile, for then males cannot time mate guarding efforts to coincide with fertility and females accordingly may be better able to seek extrapair sires when fertile. Humans may be an example (see Simpson & LaPaglia, chapter 10, this volume). In these instances, selection may operate on females to suppress incidental effects of fertility and, therefore, fertility may be “concealed.” Males, in turn, may be selected to perceive ever-more-subtle side effects of fertility status or means of detecting them. Complete suppression of incidental effects may be difficult for females to achieve, particularly if doing so adversely disrupts the system producing side effects (e.g., hormonal variations underlying ovarian function). Men act on cues associated with female fertility. They are more attentive to or possessive of female partners when partners are fertile than when partners are in the luteal phase (Gangestad, Thornhill, & Garver, 2002; Haselton & Gangestad, 2006; Pillsworth & Haselton, 2006). Men also become more attentive of mates during the fertile phase particularly when female mates are more attracted to men other than their partners (but not their partners) when fertile (Gangestad et al., 2002; Haselton & Gangestad, 2006)—i.e., men are most likely to give female partners additional attention at ovulation when female partners should least want it. This system has the signatures of being an antagonistic system, then, in which male information pick-up is detrimental to females. Though studies indicate that men can detect fertility cues in female scent, they probably do so less reliably than male baboons or chimpanzees can detect the fertile states of female baboons or chimpanzees, and possibly because women have been selected to suppress incidental effects of fertility.

DECEPTION In the 1960s and early 1970s, as cognitive psychologists turned to mathematical information theory for inspiration, terminology, and models, so too did behavioral biologists frame animal communication in terms of information transfer. An influential paper by Dawkins and Krebs (1978) dampened enthusiasm for this approach. To assume that communication involves information transfer is to assume that it is accurate and truthful. Assuming that communication is truthful, Dawkins and Krebs argued, assumes that social relationships are cooperative. When social agents have conflicting interests, individuals can benefit from communicating dishonestly. Conflicting interests pervade social relationships, even those between siblings or parents and offspring. Hence, we


should view social signals skeptically; the assumption that they convey “information” is an unwise starting point. Almost 30 years later, signaling theory has come full circle. As we have emphasized, signaling systems based on deception should be rare, despite conflicting interests of targets and perceivers (Searcy & Nowicki, 2005). Stable signaling systems do, in fact, convey “information.” Nonetheless, individuals do sometimes deceive one another. What explains deception? Where is it likely to be found?

Tolerated Dishonesty To persist, a signaling system need only be honest on average across signaling events. In some complex systems, signals may be honest in some contexts but not others (e.g., a signal may reflect quality in younger but not older individuals). Systems can persist so long as the weighted average signal is honest (Kokko, 1997). In other cases, the costs of information processing required to extract a perfectly honest signal are simply not paid for by marginal benefits. A level of dishonesty may be tolerated, despite the fact that, overall, the system is honest.

Asymmetrical Errors As Haselton (2003; Haselton & Buss, 2000) argues, perceivers can maximize benefits without maximizing accuracy. When errors have asymmetrical costs, individuals maximize utility when they make more errors of one type than another. Some inaccuracy in the signaling system may thereby evolve. Targets may be able to capitalize on perceivers’ willingness to tolerate some kinds of error more than others and engage in deceptive communication. If men are biased against missing sexual opportunities (Haselton & Buss, 2000), women may have little trouble feigning sexual interest in men to gain any one of a number of benefits.

Manipulated Incidental Effects Deception may arise in systems that involve pick-up of information through incidental effects. As noted earlier, information extraction beneficial to perceivers may be detrimental to targets in these systems and selection may act on targets to suppress incidental effects. As also noted, in some cases suppression may disrupt functionality of the system producing the incidental effect. Another way for targets to render the incidental effect useless to perceivers is to produce it when the condition it cues is absent—that is, to produce the incidental effect deceptively. This counterstrategy should be most effective if targets are harmed when perceivers detect the absence of a target condition based on incidental effects. If,


for instance, females lose when males detect absence of fertility because males can then afford to be less vigilant of partners and leave offspring care to females, females could, in theory, benefit by deceptively signaling fertility when not fertile. Of course, this system is not stable. If an incidental effect no longer provides reliable, useful information, perceivers will be selected to ignore it. Antagonistic systems of information pick-up are typically not stable.

CONCLUSION Most phenomena evolutionary psychologists study probably involve some form of social inference—inference that one interactant makes about another interactant’s condition, intention, need, relationship, state, and so on. More generally, social psychologists are interested in a wide variety of social inferences processes (e.g., those involving inference of emotional state, personality disposition, likely future behavior). Whenever an evolved social inference process is posited, it may be worth asking what kind of process evolved. Does social inference involve a signaling system? That is, have targets evolved adaptations that function to communicate information? If so, is the system an honest one, as is typically expected? What maintains its honesty? Would both targets and perceivers have benefited sufficiently for the system to evolve? If deception is claimed to exist, is deception of the sort or level that might be tolerated within an honest signaling system? Alternatively, does social inference involve perceiver pick-up of information through incidental effects, with no specialized adaptation for signaling on the part of targets? If so, what are the incidental effects? Why are they good cues? Does perceiver pick-up benefit targets as well as perceivers? Or do conflicting interests between targets and perceivers exist, such that we the system should be subject to antagonistic coevolution? If information pick-up harms targets, do signatures of antagonism (cue suppression or attempted deception) exist? If scientific psychology is truly to be rooted, in Darwin’s words, in a “new foundation” of evolutionary thinking, the forms of social inference processes we posit should obey principles derived from modern evolutionary biology.

ACKNOWLEDGMENT A version of this chapter was presented at the eighth Annual Sydney Symposium on Social Psychology, Sydney, Australia, March 14–16, 2006.

46 EVOLUTION AND THE SOCIAL MIND REFERENCES Anderson, J. L. (1988). Breasts, hips and buttocks revisited. Ethology and Sociobiology, 9, 319–324. Andrews, P. W. (2001). The psychology of social chess and the evolution of attribution mechanisms: Explaining the fundamental attribution error. Evolution and Human Behavior, 22, 11–29. Caro, T. M., & Sellen, D. W. (1990). The reproductive advantages of fat in women. Ethology and Sociobiology, 11, 51–66. Darwin, C. (1859). On the origin of species. London: Murray. Darwin, C. (1871). The descent of man and selection in relation to sex. London: Murray. Darwin, C. (1872). The expression of the emotions in man and animals. London: Murray. Dawkins, R., & Krebs, J. R. (1978). Animal signals: Information or manipulation? In J. R. Krebs & N. B. Davies (Eds.), Behavioural ecology: An evolutionary approach (pp. 282– 309). Oxford, UK: Blackwell. Deschner, T., Heistermann, M., Hodges, K., & Boesch, C. (2004). Female sexual swelling size, timing of ovulation, and male behavior in wild West African chimpanzees. Hormones and Behavior, 46, 204–215. Domb, L. G., & Pagel, M. (2001). Sexual swellings advertise female quality in wild baboons. Nature, 410, 204–206. Ellison, P. T. (2001). On fertile ground: A natural history of reproduction. Cambridge, MA: Harvard University Press. Emlen, S. T. (1976). Lek organization and mating strategies in the bullfrog. Behavioral Ecology and Sociobiology, 1, 283–313. Engelhardt, A., Pfiefer, J.-B., Heistermann, M., & Niemitz, C. (2004). Assessment of female reproductive status by male longtailed macaques, Macaca fascicularis, under natural conditions. Animal Behaviour, 67, 915–924. Frank, R. H. (1988). Passions within reason: The strategic role of the emotions. New York: Norton. Galperin, A., Gonzaga, G., Laird, K., & Haselton, M. G. (2006, June). Love as a commitment device. Poster presented at the annual meetings of the Human Behavior and Evolution Society, Philadelphia, PA. Gangestad, S. W., Thornhill, R., & Garver, C. E. (2002). Changes in women’s sexual inter-

ests and their partners’ mate retention tactics across the menstrual cycle: Evidence for shifting conflicts of interest. Proceedings of the Royal Society of London, Series B, 269, 975–982. Getty, T. (2002). Signaling health versus parasites. The American Naturalist, 159, 363–371. Gonzaga, G. Haselton, M. G., Smurda, J., Davies, M. S., & Poore, J. C. (2006). Love, desire, the suppression of thoughts of romantic alternatives. Unpublished manuscript, University of California, Los Angeles. Grafen, A. (1990). Biological signals as handicaps. Journal of Theoretical Biology, 144, 517–546. Grammer, K., Fink, B., Møller, A. P., & Thornhill, R. (2003). Darwinian aesthetics: Sexual selection and the biology of beauty. Biological Review, 78, 385–407. Haselton, M. G. (2003). The sexual overperception bias: Evidence of a systematic bias in men from a survey of naturally occurring events. Journal of Research in Personality, 37, 34–47. Haselton, M. G., & Buss, D. M. (2000). Error management theory: A new perspective on biases in cross-sex mind reading. Journal of Personality and Social Psychology, 78, 81– 91. Haselton, M. G., & Gangestad, S. W. (2006). Conditional expression of female desires and male mate retention efforts across the ovulatory cycle. Hormones and Behavior, 49, 509–518. Jasienska, G., Ziomkiewicz, A., Ellison, P. T., Lipson, S. F., & Thune, I. (2004). Large breasts and narrow waists indicate high reproductive potential in women. Proceedings of the Royal Society of London, Series B, 271, 1213–1217. Kirkpatrick, M., & Ryan, M. J. (1991). The evolution of mating preferences and the paradox of the lek. Nature, 350, 33–38. Kokko, H. (1997). Evolutionarily stable strategies of age-dependent sexual advertisement. Behavioral Ecology and Sociobiology, 41, 99– 107. Kokko, H., Brooks, R., Jennions, M. D., & Morley, J. (2003). The evolution of mate choice and mating biases. Proceedings of the


Royal Society of London, Series B, 270, 653– 664. Kokko, H., Brooks, R., McNamara, J. M., & Houston, A. I. (2002). The sexual selection continuum. Proceedings of the Royal Society of London, Series B, 269, 1331–1340. Kuukasjarvi, S., Eriksson, C. J. P., Koskela, E., Mappes, T., Nissinen, K., & Rantala, M. J. (2004). Attractiveness of women’s body odors over the menstrual cycle: The role of oral contraceptives and receiver sex. Behavioral Ecology, 15, 579–584. Leinders-Zufall, T., Brennan, P., Widmayer, P., Chandramani S. P., Maul-Pavicic, A., Jäger, M., et al. (2004). MHC Class I peptides as chemosensory signals in the vomeronasal organ. Science, 306, 1033–1037. Little, A. C., Jones, B. C., Penton-Voak, I. S., Burt, D. M., & Perrett, D. I. (2002). Partnership status and the temporal context of relationships influence human female preferences for sexual dimorphism in male face shape. Proceedings of the Royal Society of London, Series B, 269, 1095–1100. Low, B. S., Alexander, R. D., & Noonan, K. M. (1987). Human hips, breasts, buttocks: Is fat deceptive? Ethology and Sociobiology, 4, 249–257. Miller, E. M. (1996). Concealed ovulation as a strategy for increasing per capita paternal investment. Mankind Quarterly, 36, 297– 333. Pagel, M. (1994). Evolution of conspicuous estrous advertisement in old-world monkeys. Animal Behaviour, 47, 1333–1341. Pause, B. M., Krauel, K., Schrader, C., Sojka, B., Westphal, E., Müller-Ruchholtz, W., & Ferstl, R. (2005). The human brain is a detector of chemosensorily transmitted MHC class I-similarity in same- and opposite-sex relations. Proceedings of the Royal Society of London, Series B, 273, 471–478. Penn, D. J., & Potts, W. K. (1999). The evolution of mating preferences and major histocompatibility complex genes. The American Naturalist, 153, 145–164. Penton-Voak, I. S., & Perrett, D. I. (2001). Male facial attractiveness: Perceived personality and shifting female preferences for male traits across the menstrual cycle. Advances in the Study of Behavior, 30, 219–259. Pillsworth, E. G., & Haselton, M. G. (2006). Male sexual attractiveness predicts differen-

tial ovulatory shifts in female extra-pair attraction and male mate retention. Evolution and Human Behavior, 27, 247–258. Puts, D. A., Gaulin, S. J. C., & Verdolini, K. (2006). Dominance and the evolution of sexual dimorphism in human voice pitch. Evolution and Human Behavior, 27, 283– 296. Rhodes, G., Simmons, L. W., & Peters, M. (2005). Attractiveness and sexual behavior: Does attractiveness enhance mating success? Evolution and Human Behavior, 26, 186– 201. Rice, W. R., & Holland, B. (1998). The enemies within: Intragenomic conflict, interlocus contest evolution (ICE), and the intraspecific Red Queen. Behavioral Ecology and Sociobiology, 41, 1–10. Rowe, L. & Houle, D. (1996). The lex paradox and the capture of genetic variance by condition-dependent traits. Proceedings of the Royal Society of London, Series B, 263, 1415–1421. Searcy, W. A., & Nowicki, S. (2005). The evolution of animal communication: Reliability and deception in signaling systems (Monographs in Behavior and Ecology). Princeton, NJ: Princeton University Press. Singh, D. (1993). Adaptive significance of female physical attractiveness: Role of the waist-to-hip ratio. Journal of Personality and Social Psychology, 65, 293–307. Singh, D., & Bronstad, P. M. (2001). Female body odour is a potential cue to ovulation. Proceedings of the Royal Society of London, Series B, 268, 797–801. Smith, R. L. (1984). Human sperm competition. In R. L. Smith (Ed.), Sperm competition and the evolution of animal mating systems (pp. 601–660). London: Academic Press. Streeter, S. A., & McBurney, D. H. (2003). Waist–hip ratio and attractiveness: New evidence and a critique of “a critical test.” Evolution and Human Behavior, 24, 89–98. Swaddle, J. P., & Reierson, G. W. (2002). Testosterone increases the perceived dominance but not attractiveness of human males. Proceedings of the Royal Society of London, Series B, 269, 2285–2289. Thornhill, R., & Gangestad, S. W. (1999a). Facial attractiveness. Trends in Cognitive Sciences, 3, 452–460. Thornhill, R., & Gangestad, S. W. (1999b). The scent of symmetry: A human

48 EVOLUTION AND THE SOCIAL MIND pheromone that signals fitness? Evolution and Human Behavior, 20, 175–201. Thornhill, R., & Gangestad, S. W. (2006). The evolution of women’s extended sexuality, estrus, concealed estrus and permanent ornamentation. Manuscript in preparation. Thornhill, R., Gangestad, S. W., Miller, R., Scheyd, G., McCollough, J., & Franklin, M. (2003). MHC, symmetry, and body scent attractiveness in men and women. Behavioral Ecology, 14, 668–678. Thornhill, R., & Grammer, K. (1999). The body and face of woman: One ornament that signals quality? Evolution and Human Behavior, 20, 105–120. Wedekind, C., Seebeck, T., Bettens, F., & Paepke, A. J. (1995). MHC-dependent mate preferences in humans. Proceedings of the Royal Society of London, Series B, 260, 245– 249.

Williams, G. C. (1966). Adaptation and natural selection: A critique of some current evolutionary thought. Princeton, NJ: Princeton University Press. Yamazaki, K., Beauchamp, G. K., Curran, M., Baird, J., & Boyse, E. A. (2000). Parent– progeny recognition as a function of MHC odortype identity. Proceedings of the National Academy of Sciences, USA, 97, 10500–10502. Zahavi, A. (1975). Mate selection: A selection for a handicap. Journal of Theoretical Biology, 53, 205–214. Zinner, D. P., Nunn, C. L., van Schaik, C. P., & Kappeler, P. M. (2004). Sexual selection and exaggerated sexual swellings of female primates. In P. Kappeler & C. van Schaik (Eds.), Sexual selection in primates: New and comparative perspectives (pp. 71–89). Cambridge, UK: Cambridge University Press.


How the Mind Warps

A Social Evolutionary Perspective on Cognitive Processing Disjunctions DOUGLAS T. KENRICK ANDREW W. DELTON THERESA E. ROBERTSON D. VAUGHN BECKER STEVEN L. NEUBERG

Our Basic Model of How Fundamental Motives Influence Cognitive Processes An Unexpected Disjunction between Visual Attention, Frequency Estimation, and Memory An Opposite Disjunction for Outgroup Males Suppression and Amplification Disjunctions’ Functions A General Model of the Biases Underlying Disjunctions Some Empirical Implications of Considering Disjunctions in Evolutionary/Ecological Terms Conclusion f you walk down a crowded street at noon, which of the passing strangers could you pick out of a lineup an hour later? From the standpoint of common sense, and of the traditional model of attention and memory, your ability to remember other people ought to depend on initial visual attention—you’ll encode those faces you spent more time looking at, and later remember those encoded faces that managed to make it into long-term memory. Our research program on basic social cognitive processes began with just this set of apparently straightforward assumptions—that memory for faces will depend on encoding, which will in turn depend directly on initial visual




attention (e.g., Craik & Tulving, 1975). We were surprised to find that we were wrong. Let’s begin with the classic three-step model of memory that has appeared in general psychology textbooks for decades. In simplified form, this traditional view involves a series of reasonably linear steps (Atkinson & Shiffrin, 1968). First, some subset of all the information in a person’s current environment is registered in sensory memory. For example, we visually attend to some stimuli and never even rest our fovea on others: A person walking across a crowded college campus would likely scan at the height of people’s faces, as opposed to looking up at the sparrows and finches in the trees above her head (the opposite might be true if she were a birdwatcher, but then the finches, but not the faces, would get registered). At the second step, a smaller subset of the most “important” information is selected for encoding and conscious processing in short-term memory (Cowan, 1988). For example, most people in a large crowd may be scanned but never consciously registered; we fixate on a particular few, such as the man on stilts dressed in a clown costume, the fashion model, and the bluehaired grandmother carrying a poodle. From this smaller subset of information making the cut for short-term memory, a still smaller subset is presumed to get deep enough consideration to make it into long-term memory (Ranganath, Cohen, & Brozinsky, 2005). If, for example, you have a conversation with the fashion model, who turns out to be your best friend’s cousin, you will remember the conversation later, while the fleeting image of the blue-haired grandmother and her poodle are lost forever. In this chapter, we explore a number of intriguing “disjunctions”—discrepancies between early and later information processing that violate the expected linear order in interesting ways. For some categories of faces, for example, observers better remember those they looked away from; other categories of faces get preferential initial processing but are then forgotten. One goal of this chapter is to begin developing a model of when and why one might find these sorts of processing disjunctions. Toward this end, we will consider disjunctions as they reflect more generally on evolution-inspired models of cognitive processing. The central assumption driving our research program is that cognitive processing ultimately reflects a mind designed to extract and ponder information prioritized by functional relevance. We begin with the broad assumption that attention, encoding, and memory, as well as the linkages between these basic processes, are designed to serve adaptive ends. If one encounters unexpected nonlinearities, we believe it may be a mistake to adopt a default presumption that they simply reflect glitches in the system. Instead, it is worth considering whether such apparent glitches may reflect a generally functional system (cf. Fletcher, Simpson, & Boyes, 2006; Forgas, chapter 7, this volume). Following a brief examination of several data sets in which we have observed interesting disjunctions, we suggest the outlines of a conceptual model with implications for understanding when and where disjunctions will be found. Finally, we reflect on some additional empirical implications of thinking about cognitive disjunctions more generally.


OUR BASIC MODEL OF HOW FUNDAMENTAL MOTIVES INFLUENCE COGNITIVE PROCESSES With Mark Schaller and Jon Maner, we’ve been conducting a series of studies designed to explore how simple cognitive processes (e.g., attention, encoding, recognition memory) are affected by what we’ve been calling fundamental motivational states (Kenrick, Neuberg, & Cialdini, 2005; Maner et al., 2005). Under the rubric of fundamental motivational states we include affiliation, selfprotection, status-seeking, mate-search, mate retention, and familial care. We assume each of these motivational states is species-typical for Homo sapiens— involving goals that our ancestors would have needed to meet to successfully survive and reproduce in human social groups (Kenrick, Li, & Butner, 2003). We presumed there would be interesting interactions between bottom-up processes like visual scanning and top-down effects of fundamental motives. A fundamental motive is often activated by bottom-up processes—as when a social stimulus array indicates a mating opportunity, a threat to safety, or a chance to enhance one’s status. Once any powerful motivational state is activated, however, we presume it prompts increased attention to relevant features of the situation and suppresses attention to others. Figure 4.1 depicts a partial model. As shown in Figure 4.1, we presumed that activation of a mating goal would increase attention to attractive members of the opposite sex. We also presumed this activation would inspire goal-relevant interpretations. In line with evolutionbased assumptions of error management theory (Haselton & Buss, 2000), for example, we expected males in a romantic frame of mind would be especially likely to see beautiful women as feeling sexual inclinations. We also expected selective attention to attractive women would lead to overestimations of the frequency of those women in crowds of varying attractiveness. Finally, we expected men would be more able to pick those attractive women out of a lineup later. We expected both men and women in a self-protective frame of mind to attend instead to outgroup males, and to encode those men as relatively threatening. Given the phenomenon of outgroup homogeneity, we weren’t sure whether this initial attention would translate into better memory for outgroup males; instead, we thought frightened participants might make more false alarms, falsely recognizing outgroup men that they had not seen. In some ways, our findings supported our predictions (Becker, Kenrick, Guerin, & Maner, 2005; Maner et al., 2003, 2005). But, as often happens, some unexpected findings were the most thought-provoking.

AN UNEXPECTED DISJUNCTION BETWEEN VISUAL ATTENTION, FREQUENCY ESTIMATION, AND MEMORY One series of studies examined visual attention indirectly using frequency estimation (Maner et al., 2003, Exps. 1–3). Observers were briefly presented with



FIGURE 4.1 Basic model of fundamental goals’ effects on information processing. This model presumes activation of a given goal increases attention to goal-relevant stimuli, and then biases how we encode those stimuli, enhancing later memory for those stimuli. The model also presumes that activation of one goal inhibits processing of stimuli relevant to other goals, and that some goals (such as self-protection) have stronger inhibitory effects than others.

arrays of attractive and average-looking male and female faces. Because observers had limited time to process the faces, we expected that faces capturing initial attention—such as attractive members of the opposite sex—should receive greater processing and therefore be preferentially encoded into long-term memory. Thus, when asked to estimate the frequencies of various categories of faces, observers of both sexes should overestimate the number of attractive members of the opposite sex. Results showed both sexes overestimating the number of attractive female faces, suggesting that attractive female faces captured everyone’s attention (Figure 4.2). Although the same effect was observed in both sexes, it did not seem to be due to the same mechanism: Such overestimations were more likely in men not involved in committed relationships, but women who were involved in relationships.


FIGURE 4.2 Frequency estimation and attractiveness. When people are briefly exposed to

arrays of faces, both men and women overestimate attractive female faces but not attractive males. Overestimation scores were created by taking estimations of attractive faces in briefly shown arrays and subtracting estimations of attractive faces from a control condition where arrays were shown for an extended period of time (so participants had time to process all of the faces). Thus, scores greater than zero indicate attractive faces were overestimated and scores less than zero indicate attractive faces were underestimated (data drawn from Maner et al., 2003, Exp. 1, Table 1).

We were surprised to find female participants did not overestimate the number of attractive males, given that male attractiveness is a well-supported component in female mate preferences (e.g., Gangestad & Simpson, 2000). However, these men were complete strangers to the women, and a strange man simply may not reach the threshold as a mating opportunity for a woman, for reasons discussed below. The frequency estimation data suggested that people were preferentially attending to attractive women, but don’t really prove it, since frequency estimation involves a judgment that is cognitively “downstream” from attention per se. To examine visual attention directly, we turned to eye-tracking methods (Maner et al., 2003, Exp. 4). In these studies, participants are presented with arrays of different faces, and we record how long they dwell on different faces, and which ones they return to. As expected from frequency estimation results, participants of both sexes did preferentially visually attend to attractive females as compared to average looking females. This was especially true for male participants with



FIGURE 4.3 Visual attention and attractiveness. Fixation scores greater than .50 indicate

preferential attention to attractive faces. Both sexes preferentially attended to attractive female faces and women preferentially attended to attractive male faces. Had a disjunction not occurred the pattern in this graph should match the graph in Figure 4.2 (based on Figure 4 from Maner et al., 2003, Exp. 4 © American Psychological Association).

unrestricted mating orientations. Counter to the frequency estimation findings, however, women also preferentially looked at attractive over average looking males (see Figure 4.3; Figures 4.2–4.4 were constructed so without a disjunction each graph should have an identical pattern). These results were perplexing: Indirect measures suggested that attractive males did not draw women’s attention, yet direct measures—tracking where women’s eyes went—showed the opposite pattern. A final study helped resolve this paradox. This study tested participants’ memory for attractive and average members of both sexes. Results showed that both sexes, and women in particular, had accurate memory for attractive female faces but poor memory for attractive male faces (Figure 4.4). So, female responses to attractive men provided our first evidence of a disjunction between one stage of processing and another: although attractive men captured women’s initial attention, this extra visual processing did not translate into greater downstream processing—the attractive men were promptly forgotten, and their frequency was not overestimated.


FIGURE 4.4 Memory and attractiveness. Both sexes preferentially remembered attractive

females. Women had particularly poor memory for attractive males. During testing, participants were shown previously seen faces as well as novel foils and asked how confident they were that they had seen the faces before. Relative memory confidence scores are based on ratings of previously seen faces. These scores were computed by taking confidence scores for attractive faces and subtracting confidence scores for average faces. Scores above zero indicate participants were more confident in having seen attractive than average faces of that category, whereas scores below zero indicate the opposite. The effect shown in this graph probably helps understand the disjunction depicted in Figures 4.2 and 4.3: Although women preferentially attend to attractive men, they do not remember them and thus do not later overestimate them. These results are not an artifact of response bias: For novel foils, both sexes were more confident that they had not seen novel male faces than novel female faces. In other words, previously seen and novel female faces were more accurately discriminated from each other than previously seen and novel male faces (based on Figure 5 in Maner et al., 2003).

Another series of studies found a similar pattern among participants playing a version of the old Concentration game requiring them to remember the location of faces concealed behind tiles, and to match identical faces (by turning over only two tiles on any given trial). Again, we found people of both sexes especially good at processing attractive women but not attractive men. Indeed, across three studies, attractive men were matched somewhat less well than were average-looking men (Becker et al., 2005).



Furthermore, there was again an intriguing disjunction between early and later processing. In one experiment, we first flashed up the full array (24 faces) for 6 s, before proceeding with the Concentration game as usual. In this variant, women were more likely to match handsome men than average looking men or women, but only on the initial trial (immediately after getting a view of the whole “crowd”). By the end of the game, however, this initial advantage for good-looking men had been lost. Again, handsome men (or at least handsome strangers) had a very brief attentional advantage in women’s eyes, but that advantage did not persist into downstream processing (see Figures 4.5 and 4.6).

FIGURE 4.5 Results from concentration game study, initial trial. These results are from

the first trial only in a condition in which all participants first were briefly exposed to all faces. These results suggest both sexes’ attention was drawn by attractive females, and that handsome men also drew initial attention (the latter trend was significant only for female participants).


FIGURE 4.6 Results from concentration game study: Overall memory across all trials.

These results suggest any initial processing advantage for the handsome men (as shown in Figure 4.5) was lost quickly after the initial trials.

AN OPPOSITE DISJUNCTION FOR OUTGROUP MALES Our model of goal-directed cognitive processes led us to predict that activating a self-protection motive would cause people to (a) pay greater attention to other people who might be associated with heuristic danger cues, and (b) have biased



interpretations of the possible threats those individuals might pose. In particular, we expected that perceived dangers would increase attention to outgroup males. Consistent with our model, we did find that White students who are feeling threatened (after watching a scary movie) are more likely to perceive anger in the faces of Black men (but not in the faces of Black women or White targets of either sex) (Maner et al., 2005). These effects are not typical “priming effects”— in which people feeling a particular affective state perceive that same state in others (e.g., Forgas & Bower, 1987). Rather than projecting fear onto other people’s faces, frightened participants projected anger, and did so only for members of a potentially threatening outgroup. Fear also led students with implicitly negative attitudes toward Arabs to project anger onto the faces of Arab men and women. We also found that White participants in a self-protective state overestimated the number of outgroup faces in the arrays (Becker et al., 2006). Another set of eyetracker studies reveals that self-protective motivation, rather than causing White subjects to spend more time looking at outgroup males, actually led them to look away from men in general (Figure 4.7, top panel). This visual aversion also occurs if the men in the photos appear to be looking directly at the participant, and is enhanced if the face is wearing an angry expression. We would thus have expected to find that pictures of Black men, from whom visual attention has been diverted, would be especially difficult to pick out of a line-up later. But instead we have found that priming self-protection caused these nonattended outgroup males to be later remembered as well as, and sometimes better than, nonthreatening faces of ingroup members (Becker et al., 2006; Figure 4.7, bottom panel). In addition, and contrary to findings on outgroup homogeneity, we repeatedly find that Black men are remembered with especially high accuracy if they are angry (Ackerman et al., 2006; see Figure 4.8). Neutral black men, on the other hand, produce a high hit rate, but also a high false alarm rate. The punchline of these latter studies is that outgroup males all look the same, unless they’re angry, in which case they are remembered with high accuracy. White participants do not, however, remember these angry outgroup males because they look at them for a longer time; instead threatening faces seem to manifest a version of “flashbulb memory” and require less visual attention to achieve superior recognition (Brown & Kulik, 1971).

SUPPRESSION AND AMPLIFICATION These two types of disjunction can be called amplification and suppression effects. Amplification disjunctions occur when limited processing at an early stage leads to preferential “downstream” processing (illustrated in the upper line of Figure 4.9). The findings for Black and angry males illustrate an amplification


FIGURE 4.7 Attention to Black and White male faces (top); memory for these same faces

(bottom) (from Becker et al., 2006, Exp. 1).

disjunction: Fearful participants spent less time looking at the faces of outgroup males yet had better memory for them. Conversely, suppression disjunctions occur when preferential processing at an early stage does not translate into preferential processing at a later stage (illustrated in the lower line of Figure 4.9). An example of this is women’s reactions to handsome male faces—women look preferentially at these men, but do not remember them later.



FIGURE 4.8 Memory for Black and White faces. Although White participants are not

especially accurate at recognizing Black males with neutral facial expressions, they are quite accurate in recognizing rapidly presented faces of angry Black men (based on Ackerman et al., 2006). This finding provides a disjunction with other results showing that people look away from potentially threatening faces (especially outgroup males with angry expressions).

DISJUNCTIONS’ FUNCTIONS Why should these two types of disjunction occur? In retrospect, both the amplification and suppression effects we found make functional sense. Because staring at a stranger can be a threat gesture, it should have been unsurprising that people look away from potentially dangerous others—outgroup males, for example, particularly if they are angry and staring back, and particularly if other cues, such as your own feelings of fear, suggest the current situation may be dangerous. A fascinating implication of this research is that not looking does not mean not attending. Given that those individuals nevertheless pose a threat, it makes sense that the mind continues to process them even though the eyes have discreetly moved away. Thus, the amplification effect reveals a sort of a “flashbulb memory” in which a brief but important stimulus gets enhanced mental representation later. The suppression effect for handsome male strangers seems less intuitively sensible at first, but does fit well with findings on women’s criteria for mate


FIGURE 4.9 Two different types of disjunction (note: thanks to Mark Schaller for suggesting

this graphic depiction of disjunctions).

choice. Several evolutionary psychologists have provided evidence to suggest that male physical attractiveness is associated with so-called “good genes” (e.g., Gangestad, Thornhill, & Garver, 2002). Hence, it makes sense that handsome men’s faces elicit initial attention from women. Consistently, in other work, we find more visual fixations for handsome men amongst women who are ovulating, who are unrestricted, or who are in a romantic frame of mind (Maner et al., 2003; Perea et al., 2006). However, even if a woman is interested in a short-term relationship, it is unlikely that that relationship will be with a man who has not stayed around long enough to pass several levels of initial screening. Before committing to a relationship with a man, women generally require additional information, including reliable information about the man’s social status or financial status (Buunk, Dijkstra, Fetchenhauer, & Kenrick, 2002; Kenrick, Sundie, Nicastle, & Stone, 2001; Li, Bailey, Kenrick, & Linsenmeier, 2002). Clark and Hatfield (1989) found in two studies conducted across two decades that not a single woman accepted an offer of a sexual liaison with a strange man, even though about half were willing to go on a date with him. One presumes that some of these women, undergraduates at Florida State during the peak of the sexual revolution, were unrestricted, and that some were ovulating. But a total stranger, regardless of his good looks, simply does not pass the initial threshold for a woman to consider as a sexual partner. On the other hand, Clark and



Hatfield’s data also made it abundantly clear that, for most men, a total stranger is well above threshold to meet his selection criteria—with over 70% of men saying yes to an offer of sex from a woman they had never before met. Although both types of disjunctions violate the traditional linear view of information processing, they make sense in light of a model presuming information processing functions to promote survival and reproductive goals. Rather than leading us to scrap our general functional model of cognitive processing, then, these disjunctions have reinforced our view that cognitive systems are inherently adaptive.

A GENERAL MODEL OF THE BIASES UNDERLYING DISJUNCTIONS Evolutionary approaches to cognitive psychology generally presume some degree of modularity; which implies that different types of content receive different types of processing (e.g., Kenrick, Sadalla, & Keefe, 1998; Tooby & Cosmides, 1992). A functional analysis of cognition thus suggests that content is of central importance; the decision rules used for processing information about a potential mating opportunity, for example, are different from the decision rules used for processing information about a potential threat. An evolutionary perspective implies that the particular cognitive biases used by any species should reflect functional constraints imposed by typical problems their ancestors had to face. So, for example, diurnal birds (with good vision for finding food in daylight) condition nausea to the visual features of novel foods they encounter, but rats (nocturnal creatures with poor vision who find food at night) condition nausea to the taste of novel foods more easily than to visual features (Wilcoxon, Dragoin, & Kral, 1971). Further, a given bird species may use different rules for remembering locations of food stores, features of aversive foods encountered in the past, and the song of their species. The features of aversive foods are conditioned to nausea in a single trial, and are very difficult to unlearn; the locations of stored foods are repeatedly and easily learned and forgotten; and the species’ song is learned during a particular critical period by different rules depending on the social arrangement typically confronted by members of a particular species (Sherry & Schacter, 1987). In addition to different cognitive rules for learning and remembering different kinds of input, animals also have different sensory capacities and different innate templates for recognizing recurrent patterns of stimulation with functional significance. So, for example, hawks, which hunt small and fast-moving animals from high above the earth, have exquisite color vision, including two separate foveas, and several times the density of rods that humans have (Ehrlich, Dobkin, & Wheye, 1988). On the other side, rabbits, a favorite food of these raptors, have “hawk detectors”—early level pattern detectors built into their


retina (to avoid the several milliseconds’ delay associated with central processing, enough time for a speedy hawk to arrive) (Levick, 1967). Just as other animals inherit cognitive templates, so too do humans. For instance, all species have mechanisms that allow them to recognize members of their own species and even their specific mates or offspring, and we would expect humans to have reliably developing templates for recognizing attractive and unattractive members within each sex. These templates, while adaptively influenced by the developmental environment, should also have a great deal of builtin content (Lieberman, Tooby, & Cosmides, 2003). Similarly, humans may have a template for outgroup member, but this template needs to be “filled in” with a great deal of information from the environment—the look of an enemy varies by place and time, and the template must be learned and contrasted with learned features of people with whom we are familiar (cf. Hirshfeld, 1996). An evolutionary approach to cognition implies strongly that adaptive design of nervous systems did not suddenly stop with Homo sapiens, but that our species has a brain and sensory mechanisms adapted to the recurrent demands of human life. Humans don’t need early warning hawk-detection systems, but we do confront a series of special problems involved in living with other humans. For one thing, we need to be highly attentive to the grunts and groans emanating from the mouths of other humans, and to be able to recognize and make fine discriminations regarding very complex patterns within those utterances. It makes a big difference whether someone just said “No worries, mate” as opposed to “Nick’s worried, Mark!” Indeed, ample evidence suggests that the human brain is specially designed to receive and transmit linguistic information in a way that even our most intelligent primate cousins are not (Pinker, 1994). Our model of fundamental motivational systems presumes there is special and differential processing for information relevant to different social goals. Table 4.1 indicates what we think some of those biases are. We also presume there are evolutionarily significant variations in how different individuals respond to different types of information associated with these fundamental problem sets. Some of those individual differences, like sex, are innate; some, like mating strategy, depend on interactions between innate characteristics and developmental inputs, and some, like one’s current mating status or the existence of offspring, are mainly determined by experiential inputs that trigger species-typical biases.

SOME EMPIRICAL IMPLICATIONS OF CONSIDERING DISJUNCTIONS IN EVOLUTIONARY/ECOLOGICAL TERMS Although not considered in an evolutionary framework, traditional cognitive psychologists have uncovered evidence of analogous disjunctions in such phenomena as “inattentional blindness” (not consciously registering objects even though a person is looking at those objects) and “covert attention” (conscious


64 EVOLUTION AND THE SOCIAL MIND Table 4.1 Domains of Social Life Posing Recurrent Problems, with Examples of Decision Constraints, and Cognitive Biases Associated with Each Social problem domain

Evolved decision constraints (examples)

Resultant cognitive biases (examples)

Coalition formation

Exchange relationships are ultimately beneficial to the extent exchange partners (a) share our genes, (b) are good bets for future reciprocation. Men tend to compete for status more than women do.

Coalitional goals should lead to preferential attention to smiling or scornful expressions, particularly on targets not sharing our genes. Status goals should lead to preferential attention to large dominant males or attractive well-dressed females. Males should be especially prone to such biases. Attention to signs of anger, particularly on faces of males and/or outgroup members. Attention to signs of disease in unrelated others. Mating goals should increase attention to physical attractiveness in women, to status in men, and to one’s own mating relevant characteristics. Women should be attentive to signs of commitment in desirable males. Preferential attention to potential interlopers, particularly those of own sex with desirable mating characteristics, or signs of interest in one’s own mate. Preferential attention to behavior of unrelated children who are age-mates of own offspring, or to behaviors of adults likely to pose threats (e.g., low status males).



Mate choice

Outgroup members and unrelated members of own group pose recurrent sources of competition, disease, and physical threat. Mating opportunities are low cost for men, potentially higher cost for women; male commitment is key for female reproductive success.

Relationship maintenance

Costs associated with loss of mating and parenting investment, slightly different, though overlapping, for women and men.

Parental care

Human parents have high investment in biological offspring, potential conflicts with interests of unrelated children.

processing of objects without looking directly at those objects) (e.g., Carrasco & McElree, 2001; Mack, 2003). A consideration of the different domains of social life suggests other places to look for disjunctions between the different stages of cognition. For example, perhaps activation of status concerns will lead to a tendency to look away from high status males, but to remember them better than when other motives such as affiliation or family care are activated. In describing his years in Tibet, for example, Heinrich Harrer (1996) noted that everyone looked immediately at the ground if the Dalai Lama came into view. One doubts that they forgot the initial glimpse of the young god-king, however. When parental motivations are activated, on the other hand, people may look away from, but still remember, low status males (who are otherwise quickly dismissed from further processing). One might also expect that men with their relational


partners might show such a pattern for beautiful women—looking away, but covertly devoting attentional resources. One might expect amplification disjunctions for subtle cues linked to social exclusion, signs of a mate’s infidelity, potential threats to one’s own status, or disease cues in strangers, all of which are likely candidates for privileged processing (Eisenberger, Lieberman, & Williams, 2003; Faulkner, Schaller, Park, & Duncan, 2004; van Vugt & Kurzban, chapter 14, this volume). On the other side, one might expect suppression effects for information suggesting one’s own insensitivity to the needs of rejected or downtrodden others, or to information suggesting one’s own potential infidelities (“I really don’t think the attractive new lab assistant is flirting with me, dear, she’s just a naturally friendly person”). There are potentially interesting connections between these simple cognitive disjunctions and other cognitive phenomena. What kinds of social stimulus do we have difficulty keeping out of conscious working memory? One suspects some social stimuli are harder to suppress than thoughts about white bears, and these might map nicely onto the domains in Table 4.1, and include insults to one’s status, threats to one’s children, attractive offers of infidelities, others flirting with one’s mate, etc. There are also undoubtedly interesting adaptive discontinuities in judgment processes (cf. Todd, chapter 9, this volume). Further, some forms of psychopathology may be understood as individual differences in attention to, encoding of, and memory for, evolutionarily significant social situations (cf. Badcock & Allen, chapter 8, this volume). We have found theoretically meaningful individual differences linked to these cognitive biases, with males and females showing different reactions to attractive members of the opposite sex, for example, and individuals concerned about safety being more susceptible to processing biases involving potential threats. From an evolutionary perspective, other individual differences in cognitive processing might be expected based on life-history phase of the judge (different cues ought to be privileged or suppressed by people who are prepubescent, courting, young parents, or grandparents, for example), or the judge’s kinship status vis-à-vis the targets being processed (cf. Laham, Gonsalkorale, & von Hippel, 2005; Park & Schaller, 2005). Another interesting set of questions involves the neuropsychology of disjunctions. Perhaps emerging neuropsychological methods could be used to examine the possibility that certain social stimuli (such as recently encountered handsome strangers) are inaccessible to conscious processing, but nevertheless accessible to processing at other levels. Might females show physiological signs of recognition of these attractive men even as they are reporting an inability to recall them? If so, this would be a memory equivalent of the findings on “blindsight”— in which people with certain types of brain damage are unable to report seeing a stimulus, but can point correctly when asked to guess where it is in the visual field.



CONCLUSION Evolutionary models of cognitive processes are, in a sense, all about preferential treatment of certain classes of inputs, which often deviate from the standard assumptions applying to the processing of nonsense syllables or other “neutral” stimuli. Evolutionary models of cognition also typically assume that these preferential processing biases are associated with functionally relevant individual differences (Kenrick, 1994). As one example, consider the findings that, although men are good at outdoor map-following tasks that would have fit with dispersed hunting, women are better at detecting and remembering the location of objects in complex arrays, a skill critical to successful foraging (which is more often the province of women in preliterate societies) (e.g., Silverman & Eals, 1992). Likewise, detecting people who cheat on social contract rules is much easier than detecting violators of logically identical rules that are not social contracts (Cosmides & Tooby, 1992). A particularly appealing feature of an evolutionary approach to cognition is that, by emphasizing content, it can bring a whole new set of dimensions to traditional process-oriented approaches. The emphasis on domain specific qualifications to domain-general processes, which suggests numerous separate and specialized cognitive modules adapted to specific fitness problems, has a number of fruitful heuristic implications. The disjunctions we discussed here imply that the operation of domain-general cognitive processes themselves, and specifically the links between them, may be conditioned by more ancient motivational and emotional systems. Thus, while the traditional approach has yielded numerous important general descriptions of cognitive processing, adding an evolutionary perspective opens a whole new set of questions about how efficiently these basic processes work. Understanding these processes should be important to developing a comprehensive and sensible model of how humans understand and represent the social world.

REFERENCES Ackerman, J. M., Shapiro, J. R., Neuberg, S. L., Kenrick, D. T., Schaller, M., Becker, D. V., et al. (2006). They all look the same to me (unless they’re angry): From out-group homogeneity to out-group heterogeneity. Psychological Science, 17, 836–840. Atkinson, R. C. & Schiffrin, R. M. (1968). Human memory: A control system and its control processes. In K. W. Spence & J. T. Spence (Eds.), The psychology of learning and motivation (Vol. 2, pp. 89–195). New York: Academic Press. Becker, D. V., Kenrick, D. T., Guerin, S., &

Maner, J. K. (2005). Concentrating on beauty: Sexual selection and sociospatial memory. Personality and Social Psychology Bulletin, 31, 1643–1652. Becker, D. V., Maner, J. K., Robertson, T., Tybur, J., Shapiro, J., Ackerman, J., et al. (2006). A disjunction between attention and looking: Visual (in)attention to outgroup males. Manuscript in preparation. Brown, R., & Kulik, J. (1971). Flashbulb memories. Cognition, 5, 73–99. Buunk, B. P., Dijkstra, P., Fetchenhauer, D., & Kenrick, D. T. (2002). Age and gender dif-


ferences in mate selection criteria for various involvement levels. Personal Relationships, 9, 271–278. Carrasco, M., & McElree, B. (2001). Covert attention accelerates the rate of visual processing. Proceedings of the National Academy of Sciences, 98, 5363–5367. Clark, R. D. & Hatfield, E. (1989). Gender differences in receptivity to sexual offers. Journal of Psychology and Human Sexuality, 2, 39–55. Cosmides, L., & Tooby, J. (1992). Cognitive adaptations for social exchange. In J. Barkow, L. Cosmides, & J. Tooby (Eds.), The adapted mind: Evolutionary psychology and the generation of culture (pp. 277–306). New York: Oxford University Press. Cowan, N. (1988). Evolving conceptions of memory storage, selective attention, and their mutual constraints within the human information processing system. Psychological Bulletin, 104, 163–191. Craik, F. I., & Tulving, E. (1975). Depth of processing and the retention of words in episodic memory. Journal of Experimental Psychology: General, 104(3), 268–294. Ehrlich, P. R., Dobkin, D. S., & Wheye, D. (1988). Hawk-eyed. In The birder’s handbook: A field guide to the natural history of North American birds (pp. 229–231). New York: Simon & Schuster. Eisenberger, N. I., Lieberman, M. D., & Williams, K. D. (2003, October 10). Does rejection hurt? An fMRI study of social exclusion. Science, 302, 290–292. Faulkner, J., Schaller, M., Park, J. H., & Duncan, L. A. (2004). Evolved disease-avoidance mechanisms and contemporary xenophobic attitudes. Group Processes and Intergroup Relations, 7, 333–353. Fletcher, G. J. O., Simpson, J. A., & Boyes, A. D. (2006). Accuracy and bias in romantic relationships: An evolutionary and social psychological analysis. In M. Schaller, J. A. Simpson, & D. T. Kenrick (Eds.), Evolution and social psychology (pp. 189–209). New York: Psychology Press. Forgas, J. P., & Bower, G. H. (1987). Mood effects on person perception judgments. Journal of Personality and Social Psychology, 53, 53–60. Gangestad, S. W., & Simpson, J. A. (2000). The evolution of human mating: Trade-offs

and strategic pluralism. Behavioral and Brain Sciences, 23, 573–644. Gangestad, S. W., Thornhill, R., & Garver, C. E. (2002). Changes in women’s sexual interests and their partners’ mate retention tactics across the menstrual cycle: Evidence for shifting conflicts of interest. Proceedings of the Royal Society of London, Series B, 269, 975–982. Harrer, H. (1996). Seven years in Tibet. New York: Tarcher/Putnam. Haselton, M. G., & Buss, D. M. (2000). Error management theory: A new perspective on biases in cross-sex mind reading. Journal of Personality and Social Psychology, 78, 81–91. Hirschfeld, L. A. (1996). Race in the making: Cognition, culture, and the child’s construction of human kinds. Cambridge, MA: MIT Press. Kenrick, D. T. (1994). Evolutionary social psychology: From sexual selection to social cognition. In M. P. Zanna (Ed.), Advances in experimental social psychology (Vol. 26, pp. 76–122). San Diego, CA: Academic Press. Kenrick, D. T., Li, N. P., & Butner, J. (2003). Dynamical evolutionary psychology: Individual decision rules and emergent social norms. Psychological Review, 110, 3–28. Kenrick, D. T., Neuberg, S. L., & Cialdini, R. B. (2005). Social psychology: Unraveling the mystery. Boston: Allyn & Bacon. Kenrick, D. T., Sadalla, E. K., & Keefe, R. C. (1998). Evolutionary cognitive psychology: The missing heart of modern cognitive science. In C. Crawford & D. L. Krebs (Eds.), Handbook of evolutionary psychology (pp. 485–514). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. Kenrick, D. T., Sundie, J. M., Nicastle, L. D., & Stone, G. O. (2001). Can one ever be too wealthy or too chaste? Searching for nonlinearities in mate judgment. Journal of Personality and Social Psychology, 80, 462–471. Laham, S. M., Gonsalkorale, K., & von Hippel, W. (2005). Darwinian grandparenting: Preferential investment in more certain kin. Personality and Social Psychology Bulletin, 31, 63–72. Levick, W. R. (1967). Receptive fields and trigger features of ganglion cells in visual streak of rabbit retina. Journal of Physiology, 188, 285–307.


68 EVOLUTION AND THE SOCIAL MIND Li, N. P., Bailey, J. M., Kenrick, D. T., & Linsenmeier, J. A. (2002). The necessities and luxuries of mate preferences: Testing the trade-offs. Journal of Personality and Social Psychology, 82, 947–955. Lieberman, D., Tooby, J., & Cosmides, L. (2003). Does morality have a biological basis? An empirical test of the factors governing moral sentiments relating to incest. Proceedings of the Royal Society of London, Series B, 270, 819–826. Mack, A. (2003). Inattentional blindness: Looking without seeing. Current Directions in Psychological Science, 5, 180–184. Maner, J. K., Kenrick, D. T., Becker, D. V., Delton, A. W., Hofer, B., Wilbur, C. J., et al. (2003). Sexually selective cognition: Beauty captures the mind of the beholder. Journal of Personality and Social Psychology, 85, 1107–1120. Maner, J. K., Kenrick, D. T., Becker, D. V., Robertson, T. E., Hofer, B., Neuberg, S. L., et al. (2005). Functional projection: How fundamental social motives can bias interpersonal perception. Journal of Personality and Social Psychology, 88, 63–78. Park, J. H., & Schaller, M. (2005). Does attitude similarity serve as a heuristic cue for kinship? Evidence of an implicit cognitive association. Evolution and Human Behavior, 26, 158–170. Perea, E. F., Anderson, U. S., Kenrick, D. T., &

Neuberg, S. L. (2006). Women’s attention– memory disjunction for attractive male faces. Paper presented at the 18th Annual Human Behavior and Evolution Society Conference, Philadelphia, PA. Pinker, S. (1994). The language instinct. New York: Morrow. Ranganath, C., Cohen, M. X., & Brozinsky, C. J. (2005). Working memory maintenance contributes to long-term memory formation: Neural and behavioral evidence. Journal of Cognitive Neuroscience, 17, 994–1010. Sherry, D. F., & Schacter, D. L. (1987). The evolution of multiple memory systems. Psychological Review, 94, 439–454. Silverman, I., & Eals, M. (1992). Sex differences in spatial abilities: Evolutionary theory and data. In J. H. Barkow, L. Cosmides, & J. Tooby (Eds.), The adapted mind: Evolutionary psychology and the generation of culture (pp. 531–549). New York: Oxford University Press. Tooby, J., & Cosmides, L. (1992). Psychological foundations of culture. In J. Barkow, L. Cosmides, & J. Tooby (Eds.), The adapted mind: Evolutionary psychology and the generation of culture (pp. 19–136). New York: Oxford University Press. Wilcoxon, H. C., Dragoin, W. B., & Kral, P. A. (1971). Illness-induced aversions in rat and quail: Relative salience of visual and gustatory cues. Science, 171, 826–828.





Appraisals, Emotions, and Adaptation


Theories of Emotion Implications of Appraisal Theories Ambiguous Situations and Incomplete Emotions How Different are Modular and Appraisal Theories, Really? century ago emotion was regarded as quite separate from cognition— hotter, more primitive, and generally maladaptive. Reason and reflection were considered higher faculties, faculties that distinguished humans from other species, and that were more fully developed in civilized white men than in women, children, mental patients, and members of so-called “primitive cultures.” The assumption was that we evolved, both phylogenetically and ontogenetically, away from emotion and toward rational thought. Today most scholars agree that emotion is adaptive, conferring important evolutionary advantages (Cosmides & Tooby, 2000; Nesse, 2004; Tooby & Cosmides, 1990). This idea was pioneered by Arnold (1960), Tomkins (1962, 1963), and Lazarus (1991), and made credible to the scientific community by brain scientists such as Damasio (1994) and LeDoux (1996). In order to survive, an organism must respond to its environment by action. Action is more important than thinking. A bacterium that acts in response to the noxious or nutritive qualities of its environment by tumbling away or staying put can survive and multiply without thinking. A brilliant thinker who understands the universe but feels nothing has little chance of surviving: It will know that the gun or the tiger is lethal, that the mushroom is edible, that the female passing by has a perfect waist-to-hip ratio, but it won’t care. Unless its parents had strong emotions motivating them to protect and nurture it, it probably would have died in infancy. In human beings, most behavior is driven by emotions. There are of course




basic drives, such as hunger, thirst, elimination, and sex, but whenever these are frustrated, they are accompanied by emotion, and arguably it is the emotion, not the drive, that produces the motivational force (Tomkins, 1962). And human beings are motivated by an infinite variety of provocations not associated with biological drives—the desire to be appreciated, to have fun, to fit in, and the fear of failure and ostracism. Most of the mechanisms that evolutionary psychologists have proposed as enhancing inclusive fitness—escaping from predators, finding food, choosing a mate, favoring one’s kin—are fraught with emotionally-driven motivations. Emotions would only be adaptive if they motivated behavior that was appropriate to the situation. Emotions, to be functional, must be driven by an appreciation of the significance of the situation for the organism’s well being. It means, as many evolutionary psychologists have pointed out (Barrett & Kurzban, 2006; Cosmides & Tooby, 2000; Lieberman, chapter 11, this volume; Tooby & Cosmides, 1990) that information processing is essential to emotion, that emotions rarely exist without cognition, and often are the outcome of complex reasoning processes. The organism (or the species) learns what matters, which events are threats and which are opportunities, and what kinds of threat and opportunity they are. Instead of cognition slowly replacing emotion in ontogenetic or phylogenetic development, emotion evolved along with cognition, each clarifying and controlling the other, and each becoming increasingly complex. People came to fear not only tigers and snakes but ridicule and loss; to desire not only good food and mates, but social prestige and interesting activities; to feel anger not only in response to physical attack but to criticism and snubs; to feel disgust not only at putrefaction and deformity but at “immoral” social customs (Rozin & Fallon, 1987). Most human emotions involve higher-order cortical processing. Emotion is not the only mechanism that has evolved for motivating appropriate responses to situations. Most species accomplish the same end with innate neural programs that are set to respond to species-specific stimuli (the red dot on the herring gull’s bill, the silhouette of a hawk) with fixed action patterns (pecking, freezing). A particular stimulus automatically triggers a particular adaptive response. These responses are modular in the strict sense of the term: encapsulated, automatic, and domain-specific (Fodor, 1983). Emotions provide a more flexible alternative, “decoupling” the response from the eliciting stimulus (Cosmides & Tooby, 2000; Ellsworth & Scherer, 2003; Scherer, 1984). Rather than triggering a specific behavior, the situation provokes an emotion. One component of this emotion is an action tendency, a motivation to respond in a certain way, but it is a motivation, not an actual behavior, and it can be overruled. A person sees a small hairy object moving across the floor, perceives it as a spider, and feels fear. Her heart beats faster and she is motivated to run away. But then she realizes that it is not a spider at all, but a dust-bunny. Her fear subsides, and she does not run away. Even though emotions provide an adaptive motivation to immediate appropriate action, they


allow flexibility both in the interpretation of the stimulus and in the choice of response. The person realizes that the object is not a spider, so there is no need to flee. Or, if it is a spider, the initial motivation to flee may be overridden by a decision to step on it, or even by the recognition that it is a benign spider that eats garden pests and should be left alone.

THEORIES OF EMOTION There is general agreement that emotions are adaptive. The next question, about which there is considerably less agreement, is “What is the emotional repertoire in humans that enables them to recognize and deal adaptively with familiar and unfamiliar situations that have significance for their well-being?” Or, as Cosmides and Tooby (2000) put it, what is the algorithm for monitoring relevant situation-defining cues and assigning behavioral priorities? There are two current theoretical responses to this question: categorical theories and dimensional or appraisal theories. Of course there are intermediate points of view, and variations within each major theoretical perspective, and the following discussion is inevitably an oversimplification.

Categorical Theories According to categorical theories, there is a fixed set of qualitatively distinct basic emotions, such as fear, anger, and sorrow. Discrete emotions are natural kinds (Barrett, 2006). Darwin (1872/1965) analyzed the properties of each distinct category of emotion, but he was by no means the first to conceptualize emotions in this way. Most philosophers (cf. Solomon, 2000) and most ordinary people think in terms of discrete, categorically distinct emotions. The most extensive 20th-century development of this point of view is that of Silvan Tomkins, who proposed that there are nine basic emotions, each produced by a separate innate neuromotor “affective program” with its own neurophysiological, expressive, and subjective features (Tomkins, 1962, 1963). Paul Ekman and Carroll Izard are the two psychologists most responsible for perpetuating Tomkins’ theory and exploring it empirically (Ekman, 1972; Izard, 1971), but there are many other categorical theorists, some with theories of emotions in general (e.g., Levenson, 2003; Panksepp, 2000; Plutchik, 2003), and many who study a single emotion category such as fear, disgust, or anger. Most evolutionary psychologists who talk about emotion also assume, implicitly or explicitly, that emotions are categorically distinct. Their categories of emotion are domain-specific modules that detect significant situations and have associated algorithms that regulate behavioral responses (Cosmides & Tooby, 2000; Tooby & Cosmides, 1990). Unlike the psychological theorists, who usually propose a small handful of basic emotions—usually between six and ten, the evolutionary psychologists generally postulate a very large number of


modules, each designed to respond to a single kind of situation—being chased by a predator, seeing another man flirting with your wife, competing for maternal attention, being sick, and many more. Emotions are designed for coping adaptively with recurring ancestral situations in which the wrong response would diminish fitness. Figure 5.1 illustrates some possible situation-specific modules. Several criticisms have been leveled at traditional categorical theories (cf. Ortony & Turner, 1990). First, most categorical theories postulate a small number of basic emotions—somewhere between 6 and 20, but both our intuition and our lexicon include far more emotional states. Jealousy, sympathy, embarrassment, and loneliness are common emotions that are not included in most categorical theories. Two common approaches to dealing with this problem are (1) to add new categories, and (2) to postulate that the excluded emotions are blends or mixtures of the included ones. There have been moves over the past quarter-century to add contempt (Ekman, O’Sullivan, & Matsumoto, 1991) and embarrassment (Keltner, 1995) to the inner circle of basic emotions. But adding a few emotions still results in a set that falls far short of the range of emotions most people feel. The alternative is the idea that two or more emotions can combine to produce a third: Plutchik, for example, argues that love is a combination of joy and acceptance, and hatred is a combination of disgust and anger (2003). But so far the idea of emotion blends is simply a metaphor, with no evidence and not even much theory about the mechanism involved. Do two affect programs fire

FIGURE 5.1 Some examples of emotion-specific eliciting situations.


simultaneously at half blast? Is there rapid alternation between one and the other, or what? Cosmides and Tooby (2000) avoid this problem by postulating a much larger number of situationally-specific feelings and arguing that “many wellknown mental states [e.g., malaise, shock, and the appreciation of beauty] should be recognized as emotion states” (p. 112). A second problem with categorical theories is that they are nominal theories—lists of six or eight or nine unrelated emotions with no means of recognizing similarities and differences among them. Dimensional theorists since the time of Wundt have recognized that some emotions are more similar than others (Schlosberg, 1952). Disgust is more like anger than it is like joy or sorrow. Some categorical theorists now talk about “families” of related emotions (Ekman, 2003), but still without much discussion of the relations among the families. Others, like Schlosberg (1952) and Plutchik (2003) combine categories and dimensions into hybrid circumplex models, for example, with Happiness, Surprise, Fear/Suffering, Anger, Disgust, and Contempt arranged around a circle so that Happiness is close to both Surprise and Contempt. An evolutionary theorist would argue that emotions will be similar if they are responses to similar situations or if they serve similar functions. But without more precise specification of the kinds of similarity that matter for situations or functions, this doesn’t get us very far. Danger, for example, would probably be seen as a relevant dimension of similarity while the fact that two events both occurred on a Tuesday would not, but so far no systematic attempts have been made to define the relevant kinds of similarity. A third criticism of categorical theories is related to the second: They don’t deal well with incomplete emotions or with transitions between emotions. Often people say they feel “upset” or “out of sorts” or generally good or bad but can’t specify a particular emotion. Likewise categorical theories say little about how emotions change from one to another. Vague emotions and emotional transitions are part of emotional life, and for the most part categorical theories have little to say about them. To say that as the situation changes different affect programs will strengthen or weaken does little more than restate the problem. The strategy of postulating “hundreds or even thousands” of modules may solve the problem of excluded emotions, but in so doing, it exacerbates the similarity problem. Consider the small sample of modules in Figure 5.1. Each represents a different emotion-provoking situation, but some resemble each other more than others. Being smiled at and winning a race are positive experiences; the others are not. Seeing a snake and having a sick child might both arouse fear, but fear of quite different sorts—one clear and immediate with an almost automatic action tendency, the other more diffuse with no single obvious motivation to action. The sick child will also arouse sorrow and pity. Hurting one’s friend may result in guilt. Moldy food may be disgusting. The others, although certainly not identical, cluster more closely together: Someone insults me, someone cuts ahead and takes the last parking space, my cat spills my drink all over my work—all of these provoke some variety of frustration and anger. So


will a guy flirting with my woman, although in that situation we are likely to label the emotion jealousy. And when I lock myself out, I feel a somewhat similar emotion, directed at myself (Ellsworth & Tong, 2006). Even the moldy food might provoke anger, if I think my irresponsible spouse left it lying around for days instead of putting it back in the refrigerator. The idea of “hundreds or even thousands” of independent, domain-specific modules fails to provide any information about the relationships among these modules, or the ease of transition from one emotional state to another, and in fact seems to imply that they are unrelated. In more recent work Cosmides and Tooby (2000) suggest that emotional modules may vary considerably in size and specificity. At one extreme is a module like “snake present,” with a specific stimulus and response, but they also talk about “the appreciation of beauty,” and even “positive emotions.” In general, most evolutionary psychologists, including Tooby and Cosmides, seem to accept the same categories of emotion that are proposed by the categorical theorists (Cosmides & Tooby, 2000; Nesse, 2004; Pinker, 1997; Plutchik, 2003). Some focus on a single emotion, others talk about several emotions or even emotions in general, but they assume that emotions such as fear, anger, disgust, and jealousy are modules, that emotions are discrete entities, designed to respond to particular situations in particular ways, that emotions are natural kinds. The evidence for this assumption is not strong. In a recent article, Lisa Feldman Barrett (2006) has reviewed the evidence against categorical theories of emotion, and finds it persuasive. First, the correlations among the criterial components of emotion—autonomic responses, facial expressions, reported feelings, eliciting situations, and behavior—are fairly weak. John Lacey spent much of his career attempting to document the autonomic profiles of different emotions (e.g., 1967), and failed. Neuroimaging studies also show inconsistent results (Barrett, 2006). Studies that have measured two or more components of emotions (e.g., facial and autonomic responses) have generally not found strong evidence for discrete, distinct emotions. A few facial expressions correspond to a few verbal labels (Bonanno & Keltner, 2004), but this may reflect the fact that words represent discrete categories, whether or not facial expressions or emotional experiences do. A wide range of situations can produce an emotion that we would call anger, or fear, or happiness, and each of these so-called categories of emotion can produce a wide range of responses (Frijda, 2006). To reject the idea of discrete modules or categories is not to reject the idea that emotion is adaptive and in fact vitally important in human evolution. The organism must process information about its environment for signs of threats or opportunities, and must be motivated to deal with those threats or opportunities. At least one evolutionary psychologist, Nesse, rejects the modular view: As regards emotion, “the mind’s structure consists not of distinct modules, each shaped to carry out a particular task, but of jury-rigged and partly overlapping mechanisms that one way or another tend to lead to adaptive behavior most of the time” (2004, p. 1338).


Appraisal Theories In the 1980s, several different psychologists came up with quite similar theories, which are generally referred to as “appraisal theories” (Frijda, 1986; Roseman, 1984; Scherer, 1984; Smith & Ellsworth, 1985). The term “appraisal” dates back to Magda Arnold (1960), who proposed that organisms constantly appraise their environment for changes that might be significant for their well-being, and that these direct, immediate, and intuitive evaluations correspond to emotions. The appraisals are associated with central and peripheral nervous system responses, with distinctive subjective experiences, and with action tendencies. Combinations of appraisals are emotions. Appraisal theories have much in common with the views of evolutionary psychologists. Both view information processing as central to emotion. Appraisals and combinations of appraisals can be seen as situation-detecting algorithms (Cosmides & Tooby, 2000), and changes in emotion in response to changes in the situation motivate adaptive behavior (Nesse, 2004). Both emphasize the advantages of the flexibility gained from the possibility of reappraisal of the situation and reassessment of one’s behavioral options. Appraisal theories assume that during evolution emotions developed to (1) evaluate events, (2) set priorities, (3) motivate appropriate behavior, (4) communicate reactions and intentions, and (5) provide flexibility in interpretation and response (Ellsworth & Scherer, 2003). Appraisal theorists, however, do not focus on specific situations, such as the presence of a snake or a guy making a move on one’s wife. Instead, they propose a small set of more general appraisals that are particularly important in distinguishing among situations and therefore among emotions. Different appraisal theorists have somewhat different lists of the appraisals they believe to be most significant in differentiating emotions, but in general, the similarities among them are more conspicuous than the differences. Commonly proposed appraisal dimensions are (1) novelty or change; (2) intrinsic pleasantness or unpleasantness; (3) certainty or predictability; (4) goal facilitation or obstruction; (5) agency (self, other person, impersonal circumstances); (6) coping potential (easy, effortful, hopeless); and (7) compatibility with social norms or one’s own personal standards. All situations can be appraised along these dimensions, and they provide a way of assessing the similarities and differences among situations, and hence among likely emotional responses. In Figure 5.1, two situations clearly involve a change in circumstances in which another person definitely blocks my goal: “Someone insults me” and “Someone cuts ahead of me and takes the last parking space.” In both cases, appraisal theory predicts that I will feel angry. If the guy definitely flirts with my woman, I will also feel angry, but if I am uncertain about what he’s doing, or I’m just hoping she’ll choose me, I’ll feel something a little different, and I might label my emotion jealousy. I don’t need to have separate modules for P spits at me, P insults me, P snubs me, and P slaps me: They all involve a change, a human agent, negative consequences, goal


obstruction, high certainty, and probably a violation of social norms, and they all make me mad. Emotions typically begin when the organism notices a change. This is the appraisal of novelty, and creates a gateway, or state of readiness, for emotion. The appraisal of novelty, like all appraisals, is associated with brain and bodily responses (an orienting response), a change in subjective feelings, and a motivation to action, in this case, to attend closely to the novel stimulus. Any change may attract attention, and in this regard appraisal theories differ from the evolutionary theory of Tooby and Cosmides (1990), who argue that all perceptions that enhance fitness must be situation-specific. Appraisal theorists would argue that since any change of circumstances might affect the organism’s well-being, it is important to be sensitive to novelty in general. Of course many changes will turn out to be insignificant, attention will be fleeting, and the organism will return to whatever it had been doing. The next appraisal, often experienced simultaneously with novelty, is a sense of intrinsic pleasantness or unpleasantness (Zajonc, 1980). Valence is central to almost every theory of emotion ever devised, and in factor analyses of emotional states, valence almost always accounts for the largest proportion of the variance (Smith & Ellsworth, 1985). In order to survive, an organism must be able to distinguish between things that are bad for it and things that are good for it, so as to avoid harm and take advantage of opportunities. Moving toward benefits and away from harm is fundamental to all organisms that move, and may even be considered the function of movement. But human beings are vastly more complicated than the bacterium that moves along a chemical gradient in the direction of more beneficial chemicals (Nesse, 2004). Pleasure and pain can be simple or complex, innate or learned, individual or social, immediate or delayed, to name just a few significant distinctions. Valence is implicated in several different appraisals. Usually the simplest and most immediate is intrinsic attraction or intrinsic aversion, appraisals that lead to approach or avoidance. Responses to some stimuli, such as sweet and bitter tastes or smiling and frowning faces, are apparently innate, hardwired, and universal, but many, such as a taste for Beethoven or single-malt scotch, are not. In either case, the immediate sense of attraction or aversion is a response to the properties of the stimulus itself, not to its relevance to the person’s current purposes. People are also pleased, in a different way, when they perceive events as facilitating the achievement of a goal, and displeased when they perceive obstacles. Some goals, like food or love or relief from pain are universal; many are specific to particular social groups, or particular individuals, or particular occasions. A theory of emotion should include some explanation of why different people experience different emotions in the same situation, and why the same person sometimes experiences different emotions in response to apparently similar circumstances. In appraisal theories these differences are accounted for (1) by the possibility that different people appraise the same situation in different ways,


and (2) by the possibility that people are pursuing different goals (e.g., winning a fight vs. avoiding a fight). For some theorists, goal relevance is necessary condition for emotion (Nesse, 2004; Pinker, 1997), and these theorists would probably not distinguish intrinsic pleasantness from goal-conduciveness. Appraisal theorists generally keep them separate, with goal-conduciveness dependent on the person’s particular motivations at the time, unlike intrinsic pleasantness. In Figure 5.1, “winning the race” is much more obviously tied to an immediate goal than “someone smiles at me.” A downpour may elicit delight if the crops are withering but gloom and frustration if the family had been planning a day at the beach. But a genuine friendly smile reliably elicits a moment of pleasure—it is intrinsically positive in a way that most goal-related stimuli are not. The idea that there is a general appraisal that tells the organism whether its goal is becoming more or less attainable is different from the ideas of some evolutionary psychologists. For example, Cosmides and Tooby (2000) state that the mechanism for choosing the right mate is very different from the mechanism for choosing the right food (p. 99), and this is obviously true in the sense that nobody checks out the waist-to-hip ratio of the choices at a cafeteria. The specific definition of desirable attributes of course depends on the specific goal that occupies a person’s mind at the moment, as well as upon individual, contextual, and cultural differences (an Asian may choose the jellyfish over the cheese; a Westerner usually would not). Several evolutionary psychologists have relaxed the notion of strict modularity to allow for variations in goals and contexts (Barrett & Kurzban, 2006; Nesse, 2004), but they do not propose a general mechanism for recognizing whether the likelihood of reaching one’s goal is increasing or decreasing. Humans may also judge whether a behavior is compatible with important personal or social values, and this appraisal is correlated with a range of positive and negative emotional states that are weak or absent in other animals, such as shame, guilt, contempt, righteous indignation, and pride. Human beings are social animals, depending on shared norms about acceptable and unacceptable behavior. Social organization is sustained by the emotional reactions of group members to behavior that violates them. Anger and contempt can provoke group members to exclude a violator, and exclusion is perhaps the most devastating sanction people can impose on one another (Spoor & Williams, chapter 17, this volume). Certainty and control are appraisals that are often closely related. The meaning of a situation is sometimes ambiguous and the outcome unclear. Uncertainty characterizes interest and surprise. When the outcome may be unpleasant the person feels fear; when a pleasant outcome is uncertain, the person feels hope. People often vacillate between hope and fear as the perceived probability of positive or negative consequences changes. The likelihood of future outcomes often involves the person’s own ability to change the situation or to adjust to it. Lazarus (1966) was the first to clearly distinguish between the appraisal of the situation and the appraisal of one’s


ability to cope with it. According to Ellsworth and Scherer, “the major function of the . . . coping appraisal is to determine the appropriate response to an event; given the nature of the event and the resources at one’s disposal” (2003, p. 580); it allows for flexibility in choice of behavior. Appraisals of control can be fairly complex. Some events, like the weather, are intrinsically uncontrollable. Others may be controllable in principle, but whether one can actually control them depends on one’s mental, physical, material, or social resources. In dealing with competitors or predators, in deciding between fight and flight, the organism evaluates its own power relative to that of the other. Finally, coping is still a relevant dimension even when the event is uncontrollable; in this case the focus is on one’s ability to adjust to the change in circumstances. The appraisal of agency plays an important role in distinguishing among human emotions, particularly among negative emotions. An event can be appraised as caused by oneself, by someone else, or by impersonal circumstances (fate, chance). The same misfortune results in very different emotions depending upon the person’s attribution of its cause. If my child is hurt in a rockslide I feel sad. If she is hurt because someone threw a rock at her I feel angry. If she is hurt because I threw a rock over my shoulder and it hit her, I feel guilty. Perceptions of agency are crucial in distinguishing between feelings of anger, sorrow, and guilt and their associated action tendencies. Among the negative emotions in Figure 5.1, the sick child is likely to evoke sadness (circumstances). If I hurt my friend I will feel guilty (self-agency). The guy flirting with my woman, insulting me, or taking the last parking space will make me angry (other agency). Locking myself out evokes anger at myself, which shares some appraisals with anger and some with shame and guilt (Ellsworth & Tong, 2006). If I anthropomorphize my cat, I will be angry at it. If not, I will just feel frustrated or sad. Among the positive emotions, the appraisal of agency distinguishes among joy (circumstances), pride (self-agency) and gratitude (other-agency). In summary, according to appraisal theories, emotions are not like irreducible particles, but are composed of combinations of simpler but still meaningful appraisals and their associated bodily reactions and action tendencies. If one knows how a person interprets his circumstances, one can predict his emotional state. If one knows what emotion a person is feeling, one can predict how she interprets her circumstances. Any change in an appraisal corresponds to a change in emotion. Over the past quarter-century, considerable empirical support for appraisal theories has accumulated, both within and across cultures (see Ellsworth & Scherer, 2003). The list of appraisals described here is of course not exhaustive. Finer distinctions among emotions (e.g., between shame and guilt, anger and indignation) may require additional appraisals, or may depend upon the specific features of the situation (as evolutionary theorists might argue).


IMPLICATIONS OF APPRAISAL THEORIES No Fixed Number of Emotions Most appraisal theories are dimensional theories, with each appraisal able to take on any value along a continuum. Instead of six or ten or twenty separate categories of emotion, there is a potentially infinite variety of emotional states, and the capacity for subtle shades of feeling. There are no distinct boundaries between emotions. There are no such things as anger or fear or disgust, separated by a gulf from other emotions. As appraisals change, the experience of anger changes until at some point the emotion is better described by some other word in the language, such as frustration or contempt. The verbal labels are categorical; the emotional experience is not. A modular theory can account for many more varieties of emotional experience than a theory based on a small number of discrete categories, but it is still very different from a dimensional theory such as appraisal theory, which allows for an uninterrupted, continuous change in emotions as the situation or one’s interpretation of it changes. Appraisal theories can account for gradual changes in emotion as well as sudden ones.

Emotion as Process In appraisal theories the experience of emotion is not a state, but a continuous process—a river, not a series of pools. In some ways appraisals—or particular combinations of appraisals—are like algorithms for detecting significant situations, but situations are not static. The situation develops over time, and the “same” situation can be reappraised, so that the emotion changes. The great evolutionary advantage of emotion over triggering stimuli and fixed action patterns is flexibility—both in interpretation (and possibly reinterpretation) of the stimulus and in the behavioral response. In many situations the initial action tendency is powerful and automatic, and this is as it should be, particularly in dangerous situations, where mistaken activity is likely to be safer than mistaken passivity. But an action tendency is not an actual behavior, and a reappraisal (or a more complete appraisal) can check the initial impulse and result in more appropriate behavior. A strict modular perspective might have trouble with this sort of fluidity, and in fact Fodor (1983) argued that nonmodular mechanisms often replaced modular mechanisms in the evolution of higher-order cognition (and, I would argue, emotion): “Cognitive evolution would thus have been in the direction of gradually freeing certain sorts of problem-solving systems from the constraints under which input analyzers labor—hence of producing, as a relatively late achievement, the comparatively domain-free inferential capacities which apparently mediate higher flights of cognition” (p. 43). Emotional appraisals can be seen as this sort of “comparatively domain-free inferential capacity.” Many recent modular theorists have considerably relaxed Fodor’s strict criteria of modularity,


however, and in a later section I will consider the compatibility of these modified theories with appraisal theories.

Similarities Among Emotions, and Transitions Theories that postulate a set of categorically distinct “basic emotions” have no way of accounting for our intuition that some emotions are very similar to each other, while others are so different that they seem almost like opposites. In appraisal theories, emotional experiences that share many appraisals will be more similar than emotional states that share few. Negative outcomes brought about by another person will produce some form of anger, across a wide range of people and outcomes. If the event was not caused by a person but by uncontrollable circumstances, the person will feel sad or depressed. Tooby and Cosmides (1990) argue that situations that resemble each other will elicit similar emotions, and that new situations that resemble situations that existed in the Pleistocene will elicit similar emotions, but what it takes for a situation to “resemble” another is not specified. Possibly they mean that emotions that serve similar functions will resemble each other, but there is no more specificity in the definition of functional similarity than there is in the definition of situational similarity. In principle, evolutionary theorists could account for similarities and differences among emotions, if they were interested; in practice, they have not. Cosmides and Tooby speak of “algorithms that monitor for situation-defining cases” (2000, p. 410) as an essential feature of emotions, but do not say what they are: Appraisal theory can be considered an attempt to specify some of these algorithms. Similarly, categorical theories are vague about transitions from one emotion to another. Does one affect program shut down and another start up? Why? Or do they both function at half-strength for a while? Functionally-specific modular theories are mostly vague on this point too: presumably “there will be a dynamic activation and deactivation of these systems, leading to periods of transition” (R. Kurzban, personal communication), but this argument is so abstract that it generates no predictions. Appraisal theories have a much easier time accounting for transitions: Whenever an appraisal changes, the emotional experience will change in predictable ways. If the change is big enough, the experience may be described with a different emotion term. For example, if the situation is negative but the outcome is uncertain, a person will feel fear (watching from high ground, the person see the flood waters rising towards his house). But if the negative outcome occurs for certain (the house is washed away in the flood), the person will feel despair. The appraisal of certainty has changed, and the emotional consequence is predictable. If I discover that my dead dog had been poisoned by my neighbor, my grief will be replaced by anger, corresponding to the change in my appraisal of agency. Both evolutionary theories and appraisal theories predict that changes in the appraisal of the situation lead to changes in the emotional response, but appraisal theory specifies the kinds of appraisals that are important for these changes.


New Situations Most of the situations that arouse people’s emotions did not exist in the Pleistocene, and many, especially in childhood, are novel even within the span of the individual’s own lifetime. Critics argue that modules that developed to adapt to recurring conditions that existed in the Pleistocene cannot explain problem solving or appropriate emotional responses in a world where those conditions no longer recur (Chiappe & MacDonald, 2005). Evolutionary psychologists have responded to this criticism by somewhat relaxing the specificity of modularity (Barrett & Kurzban, 2006; Sperber, 1994). The stimulus for fear may not be as particular as “a tiger running towards me with fangs bared” but could be “a large powerful thing rapidly approaching me,” resulting in fear of cars, motorcycles, and snowmobiles, and provoking appropriate avoidance behavior. The crucial properties of the module are abstract, rather than concrete, and therefore generalizable to situations that the person has never experienced. So far, however, not much has been done to specify the kinds of properties that matter. Theorists provide ad hoc examples of the applicability of prehistoric modules to modern tasks—“collision-avoidance systems could be recruited in driving, strategic social cognition systems could be recruited in chess, and systems evolved for identifying objects such as tools or animals could be recruited to identify letters or words in reading”; Barrett & Kurzban, 2006), but no general theories. That is what appraisal theories attempt to do: to specify the appraisal features of the situation that reliably produce emotions. Is it novel? Is my goal blocked? How much control do I have? A phylogenetically or ontogenetically novel situation can be appraised along the same dimensions as any other situation, and similar combinations of appraisals will result in similar emotions. Tooby and Cosmides (1990) propose that new situations that “seem to resemble” situations that existed in the Pleistocene will elicit the same emotions (p. 417). Appraisal theories put some substance into the term “resemble,” specifying the dimensions of similarity that are emotionally relevant.

AMBIGUOUS SITUATIONS AND INCOMPLETE EMOTIONS A problem with categorical theories (or with theories that propose modules corresponding to “fear,” “sorrow,” “jealousy,” or other discrete emotions) is that quite often people feel emotional, but none of the basic emotion categories—or any other specific emotion that can be named—quite fits. They feel “bad,” or “upset.” It is not a blend of other emotions; it is not anything very specific; they just feel “out of sorts.” Some theorists explain these states by saying that they are not emotions at all, but moods, but that seems like an evasive semantic sleight of hand. Others might argue that the true emotion is unconscious, and that is


certainly possible, but it still doesn’t explain the nature of the person’s conscious emotional experience. According to appraisal theories, a person can become emotional without making all the appraisals that typically characterize a specific emotion. With the first appraisal, typically the appraisal of novelty, there are changes in the central and peripheral nervous system, an interruption of behavior, and a change in subjective feeling. The nature of this emotionality may be highly fluid, constantly changing as new appraisals are added or old ones revised. Or it may remain a vague sense of malaise or well-being, with no new appraisals occurring until the situation changes. Or a person may be quite close to a clear emotion, but not quite there—in a state of almost-anger, for example, if she is not sure whether the event is actually negative or not sure whether a particular person is actually responsible. One can know that a situation looks good or bad, and that it is highly uncertain, but not exactly what it means. The emotion is vague, but still enough to motivate the organism to approach, avoid, or wait. Appraisal theories reject the idea that emotions are modular, bounded categories. Rather than a single emotion of anger or sorrow there are many nuances of each— irritation, indignation, rage, depression, sudden grief, and desperation. These are names; the experience is continuous.

Differentiation A fundamental principle of evolution is differentiation. Single-celled organisms were followed by multicellular organisms, which ultimately evolved into the multitude of plant and animal species that live today. Organisms developed a huge variety of specialized capacities for sensing the vicissitudes of their environment and for responding to it. Species have developed an astonishing number of niche-specific adaptations (Hutchinson, 1978). Neither categorical nor modular theories have much to say about this aspect of evolutionary theory: the categories and modules are there, they are functional, but what was there before they developed or how they came about are unanswered questions. Nesse’s (2004) account of the phylogeny of emotions proposes that emotions, like most other capacities, became more differentiated over the course of evolution, and this view is entirely compatible with appraisal theories. Perhaps in the very beginning the survival of single-celled organisms was simply a matter of luck, but very soon they developed the ability to tumble away from noxious stimuli and to keep going in the direction of beneficial ones. There was valence in their lives. As evolution progressed, animal species developed more specific goals, and sensitivities to whether these goals were becoming more or less attainable and whether or not they had the resources to achieve them, and the variety of emotional responses proliferated. Nesse uses the venerable evolutionary metaphor of a branching tree to describe the development of differentiated emotions, and this view is also compatible with appraisal theory: As the capacity for more sophisticated appraisals develops, the tree develops new branches. Cultures and


individuals may have concerns that foster prolific elaboration of some branches, resulting in many tiny twigs of emotional nuance, while leaving other areas less fully differentiated. The branches of Nesse’s tree do not correspond closely to the commonly-proposed appraisals, but the theories are similar in that they propose an evolutionary progression of affective responses from primitive to complex. The tree is, at this stage, only a metaphor. The great virtue of this metaphor is that it captures the idea of progressive differentiation, which is one of the hallmarks of evolutionary theory. Some appraisal theorists have posited that the appraisals occur in a fixed sequence, phylogenetically, ontogenetically, and in the actual experience of emotion (Scherer, 1984). Others have used a multidimensional model, which misses this important theoretical insight, but allows greater flexibility, and can account for phenomena that are problematical for the tree. For example, in sequential tree models, the first major branch is valence, with the positive emotions irrevocably separated from the negative emotions at an early stage. But sometimes rapid vacillations between positive and negative emotions are possible, as in the case of hope and fear. In a multidimensional space where valence is one dimension among many, quick transitions between hope and fear are more easily accounted for, as is the phenomenon of ambivalence. Both metaphors are heuristic, and for now there is no reason to proclaim that one is sounder than the other, given our state of ignorance.

HOW DIFFERENT ARE MODULAR AND APPRAISAL THEORIES, REALLY? I am new to evolutionary theory, and relatively unschooled. But it seems to me that both the evolutionary theorists and their critics may exaggerate the differences that divide them. The gulf between them may be more a matter of semantics and (perhaps willful) misunderstanding than of real fundamental differences. As in politics, each side creates a caricature of the other side’s views that no one actually believes. The critics of evolutionary theory describe the evolutionary theorists’ view as involving hundreds or thousands of tiny independent modules with little coordination or intercommunication. The evolutionary theorists describe their critics as espousing one huge undifferentiated brain with no domain specificity whatsoever. No one—or hardly anyone—actually espouses either of these extreme positions. Evolutionary theorists postulate superordinate programs to coordinate the smaller modules (Cosmides & Tooby, 2000), context effects and computational resources shared by multiple systems (Barrett & Kurzban, 2006). It is wrong for critics to rely on Fodor’s (1984) strict automatic, encapsulated version of modularity in criticizing current evolutionary psychologists, since almost all of them have relaxed those criteria. And there are differences of opinion among evolutionary theorists. Sperber (1994) argues for partial modularity, and Nesse (2004) for partial differentiation among emotions. Tooby and Cosmides (1990)


flatly state that “novelty cannot in principle be a discrete selection pressure” because it is domain general (p. 410), and they seem therefore to reject the idea that organisms can have a general appraisal of novelty. Sperber argues that sensitivity to novelty exists and is “of course not domain specific” (1994, p. 50). Most evolutionary theorists seem to accept the English-language categories of fear, jealousy, disgust, anger, and others as real, and as modular, and this is a genuine, rather than a false, disagreement between evolutionary theorists and appraisal theorists. Appraisal theorists (and all other psychological theorists) believe in differentiated mental functions and capacities. With regard to emotion, I have called evolutionary psychologists categorical theorists because they talk of discrete emotions as modules (Barrett & Kurzban, 2006; Pinker, 1997; Plutchik, 2003; and even to some extent Nesse, 2004, although he argues that they are overlapping). Appraisal theorists regard these “categories” as unbounded, with a continuum of intermediate states. The evidence that such categories are real is very weak; they are consistent with our intuitions, but not with the data, despite prodigious efforts to find data to corroborate their existence (Barrett, 2006). Language, not experience, is responsible for our perception of emotions as categorical, and, as James (1890/1950) argued, it is unproductive to think of them as “psychic entities” or to worry about how to catalogue them. However, the appraisals themselves might be candidates for modules: the perception of novelty; of intrinsic valence—even in sights and tastes never before experienced; of agency; of controllability. These of course may seem to fly in the face of current evolutionary theory, because they seem like domain-general modules, but that would only be true if “domain” were defined in terms of specific content, and some evolutionary psychologists are moving away from such a limited definition of “domain” (Barrett & Kurzban, 2006). The effort at reconciliation may fizzle, and is certainly more difficult than a continuing series of attacks and counterattacks of positions no one seriously believes in, but it is an interesting new direction, and worth thinking about.

REFERENCES Arnold, M. B. (1960). Emotion and personality: Vol. 1. Psychological aspects. New York: Columbia University Press. Barrett, H. C., & Kurzban, R. (2006). Modularity in cognition: Framing the debate. Psychological Review, 116, 628–647. Barrett, L. F. (2006). Are emotions natural kinds? Perspectives on Psychological Science, 1, 28–58. Bonanno, G. A., & Keltner, D. (2004). The coherence of emotion systems: Comparing online measures of appraisal and facial

expressions, and self-report. Cognition and Emotion, 18, 431–444. Chiappe, D., & McDonald, K. B. (2005). The evolution of domain-general mechanisms in intelligence and learning. Journal of General Psychology, 132, 5–40. Cosmides, L., & Tooby, J. (2000). Evolutionary psychology and the emotions. In M. Lewis & J. Haviland-Jones (Eds.), Handbook of emotions (2nd ed., pp. 91–115). New York: Guilford Press. Damasio, A. R. (1994). Descartes’ error:


Emotion, reason, and the human brain. New York: Avon Books. Darwin, C. (1965). The expression of the emotions in man and animals. Chicago: University of Chicago Press. (Original work published 1872) Ekman, P. (1972). Universals and cultural differences in facial expressions of emotions. In J. Cole (Ed.), Nebraska Symposium on Motivation (pp. 207–283). Lincoln, NE: Nebraska University Press. Ekman, P. (2003). Emotions revealed. New York: Henry Holt. Ekman, P., O’Sullivan, M., & Matsumoto, D. (1991). Contradictions in the study of contempt: What it all about? Motivation and Emotion, 15, 293–296. Ellsworth, P. C., & Scherer, K. R. (2003). Appraisal processes in emotion. In R. J. Davidson, H. Goldsmith, & K. R. Scherer (Eds.), Handbook of affective sciences (pp. 572–595). New York/Oxford, UK: Oxford University Press. Ellsworth, P. C., & Tong, E. M. W. (2006). What does it mean to be angry at yourself? Categories, appraisals, and the problem of language. Emotion, 6, 572–586. Fodor, J. (1983). The modularity of mind. Cambridge, MA: MIT Press. Frijda, N. (1986). The emotions. Cambridge, UK: Cambridge University Press. Frijda, N. (2006). The laws of emotion. Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. Hutchinson, G. E. (1978). An introduction to population ecology. New Haven, CT: Yale University Press. Izard, C. (1971). The face of emotion. New York: Appleton-Century-Crofts. James, W. (1950). The emotions. In The principles of psychology (Vol. 2, pp. 442–485). New York: Dover. (Original work published 1890) Keltner, D. (1995). The signs of appeasement: Evidence for the distinct displays of embarrassment, amusement, and shame. Journal of Personality and Social Psychology, 68, 441– 454. Lacey, J. J. (1967). Somatic response patterning and stress: Some revisions of activation theory. In M. H. Appley & R. Trumbull (Eds.), Psychological stress: Issues in research (pp. 14– 42). New York: Appleton-Century-Crofts.

Lazarus, R. (1991). Emotion and adaptation. New York: Oxford University Press. Lazarus, R. S. (1966). Psychological stress and the coping process. New York: McGraw-Hill. LeDoux, J. E. (1996). The emotional brain. New York: Simon & Schuster. Levenson, R. W. (2003). The autonomic specificity of emotion. In R. J. Davidson, K. R. Scherer, & H. Goldsmith (Eds.), Handbook of affective sciences (pp. 212–224). New York: Oxford University Press. Nesse, R. (2004). Natural selection and the elusiveness of happiness. Philosophical Transactions of the Royal Society of London, Series B, 359, 1333–1347. Ortony, A., & Turner, T. J. (1990). What’s basic about basic emotions? Psychological Review, 97, 315–331. Panksepp, J. (2000). Emotions as natural kinds within the human brain. In M. Lewis & J. Havilland-Jones (Eds.), Handbook of emotions (2nd ed., pp. 137–156). New York: Guilford Press. Pinker, S. (1997). How the mind works. New York: Norton. Plutchik, R. (2003). Emotions and life. Washington, DC: American Psychological Association. Roseman, I. (1984). Cognitive determinants of emotion: A structural theory. In P. Shaver (Ed.), Review of personality and social psychology (Vol. 5, pp. 11–36). Beverly Hills, CA: Sage. Rozin, P., & Fallon, A. (1987). A perspective on disgust. Psychological Review, 94, 23–41. Scherer, K. R. (1984). On the nature and function of emotion: A component process approach. In K. R. Scherer & P. Ekman (Eds.), Approaches to emotion (pp. 293–318). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. Schlosberg, H. (1952). The description of facial expressions in terms of two dimensions. Journal of Experimental Psychology, 44, 229–237. Smith, C. A., & Ellsworth, P. C. (1985). Patterns of cognitive appraisal in emotion. Journal of Personality and Social Psychology, 52, 475–488. Solomon, R. C. (2000). The philosophy of emotions. In M. Lewis & J. Haviland-Jones (Eds.), Handbook of emotions (2nd ed., pp. 2–29). New York: Guilford Press. Sperber, D. (1994). The modularity of thought

88 EVOLUTION AND THE SOCIAL MIND and the epidemiology of representations. In L. A. Hirschfeld & S. A. Gelman (Eds.), Mapping the mind: Domain specificity in cognition and culture (pp. 39–67). New York: Cambridge University Press. Tomkins, S. S. (1962). Affect, imagery, consciousness: Vol. 1. The positive affects. New York: Springer. Tomkins, S. S. (1963). Affect, imagery, con-

sciousness: Vol. 2. The negative affects. New York: Springer. Tooby, J., & Cosmides, L. (1990). The past explains the present: Emotional adaptations and the structure of ancestral environments. Ethology and Sociobiology, 11, 375–424. Zajonc, R. B. (1980). Thinking and feeling: Preferences need no inferences. The American Psychologist, 35, 151–175.


The Evolutionary Bases of Social and Moral Emotions

Dominance, Submission, and True Love ROSS BUCK

Motivation, Emotion, and Communication: The DevelopmentalInteractionist View Altruism Attachment and Higher-Level Social and Moral Emotions Conclusions his chapter considers the deep evolutionary roots of motivation, emotion, and communication with the goal of examining the essential natures of cooperation and competition from the viewpoint of Developmental-Interactionist theory (Buck, 1985, 1999). We first consider definitions of motivation, emotion, and communication. I then argue that fundamental functions of motivation and emotion are revealed in behaviors of microbes, with special attention to the evolution of altruism, with empirical evidence of “paleoaltruism” demonstrating communication and individual sacrifice for group benefit in microorganisms. Finally, this chapter advances a view of the dynamic emergence via interaction of systems of primary social and moral emotions.


MOTIVATION, EMOTION, AND COMMUNICATION: THE DEVELOPMENTAL-INTERACTIONIST VIEW Developmental-Interactionist theory holds that behavior is a function of interactions over the course of development between phylogenetically-structured special-purpose processing systems and general-purpose processing systems structured 89


by experience (Buck, 1985, 1999). In human beings language imposes an additional source of behavior control functionally independent of biology, which may obscure true determinants of behavior by rationalizing acts that are actually determined by deep-seated and largely unacknowledged motives and emotions.

Defining Motivation and Emotion Definitions of “motivation” generally emphasize the activation and direction of behavior toward a goal; and definitions of “emotion” typically note the presence of peripheral physiological responses, expressive behaviors, and subjective experiences or affects (Kleinginna & Kleinginna 1981a, 1981b). DevelopmentalInteractionist theory suggests that motives and emotions imply one another: Motives constitute potential for behavior built into a system of behavior control, and emotions constitute readouts of motivational potential when that potential is activated by an effective stimulus. There are three readouts: Emotion I involves peripheral physiological responding via the autonomic, endocrine, and immune systems; Emotion II involves expressive displays such as facial expressions and pheromone releases; and Emotion III involves subjective or affective experience of desires and feelings (Buck, 1985, 1999). In this view, motivation and emotion are related similarly to energy and matter in physics. Energy is a potential that is not seen in itself, but in its manifestation in matter—in heat, light, and/or force—when activated by an effective stimulus. Thus, potential energy in an explosive or a coiled spring is not observable in itself, but is revealed when activated by lighting the fuse or releasing the spring. Similarly, motivation is not seen in itself, but rather in its manifestations in emotion: in physiological responding, expressive displays, and subjectively experienced desires and feelings. The Developmental-Interactionist view holds that biologically-based motives and emotions are always turned on. Neurochemical systems underlying the experience and expression of happiness, sadness, fear, anger, sex, hunger, thirst, etc., are always activated to some extent, like the pilot light in a gas heater. We can turn our attention to the experience of these affects, but like the feel of the shoes on our feet, this activation is typically weak and unnoticed unless an effective stimulus is presented. But the Emotion III readout of the subjective experience of all of the biological affects is always available and accessible to us. These affects function as ever-present voices of the genes: informative even when whispering. These primordial voices not so much control our behavior as cajole : They coax, wheedle, tempt us with their siren song; coloring our perceptions ranging from mild cheerfulness, annoyance, apprehension, melancholy to passionate euphoria, fury, dread, despair.

Evolutionary Roots of Motivation and Emotion The above definitions encompass primary motivational-emotional states (primes) in human beings and also animals, with the

Primes and Paleoprimes.


complexity of the state compatible with the complexity of the brain of the species in question. However, these definitions can also apply to primary motivationalemotional systems in microbes whose evolution long preceded the evolution of the brain: paleoprimes. Paleoprimes serve functions in simple creatures analogous to those served by motivation and emotion in human beings—activation, approach-avoidance, dispersion, and aggregation—and these can be observed in their behavior (Buck, 1999). Microbes are active at some times and quiescent at others. They can be observed to approach some stimuli (light, warmth) and avoid others. And, as we shall see, microbes also achieve dispersion, aggregation, and cooperation by releasing molecules (pheromones) that repel, attract, or otherwise influence others of their species. These systems in turn underlie paleosociality, involving microbes communicating and thereby completing social structures and organizations (Buck & Powers, 2006). As communication occurs in the course of interaction, social organization emerges spontaneously and effortlessly as a selforganizing system. In microbes, the systems are simple enough that sending and receiving mechanisms can be specified and the system of emergence understood at a molecular and even genetic level. These microbial systems illustrate fundamental principles of social organization that illuminate mechanisms and principles of social organization in more complex creatures, that in human beings may often be hidden by a curtain of language. More specifically, a combination of attachment motives and social comparison processes underlie a natural and effortless emergence of systems of social and moral emotions that guide every human interchange. Motivational potential inherent in attachment systems in infants is activated through early interactions with caregivers, giving rise to fundamental needs to love and to be loved which constitute the affective foundation of sociality (Baumeister & Leary, 1995). The need to be loved is fundamental to human sociality. With development, interactions with caregivers and peers teach the child rules that must be followed in order to be loved: Thus the child learns that one must meet and exceed expectations in order to be loved. The need to meet and exceed expectations therefore requires both needs to be loved and learning what those rules and expectations are and how to meet them. This learning occurs naturally, effortlessly, and largely unconsciously during the course of interactions with caregivers, peers, friends, and lovers. There is evidence that these attachment motives are related to mechanisms of physical pain (Eisenberger & Lieberman, 2005; Eisenberger, Lieberman, & Williams, 2003; Panksepp, 1991). The pain that occurs, naturally and effortlessly, when one seems to fail in these attachment motives is demonstrated by the research on ostracism reviewed by Spoor and Williams (chapter 17, this volume). This is so even when on a rational level such ostracism is clearly sham: Even though one knows rationally that one is being ostracized at random by a computer, the pain is there, and it is real.



Primary Social and Moral Emotions Developmental-Interactionist theory regards social, moral, and also cognitive emotions to be higher-level emotions in contrast to biologically-based emotions, requiring both a physiological base in neurochemical systems associated with attachment and exploration, and a rational consideration of situational and interpersonal contingencies (Buck, 1999). The neurochemical systems provide the affective “fire” to the higher-level emotions, while the contingencies determine the quality of the emotion, rather like the interaction of physiology and cognition in Schachter and Singer’s (1962) theory of emotion. Specifically, individuals are exquisitely aware of success or failure in meeting expectations and being loved, and their own success or failure may be compared, automatically and effortlessly, with the success or failure of comparison others. Possible combinations of success and failure of self and other yields eight combinations of fundamental interpersonal contingencies that correspond to eight primary social emotions in four pairs of twins. If one succeeds relative to the comparison other one tends to experience pride/arrogance, if one fails it is guilt/shame, if the other succeeds it is envy/jealousy, if the other fails it is pity/scorn. The first of these twins is associated with meeting/exceeding expectations, that second with being loved.1

Defining Communication Developmental-Interactionist theory holds that social organization emerges spontaneously and effortlessly as a self-organizing system in the course of interaction between individuals, whether they are microbes or children. The mechanism of this emergence through interaction is communication between elements, that is, between individual microbes and individual children. “Communication” is here defined, following E. O. Wilson (1975), as occurring “whenever the behavior of one individual (the sender) influences the behavior of another (the receiver) . . . behavior can be defined as communicative to the extent that it reduces uncertainty in the behavior of another” (Buck, 1984, p. 4). Communication proceeds in two simultaneous “streams:” one voluntary and symbolic, the other automatic and spontaneous (Buck, 1984; Buck & VanLear, 2002). Voluntary symbolic communication is learned, its elements are symbols that bear an arbitrary relationship to the referent, and its content consists of falsifiable statements or propositions. In contrast, spontaneous communication is biologically structured in both its sending and receiving aspects. An internal motivational/emotional state of the sender is automatically and effortlessly expressed in an evolved display, which given attention is picked up by the receiver and “known” directly via evolved preattunement as a motivational/emotional response in the receiver. The display is not a symbol, but rather is a sign of the sender’s internal state. A sign bears a natural relationship to the referent—an externally accessible aspect of the

Symbolic and Spontaneous Communication.


referent, as in smoke being a sign of fire—so that if the sign is present the referent is present by definition. Therefore symbolic communication is not propositional in that it cannot be false. These simultaneous streams of communication coexist in virtually every communicative exchange, although their relative importance varies. In formal and structured situations, such as a lecture, symbolic communication dominates but spontaneous communication plays a subsidiary albeit important role in conveying for example the charisma of the speaker and enthusiasm of the audience. The symbolic-spontaneous mix also varies with the intimacy of personal relationship of sender and receiver. All else equal, symbolic communication predominates in formal relationships, but as personal relationships develop and become more intimate, the relative importance of spontaneous communication tends to increase. The symbolic-spontaneous mix also varies over the course of development: The newborn is primarily a spontaneous communicator, but as an infant grows into a toddler and comes to learn language, symbolic communication becomes more and more important (Buck, 1984). Finally, the symbolicspontaneous mix varies along the evolutionary scale: as species increase in complexity, relatively inflexible spontaneous communication systems increasingly interact with general-purpose symbolic communication systems, so that communication becomes progressively more flexible. This progressive evolution of increased behavioral plasticity is anagenesis (Gottleib, 1984), and is compatible with R. I. M. Dunbar’s social brain hypothesis (chapter 2, this volume). Individuals have considerable “voluntary” control over the display, in that it is possible for a sender to show motivational and emotional displays that are not really present as internal states. Arthur VanLear and I termed this “pseudospontaneous communication,” because while it is voluntary from the sender’s point of view, it uses the display mechanism that can stimulate preattunements in the receiver, so if the receiver is taken in it is as if it were a veridical display. A charismatic sender, for example, can successfully “push the buttons” of an audience, and persuade by manipulating others’ emotions (Buck & VanLear, 2002). This voluntary expression of a display was termed “voluntary expression formation” in analyses of brain mechanisms of primate audiovocal communication by Jurgens (1979). The question of veridical versus manipulative displays is a fundamental issue in the evolution of communication, as we shall see (see also Gangestad & Thornhill, chapter 3, this volume).

Pseudospontaneous Communication.

Emotion, Communication, and Evolution The classical analysis of the evolution of communication stems from Charles Darwin’s theory of evolution, and particularly The Expression of the Emotions in Man and Animals (1872/1998). Darwin suggested that emotional displays can have adaptive value in social

Communication in Classical Ethology.


animals because they reveal inner states of the responder that are useful for social coordination, including for example aggressive dominance, fearful submissiveness, and sexual readiness. This implies that the inner state of the responder (sender) must be associated with external expression, and that the receiver must be able to “pick up” the expression via sensory cues: postures, facial expressions, pheromones. Darwin’s thesis requires that sending and receiving mechanisms coevolve—evolve in conjunction with one another—for the adaptive value of a system of communication to be realized. Ethologists including Lorenz and Tinbergen agreed that the communication of certain motivational-emotional states is adaptive (Hauser, 1996). Those individuals who show evidence of that state in behavior tend to be favored, and over the generations these behaviors can become “ritualized” into expressive displays. Similar reasoning was applied to the evolution of receiving mechanisms: Individuals who respond appropriately to these displays would tend to be favored, so that the perceptual systems of species members can become “preattuned” to the pickup of these displays. In this way, displays and preattunements coevolve as aspects of systems of spontaneous communication. The classical view of communication was challenged when selection was interpreted as operating, not at the level of the individual or group, but rather at the level of the gene (Hauser, 1996). Richard Dawkins and others argued that the ultimate unit of evolutionary selection is the active, germ-line replicator. A replicator is “anything in the universe of which copies are made,” an active replicator is “any replicator whose nature has some influence over its probability of being copied,” and a germ-line replicator is a “replicator that is potentially the ancestor of an indefinitely long succession of descendent replicators” (1982, p. 83). Dawkins argued that the only active replicator lasting across evolutionary timescales is the gene, so that the “selfish gene” is the unit of selection, a replicator motivated only to make copies of itself. Fitness was seen as based upon the survival, not of the individual organism or the group, but rather upon inclusive fitness : the survival of the genes. The selfish gene critique extended to the understanding of communication, and specifically to the idea that accurate communication is adaptive. Instead, it was argued that selection would actually operate against those who show veridical displays that are predictive of their true inner states and probable behaviors, and suggested that communication is actually a means by which one animal exploits another. In this regard, Dawkins and Krebs (1978) suggested an analogy between animal communication and media advertising, where the object is persuasion rather than transfer of information. This manipulative communication corresponds to pseudospontaneous communication as defined previously, where the display does not in fact truly reflect an internal state but rather is “put on” voluntarily by the sender, but it can activate the preattunements in the receiver and therefore be emotionally compelling. Krebs and Dawkins (1984) suggested that mind-reading on the part of the receiver—interpreting the actual internal The “Selfish-Gene” Critique.


state and predicting the behaviors of other animals—is the counterpart to manipulation on the part of the sender. The sender’s counterresponse to mindreading may involve concealment (a poker-face) and active deception (simulating, qualifying, or falsifying one’s display) (see Gangestad & Thornhill, chapter 3, this volume, for another view of the Dawkins & Krebs position: They point out that to evolve and stably persist a signaling system needs to be honest on average across signaling events).

ALTRUISM Kin Selection and Reciprocal Altruism The phenomenon of altruism, defined as the sacrifice of one’s own genetic fitness in favor of benefiting the genetic fitness of another, poses a fundamental problem for evolutionary theory. Dawkins (1989) argued that the “law of ruthless selfishness” governing the selection of genes implies that true altruism is impossible: “. . . Let us try to teach generosity and altruism, because we are born selfish” (p. 3). Despite this, there are apparent examples of unselfish, cooperative, and even altruistic behavior. It is widely agreed that, under conditions of kinship and/or reciprocity, mutually cooperative communication can foster the inclusive fitness of the altruist, and therefore it can be favored by selection. Because such behavior fosters the survival of the altruist’s own genes via inclusive fitness, altruism based on kin selection and reciprocity is actually selfish and does not contradict the selfish gene hypothesis (Krebs & Dawkins, 1984).

Quorum-Sensing in Bacteria Recently, evidence relevant to the evolution of altruism has come from an unexpected source: microbiology. In Animal Aggregations (1931), W. C. Allee noted self-organizing activity in simple creatures, including bacteria. Bacteria are prokaryotes—essentially bags of DNA—that lack the nuclei, mitochondria, and organelles associated with the eukaryotic cells that make up all complex multicelled life. Despite their relative simplicity, recent studies have discovered surprising complexity and sophistication in the behavior of bacteria, including intra- and intercellular communication resulting in behavior suggestive of intelligence and memory (Hellingwerf, 2005), cooperation and altruism (Griffin, West, & Buckling, 2004; Kreft, 2004a, 2004b), and even “social intelligence” (Ben-Jacob, Becker, Shapira, & Levine, 2004). In addition, there is evidence that prokaryotes lived socially virtually from the beginning. The most ancient known organisms are stromatolites: fossilized colonies of cyanobacteria that self-organize and self-configure to create a spatially-bounded working community that recovers from damage. Recent studies have elucidated mechanisms by which such bacterial social


behavior is coordinated. Many if not most species of bacteria exhibit quorumsensing : mechanisms for recruiting the mass production of molecules or engaging in other collective activities beneficial to the bacteria. The bacteria are quiescent until a critical mass of individuals—a “quorum” of millions or billions—has assembled, and they then produce the molecule en mass in a useful concentration (Swift, Throup, Williams, Salmond, & Stewart, 1996; Waters & Bassler, 2005). An example is the marine bioluminescent bacterium Vibrio fischeri. This bacterium lives freely in a planktonic state, and also exists in symbiotic relationship with certain fish and squid (i.e., the Hawaiian squid Euprymna scolopes), causing luminescence that functions to attract food and camouflage the squid in moonlight (Waters & Bassler, 2005). In the laboratory, a growing colony of V. fischeri remains dark until a relatively high density of individuals is achieved, at which point luminescence increases rapidly. This phenomenon is caused by the action of signal molecules, which increase in concentration with an increasing number of individuals. The signal molecule responsible for the activation of luminescence in V. fischeri was identified in 1981, and the genetic system analyzed in 1983 (Eberhard et al., 1981; Engebrecht, Nealson, & Silverman, 1983). When the level of signal molecules reaches a threshold, they enable proteins called LuxR to bind to specific genes within the individual cells. A molecular apparatus “turned on” by the interaction of LuxR and genes generates the light simultaneously in many individual bacteria (Fuqua, Winans, & Greenberg, 1996; Greenberg, 1997; Waters & Bassler, 2005). Interest in quorum sensing within microbiology has exploded in the past decade: “the number of known regulatory systems and the diversity of phenomena regulated are growing dramatically, and it now appears that most bacteria possess at least one quorum-sensing system” (Redfield, 2002, p. 365). While the functions of these systems are extraordinarily diverse, the systems by which quorum-sensing is based are surprisingly uniform. Quorum-sensing works by a bacterium’s release of an autoinducer molecule into the environment. These constitute signals or displays, and typically involve amino acids or peptides functioning as pheromones (Gallio, Sturgill, Rather, & Kylsten, 2002). The bacterium also has the capacity to sense the concentration of this autoinducer in the environment. If the concentration exceeds a threshold, the expression of genes within the bacterium is altered, producing a variety of effects: motility, swarming, pigment formation, etc. (Redfield, 2002). The production by a bacterium of an autoinducer, and its responsiveness to the environmental concentration of the autoinducer, has all of the qualities of spontaneous communication defined previously. The display (autoinducer production) and preattunement (responsiveness) are biologically based, the autoinducer is a sign of the referent (concentration of individuals), and the communication is in no way intentional and is nonpropositional.


Altruism and Quorum-Sensing The question whether quorum sensing represents a system of communication to promote collective action useful to the group is relevant to the question of cooperation and altruism. In everyday language, altruism is typically related to a concern for the welfare of others and the common good. In the context of evolutionary theory, these are not considered: “Evolutionary altruism . . . does not require memory of past interactions, recognition of individuals, sophisticated interactions or behavioral repertoires, or direct interactions between individuals. It is therefore the simplest form of altruism” (Kreft, 2004a, p. 2751). There is evidence of individual sacrifice for group benefit in microorganisms (Walters & Bassler, 2005). One example involves the slime mold Dictyostelium discoideum. At one stage in its life cycle, slime molds exist as unicellular amoebae feeding on bacteria. At this stage the amoebae exhibit positive chemotaxis (attraction) vis-à-vis their prey and negative chemotaxis (repulsion) vis-à-vis one another (Lackie, 1986). Thus, this simple organism demonstrates a sort of “threat display” that may fulfill functions analogous to those ascribed to territorial displays in more complex creatures (Eibl-Eibesfeldt, 1975). As bacteria in the environment are consumed, the amoebae begin to starve, the negative chemotaxis to other amoebae ceases, and a positive chemotaxis begins. Within 4–6 hours, the individual amoebae begin to move toward aggregation centers, which apparently contain those individuals whose positive chemotaxic systems were first “turned on” by starvation. The aggregation center next forms a multicelled slug, or grex, in which the cells are derived from individual amoebae. The grex moves in a looping motion like an inchworm caterpillar from the domain of the individual amoebae in damp forest litter toward the light in a journey that may take many days. The sensory analysis of the environment presumably takes place in the front tip of the grex, and it is the cells in this area that become anchored and “altruistically” die to form the cellulose stalk of a fruiting body (Strassmann, Zhu, & Queller, 2000). The stalk formation is altruistic on the part of these individuals in the technical evolutionary sense. They “surrender . . . personal genetic fitness for the enhancement of personal genetic fitness in others” (Wilson, 1975, p. 106), or behave “to increase another such entity’s welfare at the expense of its own” (Dawkins, 1976, p. 4). Cells at the rear of the grex form a mass that climbs to the top of the stalk to become individual spores, which are released into the environment. Given favorable conditions, they germinate into individual amoebae, and begin the life cycle again. A similar life cycle occurs in the soil-dwelling bacterium Myxococcus xanthus involving quorum sensing in the development of fruiting bodies (Clarke, 1981; Losick & Kaiser, 1997; Shimkets, 1999). In both cases, “spore development requires a large percentage of the population to undergo a lethal differentiation event that leads to structures whose function is to promote spore generation and dispersion” (Waters & Bassler, 2005, p. 336). These indeed appear to be


examples of altruism—in the technical evolutionary sense—at the microbial level: How can this be reconciled with the powerful and compelling gene-selectionist account of evolution?

The Communicative Gene Hypothesis The issue of the possibility of “true” altruism turns on the issue of the unit of selection: whether the unit of selection is the individual gene, or whether evolution can involve the selection of units above the level of the individual gene. A possible solution is that communicative relationships between genes can be replicators; units of selection in evolution (Buck, 2002; Buck & Ginsburg, 1991, 1997). This retains a gene-selectionist position without a largely unexamined accompanying assumption of genetic atomism: Genes are selected in isolation from other genes: “selection purely at the level of the individual gene” (Dawkins, 1989, pp. 84– 85). Buck and Ginsburg (1991) noted that genes do not function alone; rather, genes function in company with other genes, and more specifically, genes function by communicating with other genes. This is not a controversial contention, but its implications regarding the unit of selection have perhaps not been fully appreciated. A critical postulate of the communicative gene view is that in any system of interacting elements, communication involves both individual elements and the unique relationship between those elements relative to other elements. This analysis suggests that communicative relationships can be replicators in Dawkins’s (1982) sense. Communicative relationships arguably meet the criteria for being active germ-line replicators. A replicator is anything of which copies are made; an active replicator influences the probability of being copied; a germ-line replicator is an ancestor of descendant replicators; and replicators exist across evolutionary time scales. Communicative relationships and can be copied via the selection of the phenotype communication; the nature of such relationships can influence the probability of being copied; and these relationships can exist across evolutionary timescales. Therefore, communicative relationships between genes can be active, germ-line replicators. Moreover, communicating genes are not necessarily within the same cell or organism. The quorum-sensing example demonstrates that genes in different individual bacteria can communicate via autoinducer molecules functioning as signs of population density: It involves displays in the sender and preattunements in the receiver.

Communication and the Unit of Selection.

Communicative Relationships as Units of Selection. There are many examples of specific communicative relationships that have existed across evolutionary timescales; indeed, the ritualized displays associated the classical ethological view meet this criterion. Specific displays associated with dominance, submission, warning, courting, and bonding have existed across evolutionary timescales and in many species: For example, Livingstone, Harris-Warrick, and


Kravitz (1980) demonstrated that serotonin injections in crayfish and lobsters produce characteristic dominance postures. Also, injections of octopamine (the phenol analogue of norepinephrine) produce subordinate postures; and the mating pheromone in the single-celled yeast Saccharomyces cerevisiae is a peptide molecule that closely resembles GnRH involved in mating in mammals, including human beings (Loumaye, Thorner, & Catt, 1982). Both examples imply a conservation of sending and receiving mechanisms across an enormous span of time. The social brain hypothesis (Dunbar, 1993, chapter 2, this volume) is quite consistent with the communicative gene hypothesis, as is evidence that social group size correlates with relative neocortex volume in primates. Indeed, communication arguably may be the mechanism of the relationship between group size and brain size: The social brain is the communicative brain.

ATTACHMENT AND HIGHER-LEVEL SOCIAL AND MORAL EMOTIONS The foregoing arguments imply that cooperative as well as competitive tendencies have been built into the genome from the beginning. Indeed, cooperative tendencies are manifested in neurochemical systems in the brain associated with powerful prosocial motives and emotions. These powerful prosocial emotions are hiding in plain sight: While they are involved in motivating much of the behavior of interest to social psychology, they tend to be taken for granted and are rarely recognized to be emotions per se. Just as effectance motives and needs for understanding underlie tendencies toward cognitive consistency, attribution processes, and attitude formation and change, attachment motives and needs to be loved underlie tendencies toward social referencing, modeling and imitation, conformity, and obedience (Buck, 1976/1998, 2004). They are involved in every human action and interaction.

The Dynamics of Social and Moral Emotions There is evidence in the extensive research stemming from attachment theory (Bowlby, 1969/1982) that securely attached persons are relatively certain of being loved, while persons with attachment anxiety worry that they are not loved and persons with avoidant attachment distrust others and in effect eschew love (e.g., Mikulincer, 1998; Mikulincer & Shaver, 2003, 2005; Mikulincer, Shaver, Gillath, & Nitzberg, 2005). Secure attachment therefore is associated with relatively less attention to being loved and greater attention to meeting/exceeding expectations; anxious attachment is associated with a greater attention to being loved; and avoidant attachment is associated with relatively low prosocial needs and therefore relatively weak social emotions. Therefore, based upon the differentiation of primary

Attachment and Social Emotions.


social emotions suggested previously, a secure person would be expected to experience pride, guilt, envy, and pity in situations where an anxious person would experience arrogance, shame, jealousy, and scorn. Moreover, an avoidant person would tend to not experience any of these emotions strongly with the extreme being a psychopath with no need to be loved who is incapable of experiencing any of the social or moral emotions. In this analysis, attachment is viewed as both a trait and a state. Most children can be classified as having a specific attachment style, whereas most adults show a mixed style. This may be because as social development proceeds, attachment security can increasingly vary with the personal relationship: We may be secure that we are loved by some persons but anxious about the love of others. So for example, we tend to feel pride in comparison with the former and arrogance in comparison with the latter. Also, we can learn to avoid attachment with some persons. It is all too easy to be taught that certain persons are enemies who do not deserve our love, to whom social and moral emotions are irrelevant. The ability of even normal human beings to become situational psychopaths—to destroy others without moral compunction—is all too apparent, as Hannah Arendt showed in her analysis of the banality of evil (1951/1973). This analysis implies that the eight primary social emotions are interrelated: That for example when a person is proud they tend to pity others, and to not to feel guilt or envy of others. Buck, Nakamura, Vieira, and Polonsky (2005) tested this by giving scenarios about comparative success and failure to University students in the United States and Japan. This study found support for the hypothesized relationships between primary social emotions in virtually every case, whether labeled in English or Japanese. This supported the hypothesis of universal labeling—that the words for the eight primary social emotions could be found in all languages—and the hypothesis of universal dynamics—that they would be interrelated similarly in all languages. Future studies might measure social comparison processes in terms of dispositional tendencies to engage in social comparisons (Buunk & Gibbons, 2005; Gibbons & Buunk, 1999). Buunk et al. (chapter 13, this volume) found this tendency to be positively correlated with jealousy, and we would expect it to be similarly related to other social emotions and also to moral emotions. Primary Moral Emotions. This analysis has been extended to the analysis of eight primary moral emotions, where success or failure for self and other is combined with the judgment that the outcome is just or unjust. When one’s success is just the result is triumph, when unjust it is modesty; when one’s failure is just the result is humiliation, when unjust it is indignation; when the other’s success is just the result is admiration, when unjust it is resentment; when the other’s failure is just the result is contempt, when unjust it is sympathy. The primary morals are related to the primary social emotions: For example, envy and jealousy can go with either admiration or resentment depending upon whether the other’s success is seen as justified.


Dominance-Submission Versus Civility The dynamics of social and moral emotions are illustrated in Figure 6.1, which illustrates the social and moral emotions in a situation of dominance and submission. The successful figure on the left exudes triumph, pride, and arrogance; and regards the relatively inept figure on the right with contempt and scorn. The unsuccessful figure feels guilt, shame, humiliation, and indignation; and regards the other with envy, jealousy, and resentment. These can be strong acute emotions; they can be fleeting as one might feel as one sees a person driving by in an expensive car, or they can be chronic, unremitting, and grinding; leading to a lack of authentic social contact and communication, the exacerbation of stress, unhappiness, and depression—perhaps in both parties—whose true causes may go unrecognized. On the other hand, if interactants show each other civility, a pattern of positive interpersonal emotions can result. Even despite differences in wealth and status, it is possible to have a relationship of mutual trust and respect in which each regard the other as following the rules fairly and with a sense of justice. Each can then regard their own successes with modesty, and respond to the other with a sense of mutual gratitude for following the rules and admiration for their deserved

FIGURE 6.1 In a dominant–submissive relationship, conflict can be exacerbated by moral

considerations: The perception of unfairness can facilitate conflict.


success. These are the ingredients for authentic communication and the powerful stress-buffering that can come with social support. This view of primary social and moral emotions implies that sociality and morality arise spontaneously and effortlessly from interaction: a spontaneous restructuring of socioemotional experience analogous to Piaget’s (1971) assimilation and accommodation process in the realm of cognitive development. From this point of view, organized religion is not necessary to morality, and indeed can potentially get in the way of morality, even promoting the situational psychopathy of religious intolerance and conflict. This suggests that kin selection and reciprocity, rather than being the bases of altruism, are actually mechanisms to restrict loving and altruistic feelings to kin and comrade: Rather than underlying altruism, kin selection and reciprocity are at the roots of xenophobia (Buck & Ginsburg, 1991, 1997).

CONCLUSIONS Social emotions are present in every human relationship and interaction, including those in memory and the imagination; and moral emotions are present whenever equity considerations are relevant. This analysis suggests that their evolutionary roots go back to the beginnings of life, and indeed there is evidence that prosocial emotions are associated with peptides such as the endorphins, oxytocin, and vasopressin: direct genetic products. Their potential is activated and directed in processes of communication; their power is based in early love and nurturance, and they are shaped and molded in the simplest and most natural of childish play. If critical communicative experiences—reflecting and at the same time generating true love—are absent or aberrant in the course of development, social and moral emotions can be scarred for life. Apparently trivial signals of politeness serve the vital function of reminding interactants that each is cooperating and playing by the rules, thereby fostering mutual respect, trust, and rapport. But, the slightest sign—a mistimed gesture, a glance that is a fraction too long or too short, a facial nuance—can arouse strong social and moral emotions associated with competitive dominance and submission whose origin may go unrecognized, or may be “unconscious.” Social and moral emotions are hiding in plain sight, so omnipresent and fundamental that they are often taken for granted, reflecting the evolved essence of sociality in a Yin and Yang of cooperation and competition. NOTE 1. “Jealousy” is defined more broadly here than in cases where jealousy is regarded as explicitly sexual. For example, Buunk, Massar, and Dijkstra (chapter 13, this

volume) define jealousy as involving a sexual rival interested in one’s partner, or in whom one’s partner is interested.


ACKNOWLEDGMENTS Students and colleagues involved in the research reported here include R. Thomas Boone, Rebecca Ferrer, Benson Ginsburg, David A. Kenny, Makoto Nakamura, Maxim Polonsky, Stacie Renfro Powers, Christian Rauh, Elliott Ross, and Edward T. Vieira, Jr. This research was funded in part by the following grants: The Russell Sage Foundation: “The Communication of Trustworthiness: Affect expression as the mechanism for building trust and cooperation” (with R. Thomas Boone); The EJLB Foundation, Canada Council: “Emotional Expression and Communication in Schizophrenic Patients” (with Elliott Ross); The Harry Frank Guggenheim Foundation: “The Affective Bases of Social Organization: Communicative genes in aggression and attachment” (with B. Ginsburg).

REFERENCES Allee, W. C. (1931). Animal aggregations: A study in general sociology. Chicago: University of Chicago Press. Arendt, H. (1973). The origins of totalitarianism. New York: Harcourt. (Original work published 1951) Baumeister, R. F., & Leary, M. R. (1995). The need to belong: Desire for interpersonal attachments as a fundamental human motivation. Psychological Bulletin, 117, 497–529. Ben-Jacob, E., Becker, I., Shapira, Y., & Levine, H. (2004). Bacterial linguistic communication and social intelligence. Trends in Microbiology, 12, 366–372. Bowlby, J. (1982). Attachment and loss: Vol. I. Attachment. New York: Basic Books. (Original work published 1969) Buck, R. (1984). The communication of emotion. New York: Guilford Press. Buck, R. (1985). Prime theory: An integrated view of motivation and emotion. Psychological Review, 92, 389–413. Buck, R. (1988). Human motivation and emotion (2nd ed.). New York: Wiley. (Original work published 1976) Buck, R. (1999). The biological affects: A typology. Psychological Review, 106(2), 301–336. Buck, R. (2002). The genetics and biology of true love: Prosocial biological affects and the left hemisphere. Psychological Review, 109, 739–744. Buck, R. (2004). The gratitude of exchange and the gratitude of caring: A developmentalinteractionist perspective of moral emotion. In R. A. Emmons & M. McCullough (Eds.), The psychology of gratitude (pp. 100–122). New York: Oxford University Press.

Buck, R., & Ginsburg, B. E. (1991). Emotional communication and altruism: The communicative gene hypothesis. In M. Clark (Ed.), Review of personality and social psychology: Vol. 12. Altruism (pp. 149–175) Newbury Park, CA: Sage. Buck, R., & Ginsburg, B. E. (1997). Communicative genes and the evolution of empathy. In W. Ickes (Ed.), Empathic accuracy (pp. 17–43). New York: Guilford Press. Buck, R., Nakamura, M., Vieira, E. T., Jr., & Polonsky M. (2005). A developmentalinteractionist theory of biological and higher level emotions: A cross-national comparison of America and Japan. Manuscript submitted for publication. Buck, R., & Powers, S. R. (2006). The biological foundations of social organization: The dynamic emergence of social structure through nonverbal communication. In V. Manusov & M. Patterson (Eds.), The Sage handbook of nonverbal communication (pp. 119–138). Thousand Oaks, CA: Sage. Buck, R., & VanLear, C. A. (2002). Verbal and nonverbal communication: Distinguishing symbolic, spontaneous, and pseudospontaneous nonverbal behavior. Journal of Communication, 52, 522–541. Buunk, B. P., & Gibbons, F. X. (2005). Social comparison orientation: A new perspective on those who do, and those who don’t compare with others. In S. Guimond (Ed.), Social comparison processes and levels of analysis (pp. 15–32). Cambridge, UK: Cambridge University Press. Clarke, C. H. (1981). Motility and fruiting in

104 EVOLUTION AND THE SOCIAL MIND the bacterium Myxococcus xanthus. In J. M. Lackie & P. C. Wilkinson (Eds.), Biology of the chemotactic response (pp. 155–171). Cambridge, UK: Cambridge University Press. Darwin, C. (1998). The expression of the emotions in man and animals (3rd ed.). New York: Oxford University Press. (Original work published 1872) Dawkins, R. (1976). The selfish gene. New York: Oxford University Press. Dawkins, R. (1982). The extended phenotype: The gene as the unit of selection. Oxford, UK: Oxford University Press. Dawkins, R. (1989). The selfish gene: New edition. New York: Oxford University Press. Dawkins, R., & Krebs, J. R. (1978). Animal signals: Information or manipulation? In J. R. Krebs & N. B. Davies (Eds.), Behavioural ecology: An evolutionary approach (pp. 282– 309). Oxford, UK: Blackwell Scientific. Dunbar, R. I. M. (1993). The co-evolution of neocortical size, group size, and language in humans. Behavioral and Brain Sciences, 16, 681–694. Eberhard, A., Burlingame, A. L., Eberhard, C., Kenyon, C. L., Nealson, K. H., & Oppenheimer, N. J. (1981). Structural identification of autoinducer of Photobacterium fischeri luciferase. Biochemistry, 20, 2444– 2449. Eibl-Eibesfeldt, I. (1975). Ethology: The biology of behavior (2nd ed.), New York: Holt, Rinehart, & Winston. Eisenberger, N. I., & Lieberman, M. D. (2005). Why it hurts to be left out: The neurocognitive overlap between physical and social pain. In K. D. Williams, J. P. Forgas, & W. von Hippel (Eds.), The social outcast: Ostracism, social exclusion, rejection, and bullying (pp. 109–130). New York: Psychology Press. Eisenberger, N. I., Lieberman, M. D., & Williams, K. D. (2003). Does rejection hurt? An fMRI study of social exclusion. Science, 302, 290–292. Engebrecht, J., Nealson, K., & Silverman, M. (1983). Bacterial bioluminescence: Isolation and genetic analysis of functions from Vibrio fischeri. Cell, 32, 773–781. Fuqua, C., Winans, S. C., & Greenberg, E. P. (1996). Census and consensus in bacterial ecosystems: The LuxR-LuxI family of

quorum-sensing transcriptional regulators. Annual Reviews of Microbiology, 50, 727. Gallio, M., Sturgill, G., Rather, P., & Kylsten, P. (2002). A conserved mechanism for extracellular signalling in eukaryotes and prokaryotes. Proceedings of the National Academy of Sciences, 99(19), 12208–12213. Gibbons, F. X., & Buunk, B. P. (1999). The INCOM: Development of a measure of social comparison orientation in two countries. Journal of Personality and Social Psychology, 76, 129–142. Gottleib, G. (1984). Evolutionary trends and evolutionary origins: Relevance to theory in comparative psychology. Psychological Review, 91, 448–456. Greenberg, E. P. (1997). Quorum-sensing in gram-negative bacteria. ASM News, 63(July), 371. Griffin, A. S., West, S. A., & Buckling, A. (2004). Cooperation and competition in pathogenic bacteria. Nature, 430(7003), 1024–1027. Hauser, M. D. (1996). The evolution of communication. Cambridge, MA: MIT Press. Hellingwerf, K. J. (2005). Bacterial observations: A rudimentary form of intelligence? Trends in Microbiology, 13(4), 152–158. Jurgens, U. (1979). Neural control of vocalization in nonhuman primates. In H. L. Steklis & M. J. Raleigh (Eds.), Neurobiology of social communication in primates (pp. 11–44). New York: Academic Press. Kleinginna, P. R., & Kleinginna, A. M. (1981a). A categorized list of motivation definitions, with suggestions for a consensual definition. Motivation and Emotion, 5, 263– 291. Kleinginna, P. R., & Kleinginna, A. M. (1981b). A categorized list of emotion definitions, with suggestions for a consensual definition. Motivation and Emotion, 5, 348–379. Krebs, J. R., & Dawkins, R. (1984). Animal signals: Mind-reading and manipulation. In J. R. Krebs & N. B. Davies (Eds.), Behavioural ecology: An evolutionary approach (2nd ed., pp. 380–402). Sunderland, MA: Sinauer. Kreft, J.-U. (2004a). Biofilms promote altruism. Microbiology, 150, 2751–2760. Kreft, J.-U. (2004b). Conflicts of interest in biofilms. Biofilms, 1, 265–276.


Lackie, J. M. (1986). Cell movement and cell behaviour. London: Allen & Unwin. Livingstone, M. S., Harris-Warrick, R. M., & Kravitz, E. A. (1980). Serotonin and octopamine produce opposite postures in lobsters. Science, 208(4439), 76–79. Losick, R., & Kaiser, D. (1997). Why and how bacteria communicate. Scientific American, 276(2), 68–73. Loumaye, E., Thorner, J., & Catt, K. J. (1982). Yeast mating pheromone activates mammalian gonadotropins: Evolutionary conservation of a reproductive hormone? Science, 218(4579), 1324–1325. Mikulincer, M. (1998). Attachment working models and the sense of trust: An exploration of interaction goals and affect regulation. Journal of Personality and Social Psychology, 74, 1209–1224. Mikulincer, M., & Shaver, P. (2003). The attachment behavioral system in adulthood: Activation, psychodynamics, and interpersonal processes. In M. P. Zanna (Ed.), Advances in experimental social psychology (Vol. 35, pp. 53–152). San Diego, CA: Academic Press. Mikulincer, M., & Shaver, P. (2005). Mental representation of attachment security: Theoretical foundation for a positive social psychology. In M. W. Baldwin (Ed.), Interpersonal cognition (pp. 233–266). New York: Guilford Press. Mikulincer, M., Shaver, P., Gillath, O., & Nitzberg, R. A. (2005). Attachment, caregiving, and altruism: Boosting attachment security increases compassion and helping.

Journal of Personality and Social Psychology, 89, 817–839. Panksepp, J. (1991). Brain opioids: A neurochemical substrate for narcotic and social dependence. In S. Cooper (Ed.), Theory in psychopharmacology.(pp. 149–175). London: Academic Press. Piaget, J. (1971). Piaget’s theory. In P. Mussen (Ed.), Handbook of child development (Vol. I, pp. 703–732). New York: Wiley. Redfield, R. J. (2002). Is quorum sensing a side effect of diffusion sensing? Trends in Microbiology, 10, 365–370. Schachter, S., & Singer, J. (1962). Cognitive, social, and physiological determinants of emotional state. Psychological Review, 69, 379–399. Shimkets, L. J. (1999). Intracellular signaling during fruiting-body development of Myxococcus xanthus. Annual Review of Microbiology, 53, 525–549. Strassmann, J. E., Zhu, Y., & Queller, D. C. (2000). Altruism and social cheating in the social amoeba Dictyostelium discoideum. Nature, 408, 965–967. Swift, S., Throup, J. P., Williams, P., Salmond, G. P., & Stewart, G. S. (1996). Quorum sensing: A population-density component in the determination of bacterial phenotype. Trends in Biochemical Sciences, 21, 214–219. Waters, C. M., & Bassler, B. L. (2005). Quorum sensing: Cell-to-cell communication in bacteria. Annual Review of Cell and Developmental Biology, 21, 319–346. Wilson, E. O. (1975). Sociobiology: The new synthesis. Cambridge, MA: Belknap.


The Strange Cognitive Benefits of Mild Dysphoria

On the Evolutionary Advantages of Not Being Too Happy JOSEPH P. FORGAS

Introduction The Evolutionary Functions of Affect Contemporary Cognitive Approaches The Empirical Evidence The Interpersonal Benefits of Negative Affect Conclusions

INTRODUCTION ne of the greatest puzzles about human nature concerns the fascinating and still poorly understood interplay between thinking and feeling, rational and emotional ways of dealing with the world around us. Affect is a ubiquitous and powerful phenomenon in our lives, yet research on human affectivity has been neglected until recently. Of the three basic faculties of the human mind that dominated philosophy and empirical psychology for the last few hundred years—cognition, affect and conation—affect remains arguably the last and least well understood (Hilgard, 1980). What is the function of affective states? In particular, is there an identifiable adaptive advantage that humans derive from experiencing moods? It seems intriguing that despite our apparently never-ending quest for happiness and satisfaction, the human emotional repertoire is nevertheless heavily skewed towards negative feelings. Four of the six deeply ingrained basic emotions




identified in humans with distinct physiological substrates are negative ones— fear, anger, disgust and sadness—suggesting that these emotions were adaptive in the highly dangerous and precarious ancestral environment, preparing the organism for flight, fight or avoidance in the face of danger. But what about sadness? The possible adaptive functions of sadness in particular remain puzzling and poorly understood. Even though sadness is clearly bothersome and provides no hedonic benefit, it remains one of the most enduring and common affective states (Ciarrochi, Forgas & Mayer, 2006). Indeed, throughout human history much effort has been extended in controlling sadness and dysphoria, and this never-ending quest remains a major objective in contemporary clinical practice. This chapter will suggest that evolutionary pressures probably shaped the development of all affective responses, including sadness in a way that is highly sensitive to situational requirements, and operates by spontaneously triggering different information processing strategies that appear to be highly adaptive to the requirements of different social situations. Interestingly, much recent research on the functions of affect also focused on the beneficial consequences of positive affect. It has been variously suggested that feeling good promotes creativity, flexibility, co-operation, integrative thinking, successful negotiation, work motivation, relationship satisfaction and a host of other desirable outcomes (Forgas, 1994, 1998, 2002; Forgas & George, 2001). In contrast, most experimental and clinical work emphasized the need to limit, control and avoid negative affectivity (Ciarrochi et al., 2006; Clark & Isen, 1982). If negative affect like sadness is so universally undesirable, what then accounts for its surprising ubiquity? This chapter will describe a series of empirical studies that demonstrate that negative moods such as sadness may confer significant adaptive advantages. In particular, negative affect promotes a more attentive, accommodating thinking style that produces superior outcomes whenever a cognitive or social task requires detailed, externally oriented, inductive thinking. The objective of this chapter is thus to combine evolutionary theorizing and experimental research on affect and cognition, and so contribute to the age-old quest to understand the relationship between the rational and the emotional aspects of human nature (Hilgard, 1980; see also, in this volume, Badcock & Allen, chapter 8; Buck, chapter 6).

THE EVOLUTIONARY FUNCTIONS OF AFFECT The influence of feelings on thinking and behavior has long fascinated writers, artists and laypersons alike. Ever since Plato, many theorists saw affect as a potentially dangerous, invasive force that subverts rational judgement and action. This idea gained its most powerful expression in Freud’s psycho-dynamic theories early last century. According to this view, affect can “take over” thinking and behaviour unless psychological resources are deployed to control these


impulses. For example, attempts to suppress negative affect such as fear can “facilitate the tendency to project fear onto another social object” (Feshbach & Singer, 1957, p. 286). However, the last few decades saw something like an “affective revolution” in psychology, neuroanatomy and psychophysiology producing a radically different view. Rather than viewing affect as a dangerous force, recent evidence suggests that affect is a useful and even essential component of adaptive responding to social situations (Adolphs & Damasio, 2001; Damasio, 1994; Ito & Cacioppo, 2001). The idea that affect is an integral aspect of social thinking and memory was first advanced in the 1980s by Gordon Bower (1981). Others, such as Robert Zajonc (1980, 2000) argued that affect also functions as an independent and primary force in responding to social situations, consistent with the view that affect constitutes a basic and universal human response system rooted in the evolutionary past (Darwin, 1872/1965). We now know that affect also plays a crucial role in how people organize and represent their daily social experiences (Forgas, 1979; Pervin, 1976). It seems that social “stimuli can cohere as a category even when they have nothing in common other than the emotional responses they elicit” (Niedenthal & Halberstadt, 2000, p. 381). Affective reactions seem to define the way people mentally represent common social episodes (Forgas, 1979), and as Pervin (1976) noted four decades ago, “what is striking is the extent to which situations are described in terms of affects (e.g. threatening, warm, interesting, dull, tense, calm, rejecting) and organized in terms of similarity of affects aroused by them” (p. 471). Thus, affective reactions play a universal, ubiquitous and powerful role in how people think and behave in social situations. Recent psychological theorizing highlights several adaptive functions associated with feelings. According to one view, the basic function of affective states is to provide feedback signals about progress in goal achievement. According to another theory, emotional states evolved to trigger specific behavioral responses appropriate to the situations that elicit them. Thus, emotional appraisals (Smith & Kirby, 2001) involve spontaneous cognitive processes that produce the appropriate affective response in a given situation. Indeed, an entire “rule system” of appropriate emotional reactions can be constructed that encapsulates these “affect rules” (Ortonyi, Clore & Collins, 1988). Many emotional reactions to situational challenges are fast, effective and precede systematic evaluations. A good example is the extremely negative “pain affect” elicited when an individual faces social ostracism (Spoor & Williams, chapter 17, this volume), triggering immediate neurological and psychological reactions that motivate adaptive responses. It is not too far-fetched to suggest that in early evolutionary history, such wired-in emotional reactions were likely to provide distinct survival advantages (Frijda, 1986). Individuals who detected and responded to threats and other social and environmental challenges most rapidly and effectively could derive a fitness advantage over those who did not (e.g., Blascovich & Mendes, 2000).



Extensive research now documents the functions of the adaptive affective response system (Lerner & Keltner, 2001), suggesting that affective reactions operate like domain-specific adaptations that may meet the requirements for special design (Tooby & Cosmides, 1992; see also von Hippel, Haselton & Forgas, chapter 1, this volume).

CONTEMPORARY COGNITIVE APPROACHES The advent of contemporary cognitive theories produced a more finely grained understanding of affective influences on thinking and behaviour. Affect can profoundly influence the content of thinking and memory according to Bower’s (1981) associative network model, as affective states can selectively prime related thoughts and ideas to be used in constructive cognitive tasks. Such moodcongruence influences many cognitive tasks (Bower, 1981; Clark & Isen, 1982; Forgas & Bower, 1987), where the open, constructive use of memory-based information is required (Fiedler, 2001; Forgas, 1995, 2002). More importantly, affect can influence not only the content of thinking (influencing what people think), but also the process of cognition, that is, how people think. It was first thought positive affect simply leads to more lazy, heuristic and more superficial processing strategies, whereas negative affect triggers a more effortful, systematic, analytic and vigilant processing style (Clark & Isen, 1982; Schwarz, 1990). Such affect-induced processing differences were originally explained as motivational differences between happy and sad individuals. According to the mood-maintenance hypothesis (Clark & Isen, 1982), those in a positive mood may refrain from effortful thinking to maintain this pleasant state. In contrast, those in a negative mood might engage in vigilant, effortful processing as an adaptive response to improve an aversive state. Others such as Schwarz (1990) and Wegener and Petty (1994) offered a kind of functionalist “cognitive tuning” account, suggesting that positive and negative affect have a signalling/ tuning function, informing the person of whether a relaxed, effort minimizing (in positive affect) or a vigilant, effortful (negative affect) processing style is appropriate. More recent theories, however, suggest a more complex pattern, showing that positive affect can also produce distinct processing advantages. Happy people often adopt a more open, creative and inclusive thinking style, use broader categories, show greater mental flexibility and can perform just as effectively on secondary tasks (Bless, 2001; Bless & Fiedler, 2006; Fiedler, 2001). According to recent theories by Bless (2001) and Fiedler (2001; Fiedler & Bless, 2001), the evolutionary significance of affective states is not simply to influence processing effort. Rather, they trigger equally effortful, but qualitatively different processing styles. Thus, positive affect recruits a more assimilative, schema-based, top-down processing style, as pre-existing knowledge guides information processing. In contrast, negative affect produces a more accommodative, bottom-up and externally focused processing strategy where attention


to situational information drives thinking (Bless, 2001; Fiedler, 2001). These processing styles can be equally vigilant and effortful, yet produce qualitatively different outcomes. Both positive and negative affect can thus produce adaptive, functional advantages in the right circumstances. Much has been written about the beneficial effects of positive affect (Forgas, 1998; Forgas & George, 2001). Much less is known about the adaptive advantages of dysphoria. The following experiments will explore the subtle cognitive advantages of feeling bad.

THE EMPIRICAL EVIDENCE This section will review several experiments that demonstrate the adaptive consequences of negative affect in such areas as judgemental errors, eyewitness accuracy, interpersonal communication and detection of deception. One study by Sinclair and Mark (1992) found that sad mood may improve accuracy in person perception, as heuristic shortcuts such as primacy effects are more common in happy mood. Those in sad mood were less influenced by primacy manipulations, and paid balanced attention to both positive and negative information in their impressions. It seems that people spontaneously reduce positive affect in anticipation of demanding and difficult social tasks, such as interacting with a stranger, by selectively reading sad rather than happy articles (Parrott, 1993). Thus, negative affect may not only confer processing advantages, but people seem to adopt subconscious strategies to reduce euphoria to fit the requirements of the situation. Of course, the kind of accommodative processing promoted by negative affect will not always improve the accuracy of judgements. For example, Ambady and Gray (2002) found that sadness and depression impaired people’s ability to correctly interpret brief cues predictive of social behaviours. Thus, when the task requires heuristic processing, positive mood may confer processing advantages. In terms of an earlier dichotomy identified in the person perception literature, positive mood seems to promote stereotype accuracy, whereas negative mood improves differential accuracy.

Is This True . . .? Mood Effects on Interpersonal Scepticism How do we know if the information we come across in everyday life is true or false? Much of what we know about the world is second hand knowledge. Deciding whether to accept or reject social information is a critical decision in everyday life. Accepting invalid information as true (false positives, excessive gullibility) can be just as dangerous as rejecting information that is valid (false negatives, excessive scepticism). Credibility judgements can be influenced by a variety of factors, such as information quality, prior knowledge and heuristic cues such as source credibility and attractiveness (e.g., Petty, DeSteno, & Rucker, 2001). In



several recent experiments we found that moods also have a significant influence on accepting or rejecting information. Many claims can potentially be evaluated against objective evidence. For example, trivia questions, urban myths and rumours are open to checking, but are in practice difficult to test (e.g., power lines cause leukaemia; AIDS originated in Cameroon; the CIA murdered Kennedy; etc.). A second kind of scepticism, interpersonal scepticism, concerns the acceptance or rejection of interpersonal messages that are by their very nature ambiguous and not open to objective validation. For example, deciding whether a smile, or a denial, is genuine involves this kind of credibility judgement. Several experiments found that induced mood states have a significant influence on both kinds of credibility judgement: (a) the acceptance or rejection of factual claims (factual scepticism), and (b) the acceptance or rejection of preferred interpersonal representations (interpersonal scepticism) (East & Forgas, 2006). For example, in one study we asked participants who were induced into positive, neutral and negative moods to judge the probable truth of a number of urban legends and rumours. Mood did have a significant influence on scepticism, but only for claims that were new and not previously encountered by respondents, suggesting that familiarity is an important moderator of mood effects on scepticism. A follow-up experiment explicitly manipulated the familiarity of a variety of factual claims taken from trivia games. Some were familiar (presented to judges several weeks before), and some were entirely new. Participants (N = 135) induced into a positive or negative mood by watching affectively laden videos rated previously seen items as more credible, and happy mood also significantly increased the tendency to accept familiar items as true. Negative mood in turn produced greater scepticism, consistent with the hypothesis that negative affect triggers a more externally focused and accommodative thinking style. Will mood still influence credibility judgements when previous exposure to factual claims also includes explicit feedback about their actual truth or falsity? In one experiment participants (N = 118) judged the truth of 25 true and 25 false general knowledge trivia statements, and were also told whether each item was true. Two weeks later, after a positive or negative mood induction, they rated the credibility of some familiar statements from the earlier session, as well as some completely new statements. Results showed that only sad participants were able to correctly distinguish between true and false claims they had seen previously. Happy participants seemed unable to remember the truth of claims, and were more likely to rate all previously seen, familiar information as true, even if they were told previously that the information was false. This pattern confirms that happy mood increased and sad mood reduced the tendency to rely on the “what is familiar is true” heuristic, whereas negative mood conferred an adaptive advantage by promoting a more accommodative, systematic processing style (Fiedler & Bless, 2001). Overall, negative mood

Mood Effects on Factual Scepticism.


increased, and positive mood decreased the degree of scepticism people display when assessing the truth of ambiguous factual claims. This effect seems due to negative mood reducing, and positive mood increasing the tendency to use perceived familiarity as an indication of truthfulness. Mood may also influence people’s tendency to accept or reject interpersonal communications as genuine or false. Negative moods might produce more critical and sceptical judgements, while happy people may accept interpersonal messages at “face value”, as genuine and trustworthy. In one experiment, we asked happy and sad participants (N = 90) to judge the genuineness of people displaying positive, neutral and negative facial expressions. As predicted, those in a negative mood were significantly less likely to accept facial expressions as genuine than those in the neutral or happy condition. Curiously, happy participants were more confident in their judgement about the genuineness of the facial expressions than were the other groups. In another study, instead of positive and negative facial displays, the six basic emotions were used as targets (i.e., anger, fear, disgust, happiness, surprise and sadness; Ekman, 1972). Once again, negative mood reduced, and positive mood increased people’s tendency to accept the facial displays as genuine, consistent with the more attentive and accommodative processing style associated with negative moods.

Mood Effects on Interpersonal Scepticism.

Do these mood effects also occur in realistic situations involving both verbal and nonverbal communication? To explore this, we asked happy or sad participants to accept or reject the videotaped statements of targets who were interrogated after a staged theft, and were either guilty, or not guilty. The targets were instructed to either steal, or leave in place a movie pass in an empty room, unobserved by anyone, and then deny taking the movie ticket. So some targets were lying and some were telling the truth when denying the theft. Those in a positive mood were more likely to accept denials as truthful. Sad participants made significantly more guilty judgements, and were significantly better at correctly detecting deceptive (guilty) targets (Figure 7.1). Negative affect thus produced a significant advantage at accurately distinguishing truths from lies in the observed interviews. A signal detection analysis also confirmed that sad judges were more accurate in detecting deception (identifying guilty targets as guilty) than were neutral or happy judges, consistent with the predicted mood-induced processing differences. In summary, negative affect seems to increase scepticism both about factual, and about interpersonal messages, and those in a negative mood were also significantly better able to detect deception. These results are conceptually consistent with recent affect-cognition theories showing that negative affect generally produces a more situationally oriented, accommodative and inductive cognitive style that provides an adaptive advantage when it comes to accurately detecting deception. This conclusion is also consistent with some earlier claims about

Mood Effects on the Detection of Deception.



FIGURE 7.1 The effects of mood and the target’s veracity (truthful, deceptive) on judgements of guilt of targets accused of committing a theft (average percentage of targets judged guilty in each condition) (after East & Forgas, 2006).

“depressive realism”, and recent research by Lane and DePaulo (1999), who found that dispositionally dysphoric individuals might have an advantage at detecting specific types of lies, such as false reassurances.

Negative Affect Reduces Judgemental Errors Interpreting the behaviour of others is a critical and demanding cognitive task in everyday life (Heider, 1958). The fundamental attribution error (FAE) or correspondence bias refers to a pervasive tendency by people to see intentionality and internal causation and underestimate the impact of situational forces in their judgements of others (Gilbert & Malone, 1995). The FAE largely occurs because, all things being equal, observers pay disproportionate attention to the most conspicuous information in the focus of their attention—the actor—and fail to adequately process information about situational constraints (Gilbert, 1991). If the detailed processing of situational information is facilitated, for example, by a negative mood state, the incidence of the FAE may be reduced. These experiments explored the possibility that good moods can increase, and bad moods can reduce the FAE (Forgas, 1998). Past work shows that moods can influence attribution strategies. Happy persons tend to identify stable, internal causes when doing well, and blame unstable, external causes for doing badly in achievement situations (Forgas, Bower & Moylan, 1990). In contrast, sad people make more internal and stable


attributions for their failures than for their successes. Moods can even influence explanations for deeply involving events, such as relationship conflicts with one’s intimate partner (Forgas, 1994). It was expected here that the more accommodative processing promoted by negative mood should enhance the processing of situational information, and so reduce the incidence of incorrect internal attributions (Gilbert & Malone, 1995). Further, in terms of Jones and Davis’ (1965) theory of correspondent inferences, these mood effects should be most pronounced when the behaviour of the actor is particularly informative and salient as it deviates from popular expectations. In one experiment, happy or sad participants (N = 96) were asked to read and make attributions about the writer of an essay advocating a popular or unpopular position (for or against nuclear testing) which they were told was either assigned, or was freely chosen. As also found by Jones and Harris (1967), essay content influenced attributions even when the essay was assigned. Happy persons were more likely, and sad people were less likely than controls to commit the FAE and incorrectly infer attitude differences based on coerced essays. It seems that the accommodative processing style recruited by negative mood significantly reduced the FAE, especially when correspondent inferences could be readily based on highly salient and captivating information (an unpopular essay; Gilbert, 1991). Similar effects can also occur in real life. In a field study, participants (N = 120) who were feeling good or bad after seeing happy or sad movies were asked to read and make attributions about the writers of popular and unpopular essays arguing for, or against recycling (cf. Forgas & Moylan, 1987). Once again, those in a negative mood after seeing sad films were significantly less likely to commit the FAE. In other words, positive affect increased and negative affect decreased the FAE, especially when the essays were highly salient because they advocated unpopular positions. Are these effects indeed due to the more attentive processing of situational information in negative mood? To test this, happy or sad participants (N = 84) again made attributions based on freely chosen or coerced essays advocating popular or unpopular positions (for or against environmentalism; Forgas, 1998, Exp. 3). Their subsequent recall of essay details was also assessed as a measure of mood-induced differences in information processing style. Once again, negative mood significantly reduced the incidence of the FAE, especially when the essays advocated unpopular positions. Paradoxically, happy persons were more confident in their judgements, indicating that judges had no introspective awareness of mood effects on processing strategies and attributions. Recall memory data confirmed that those in a negative mood remembered significantly more than did others, demonstrating a direct association between mood and the amount of processing the stimulus information received. A mediational analysis further confirmed that processing strategy was a significant mediator of mood effects on attributions.



Thus, mild negative moods produced improved judgements, reducing the incidence of the fundamental attribution error, both in laboratory and in real-life settings. These effects were directly due to the more detailed and accommodative processing style associated with dysphoria, consistent with the suggested evolutionary benefits of negative affect in conferring cognitive advantages when dealing with complex social information. These results then lend support to evolutionary explanations that emphasize the adaptive, functional significance of affective states.

Dysphoria Improves Eyewitness Memory Can mood also influence the accuracy of eyewitness recollections? In a series of three experiments we found that positive affect promoted, and negative affect inhibited the incorporation of false details into eyewitness memories. Affect may influence eyewitness memory (1) when the event is first witnessed (encoding stage), (2) when misleading information is encountered later on (post-event stage) and (3) when the information is retrieved (retrieval stage). These studies explored mood effects at Stage 2, on the incorporation of false information into memories (Forgas, Vargas & Laham, 2005). We expected that good moods can promote, and bad moods can inhibit the incorporation of false information into eyewitness memory, consistent with the information processing consequences of these affective states. In the first experiment (N = 96), participants viewed pictures showing a car crash scene (negative event), and a wedding party scene (positive event). One hour later, allegedly as part of an unrelated study, they received an autobiographical mood induction (recall happy or sad events from their past), and completed a short questionnaire about the scenes that either contained, or did not contain misleading information (e.g., set in italics here: “Did you see the overturned car next to the broken guard rail ?”, “Did you see the fireman holding a fire hose ?”). After a further 45-minute interval filled with other tasks, the accuracy of their eyewitness memory for the scenes was tested. As predicted, exposure to misleading information significantly reduced eyewitness accuracy, and positive mood increased, and negative mood decreased this tendency. In fact, negative mood almost completely eliminated this common “misinformation effect”. A signal detection analysis confirmed that negative affect when exposed to false details significantly improved and positive mood impaired memory performance. In a second experiment (N = 144), students in a lecture theatre witnessed a staged 5-minute aggressive encounter between a lecturer, and a female intruder (Forgas et al., 2005, Exp. 2). One week later eyewitnesses to this episode received a mood induction (viewed short 10-minute video-films), and then responded to a brief questionnaire about the episode that contained planted, misleading information (set in italics here: “Did you see the lecturer removing his microphone, as the woman wearing a light jacket moved towards him?”, “Can you remember the young woman fiddling with her scarf as the lecturer gave her


something from his wallet?”). After a further 45-minute interval, the accuracy of their eyewitness memory for the episode was tested. Those in a positive mood while receiving the misleading information were more likely subsequently to report it as true (Figure 7.2). In contrast, negative affect seems to have all but eliminated this source of error in eyewitness memory. Signal detection analyses confirmed that negative affect improved the ability to discriminate between correct and misleading details. Interestingly, those in the positive mood, although actually less accurate, were more confident in their accuracy, suggesting that there was no meta-cognitive awareness of these mood effects. Can people suppress the impact of their moods when instructed to do so? In a third study, participants (N = 80) saw 5-minute videotapes showing (a) a robbery in a convenience store, and (b) a wedding scene. After a 45-minute interval they received an audio-visual mood induction and then completed a short questionnaire that either did, or did not contain misleading information about the event. Some were also instructed to “disregard and control their affective states”. Finally, the accuracy of their eyewitness memory for the two events was tested. Participants also completed the Snyder self-monitoring scale, and the Crowne-Marlowe social desirability scale during a separate testing session at the beginning of the semester. Exposure to misleading information again reduced eyewitness accuracy, and did so most when people where in a happy rather than a sad mood. A signal detection analysis again confirmed the beneficial effects of negative affect for memory performance. Instructions to control affect did not reduce this mood effect, but rather, produced an overall conservative response bias. Interestingly,

FIGURE 7.2 The interaction between mood and the presence or absence of misleading

information on recognition (Experiment 2): Positive mood increased, and negative mood decreased the influence of misleading information on subsequent eye-witness reports (false alarms) (after Forgas et al., 2005).



individuals who scored high on self-monitoring and social desirability were better able to suppress mood effects when instructed to do so than were others. These experiments offer convergent evidence that negative moods can have significant adaptive effects on cognitive performance, by reducing people’s susceptibility to misleading information and thus improving eyewitness accuracy. Paradoxically, happy mood reduced accuracy yet increased confidence, suggesting that people were unaware of the consequences of their mood states for their thinking and memory. Instructions to suppress affect were only effective for participants who scored high on self-monitoring, and social desirability. These results are consistent with affect-cognition theories that predict that good and bad moods should have an asymmetric effect on processing strategies and outcomes (Bless, 2001; Fiedler & Bless, 2001; Forgas, 1995, 2002). Within an evolutionary approach to social cognition advocated here, our results suggest that both good and bad mood can have a significant impact on eyewitness memories, due to the kind of information processing strategies they generate, and these findings may have a number of applied implications for forensic, organizational and clinical psychology.


THE INTERPERSONAL BENEFITS OF NEGATIVE AFFECT Could negative affect also confer identifiable benefits when it comes to effective interpersonal communication, such as the production of persuasive messages? Despite extensive research on responding to persuasion (Petty et al., 2001) there has been little work on how such messages are produced. What role does everyday mood play in the production of persuasive messages? It was expected that accommodative processing promoted by negative affect should produce more concrete and factual thinking and result in the production of superior persuasive messages. This prediction is also consistent with much early theorizing about rhetorical effectiveness going back to Aristotle (Cooper, 1932), as well as psychological research suggesting that “expository information that is concrete . . . tends to be interesting and well recalled” (Sadowski, 2001, p. 263). In the first experiment (Forgas, in press, Exp. 1), participants (N = 59) received an audio-visual mood induction, and were then asked to produce persuasive arguments for or against an increase in student fees, and Aboriginal land rights. They produced an average of seven arguments, and each argument was rated by two raters blind to the manipulations for overall quality, persuasiveness, level of concreteness and valence (positive–negative). Those in a negative mood produced arguments on both issues that were of significantly higher quality and more persuasive than the arguments produced by happy participants. This mood effect was largely due to the greater specificity and concreteness of arguments produced in a negative mood. A mediational analysis showed that it was moodinduced variations in argument concreteness that influenced argument quality.


In a further experiment, happy or sad participants (N = 125) were asked to produce persuasive arguments for or against Australia becoming a republic, and for or against a radical right-wing party. Two raters (r = .91) assessed each argument in terms of (a) persuasiveness and argument quality, (b) valence (the use of positive or negative contents) and (c) self-relevance (the extent to which participants used personal, self-relevant themes). Sad mood again resulted in higher quality and more persuasive arguments (see Figure 7.3), consistent with the theoretical prediction that negative mood should promote a more careful, systematic, bottom-up processing style that is more attuned to the requirements of a particular situation (Bless, 2001; Bless & Fiedler, 2006; Fiedler, 2001; Forgas, 2002). However, the ultimate significance of these findings depends on whether the arguments produced by happy and sad participants indeed differ in actual persuasive power, as distinct from ratings of persuasiveness produced by trained raters. In Experiment 3 the arguments produced by happy or sad participants were presented to a naive audience of 256 undergraduate students. Their baseline

FIGURE 7.3 Mood effects on the quality and concreteness of the persuasive messages

produced: Negative affect increases the degree of concreteness of the arguments produced, and arguments produced in negative mood were also rated as more persuasive (after Forgas, in press, Exp. 2).



attitudes on the four issues were assessed at the beginning of the term. After reading one of the pro- or contra-persuasive arguments on one of the issues written by one of the happy or sad participants in Experiments 1 and 2, their attitude on all four issues was again assessed. Observed changes in attitudes in response to the persuasive arguments were assessed against the baseline measurement obtained earlier. Results showed that arguments written by negative mood participants in Experiments 1 and 2 were significantly more successful in producing a real change in attitudes than were arguments produced by happy participants. Attitudes were also more likely to change when the arguments advocated a popular rather than an unpopular position, and negative mood arguments were especially successful in producing attitude change when they advocated a popular position. Finally, in Experiment 4 persuasive attempts by happy and sad people were directed at a “partner” to volunteer for a boring experiment using e-mail exchanges (Forgas, in press). The motivation to be persuasive was also manipulated by offering some of them a significant reward if successful (movie passes). Mood again had a significant effect on argument quality: People in a negative mood produced higher quality persuasive arguments than did the neutral group, who in turn did better than the positive group. However, the offer of a reward reduced mood effects on argument quality, confirming a key prediction of the Affect Infusion Model (Forgas, 1995, 2002), that mood effects on information processing—and subsequent social influence strategies—are strongest in the absence of motivated processing. A mediational analysis again confirmed that negative mood induced more accommodative thinking, and more concrete and specific arguments, as predicted. This series of experiments thus confirms that persuasive arguments produced in negative mood are not only of higher quality as judged by raters, but are also significantly more effective in producing genuine attitude change in people. Arguments produced in negative mood were more effective, because they contained more concrete details and more factual information (Cooper, 1932). Such messages are seen by people as more interesting and more memorable (Sadowski, 2001). However, when motivation to be effective is already high, mood effects tend to diminish, as predicted by the Affect Infusion Model (Forgas, 2002). These results are consistent with other studies suggesting that negative affect typically promotes a more concrete, accommodative, externally focused information processing style that also can reduce the incidence of judgemental errors and improve eye-witness memory (Forgas, 1998; Forgas et al., 2005). This kind of concrete, accommodative processing also has direct benefits when it comes to the effective use of social influence strategies, such as the production of persuasive arguments. This finding may have interesting applied implications, for example in industrial and organizational settings where encounters involving persuasive communication are very common (Forgas & George, 2001). Managing successful relationships and resolving personal conflicts also involves a great deal of persuasive communication, often in situations that are affectively charged. It is


an intriguing possibility that mild negative affect may actually promote a more concrete, accommodative and ultimately, more successful communication style in intimate relationships.

CONCLUSIONS In contrast with the overwhelming emphasis on the benefits of positive affect in the recent literature, these results highlight the potentially adaptive and beneficial consequences of negative mood (Forgas & George, 2001). Positive affect is not always desirable (Sinclair, 1988). People in a negative mood are less prone to judgemental errors (Forgas, 1998), are more resistant to eye-witness distortions (Forgas et al., 2005) and are less likely to adopt dysfunctional self-handicapping strategies (Alter & Forgas, in press). To this list we may now add another caveat: People in a negative mood may also be better at producing high-quality and effective persuasive messages. Just how robust and reliable are these effects? Given the consistency of the results across a number of different experiments, different populations and different mood inductions, we can be reasonably confident of the reliability of these effects. Future experiments may also provide additional insights into the precise nature of the adaptive mechanisms responsible for these effects. Dealing with social information is necessarily a complex and demanding cognitive task that requires a degree of elaborate processing. The empirical studies presented here suggest that in many situations, negative affect such as sadness may increase, and positive affect decrease the quality and efficacy of cognitive processes and interpersonal behaviours. Much has been learned about the way affective states influence memory, thinking and judgements in recent years, yet not enough is known about the evolutionary mechanisms that are responsible for the way we respond to various affective states.

Negative Affect: An Evolutionary Adaptation? Our findings are broadly consistent with the notion that over evolutionary time, affective states became adaptive, functional triggers to elicit information processing patterns that are appropriate in a given situation. However, one recurring problem in applying evolutionary principles to understanding social cognition is that such interpretations are usually post hoc, and notoriously hard to prove. How do we really know if an experimentally demonstrated phenomenon, such as the beneficial influences of negative affect on social information processing demonstrated in these studies, is indeed an evolutionary adaptation, or merely the side effect of an adaptation, or perhaps even just error? There are some commonly accepted criteria, but no hard and fast rules. Discerning the evolutionary roots of very specific effects can be very difficult, as Halberstadt (chapter 15, this volume) shows. Any phenomenon claiming to be



evolutionary in origin needs to be culturally universal. Although few explicitly cross-cultural studies have so far been carried out on mood effects on information processing, there is reason to believe that these effects, as are indeed most fundamental cognitive phenomena, are not culture dependent. The convergent validation of this effect in a variety of different cognitive tasks, using a variety of mood induction procedures, and different subject populations also suggests that the effect is real and universal. Some evidence from neuropsychology and in particular from fMRI studies should be helpful to bolster the case for evolutionary origins, and we are currently engaged in such research. At this point, however, we must accept that the case for the evolutionary nature of mood effects on thinking is not yet made. This is not necessarily a major problem, however, as applying an evolutionary frame of thinking to social cognitive phenomena can be beneficial in a variety of ways. Taking an evolutionary perspective helps us to realize that the phenomena we study have biological roots, and offers an important and productive link between cognitive theorizing and the neurosciences (Tooby & Cosmides, 1992). Perhaps evolutionary psychology at this stage is more a “meta-theory”, a way of thinking about psychology, than a strictly testable theory (Ketelaar & Ellis, 2000). Nevertheless, evolutionary principles help to link and integrate a variety of otherwise disconnected findings, and thus help to bring order and connectedness into our field, as the contributions to this volume amply demonstrate. Further research on the nature of affective influences on complex interpersonal behaviours should be of considerable theoretical, as well as applied relevance to researchers interested in evolutionary approaches to social cognition.

REFERENCES Adolphs, R. & Damasio, A. (2001). The interaction of affect and cognition: A neurobiological perspective. In J. P. Forgas (Ed.), The handbook of affect and social cognition (pp. 27–49). Mahwah, NJ: Lawrence Erlbaum Associates, Inc. Alter, A., & Forgas, J. P. (in press). On being happy but fearing failure: The effects of mood on self-handicapping strategies. Journal of Experimental Social Psychology. Ambady, N., & Gray, H. M. (2002). On being sad and mistaken: Mood effects on the accuracy of thin slice judgments. Journal of Personality and Social Psychology, 83, 947– 961. Blascovich, J., & Mendes, W. B. (2000). Challenge and threat appraisals: The role of affective cues. In J. P. Forgas (Ed.), Feeling and thinking: The role of affect in social cognition

(pp. 59–82). New York: Cambridge University Press. Bless, H. (2001). Mood and the use of general knowledge structures. In L. L. Martin (Ed.), Theories of mood and cognition: A user’s guidebook (pp. 9–26). Mahwah, NJ: Lawrence Erlbaum Associates, Inc. Bless, H., & Fiedler, K. (2006). Mood and the regulation of information processing and behavior. In J. P. Forgas (Ed.), Hearts and minds: Affective influences on social cognition and behaviour (pp. 65–84). New York: Psychology Press. Bower, G. H. (1981). Mood and memory. The American Psychologist, 36, 129–148. Ciarrochi, J. V., Forgas, J. P., & Mayer, J. (Eds.). Emotional intelligence in everyday life (2nd ed.). New York: Psychology Press. Clark, M. S., & Isen, A. M. (1982). Towards


understanding the relationship between feeling states and social behavior. In A. H. Hastorf & A. M. Isen (Eds.), Cognitive social psychology (pp. 73–108). New York: Elsevier-North Holland. Cooper, L. (1932). The rhetoric of Aristotle. New York: Appleton-Century. Damasio, A. R. (1994). Descartes’ error. New York: Grosste/Putnam. Darwin, C. (1965). The expression of emotions in man and animals. Chicago: University of Chicago Press. (Original work published 1872) East, R., & Forgas, J. P. (2006). Mood effects on factual and interpersonal scepticism. Unpublished manuscript, University of New South Wales, Sydney, Australia. Ekman, P. (1972). Universals and cultural differences in facial expressions of emotion. In J. Cole (Ed.), Nebraska Symposium on Motivation (Vol. 19, pp. 207–283). Lincoln, NE: University of Nebraska Press. Feshbach, S., & Singer, R. D. (1957). The effects of fear arousal and suppression of fear upon social perception. Journal of Abnormal and Social Psychology, 55, 283–288. Fiedler, K. (2001). Affective influences on social information processing. In J. P. Forgas (Ed.), The handbook of affect and social cognition (pp. 163–185). Mahwah, NJ: Lawrence Erlbaum Associates, Inc. Fiedler, K., & Bless, H. (2001). The formation of beliefs in the interface of affective and cognitive processes. In N. Frijda, A. Manstead, & S. Bem (Eds.), The influence of emotions on beliefs (pp. 210–246). New York: Cambridge University Press. Forgas, J. P. (1979). Social episodes: The study of interaction routines. London/New York: Academic Press. Forgas, J. P. (1994). Sad and guilty? Affective influences on explanations of conflict episodes. Journal of Personality and Social Psychology, 66, 56–68. Forgas, J. P. (1995). Mood and judgment: The Affect Infusion Model (AIM). Psychological Bulletin, 116, 39–66. Forgas, J. P. (1998). Happy and mistaken? Mood effects on the fundamental attribution error. Journal of Personality and Social Psychology, 75, 318–331. Forgas, J. P. (2002). Feeling and doing: Affective influences on interpersonal behavior. Psychological Inquiry, 13, 1–28.

Forgas, J. P. (in press). When sad is better than happy: Mood effects on the effectiveness of persuasive messages. Journal of Experimental Social Psychology. Forgas, J. P., & Bower, G. H. (1987). Mood effects on person perception judgements. Journal of Personality and Social Psychology, 53, 53–60. Forgas, J. P., Bower, G. H., & Moylan, S. J. (1990). Praise or blame? Mood effects on attributions for success or failure. Journal of Personality and Social Psychology, 59, 809–819. Forgas, J. P., & George, J. M. (2001). Affective influences on judgments and behavior in organizations: An information processing perspective. Organizational Behavior and Human Decision Processes, 86, 3–34. Forgas, J. P., & Moylan, S. J. (1987). After the movies: The effects of transient mood states on social judgments. Personality and Social Psychology Bulletin, 13, 478–489. Forgas, J. P., Vargas, P., & Laham, S. (2005). Mood effects on eyewitness memory: Affective influences on susceptibility to misinformation. Journal of Experimental Social Psychology, 41, 574–588. Frijda, N. (1986). The emotions. Cambridge, UK: Cambridge University Press. Gilbert, D. T. (1991). How mental systems believe. The American Psychologist, 46, 107–119. Gilbert, D. T., & Malone, P. S. (1995). The correspondence bias. Psychological Bulletin, 117, 21–38. Heider, F. (1958). The psychology of interpersonal relations. New York: Wiley. Hilgard, E. R. (1980). The trilogy of mind: Cognition, affection, and conation. Journal of the History of the Behavioral Sciences, 16, 107–117. Ito, T., & Cacioppo, J. (2001). Affect and attitudes: A social neuroscience approach. In J. P. Forgas (Ed.), The handbook of affect and social cognition (pp. 50–74). Mahwah, NJ: Lawrence Erlbaum Associates, Inc. Jones, E. E., & Davis, K. E. (1965). From acts to dispositions. In L. Berkowitz (Ed.), Advances in experimental social psychology (pp. 124–167). New York: Academic Press. Jones, E. E., & Harris, V. A. (1967). The attribution of attitudes. Journal of Experimental Social Psychology, 3, 1–24.


124 EVOLUTION AND THE SOCIAL MIND Ketelaar, T., & Ellis, B. J. (2000). Are evolutionary explanations unfalsifiable? Evolutionary psychology and the Lakatosian philosophy of science. Psychological Inquiry, 11, 1–21. Lane, J. D., & DePaulo, B. M. (1999). Completing Coyne’s cycle: Dysphorics’ ability to detect deception. Journal of Research in Personality, 33, 311–329. Lerner, J. S., & Keltner, D. (2001). Fear, anger, and risk. Journal of Personality and Social Psychology, 81, 146–159. Niedenthal, P., & Halberstadt, J. (2000). Grounding categories in emotional response. In J. P. Forgas (Ed.), Feeling and thinking: The role of affect in social cognition (pp. 357– 386). New York: Cambridge University Press. Ortonyi, A., Clore, G. L., & Collins, A. (1988). The cognitive structure of emotions. New York: Cambridge University Press. Parrott, W. G. (1993). Beyond Hedonism: Motives for inhibiting good moods and for maintaining bad moods. In D. M. Wegner & J. W. Pennebaker (Eds.), Handbook of mental control (pp. 278–305). Englewood Cliffs, NJ: Prentice Hall. Pervin, L. A. (1976). A free response description approach to the study of person situation interaction. Journal of Personality and Social Psychology, 34, 465–474. Petty, R. E., DeSteno, D., & Rucker, D. (2001). The role of affect in attitude change. In J. P. Forgas (Ed.), The handbook of affect and social cognition (pp. 212–236). Mahwah, NJ: Lawrence Erlbaum Associates, Inc. Sadowski, M. (2001). Resolving the effects of concreteness on interest, comprehension and learning. Educational Psychology Review, 13, 263–281. Schwarz, N. (1990). Feelings as information: Informational and motivational functions of

affective states. In E. T. Higgins & R. Sorrentino (Eds.), Handbook of motivation and cognition (Vol. 2, pp. 527–561). New York: Guilford Press. Sinclair, R. C. (1988). Mood, categorization breadth and performance appraisal: The effects of order of information acquisition and affective state on halo, accuracy, information retrieval and evaluations. Organizational Behavior and Human Decision Processes, 42, 22–46. Sinclair, R. C., & Mark, M. M. (1992). The influence of mood state on judgment and action. In L. L. Martin & A. Tesser (Eds.), The construction of social judgments (pp. 165– 193). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. Smith, C. A., & Kirby, L. D. (2001). Affect and cognitive appraisal processes. In J. P. Forgas (Ed.), The handbook of affect and social cognition (pp. 75–94). Mahwah, NJ: Lawrence Erlbaum Associates, Inc. Tooby, J., & Cosmides, L. (1992). The psychological foundations of culture. In J. H. Barkow & L. Cosmides (Eds.), The adapted mind: Evolutionary psychology and the generation of culture (pp. 19–136). London: Oxford University Press. Wegener, D. T., & Petty, R. E. (1994). Mood management across affective states: The hedonistic contingency hypothesis. Journal of Personality and Social Psychology, 66, 1034– 1048. Zajonc, R. B. (1980). Feeling and thinking: Preferences need no inferences. The American Psychologist, 35, 151–175. Zajonc, R. B. (2000). Feeling and thinking: Closing the debate over the independence of affect. In J. P. Forgas (Ed.), Feeling and thinking: The role of affect in social cognition (pp. 31–58). New York: Cambridge University Press.


Evolution, Social Cognition, and Depressed Mood

Exploring the Relationship Between Depression and Social Risk Taking PAUL B. T. BADCOCK NICHOLAS B. ALLEN

Introduction Darwinian Models of Depressed Mood Theories of Resource/Energy Conservation Social Theories of the Evolution of Depression The Social Risk Hypothesis: An Integrative View Depression and Cognition About Social Risk Depression and Reduced Social Risk Taking Recent Studies on Depression and Risk Propensity Discussion and Conclusions

INTRODUCTION he idea that the human capacity for depressed mood evolved in response to specific adaptive problems is by no means new. The argument enjoys a six-decade history (McGuire & Troisi, 1998), and has drawn contributions from a broad range of theorists (see Allen & Badcock, 2003). Generally, explanatory frameworks that have emerged from this expanding body of literature can be meaningfully grouped into two broad categories: explanations that emphasize the inherently social nature of the selection pressures that have shaped the human capacity for depression, and those that also focus upon non-social influences (Gilbert, 2006). We shall now consider each of these schools of thought in turn.




DARWINIAN MODELS OF DEPRESSED MOOD While evolutionary approaches have been proposed to explain both the ubiquitous human capacity for depressed mood states, as well as the emergence in some individuals of clinically significant depressed states (see Allen & Badcock, 2006, for a review of both types of models), in this chapter we shall focus on explanations for depressed mood only. Although the concept of adaptation should not be reflexively applied to biological and behavioural features without sound logical reasons, we propose that mood states are appropriate targets for an evolutionary analysis. First, they are ubiquitous human capacities indicating a considerable degree of specialization (Cosmides & Tooby, 1994). Second, they are activated by specific contexts, suggesting that their input is specialized (Oatley, 1992). Finally, mood states are characterized by complex but co-ordinated sets of output in terms of physiology, overt behaviour and conscious experience. While there has been some very useful work on the distinct adaptive functions of other negative mood states like anxiety (Öhman, Dimberg & Ost, 1985), the adaptive significance of depressed mood has proven more elusive. For example, there is considerable evidence that the essential psychological “theme” of anxiety is threat, whereas the “theme” for depression is loss (Clark & Beck, 1989) or defeat (Price, 1972). Despite such distinctions, the argument that anxiety is an adaptive response to threat (facilitating early detection and responses) is more widely accepted than any evolutionary claim regarding depression. Indeed, it is not even clear what kind of loss is most relevant to depression (McGuire, Troisi & Raleigh, 1997). Nesse (1998) has argued that circumstances involving the loss (or the threat thereof ) of a reproductive resource are likely to shape negative mood states, whereas circumstances involving the gain (or potential gain) of such resources are likely to shape positive ones. Thus, an evolutionary understanding of depression should depend on: (1) identifying recurrent situations in the ancestral environment typically associated with biologically significant loss; (2) describing the selection pressures in such situations (i.e., the particular social-reproductive goals that they would have threatened); and (3) isolating the features of depressed mood that would have enabled the organism to cope with these pressures (Nesse, 1990). While this approach has engendered a range of propositions, three prevailing schools of thought characterize the literature: the conservation of resources, social competition and attachment theories of depression.

THEORIES OF RESOURCE/ENERGY CONSERVATION Conservation of resource theories assert that the inhibition of appetitive functions associated with depression (i.e., low levels of energy, pleasure and appetitive motivation) is likely to be adaptive by allowing an individual to conserve resources and later redirect them towards more productive endeavours. These


theories tend to emphasize both social and non-social outcomes, and as such, do not usually consider social processes to be uniquely influential in the evolution of depressive mechanisms (Gilbert, 2006). According to such views, depressed mood is instigated by a low rate of positive reward or insufficient control over rewards or punishments. Seligman’s (1975) learned helplessness theory, for example, was founded on studies of animals exhibiting helpless behaviour when subjected to uncontrollable aversive events. Nesse’s (2000) resource allocation model concentrates more on low rates of rewarding outcomes. Here, depressed mood represents an adaptive response to the propitiousness of situations by adjusting resource allocation (e.g., energy and investment) to inhibit investments in poor pay-off activities. In a similar vein, incentive disengagement theory (Klinger, 1975) proposes that depressed states disengage an organism from unobtainable incentives or goals, whilst Leahy’s (1997) “sunk costs” model suggests that depression occurs when people persevere too long with behaviours that result in insufficient or diminishing rewards. Consistent with such models, Champion and Power (1995) have argued that depression-prone people tend to over-invest in a limited number of goals and, when such goals fail, there is a collapse in an individual’s incentive and motivational systems.

SOCIAL THEORIES OF THE EVOLUTION OF DEPRESSION The Evolution of the Social Brain The social brain hypothesis (see Dunbar, chapter 2, this volume) asserts that group living results in strong evolutionary pressures that favour the selection of cognitive capacities that facilitate an individual’s successful navigation of competitive and highly complex social environments. A seminal chapter by Humphrey (1976) has pointed out that the complexity of social relations within primate groups poses particularly difficult adaptive problems for the members of such groups. Given the likely impact of these processes on fundamental neural design, it is not surprising that the attention of neuroscientists has recently been drawn to the way in which social information and behaviour is processed by the human brain (Adolphs, 2003). Although there are substantial adaptive advantages to group living—most obviously, better protection from predators, improved success as predators, and more ready access to mates—there is also genetic competition between group members, meaning that those attributes best suited to never-ending Machiavellian games of “social chess” or “plot and counter plot” will be positively selected. Importantly, the pursuit of adaptive goals within social contexts is not without risk. For example, co-operating in a group of exploiters might mean that they benefit at one’s own expense, and competing for status and access to resources against more powerful conspecifics is likely to elicit constant attacks, loss of status and/or humiliation. It is crucial, then, for individuals to evaluate



risks against benefits whenever they pursue a particular social role or goal. Additionally, risks of attachment loss (separation) for a young child will be calculated and coped with in a different way to, for example, risks encountered when competing for sexual partners. This means that humans need to monitor the social roles they are engaged in, and increase certain behaviours when opportunities exist, or inhibit them when they entail too much risk. It is the latter circumstance that has been the subject of social theories of the evolution of depression.

Social Competition Theories One of the earliest and most influential approaches to the evolution of depression has emphasized the role of social competition in shaping the human capacity for depressed mood. A central claim of evolutionary theory is that an individual’s access to reproductive resources will vary according to his or her position, or rank, within the wider social group (Buss, 1999). Accordingly, humans are powerfully motivated to acquire status through competitive encounters with others (Buss, 1991). The first author to apply this argument to depression was Price (1967), who has developed a Darwinian explanation of depressive states based on ritual agonistic (or fighting) behaviours. Price has argued that in any agonistic encounter between competitors, the winning contestant will typically adopt an escalating strategy to increase its chances of success (i.e., continuing to participate in the contest, possibly threatening or attacking its opponent); the losing contestant will adopt a de-escalating strategy, characterized by subordinate or yielding behaviour (Price, 1998). This latter strategy represents a withdrawal from the fight, reducing the risk of physical incapacity or death by sending “no-threat” signals to de-activate the aggressive behaviour of the attacker. Here, depression is conceptualized as an evolved, involuntary de-escalating strategy—enabling the individual to acknowledge defeat in ritual agonistic encounters, and adapt to the corresponding loss of social rank (Price, Sloman, Gardner, Gilbert, & Rohde, 1994). However, while fighting and intimidation are effective competitive strategies for most species, this is not always the case with humans and other primates (de Waal, 1989). Instead, resource acquisition and the formation of important social defences rely heavily upon an individual’s ability to elicit help from others. Thus, ritual agonistic behaviours have been replaced, at least partially, by competition through attraction—such that an individual’s fitness prospects will depend largely upon his or her social value, prestige and attractiveness (Barkow, 1989). Gilbert and colleagues (e.g., Gilbert, Allan & Trent, 1995) embrace this argument in their notion of Social Attention Holding Power (SAHP). SAHP refers to “the ability to elicit positive attention and social rewards, in the form of approval, praise, acceptance, respect, admiration, desire, etc.” (Gilbert, 1997, p. 118). Gilbert suggests that humans compete for status through bestowing benefits on others to maximize SAHP, and that status differences are


attributable to differences in the degree and quality of attention conferred by others (Buss, 1999). The biological significance of high SAHP has been highlighted by recent research demonstrating that rhesus macaques sacrifice a fluid reward in order to view images of high status conspecifics, but require an overpayment of fluid rewards in order to view images of low status conspecifics (Deaner, Khera & Platt, 2005). These findings suggest that even in non-human primates, social attention is differentially allocated according to status. Importantly, agonistic strategies to cope with conspecific challenges have not disappeared, but are over-laid with SAHP-oriented systems. Consequently, not only may aggression be a common response to being deprived of the approval, support, admiration, respect or love to which one thinks oneself is entitled (Baumeister, Smart & Boden, 1996), but loss of control over such signals may activate the defensive, subordinate and social defeat responses of loss of confidence, anxiety and depression.

The Attachment Theory of Depression Central to the attachment theory of depression is the proposed fitness consequences of affective bonding. Indeed, according to Buss (1991), mate retention, reciprocal dyadic alliance formation and coalition building are among the principal social selection pressures that have shaped human evolution. One of the first proponents of the attachment model of depression was Bowlby (1969), who forwarded an evolutionary argument for interpersonal attachments based on parent–child interactions. Given the gradual maturation and protracted helplessness of human infants, the survival and emergent reproductive success of one’s offspring necessitates an intensive contribution of time, commitment, energy and resources (Ingram, Miranda & Segal, 1998). Bowlby (1988) argues that affective bonding ensures that a parent will provide the necessary commitments to safeguard the survival, and latent reproductive success, of his or her child(ren), and thus, the survival and perpetuation of that parent’s genes. It has also been suggested that affectional bonding between adult sexual partners has evolved to ensure the contribution of protection and resources from the father during the mother’s pregnancy, and periods when the female must focus on the care of offspring (Ainsworth, 1991). Affiliative relationships are also seen to play a central role in the formation of social defences and the acquisition of resources, as they enable the individual to elicit assistance from conspecifics (Ainsworth, 1991). In specific relation to depression, the attachment model suggests that behaviours designed to maintain proximity to caregivers are instigated when significant affectional bonds are threatened (Gilbert, 1992). The model attributes depressive onset to the loss or dissolution of significant interpersonal relationships (Ingram et al., 1998), and has germinated several hypotheses concerning the adaptive function of depressive states. Some, for example, have



suggested that depression inhibits exploratory or risk-laden activities in the absence of secure attachment bonds, and instigates appeasement-related behaviours designed to maintain relationships (Gilbert, 1992). Others have argued that the depressive response serves as a distress call (Frijda, 1994), provokes a search for the lost relationship (Averill, 1968), or motivates the sufferer to avoid further deterioration of pre-existing bonds (Ingram et al., 1998).

THE SOCIAL RISK HYPOTHESIS: AN INTEGRATIVE VIEW While the perspectives outlined above provide important, complementary insights into the adaptive significance of depression, we have suggested elsewhere that they are not mutually exclusive (Allen & Badcock, 2003). The conservation of resources views, while explaining the inhibition of reward-seeking behaviour that is prominent in depression, do not explicitly account for the features of depressed mood that are inextricably associated with social cognition and behaviour, especially self-depreciation and the fact that depression is specifically related to withdrawal from social contexts. They also do not explain why precipitants of depression are typically social rather than asocial in nature (e.g., Monroe, Rohde, Seeley & Lewinsohn, 1999). Moreover, although the idea that depression evolved as a response to the loss or dissolution of attachment bonds points to the social origins of depressive phenomena, its explanatory power is limited by the fact that only a proportion of human relationships (especially amongst adults) can be validly characterized as attachment relationships (Haslam, 1994), and that the contexts that elicit depression often involve humiliation and entrapment rather than interpersonal loss (Brown, Harris & Hepworth, 1995). Furthermore, the idea that depressed states motivate an individual to re-establish lost relationships or seek out new ones does not fit well with the demobilization and social withdrawal that characterize depression. Finally, while the argument that depression evolved from more primitive subordinate strategies explains many of its features and has led to some supportive research findings (e.g., Gilbert & Allan, 1998), previous accounts have not explicitly explored the way in which this rankoriented strategy has been adapted to the human social environment, especially with respect to the diversity of human social relationships (Haslam, 1994), and the dramatic advances in social cognition associated with evolution whereby humans (and possibly some other primates) developed the capacity to think about the mental states of others (Byrne & Whiten, 1986). We have proposed that many of the important insights of these approaches can be integrated and extended by an analysis of social risk assessment in depression (Allen & Badcock, 2003). Wiggins and Trapnell (1996) have argued that all forms of interpersonal relatedness can be understood in terms of two fundamental dimensions: agency (or power) and communion (affiliation). These dimensions of relatedness are emphasized by the social competition and attachment views, respectively.


Notably, both of these domains of interpersonal behaviour present social risks— that is, risks to one’s social circumstances, well-being and reputation. The dimension of agency, for example, presents risks of defeat, humiliation and entrapment, whereas the dimension of affiliation presents risks of rejection and shunning. Our social risk hypothesis (Allen & Badcock, 2003) suggests that depressive phenomena can be interpreted as a defensive psychobiological response to increased risk within either one of these interpersonal domains.

Avoiding Social Exclusion Many theorists have argued that depressed states are fundamentally related to reductions of positive affect (anhedonia being a key defining feature), and that the regulators of positive affect are embedded in social cognition and behaviour (Allen & Badcock, 2003; Joiner & Coyne, 1999). Central to our functional or evolutionary view of depression is the proposition that there are various biological processes that guide individuals to enact certain social roles. There are many clues in the research literature that suggest that social processes (both in terms of social cognition and interpersonal behaviour) play a crucial role in the aetiology and maintenance of depressed states. Critical empirical observations include findings that depression is often precipitated by interpersonal events (as noted above), and that interpersonal processes typically mediate the exacerbation or resolution of depressive episodes (Joiner & Coyne, 1999). Stressful interpersonal contexts are amongst the most reliable precipitants of depressed states (Monroe et al., 1999) and certain interpersonal behaviours, such as excessive reassurance seeking, are strong and specific predictors of risk for depression (Joiner & Metalsky, 2001). Our model of the function of depressed states seeks to explain why there is such a close link between social cognition, social behaviour and depressive phenomena. The social risk hypothesis of depression (Allen & Badcock, 2003) suggests that depressed mood (i.e., down-regulation of positive affect and confident engagement in the world) evolved to facilitate a risk-averse approach to social interaction in situations where individuals were typically at risk of exclusion from social contexts (i.e., dyadic relationships or groups) that were vital to dealing with adaptive, socio-reproductive challenges. As we have already noted, evolutionary models emphasize that an individual’s access to non-plentiful, fitness-enhancing resources depends largely upon his or her position in a social context. Indeed, given that human groups contain concentrations of particular reproductive resources, including potential mates, kin to whom altruism can be directed, and non-kin with whom to exchange resources, the effect of social exclusion on various proximal adaptive tasks can be considerable (Buss, 1990). Furthermore, in the ancestral environment, social exclusion may have threatened one’s survival by excluding the individual from group-based benefits such as protection from predators and foraging for food.



Consequently, a critical matter for the individual is to detect when the danger of exclusion from currently beneficial social relationships is high. The argument that individuals are highly sensitive to how they are perceived and valued by others, and that this sensitivity is based on an evolved human drive for social belonging, has previously been forwarded by others (e.g., Baumeister & Leary, 1995), and has also received empirical support (see Spoor & Williams, chapter 17, this volume). The social risk hypothesis maintains that the loss or dissolution of significant interpersonal relationships, and/or experiences implicative of low status (such as defeat or humiliation), can be categorized more broadly as signals that throughout evolutionary history have been associated with the kind of lowered social value that could lead to ostracism from important social contexts. Indeed, we would argue that precipitants of depression do not just involve loss or defeat, but rather, any socially relevant experience that indicates (or has indicated through evolutionary history) to an individual that his or her ability to successfully negotiate important social contexts is critically low. Such precipitants may include, for example, negative interpersonal experiences (such as losses or rejections); the failure of an important goal; loss of social rank or status; and/or perceptions of a lack of control in social situations (as is the case with experiences of entrapment). According to this view, the psychological theme for depression is not so much about defeat or loss, but more about inadequacy. Finally, once the depression mechanism is activated, we have argued that the usual opportunistic social investment strategy of individuals shifts towards a risk-averse strategy—namely, that of depressed mood (Allen & Badcock, 2003).

Actions of the Depression Mechanism within the Social Ecology The social risk hypothesis suggests that the adaptive function of the depressive state is to protect an individual’s fitness prospects by minimizing behaviours that put social connections at risk, and ensuring the reduction and avoidance of further threats to reproductive opportunities. Arguably, under the threat of social exclusion, those who were able to cautiously increase their social value, while minimizing the risk of further reductions, were more likely to have preserved their participation in adaptive social contexts, and were favoured by natural selection. On the other hand, those who did not adjust their social behaviour in the face of low social value were at even greater risk of exclusion from critical social alliances. Thus, we have argued that the depression mechanism controls aspects of both social perception and behaviour to reduce the likelihood of further, critical reductions in an individual’s social value (Allen & Badcock, 2003). The mechanism affects social-perceptual processes in that the individual becomes hypersensitive to indications of social risk. In terms of social behaviour, the mechanism influences both communicative behaviour (signalling in order to reduce threats and to elicit safe forms of support), and resource acquisition behaviours (a


general reduction in behavioural propensities towards high-risk investments that may result in interpersonal conflict or competition).

DEPRESSION AND COGNITION ABOUT SOCIAL RISK There exists a wealth of research to support the idea that depressed mood is associated with an increased sensitivity to indicators of social risk. For example, in a study requiring depressed versus non-depressed participants to evaluate the potential risks and benefits associated with a range of domain-specific decisionmaking scenarios, Pietromonaco and Rook (1987) found that depressed individuals were considerably more sensitive to risks in the social domain. Furthermore, Mathews, Ridgeway and Williamson (1996) report that depressed individuals display attentional vigilance towards socially threatening words. Others have found that mild-to-moderately depressed participants seek out, and are particularly vigilant and sensitive to, social information (e.g., Gleicher & Weary, 1991). Also, depression has been found to be associated with an increased sensitivity to negative interpersonal and/or achievement related experiences (Cole, 1990), and heightened and reduced anticipation of negative and positive life events, respectively (MacLeod & Byrne, 1996). In a more direct test of this proposal, Badcock and Allen (2003) examined the effect of experimentally induced depressed mood upon reasoning about social risk through use of a novel version of the Wason card selection task (Wason, 1966). The response on the Wason selection task acts as a measure of an individual’s reasoning about the violation of a conditional rule. The tasks designed by Badcock and Allen required participants to detect violations of rules that predicted positive social outcomes from taking social risks (i.e., they were asked to detect the possibility of negative social outcomes). It was found that participants in an induced depressed mood state reasoned more adequately about risks related to social competition (i.e., they were more able to detect violation of the rule “if I invest the resources in competing, then I will be successful”) than those in a neutral mood state. This mood-facilitation effect was not observed for reasoning about other types of content. This provides some empirical support for the notion that depression is associated with inferential reasoning biases that result in greater sensitivity to (i.e., the likelihood of detection of ) social risks.

DEPRESSION AND REDUCED SOCIAL RISK TAKING An issue that we have explored in more detail in recent studies is the proposal that depression is associated with a reduction in an individual’s propensity to take social risks, but does not affect risk propensity for other types of risk, such as physical or financial ones. This hypothesis enjoys some support from a range of previous research findings (Forgas, 1995, 2002), for example, has shown that



while positive affect is linked to confident and assertive approach behaviours in interpersonal contexts, negative affect is associated with more avoidant and defensive social behaviours. Consistent with this view, he has demonstrated that individuals in a sad mood pay more attention to the requirements of interpersonal situations, and process their responses in a more careful, bottom-up fashion (Forgas, 1998, 2002). Further evidence for a link between depression and reduced social risk taking has been provided by Pietromonaco and Rook (1987). Administering a range of decision-making scenarios pertaining to a variety of different risk domains, they found that “depressed subjects reported less willingness than did the nondepressed subjects to take actions that might expose them to social risks (such as embarrassment or conflict)” (p. 405). Another relevant research area is found in studies of the structure of selfrated mood. Researchers have argued that while high levels of negative affect are associated with a variety of distressed states, low levels of positive, or appetitive, emotional states are depression-specific (Clark & Watson, 1991). Relevant to the association between depression and risk taking in social contexts, Watson (2000) has demonstrated that positive affect is strongly associated with the desire to engage in social activity. Collectively, these finding provide some support for the notion that depressed mood is associated with reduced social risk taking. We have argued, moreover, that this relationship is attributable to the influence of depression on levels of positive affect, causing low levels of pleasure and energy, alongside reductions in an individual’s self-assurance and motivation to engage in social activities (Allen & Badcock, 2003).

RECENT STUDIES ON DEPRESSION AND RISK PROPENSITY We recently extended our theoretical work on the effect of depression on social risk taking in a series of studies (Badcock & Allen, 2006). In the first of these, we compared clinically depressed participants with anxious and non-psychiatric controls. We have argued that evidence from clinical populations can be used to test predictions concerning the design of the depression mechanism, but not those regarding its ecological functions, as we believe that in clinical manifestations, the depression mechanism operates beyond its range of adaptive functioning and results in patently maladaptive behaviour (Allen & Badcock, 2003). In our first study, risk propensity was measured using three dichotomousresponse risk propensity measures added to the social risk (Badcock & Allen, 2003) and cheater-detection (Cosmides & Tooby, 1992) Wason selection tasks. After each task, participants were asked to select which behavioural strategy (i.e., “take the risk” vs. “not take the risk”) they would be likely to adopt. Although there were no significant differences between the risk propensity responses of the depressed and anxious groups, a significantly greater proportion of depressed participants selected the risk-averse option for both the attachment and social


competition measures than those in the control group, providing partial support for the hypothesis. As expected, there were no significant differences between the depressed versus anxious and depressed versus control groups in terms of responses for the cheater-detection risk-taking measure, indicating that group differences were specific to risks associated with attachment and social status. In order to determine whether dimensions of symptomatology, rather than diagnostic categories alone, were associated with reduced social risk taking, we also examined relationships between symptom dimensions linked to positive affect (i.e., high levels of positive emotions such as pride, excitement or joy) and negative affect (e.g., sadness, fear, irritability) and risk propensity. Analyses revealed that positive affect significantly enhanced prediction for both the attachment and social competition risk propensity measures, indicating that this predictor was reliably, negatively and independently associated with participants’ selection of the risk-averse option for both social risk measures. These results support our hypothesis that participants scoring low on positive affect (a symptom profile uniquely associated with depression; Clark & Watson, 1991) would be more likely to select the risk-averse option for both the attachment and social competition risk-taking measures. Finally, neither positive nor negative affect predicted the risk-averse response for the cheater detection task. Our findings suggest that, as participants’ levels of positive affect increased, they were less likely to select the risk-averse option for both social risk-taking measures. Also, negative affect did not reliably enhance prediction of participants’ responses on any of the risk-taking measures, and positive affect was not significantly associated with responses on the cheater-detection measure. Since low positive affect is a defining feature of depression (e.g., Clark & Watson, 1991), these findings appear to indicate that depression specifically reduces an individual’s propensity to undertake agency- and affiliation-oriented risks. Furthermore, given the absence of a significant relationship between negative affect and risk-averse responses, results suggest that reduced social risk taking is particular to depressed mood. A potential limitation, however, arises from the rudimentary measures of risk propensity added to the selection tasks. Indeed, since these measures were neither pilot-tested, nor correlated with an alternate measure of risk propensity, their validity remains uncertain. To address this issue, we conducted a second study with a larger, non-clinical sample, and more extensive, validated assessments of risk propensity. Moreover, given the argument and evidence that depression has its roots in social contexts (see Allen & Badcock, 2003), our social risk hypothesis predicts that low positive affect (as a depression-specific dimension of mood) should be associated with a reduced propensity to take social risks, but should not affect propensities for other types of risk, such as physical or financial ones. To test these hypotheses, our second study investigated relationships between mood and risk taking in a variety of hazard domains, using two measures of risk propensity that independently assess an individual’s propensity to take social, financial, health- and accident-related risks (see Rohrmann, 2004).



We found that after controlling for the effects of individual personality differences, negative affect and self-esteem, positive affect was significantly and positively related to social risk taking. Also, and as expected, relationships between positive affect and participants’ responses for the non-social risk-taking dimensions were non-significant and statistically trivial. Once again, given that low positive affect is a defining feature of depression (e.g., Clark & Watson, 1991), such results also suggest that depressed mood specifically reduces an individual’s propensity to exhibit risky social behaviours. It also bears mentioning that self-esteem was found to moderate this relationship, such that as participants’ levels of self-esteem increased, so too did the strength of the relationship between positive affect and social risk propensity. Finally, in a third study, we aimed to replicate these findings by analysing relationships between participants’ current levels of mood and their social, financial, accident- and health-related risk propensities. However, revealing significant relationships among variables does not imply that these relationships are causal. Accordingly, experimental work is required before causal inferences can be made. Thus, an additional aim of this study was to test for a causal link between depression and reduced social risk taking through use of the experimental manipulation of mood. In this final study, a musical mood induction was used. Here, participants are exposed to mood-suggestive music, and are asked to employ this music to assist their own efforts to get into a depressed, elated or neutral mood (Martin, 1990). Thus, the current study tested whether depressed states are causally associated with reduced social risk taking by exposing participants to one of two musical mood induction procedures (depressed vs. neutral), followed by administration of two assessments of domain-specific risk propensity (Rohrmann, 2004). Again, results supported the hypothesis that individual differences in selfrated positive affect at the beginning of the experiment were significantly, positively and uniquely associated with participants’ social risk propensities. These results provided a partial replication of those in the previous study. By contrast, analyses of the two groups’ mean ratings failed to support the hypothesis that participants exposed to the depressed mood induction would exhibit significantly lower mean scores on measures of social risk taking than those in the neutral mood condition. Moreover, results of equivalence tests indicated that effect sizes for all non-social risk propensity scales were non-trivial. In general, then, results of between-group analyses were inconsistent with our hypotheses, and render the nature of the causal relationship between depression and social risk taking somewhat unclear. A possible explanation, however, is that reduced social risk taking only results from more stable, naturally occurring depressed mood states, rather than the sort produced in an artificial laboratory setting. This might explain why participants’ mood states prior to experimentation were a reliable predictor of social risk propensity, whereas induced mood states were not. Expressed another


way, the depressed mood induced by the musical procedure may not have been sufficiently intense or stable to produce the expected results. With this in mind, experimental mood manipulations may not represent an ideal research avenue for an empirical test of our model. Another possibility is that future research requires alternate, more effective mood manipulation procedures, such as moodsuggestive films or stories (Westermann, Spies, Stahl & Hesse, 1996). Clearly, this is an issue that necessitates further investigation.

DISCUSSION AND CONCLUSIONS In this chapter, we have reviewed major approaches to the evolution of the human capacity for depressed mood, and have specifically highlighted recent social cognitive research designed to test one of the central proposals emerging from our integrative social risk hypothesis of depression (Allen & Badcock, 2003). Taken as a whole, the results of these studies provide consistent support for our prediction that depressed mood (particularly the down-regulation of positive mood states associated with depression) is related to a reduction in an individual’s propensity to take social risks. The findings presented in this chapter have a range of implications with respect to evolutionary conceptions of depression. First, our results do not fit particularly well with Nettle’s (2004) individual difference model of clinical depression. According to Nettle, depressive vulnerability is attributable to excessively high levels of neuroticism. Although we certainly agree with this contention, a further tenet of his argument is that increasing neuroticism was selected for in the evolutionary environment because it promotes striving behaviours in interpersonal contexts, and causes people to seek out desirable social outcomes and avoid negative ones. On the basis of these arguments, one might expect neuroticism to be in some way related to risk propensity, particularly in the social domain. However, regression analyses of relationships between participants’ neuroticism scores derived from the Zuckerman-Kuhlman Personality Test (Zuckerman, Kuhlman, Joireman, Teta & Kraft, 1993) and their responses for the social, financial, accident- and health-related risk-taking scales failed to indicate a significant association between neuroticism and any of these variables. Notably, while evidence for a connection between low positive affect and reduced social risk taking points to the centrality of social contexts to depressive phenomena, and thus lends some support for the attachment view of depression, our results also conflict with some of the predictions emerging from this model. In particular, our findings emphasize the previously established link between low mood and social withdrawal (Goldberg & Huxley, 1992), and are therefore inconsistent with the notion that depressed mood motivates a person to re-establish lost relationships or seek out new ones (e.g., Averill, 1968). On the other hand, the argument that depression represents an adaptive strategy that reduces risk taking is not new, and is particularly emphasized by the



resource conservation views (Allen & Badcock, 2003). The findings presented above not only provide support for this line of reasoning, but also introduce an important caveat by suggesting a high level of specificity between depression (i.e., low positive affect) and reduced risk taking in social contexts in particular. Also, results were partially supportive of the social competition model spearheaded by Price and colleagues (Price et al., 1994). According to this view, one of the adaptive operations of the depression mechanism is to inhibit competitive, confident and/or assertive interpersonal behaviours. From this perspective, our demonstration of a significant relationship between low positive affect and reduced social risk propensity is not surprising. However, an important addendum is required. As noted above, the social competition view neglects to explicitly examine the ways in which the proposed rank-oriented strategy has been adapted to accommodate the full diversity of human social relationships (Haslam, 1994), and in particular, those not directly related to the negotiation of status hierarchies. To elaborate, it is worthwhile returning to Wiggins and Trapnell’s (1996) argument that all forms of interpersonal relatedness are reducible to the fundamental dimensions of agency (or power) and communion (affiliation). Although the social competition model apparently anticipates reduced social risk taking amongst depressives in the former domain, the model fails to extend this prediction to the latter. More specifically, despite the fact that this perspective could probably be extended or adapted to accommodate the negotiation and/or management of affiliation-oriented relationships, reduced social risk taking in this domain is neither directly nor explicitly predicted by the model (Allen & Badcock, 2003). Conversely, the social risk hypothesis emphasizes that depressive phenomena constitute a defensive psychobiological response to increased risk within either one of these domains, and as such, should be associated with reduced risk taking in both (Allen & Badcock, 2003). Since the social risk propensity items used in our research represented risks associated with either agency or affiliation (e.g., asking someone out on a date vs. applying for a job), the demonstration of significant, positive relationships between positive affect and participants’ responses for all social risk-taking scales appears to provide strongest support for our model. It is worthwhile pointing out that numerous proximate explanations have been offered for a link between depression and reduced social risk taking. Pietromonaco and Rook (1987) briefly review these arguments in their own study. First, they assert that a history of problematic social encounters among depressives, alongside their reduced sensitivity to the receipt of positive social feedback, is likely to sensitize them to social risks. Second, such sensitivity is likely to be exacerbated by realistic self-appraisals of limited social skills (Libert & Lewinsohn, 1973). Third, depressive sensitivity to, and avoidance of, social risks may be partly attributable to a self-regulation strategy that protects fragile feelings of self-worth from additional exposure to embarrassment, rejection or conflict (Pietromonaco & Rook, 1987). That is, depressives are likely to avoid social risks because they are motivated to protect their low self-esteem.


Such explanations are by no means incompatible with our model. Rather, while the above represent proximate arguments for reduced social risk taking among depressives, the social risk hypothesis provides an ultimate explanatory framework that successfully integrates them. For example, given the proposal that depressed mood represents an adaptive response to important social disruptions, our social risk hypothesis clearly predicts a past history of problematic interpersonal encounters amongst depressives. Furthermore, reduced sensitivity to positive social feedback and a pessimistic, error-avoidant appraisal of limited social skills are arguably important features of a cautious, risk-minimizing approach to interpersonal contexts (Allen & Badcock, 2003). Such features are likely to operate adaptively by contributing to the inhibition of confident and/or assertive social behaviours in the face of threats of social exclusion. Finally, in light of the argument that self-esteem alerts an individual to threats of social exclusion (Allen & Badcock, 2003; Leary, Tambor, Terdal & Downs, 1995), the motivation to protect low self-esteem can be readily interpreted as the proximate result of an underlying adaptive strategy that promotes social inclusion (see Leary et al., 1995). It follows, then, that low self-esteem may act as a leading indicator of the possibility of social exclusion, but reduced social risk taking only results once the depression mechanism (i.e., low positive affect) is activated. This conjecture is consistent with our finding that self-esteem moderates the relationship between positive affect and social risk taking, without being related to social risk taking directly. Taken together, the findings presented herein not only provide empirical support for some important aspects of the social risk hypothesis, but in a more general sense, exemplify how evolutionary hypotheses can motivate a program of research. Indeed, the investigation of relationships between mood and domainspecific biases in social reasoning and risk propensity would not have been contemplated without the social risk hypothesis. Furthermore, whilst our studies have provided support for some of the specific predictions derived from our social risk hypothesis and failed to support others, they will also form a basis for further empirical studies and, ultimately, important refinements (or the refutation) of our model. REFERENCES Adolphs, R. (2003). Cognitive neuroscience of human social behaviour. Nature Reviews Neuroscience, 4(3), 165–178. Ainsworth, M. D. S. (1991). Attachments and other affectional bonds across the life cycle. In C. M. Parkes, J. Stevenson-Hinde, & P. Marris (Eds.), Attachment across the life cycle (pp. 33–51). London: Routledge. Allen, N. B., & Badcock, P. B. T. (2003). The social risk hypothesis of depressed mood:

Evolutionary, psychosocial, and neurobiological perspectives. Psychological Bulletin, 129, 887–913. Allen, N. B., & Badcock, P. B. T. (2006). Darwinian models of depression: A review of evolutionary accounts of mood and mood disorders. Progress in NeuroPsychopharmacology and Biological Psychiatry, 30, 815–826. Averill, J. R. (1968). Grief: Its nature and


140 EVOLUTION AND THE SOCIAL MIND significance. Psychological Bulletin, 70, 721– 748. Badcock, P. B. T., & Allen, N. B. (2003). Adaptive social reasoning in depressed mood and depressive vulnerability, Cognition and Emotion, 17, 647–670. Badcock, P. B. T., & Allen, N. B. (2006). Mood, personality, and domains of risk taking. Manuscript in preparation. Barkow, J. H. (1989). Darwin, sex and status: Biological approaches to mind and culture. Toronto, Canada: University of Toronto Press. Baumeister, R., Smart, L., & Boden, J. (1996). Relation of threatened egotism to violence and aggression: The dark side of high selfesteem. Psychological Review, 101, 5–33. Baumeister, R. F., & Leary, M. R. (1995). The need to belong: Desire for interpersonal attachments as a fundamental human motivation. Psychological Bulletin, 117, 497– 529. Bowlby, J. (1969). Attachment and loss: Vol. 1. Attachment. New York: Basic Books. Bowlby, J. (1988). A secure base: Parent–child attachment and healthy human development. New York: Basic Books. Brown, G. W., Harris, T. O., & Hepworth C. (1995). Loss, humiliation and entrapment among women developing depression: A patient and non-patient comparison. Psychological Medicine, 25, 7–21. Buss, D. M. (1990). The evolution of anxiety and social exclusion. Journal of Social and Clinical Psychology, 9, 196–201. Buss, D. M. (1991). Evolutionary personality psychology. In M. R. Rosenzweig & L. W. Porter (Eds.), Annual Review of Psychology (Vol. 42, pp. 459–491). Palo Alto, CA: Annual Reviews. Buss, D. M. (1999). Evolutionary psychology: The new science of the mind. Needham Heights, MA: Allyn & Bacon. Byrne, R. W., & Whiten, A. (1986). Machiavellian intelligence: Social expertise and the evolution of intellect in monkeys, apes, and humans. Oxford, UK: Oxford University Press. Champion, L. A., & Power, M. J. (1995). Social and cognitive approaches to depression. British Journal of Clinical Psychology, 34, 485–503. Clark, D. A., & Beck, A. T. (1989). Cognitive theory and therapy of anxiety and depression.

In P. C. Kendall & D. Watson (Eds.), Anxiety and depression: Distinctive and overlapping features. San Diego, CA: Academic Press. Clark, L. A., & Watson, D. (1991). Tripartite model of anxiety and depression: Psychometric evidence and taxonomic implications. Journal of Abnormal Psychology, 100, 316– 336. Cole, D. A. (1990). Relation of social and academic competence to depressive symptoms in childhood. Journal of Abnormal Psychology, 99, 422–429. Cosmides, L., & Tooby, J. (1992). Cognitive adaptations for social exchange. In L. Cosmides, J. Tooby, & J. Barkow (Eds.), The adapted mind: Evolutionary psychology and the generation of culture (pp. 163–228). New York: Oxford University Press. Cosmides, L., & Tooby, J. (1994). Beyond intuition and instinct blindness: Toward an evolutionarily rigorous cognitive science. Cognition, 50, 41–77. Deaner, R. O., Khera, A. V., & Platt, M. L. (2005). Monkeys pay per view: Adaptive value of social images by Rhesus Macaques. Current Biology, 15, 543–548. De Waal, F. M. B. (1989). Peacemaking among primates. Cambridge, MA: Harvard University Press. Forgas, J. P. (1995). Mood and judgment: The affect infusion model (AIM). Psychological Bulletin, 117, 39–66. Forgas, J. P. (1998). Happy and mistaken? Mood effects on the fundamental attribution error. Journal of Personality and Social Psychology, 75, 318–331. Forgas, J. P. (2002). Feeling and doing: Affective influences on interpersonal behaviour. Psychological Inquiry, 13, 1–28. Frijda, N. H. (1994). Emotions are functional, most of the time. In P. Ekman & R. Davidson (Eds.), The nature of emotion: Fundamental questions (pp. 97–177). New York: Oxford University Press. Gilbert, P. (1992). Depression: The evolution of powerlessness. New York: Guilford Press. Gilbert, P. (1997). The evolution of social attractiveness and its role in shame, humiliation, guilt and therapy. British Journal of Medical Psychology, 70, 113–147. Gilbert, P. (2006). Evolution and depression: issues and implications. Psychological Medicine, 36, 287–297.


Gilbert, P., & Allan, S. (1998). The role of defeat and entrapment (arrested flight) in depression: An exploration of an evolutionary view. Psychological Medicine, 28, 584–597. Gilbert, P., Allan, S., & Trent, D. (1995). Involuntary subordination or dependency as key dimensions of depressive vulnerability. Journal of Clinical Psychology, 51, 740–752 Gleicher, F., & Weary, G. (1991). The effect of depression on the quantity and quality of social inferences. Journal of Personality and Social Psychology, 51, 140–148. Goldberg, D., & Huxley, P. (1992). Common mental disorders: A bio-social model. London: Tavistock/Routledge. Haslam, N. (1994). Categories of social relationship. Cognition, 53, 59–90. Humphrey, N. K. (1976). The social function of intellect. In P. P. G. Bateson & R. A. Hinde (Eds.), Growing points in ethology (pp. 303– 317). Cambridge, UK: Cambridge University Press. Ingram, R. E., Miranda, J., & Segal, Z. V. (1998). Cognitive vulnerability to depression. New York: Guilford Press. Joiner, T. E., & Coyne, J. C. (1999). The interactional nature of depression: Advances in interpersonal approaches. Washington DC: American Psychological Association. Joiner, T. E., Jr., & Metalsky, G. I. (2001). Excessive reassurance seeking: Delineating a risk factor involved in the development of depressive symptoms. Psychological Science, 12, 371–378. Klinger, E. (1975). Consequences of commitment to and disengagement from incentives. Psychological Review, 82, 1–25. Leahy, R. L. (1997). Depression and resistance: An investment model of decision-making. The Behavior Therapist, 20, 3–6. Leary, M. R., Tambor, E. S., Terdal, S. K., & Downs, D. L. (1995). Self-esteem as an interpersonal monitor: The sociometer hypothesis. Journal of Personality and Social Psychology, 68, 519–530. Libert, J., & Lewinsohn, P. M. (1973). The concept of social skill with special reference to the behavior of depressed persons. Journal of Consulting and Clinical Psychology, 40, 304–312. MacLeod, A. K., & Byrne, A. (1996). Anxiety, depression, and the anticipation of future

positive and negative experiences. Journal of Abnormal Psychology, 105, 286–289. Martin, M. (1990). On the induction of mood. Clinical Psychology Review, 10, 669– 697. Mathews, A., Ridgeway, V., & Williamson, D. A. (1996). Evidence for attention to threatening stimuli in depression. Behaviour Research and Therapy, 34, 695–705. McGuire, M. T., & Troisi, A. (1998). Prevalence differences in depression among males and females: Are there evolutionary explanations? British Journal of Medical Psychology, 71, 479–491. McGuire, M. T., Troisi, A., & Raleigh, M. M. (1997). Depression in evolutionary context. In S. Baron-Cohen (Ed.), The maladapted mind (pp. 255–282). Hove, UK: Psychology Press. Monroe, S. M., Rohde, P., Seeley, J. R., & Lewinsohn, P. M. (1999). Life events and depression in adolescence: Relationship loss as a prospective risk factor for first-onset of major depressive disorder. Journal of Abnormal Psychology, 108, 606–614. Nesse, R. M. (1990). Evolutionary explanations of emotions. Human Nature, 1, 261–289. Nesse, R. M. (1998). Emotional disorders in evolutionary perspective. British Journal of Medical Psychology, 71, 397–415. Nesse, R. M. (2000). Is depression an adaptation? Archives of General Psychiatry, 57, 14– 20. Nettle, D. (2004). Evolutionary origins of depression: A review and reformulation. Journal of Affective Disorders, 81, 91–102. Oatley, K. (1992). Best laid schemes: The psychology of emotions. New York: Cambridge University Press. Öhman, A., Dimberg, U., & Öst, L. G. (1985). Animal and social phobias: Biological constraints on learned fear responses. In S. Reiss & R. R. Bootzin (Eds.), Theoretical issues in behaviour therapy (pp. 123–175). New York: Academic Press. Pietromonaco, P. R., & Rook, K. S. (1987). Decision style in depression: The contribution of perceived risks versus benefits. Journal of Personality and Social Psychology, 52, 399– 408. Price, J. (1967). The dominance hierarchy and the evolution of mental illness. Lancet, 7502, 243–246.


142 EVOLUTION AND THE SOCIAL MIND Price, J. (1998). The adaptive function of mood change. British Journal of Medical Psychology, 71, 465–477. Price, J. S. (1972). Genetic and phylogenetic aspects of mood variations. International Journal of Mental Health, 1, 124–144. Price, J. S., Sloman, L., Gardner, R., Gilbert, P., & Rohde, P. (1994). The social competition hypothesis of depression. British Journal of Psychiatry, 164, 309–315. Rohrmann, B. (2004). Risk Attitude Scales: Concepts and questionnaires. Project Report, Department of Psychology, University of Melbourne. Retrieved November 8, 2006, from http://www.rohrmannresearch.net/ pdfs/rohrmann-racreport.pdf Seligman, M. E. P. (1975). Helplessness: On depression development and death. San Francisco: Freeman. Wason, P. (1966). Reasoning. In B. M. Foss

(Ed.), New horizons in psychology (pp. 135– 151). Harmondsworth, UK: Penguin. Watson, D. (2000). Mood and temperament. New York: Guilford Press. Westermann, R., Spies, K., Stahl, G., & Hesse, F. W. (1996). Relative effectiveness and validity of mood induction procedures: A meta-analysis. European Journal of Social Psychology, 26, 557–580. Wiggins, J. S., & Trapnell, P. D. (1996). A dyadic-interactional perspective on the fivefactor model. In J. S. Wiggins (Ed.), The five factor model of personality (pp. 88–162). New York: Guilford Press. Zuckerman, M., Kuhlman, D. M., Joireman, J., Teta, P., & Kraft, P. (1993). A comparison of three structural models for personality: The Big Three, the Big Five, and the Alternative Five. Journal of Personality and Social Psychology, 65, 757–768.





Coevolved Cognitive Mechanisms in Mate Search Making Decisions in a Decision-shaped World


Searching for a Space The Big Picture: Ecological Rationality Sequential Decision Making in Mate Choice Strategies for Mutual Mate Search Summary and Connections

SEARCHING FOR A SPACE magine that you are driving to a movie theater to watch a film. You look for a place to park as you approach the cinema, and show time is coming up. You feel some time pressure not to miss all the previews, and you are motivated to try to minimize your total travel time, including driving and then walking from wherever you park. What is the best strategy to follow in this case? Should you park in the first space you come to? Or should you immediately drive all the way to the theater, then turn around and take the first space you find as you head away again? But what if everyone else driving to the movies does the same thing? How should you choose a parking space in the face of everyone else’s choices so as not to have to skip buying Raisinets before the film starts? The search for a parking space did not present much of a concern to our ancestors in the course of human evolution, but this modern trial illustrates a common aspect of the situations that challenged our evolving minds: The problem here and in many other social domains is how to make decisions in




environments that are shaped by the decisions of ourselves and others. What makes such tasks particularly tricky is that whatever decision mechanism we use must work within some rather stringent bounds. For instance, in this setting, we have limited time to make the parking choice as we drive along; we have limited knowledge of what spaces may lie ahead, perhaps some prior expectations and perhaps we can see a few upcoming spots, but not beyond the next parked SUV; and we have limited cognitive capacity to put together our current experience of who is parking where with our past expectations to come up with a decision about whether to take an empty spot we find or to keep driving and hope that a better one turns up later. And even if we did have Total Information Awareness satellites to tell us that, indeed, the spot right in front of the theater is still empty right now, that could change by the time we drove there—so in this dynamic, uncertain, self-creating environment, more information would not necessarily be better. How, and how well, can decision makers operate in the face of such challenges? In this chapter, I will show how some simple strategies can allow individuals to behave adaptively even in complex domains, focusing specifically on the important adaptive task of choosing a mate. Part of the power of these decision mechanisms comes from their fit to the structure of information in their particular environment. But because individuals actually change their environment—and the environment faced by others—through the decisions they make, their strategies must be fit to the structure of that changed, individualconstructed environment. This coadaptation or coevolution of decision mechanisms and environment structure leads to distinct strategies being most appropriate for different problems. Many of these problems, such as mate choice, involve strategic interactions where individuals alter the environment of other individuals in their own social network, putting this topic at the center of social psychology; and yet, despite the complexity of the interactions involved, simple decision heuristics can lead to good outcomes, contrary to common assumptions in social cognition research (e.g., Nisbett & Ross, 1980).

THE BIG PICTURE: ECOLOGICAL RATIONALITY We will first consider the big picture within which this research fits, driven by the following big question: How can good decisions be made by real minds operating in an uncertain world? This is a mystery, because humans and other animals must make decisions within the rather severe bounds that our minds and the world impose on us. As mentioned earlier, these bounds include the limited time that we have to make decisions before an opportunity may be gone, the limited and uncertain information we can access within that time, and the limited ability we have to process that information, owing to constraints of memory, processing speed, and the amount of complexity we can deal with. To work within these bounds and still behave adaptively, agents can rely on


simple “fast and frugal” heuristics (Gigerenzer & Goldstein, 1996; Gigerenzer, Todd, & the ABC Research Group, 1999)—decision rules that use a small amount of time, information, and processing to come up with what are usually good choices, when they are employed in the proper environments. This use in appropriate environments is key to the heuristics’ successful application, because it allows them to exploit the fact that information in the world is typically structured in useful ways. For example, if you ask what authors, or places, or products are widely recognized in a given society, you will find systematic patterns relating recognition knowledge to the publication rate, or population size, or prevalence of those things, rather than a random or uniform distribution of what is recognized (Gigerenzer et al., 1999; Goldstein & Gigerenzer, 2002). This structure can then be capitalized on by simple heuristics that employ recognition as a cue in making choices, for instance what paper to cite or what brand to buy. In fact, by counting on certain information structures to be present in the environment, decision heuristics can be correspondingly simpler, effectively letting the world do some of the work for them. Using simple heuristics in environments to which they fit can enable decision-making agents to achieve what Herbert Simon (1990) called bounded rationality. In contrast to the largely unachievable dream of unbounded rationality, which assumes optimal processing of all available information without concern for computational or informational costs, Simon saw humans as exhibiting a bounded form of rationality emerging from the interaction of two forces: the cognitive capabilities of the agent and the structure of the task environment. These two components should fit together like the two blades of a pair of scissors for adaptive, or boundedly rational, behavior to be produced—that is, mind and environment should be closely matched if decision outcomes are to be useful. This perspective aligns well with that of evolutionary psychology, which adds the assumption that the close mind–environment fit has been achieved by evolution honing the former to match the latter. This chapter in turn focuses on how minds shape their own environments, particularly in social domains, so that the adaptive forces flow in both directions between the organisms and their world. Gigerenzer and colleagues (1999) have taken up the challenge of identifying the particular decision mechanisms that can produce bounded rationality in the presence of particular structures of information in the environment. They call this research program ecological rationality, to emphasize the importance of considering both environmental information structure and psychological information-processing mechanisms, and how the former enables and constrains the latter to yield adaptive decisions. Their strategy for studying the ecological rationality of particular decision mechanisms proceeds through a sequence of steps that largely follows the research plan for evolutionary psychology set out by Cosmides and Tooby (1987), including analysis of the environment, simulation of proposed heuristic mechanisms, mathematical analysis of the information structures in which they will and will not work well, and empirical investigation


of when people actually use these heuristics (see Todd, 2000, for more connections). But so far the focus of this work has been on environments whose structure could be specified independently of the decision makers acting in it. This is a simple starting point, but often the world does not work this way, and does not oblige us with such independence or stability. In many real situations instead, the environment is shaped by the decisions of the agents acting in it. This was the case in the initial example of the parking situation—the pattern of occupied and free spots, that is, the structure of the environment, is created just through the action of the decision makers who have arrived before you. Imagine introducing a burr or bump onto the “agent” blade of Simon’s scissors—over time it will carve a matching channel in the environment blade, shaping the environment to fit the agents’ cognitive mechanisms more closely. And in turn the bump on the cognitive blade will be worn down by the counteracting force of the environment blade—so both the cognitive system and the environment can coadapt or coevolve to shape each other. (See Gangestad & Thornhill, chapter 3, this volume, for examples of coadaptation when the cognitive system belongs to a signal-perceiver and the environment comprises a target signal-producer.) Examples of the coconstruction of decision mechanism and environment can be found in different classes of heuristics. For instance, repeated recognitionbased decisions, such as when large numbers of scholars choose whom to cite in a paper or shoppers decide what music to buy based on what they recognize, can result in some things in the world, such as some authors or bands, becoming much more recognized, and chosen, than others (Todd & Heuvelink, 2006). The repeated application of the recognition heuristic can lead the world to become J-shaped in the sense of a power law describing the rates of choices among options. This environment structure in turn can affect the efficacy of the recognition heuristic itself, making it more beneficial to use. Such strategic interactions have long been the province of game theory, but here we take a more psychologically-oriented approach to looking at the particular informationprocessing mechanisms that interacting individuals may adaptively use. We next turn to the case of sequential decision heuristics, focusing on the domain of mate choice, for a particular example of how decision mechanisms can change their environments, and how those changes can correspondingly influence which decision heuristics may function well in the newly-altered environments.

SEQUENTIAL DECISION MAKING IN MATE CHOICE The task of mate choice can be broadly thought of as incorporating three steps: first, assessing the relevant cues of mate quality of an individual (see Gangestad & Thornhill, chapter 3, this volume, for several such cues), then processing those cues somehow into an overall judgment of the individual’s mate quality, and finally using that judged quality in the process of searching through a sequence of


individuals to decide whom to court and whom to pass by. Decision heuristics involving limited cue processing can be used in the first and second steps; here we will assume that the outcome of these two steps is that all cues have been collapsed into a single criterion value of mate quality. What heuristics can be used to guide mate search through a sequence of potential mates with different mate quality values? The mate choice domain for humans has a particular constellation of features, which can vary somewhat from culture to culture, not to mention from other species. But there are some typical basic underlying commonalities that are also found in other forms of sequential search (Schotter & Braunstein, 1981). In many search domains where there is competition for specific alternatives, such as searching for a mate, buying unique items like antiques or houses, or looking for a job or job candidate, once you have passed by an alternative and decided not to pick it, there may be no chance of changing your mind and returning to that alternative later, because someone else will have bought the house you rejected or married the person you spurned. Also in such situations, you probably will not know the range of possible alternatives ahead of time so that you have to learn about this distribution as you search. To find out what kind of approach is appropriate for searching in such an environment, we begin with a problem of this form that has been well-studied in probability theory, where it is known as the secretary problem in the job search domain, or the dowry problem in the mate search domain (Ferguson, 1989). As the dowry problem, it goes like this: A sultan wishes to test the wisdom of his advisor, who is seeking a wife. The sultan arranges to have 100 women from the kingdom brought before the advisor in succession, and all the advisor has to do is to choose the woman with the highest dowry. If he chooses correctly, he gets to marry that woman and keep his job with the sultan. The advisor can see one woman at a time and ask her dowry; then he must decide immediately if he thinks she is the one with the highest dowry out of all 100 women, or else let her pass by and go on to the next woman. He cannot return to any of the women he has already seen—once he lets a woman go by, she becomes unavailable. Moreover, the advisor has no idea of the range of dowries before he starts his search. What strategy can he use to have the greatest chance of selecting the one woman with the highest dowry? For such challenging search problems, Simon (1990) suggested the satisficing approach of setting an aspiration level somehow and then searching until an option is encountered that exceeds it. The optimal way to set an aspiration level in the dowry problem is to search through the sequence of options for some time without making any final choice, so that enough information can be gathered about the available values to make a good decision, but not so long that the highest value is passed by in this initial information-gathering stage. The length of initial search that optimizes this balance is to look at N/e of the available alternatives, where N is the number of alternatives and e ≈ 2.718 is the base of the natural logarithm system, which comes out to 37% of the sequence of


alternatives (Ferguson, 1989). In other words, the optimal approach is to follow the 37% rule: In Phase 1 of search, look at 37% of the alternatives; then set the aspiration level to equal the highest value seen among all those alternatives; and then continue search in Phase 2 until an alternative is found that exceeds the aspiration level. Note that this type of strategy has minimal cognitive and information requirements: There is no need to remember a distribution or calculate statistics over multiple values; in fact, the searcher only needs to remember a single value as the aspiration level, and make a simple comparison between it and every currently-perceived value in succession. But this optimal rule, while it is the best possible under the circumstances, still does not do that well—the chance of picking the highest dowry is only about 1 in 3, which means there is a good chance that the sultan’s advisor ends up single and jobless. Luckily, real human mate choice is seldom like this, and instead our outcome criteria are usually a bit more lenient—rather than having to find the one best, a more reasonable version of this scenario would say the advisor just has to live with whatever choice he makes, and so he should aim to maximize the expected mean value of the dowry he stops at. To do this, he only needs to look at nine women, rather than 37, and set his aspiration level to the highest among those nine before stopping search at the next woman who exceeds that aspiration (Todd & Miller, 1999). If no other dowry exceeds that aspiration, then he settles with the last woman he encounters. This best performing rule can be seen in the top line in Figure 9.1, which shows the length of search in Phase 1 before setting the aspiration level, along the x axis, plotted against the mean dowry value

FIGURE 9.1 Performance in terms of mean mate value selected (out of 100) versus length of Phase 1 search for two one-sided sequential search rules, one with and one without competition.


selected, assuming dowries from 1 to 100, along the y axis. The best performance reached yields a mean selected value of 92. Searching through less than nine initial samples does not give enough information to set quite as good an aspiration level, while going through more than nine candidates before turning to the second phase of search increases the chances of missing a high dowry in the initial phase. So setting an aspiration level after seeing just nine options in this scenario is a strategy that is appropriate for a lone searcher to use, living in the adolescent male fantasy world of one man and 100 available partners. But of course, real life is seldom like this simple lone-searcher scenario—other people come along and mess things up. This leads back to our central question: What happens when the decisions of others (and oneself ) affect the structure of the world? In particular, we can ask how the best strategy in this scenario changes when competition is introduced. Now, rather than one wise-man advisor, imagine 100 wise women, each searching through the same set of 100 male candidates, trying to select the best one according to some criterion. If every woman was looking for exactly the same thing in a man, then the best any of them could do would be to take the first man they see, and hope they get lucky—so much direct competition creates a zero-sum game where the best average payoff comes from making a random (and fast, if there are time costs) selection. But in real mate choice, people’s criteria differ to a greater or lesser extent. So to make things somewhat more realistic and easier for everyone, we can go to the opposite extreme and give all the women a different, independent and uncorrelated, criterion to search for. Thus wise woman 1 is looking for the man with the bluest eyes, wise woman 2 seeks the man with the longest ears, etc. Hence, while they are all competing for someone among the same pool of candidates, they are each looking for something different. In this situation of indirect competition, how should each woman search to maximize the criterion she alone seeks? In the bottom line in Figure 9.1, we see that even indirect competition makes things more challenging for the individual searchers: To maximize the mean value that they select, they have to act more quickly than before when there was no competition, now checking out only three or four potential mates in Phase 1 search before setting their aspiration level. If any searcher takes longer than this, then someone else is more likely to take the candidate that she is most attracted to. And this competition also means that the searchers’ overall performance falls—the highest mean value they can hope for is around 80, instead of around 92 before. Again, in actual mate search, there will certainly be some correlation among people’s preferences, so a more realistic scenario would fall between this case and the full direct competition case where everyone has the same preferences. This indicates that an even shorter Phase 1 search would be called for. Thus, as the search problem is made more complex by taking into account the decisions being made by others, the most appropriate strategy is to use a satisficing mechanism that sets its aspiration level after shorter amounts of initial


Phase 1 search—in fact, less than half as much search as is appropriate for a lone searcher. While the direction of this strategy change is not surprising, the magnitude was unexpected. But even greater, and qualitative, changes in strategy are called for when we take into account the other half of the decisions made in the mating domain that have been missing in this scenario.

STRATEGIES FOR MUTUAL MATE SEARCH Those other decisions are being made not by our competitors, but by those individuals whom we are searching through. In some evolutionarily-important search domains, the options that people are searching through have little choice themselves as to whether or not they will be chosen, such as in habitat choice or food-patch selection. The challenge in mate choice though is that few of us are sultans, able to line up a selection of potential mates and one-sidedly declare which one we will have. For most of us, mate choice is two-sided, and mate search is mutual, which means that searchers are being searched at the same time. The empirical manifestations of such a mutual process, as it operates in human mate choice at least, is that most people find a mate who is typically somewhat matched to them on attractiveness and other dimensions (Kalick & Hamilton, 1986) after a reasonably short search. What simple decision algorithms can create such outcomes given the decisions that everyone else is making, competitors and choosy potential mates alike? To find out, we can turn to computer simulations, which allow us to test how different decision strategies would work when employed by a population of individuals searching for partners. Such simulation studies are useful for testing the implications of proposed psychological mechanisms and for generating testable predictions about the behaviors they would lead to, forming “runnable thought-experiments” that can bolster our understanding of possible social interactions (as in mutual mate search) that are beyond our intuitive abilities to predict. We set up a simulation similar to a classroom demonstration called the Pairing Game (Ellis & Kelley, 1999) in which two sets of individuals with numbers on their foreheads must wordlessly find their numeric match from the other set (see Fletcher & Overall, chapter 12, this volume, for more details on this game, and for further consideration of how individuals can find, or end up with, similar partners). In this model (Todd & Miller, 1999), we simulate 100 males and 100 females, each with some attractiveness value drawn from a uniform distribution from 0 to 100. As in the Pairing Game and in real life, individuals do not innately know their own attractiveness value, but they can see the values of all potential mates they encounter. Individuals meet in male–female pairs, assess each other, and decide somehow whether or not to make a proposal to each other. This meeting and assessing process happens in two phases. In the first “adolescent” phase, proposals and rejections do not result in actual pairing. But they can be used to set or adjust an aspiration level that will determine to


whom later proposal offers are made. In the following “adult” phase, the aspiration level set during the adolescent phase is fixed and used to make decisions during the rest of the search. These proposal and rejection decisions are now “real,” in that mutual proposals result in a pair being made and that couple leaving the simulation. It is this necessity for mutual agreement that makes this scenario different from the one-sided case described above—the decisions of potential mates play a critical role here in determining one’s mating fate, so one’s decision strategy should take this into account. The degree of competition is also different from the one-sided search cases. Here, as before, everyone shares the same assessment of the attractiveness of others, which is an indication of direct competition. But because everyone is different in terms of their own attractiveness, and must find a mate who also wants them, everyone’s goals are by definition different—so the level of competition here is somewhere between a direct and an indirect situation. How well do different search strategies fare in this setting? A simple strategy that only uses half of the available information, ignoring the decisions made by potential mates, is the one-sided strategy discussed in the previous section, now applied in the mutual setting. In this case, each individual goes through the adolescent period and just sets his or her aspiration level at the highest mate value seen among the potential mates that he or she meets. What happens when this “ignorant” strategy is used in the mutual search case is that most everyone quickly ends up with very high aspiration levels, and thus only those with very high mate values will find willing mates—who must also have very high mate values. As a consequence, unrealistically very few pairs are formed using this strategy with anything other than an extremely short Phase 1 search. This is shown in the bottom line in Figure 9.2, plotting the mean number of pairs formed in the population against the adolescent learning period of Phase 1 search used by this ignorant strategy. These pairs are well-matched, in terms of having small within-pair differences in mate value, but that is because only the very high-valued individuals find mates. So ignoring the decisions made by others, and trying to get the best mate possible without regard for one’s own attractiveness on the market, results in unrealistically poor outcomes for most searchers. In contrast, if individuals magically knew their own mate value and used that as their aspiration level, then most of the population could quickly find a well-matched partner. But as mentioned earlier, the problem here is that individuals do not have built-in knowledge of their relative ranking in the current mate market, so if they wanted to use it in their decisions, they must infer it or learn it. A reasonable approach to this problem could be to use the assessments that others make about oneself as a cue about one’s own mate value, which, after all, the others can see. So one could raise one’s self-appraisal, and hence one’s aspiration level, every time an offer is received and lower it after every rejection. This also fits with intuitions about how romantic successes and failures can induce self-esteem to go up and down, which in turn can affect how high or low people aim in their next romantic endeavors. (Another potential source of


FIGURE 9.2 Mean number of mated pairs formed (out of 100) versus length of Phase 1

search for two different mutual search rules.

information which we do not include here is feedback from, and comparisons to, one’s competitors.) To specify a decision mechanism in more detail, all individuals start with an initial aspiration level of 50, which corresponds to assuming oneself to be just average. Then, during the adolescent learning period, for every proposal from someone more attractive than one’s current aspiration level, raise one’s aspiration level to be partway to the other’s attractiveness value. Any proposals from someone less attractive than one’s aspiration level are somewhat to be expected, and so will not have any effect. Just the reverse happens for rejections: For every rejection from someone below one’s current aspiration level, lower the aspiration level toward the other’s attractiveness. As individuals’ aspiration levels change over the course of the adolescence period, they also influence the learning of everyone else’s aspiration levels via the combined effect of the proposals and rejections made. With this simple rule, taking into account all of the decisions made by others, many more pairs are formed, at least if the adolescence period is not too long (top line, Figure 9.2). The within-pair attractiveness difference is also smaller for this rule than for the ignorant rule, meaning closer matches for the same amount of search. This simple rule’s performance does not exactly meet the observed realities of human mate search, particularly because too few pairs are formed (e.g., here only around half of the population finds a mate). However, it goes in the right direction, and modifications of this type of mutual search rule can come much closer to human behavior (Simão & Todd, 2003). What forms of empirical evidence can we find to assess whether or not people are actually using such simple sequential search mechanisms when


looking for a mate? A variety of indirect evidence is at least consistent with these aspiration-adjustment mechanisms. For instance, individual self-esteem goes up and down with dating success or failure, and this could be the basis of an aspiration level for further search (Kenrick, Groth, Trost, & Sadalla, 1993; Kirkpatrick & Ellis, 2001). People searching for a match in an online dating setting want a small set of potential dates to look through (even though they may profess ahead of time that they want many options—Lenton, Fasolo, & Todd, 2005). In addition, the behavior of these proposed search mechanisms can be assessed against population-level outcome measures. For instance, demographers have long puzzled over a frequently-observed skewed bell-shaped pattern in the distribution of ages at which people first get married (Coale, 1971). When we created an agent-based demographic model of a population of males and females looking for marriage partners by using the mutual sequential search heuristics described earlier, we found that the “ages” at which the individuals got married basically matched the observed demographic data (Todd, Billari, & Simão, 2005). The most compelling evidence regarding mechanisms being used, though, would come from direct observation of the sequential mate search process as it transpires. The challenge is that this search typically takes an extended period of time, happening over months and years, which would require a detailed longitudinal study. What would be very handy is a way to watch the sequential mate search process that people go through, somehow distilled down into an easilyobservable sped-up version of reality. Just such an opportunity is afforded by the phenomenon of speed dating: a commercially-sponsored occasion in which several men and women seeking dates sequentially meet and assess each other within the span of an evening. Researchers have begun using speed dating as a source of data about the mate choices that people make (Kurzban & Weeden, 2005), and we are using data from events conducted by the FastDating firm in Germany to explore the decision mechanisms involved. What happens at these events is that 20–25 men and an equal number of women gather in a large room one evening, and all the women sit down at separate tables. Each man then sits down across from a particular woman, and the pair get to talk for 5 minutes about whatever they want. At the end of this time, the organizer rings a bell and all the men and women circle a response on a card they are carrying to indicate whether or not they would like to interact further with this person they have been talking to. Then all the men stand up and shift to the next woman in line at the next table, and the process repeats. After everyone has talked to every member of the opposite sex in this way, the organizer compares everyone’s cards to see which pairs expressed mutual interest in each other, and the members of those pairs are sent each other’s contact information. This speed dating setup is quite similar to the sequential mate search situation we have been exploring, because everyone sees a succession of potential dates without knowing exactly who is coming later, and must decide after each 5minute meeting whether or not they are interested in that person. However,


there are also a number of differences from the dowry-problem scenario (and, presumably, from much of traditional mate search): Here, the men and women gathering for the evening’s event actually have some time before it starts to mill around and meet each other initially, and thus to get an idea of the range of potential dates that they will be talking to; and the decisions that are made after talking with someone can be erased and changed later in the evening after meeting more candidates. (Of course, the other major difference is that in the dowry problem, only a single choice can be made, while here, individuals can indicate interest in as many others as they wish, which will modify the search mechanism used somewhat.) To address some of these differences and create a scenario that is closer to the extended, low-knowledge situation in traditional mate search, we are developing our own new speed dating setup in which men and women will be kept isolated from each other until their appointed 5-minute meeting, and must make their decisions on the spot without the possibility of changing them later. We also hope to gather data throughout the evening about how each individual’s aspiration level changes as a consequence of their experience and feedback in each minidate, to help identify what cognitive mechanisms are being used to decide on dating interest given the decisions of everyone else.

SUMMARY AND CONNECTIONS To summarize the above results on how sequential mate search must be adjusted to take into account the decisions made by others, a solo searcher (who has no others to account for) looking to maximize his or her mean selected mate value should set an aspiration level to the highest value seen during an initial short trial period. When indirect competition is introduced, so that there are multiple searchers seeking different things among the same set of options, each searcher should adjust for the decisions being made by his or her competitors by making a quicker choice, shortening the length of the initial trial period. And when mutual choice is included, so that searchers must also be sought in order to succeed, searchers should adjust their aspiration level during adolescence toward the mate value of the successive potential mates they encounter, conditional on their current aspiration level and on getting a proposal or rejection. In this case, rather than just switching to a new behavior based on the fact that others are also making choices, individuals can actually use some of the decisions made by others to inform and adjust their own decision-making strategies. The impact of other mutual mate searchers can be seen not only in the sequential search strategies individuals use, but also in their weighting and processing of the separate cues of the quality of potential mates they encounter (the first two steps of the mate choice task indicated earlier). Again, if a solo searcher had no competition to contend with, he or she could set an aspiration level for traits such as wealth and status, physical attractiveness, and degree of parental investment as high as desired. But in mutual mate choice, one’s aspirations must


be tempered by what one has to offer in return—someone with a portfolio of only low trait values who nonetheless demands high values in a mate will end up disappointed and alone. Buston and Emlen (2003) found evidence for exactly this sort of self-sensitive aspiration setting, in that individuals who had higher overall self-appraisals (created by summing their self-reported levels on 10 traits of mate quality) had choosier preferences for mates (again found by summing the preferred values of the same 10 traits). There is currently some disagreement as to how the trait values of self and other are ultimately processed to yield an overall judgment of attractiveness—whether individuals seek mates with the same trait levels as themselves, as indicated by Buston and Emlen’s questionnaire data, or seek mates with complementary values such as male status against female attractiveness, as indicated by evolutionary theory (Buss, 1989) and data from actual choices made in the speed-dating context described earlier (Todd, Penke, Fasolo, & Lenton, 2005; see also Fletcher & Overall, chapter 12, this volume, on what traits people use and how). But in either case, to be successful individuals must adjust their mate choice strategies to fit with how others are choosing. Finally, what of parking? After all, once one has found a mate, and a job, and a place to live, finding a good parking place may be one of the most challenging sequential—and social—choice problems left to tackle. Do any of the search strategies we may have evolved for use in other domains find application in this modern task? Parking is not so much like mutual mate search, in that parking spaces cannot veto our decisions and tell us we cannot park there. But parking search is similar to the one-sided search situation we described earlier. Imagine a long road leading to a destination, with cars parked along one side of it, so that as we drive toward the destination we encounter a sequence of possible parking spaces. As in other forms of sequential search, we do not know what opportunities lie ahead, and, while we can turn around and go back to a spot we passed before, we cannot be certain that it will still be available. One unique aspect here is that the parking spaces we encounter keep getting better in quality over time as we drive toward the destination, which is not usually the case for other forms of search (though it might be, if the search can be focused on more productive options over time). So now for the parking search problem, what sorts of rules might be appropriate, that is, able to find spaces close to the destination preferably without turning around? Our discussion earlier of one-sided search suggests the satisficing approach: Set an aspiration level, and take the next better option encountered after that. In this case, that means passing by some fixed number of parking spots, whether they are empty or occupied, and then taking the first available space we come to thereafter. This type of fixed-distance rule is in fact the optimal approach for an infinite parking lane filled with a constant density of spaces (MacQueen & Miller, 1960). But our interest here is in what strategies work well when the pattern of spaces is not necessarily constant, but rather is created by the decisions of other drivers parking. To find out, we put evolution to


work, allowing a variety of parking-search strategies to compete and evolve over time in a simulated version of the single-lane world described above (Hutchinson, Fanselow, & Todd, 2005). We used different types of distance-based strategies (such as the fixed-distance satisficing rule) along with density-based ones (such as rules that take the next available parking space as soon as enough of the last few spots passed have been occupied). Those strategies that found closer parking places on one day’s worth of parking were used more often in the population of drivers on the next parking day, and slight mutations were introduced into the evolutionary process to allow the full space of strategies to be explored. After just a few generations of evolving parking strategies, only two rules emerge as the winners in a mixed equilibrium: the fixed-distance heuristic, used by about 80% of the population, and a density-based linear-operator heuristic, used by the rest. Thus, while it is best for everyone to use the same fixed-distance heuristic if they are all in a static environment that they cannot affect, this is not the best option when the environment is created by the drivers themselves—in that case, some individuals do better by using a density-based mechanism, which can take advantage of the environment structure created by the fixed-distance users. Whether or not such a mixed strategy is used by real drivers, and whether it is also appropriate in other domains such as mate search, remains to be explored. Many other questions about the decision mechanisms that are best suited to environments they themselves shape are also open for exploration. In addition to developing models of such mind–environment coadaptation in other domains and looking for evidence of the process in operation, the most important questions center on whether we can develop a theory of the principles underlying such coadaptation that will allow us to predict when it will occur and what form it will take in different domains. Doing so will require a three-way focus: on the constraints and structure of the information-processing mechanisms that individuals bring to bear on the environmental challenges they face, on the constraints and structure that the environment imposes on the information available to individuals, and on the way these two sets of constraints interact with and shape each other over time. REFERENCES Buss, D. M. (1989). Sex differences in human mate preferences: Evolutionary hypotheses tested in 37 cultures. Behavioral and Brain Sciences, 12, 1–49. Buston, P. M., & Emlen, S. T. (2003). Cognitive processes underlying human mate choice: The relationship between selfperception and mate preference in Western society. Proceedings of the National Academy of Sciences of the USA, 100, 8805–8810.

Coale, A. J. (1971). Age patterns of marriage. Population Studies, 25, 193–214. Cosmides, L., & Tooby, J. (1987). From evolution to behavior: Evolutionary psychology as the missing link. In J. Dupré (Ed.), The latest on the best: Essays on evolution and optimization (pp. 277–306). Cambridge, MA: MIT Press/Bradford Books. Ellis, B. J., & Kelley, H. H. (1999). The pairing game: A classroom demonstration of the


matching phenomenon. Teaching of Psychology, 26, 118–121. Ferguson, T. S. (1989). Who solved the secretary problem? Statistical Science, 4, 282– 296. Gigerenzer, G., & Goldstein, D. G. (1996). Reasoning the fast and frugal way: Models of bounded rationality. Psychological Review, 103, 650–669. Gigerenzer, G., Todd, P. M., & the ABC Research Group. (1999). Simple heuristics that make us smart. New York: Oxford University Press. Goldstein, D. G., & Gigerenzer, G. (2002). Models of ecological rationality: The recognition heuristic. Psychological Review, 109, 75– 90. Hutchinson, J., Fanselow, C., & Todd, P. M. (2005). Car parking as a game between simple heuristics. In P. M. Todd, G. Gigerenzer, & the ABC Research Group (Eds.), Ecological rationality: Intelligence in the world. Manuscript in preparation. Kalick, S. M., & Hamilton, T. E. (1986). The matching hypothesis reexamined. Journal of Personality and Social Psychology, 51, 673– 682. Kenrick, D. T., Groth, G. E., Trost, M. R., & Sadalla, E. K. (1993). Integrating evolutionary and social exchange perspectives on relationships: Effects of gender, self-appraisal, and involvement level on mate selection criteria. Journal of Personality and Social Psychology, 64, 951–969. Kirkpatrick, L. A., & Ellis, B. J. (2001). An evolutionary-psychological approach to selfesteem: Multiple domains and multiple functions. In G. Fletcher & M. Clark (Eds.), The Blackwell handbook of social psychology: Vol. 2. Interpersonal processes (pp. 411–436). Oxford, UK: Blackwell. Kurzban, R., & Weeden, J. (2005). HurryDate: Mate preferences in action. Evolution and Human Behavior, 26(3), 227–244. Lenton, A. P., Fasolo, B., & Todd, P. M. (2005).

When less is more in “shopping” for a mate: Expectations vs. actual preferences in online mate choice. Manuscript submitted for publication. MacQueen, J., & Miller, R. G., Jr. (1960). Optimal persistence policies. Operations Research, 8, 362–380. Nisbett, R., & Ross, L. (1980). Human inference: Strategies and shortcomings of social judgment. Englewood Cliffs, NJ: Prentice Hall. Schotter, A., & Braunstein, Y. M. (1981). Economic search: An experimental study. Economic Inquiry, 19, 1–25. Simão, J., & Todd, P. M. (2003). Emergent patterns of mate choice in human populations. Artificial Life, 9, 403–417. Simon, H. A. (1990). Invariants of human behavior. Annual Review of Psychology, 41, 1–19. Todd, P. M. (2000). The ecological rationality of mechanisms evolved to make up minds. The American Behavioral Scientist, 43(6), 940–956. Todd, P. M., Billari, F. C., & Simão, J. (2005). Aggregate age-at-marriage patterns from individual mate-search heuristics. Demography, 42(3), 559–574. Todd, P. M., & Heuvelink, A. (2006). Shaping social environments with simple recognition heuristics. In P. Carruthers (Ed.), The innate mind: Culture and cognition (pp. 165–180). Oxford, UK: Oxford University Press. Todd, P. M., & Miller, G. F. (1999). From pride and prejudice to persuasion: Satisficing in mate search. In G. Gigerenzer, P. M. Todd, & the ABC Research Group (Eds.), Simple heuristics that make us smart (pp. 287–308). New York: Oxford University Press. Todd, P. M., Penke, L., Fasolo, B., & Lenton, A. P. (2005). What we seek and what we find: Reconsidering the cognitive processes underlying human mate choice. Manuscript in preparation.


An Evolutionary Account of Strategic Pluralism in Human Mating Changes in Mate Preferences Across the Ovulatory Cycle


Strategic Pluralism and Human Mating: Patterned Changes in Women’s Mate Preferences Across the Ovulatory Cycle Basic Evolutionary Concepts Mating Strategies in Humans Study 1 Study 2 Broader Theoretical Considerations

STRATEGIC PLURALISM AND HUMAN MATING: PATTERNED CHANGES IN WOMEN’S MATE PREFERENCES ACROSS THE OVULATORY CYCLE esearchers have begun to study how women perceive and evaluate men as potential mates at different points of their reproductive cycle. One of the reasons behind the rising interest in testing for patterned changes in women’s mate preferences across the ovulatory cycle is clear. According to evolutionary-based models of mating (e.g., Gangestad & Simpson, 2000), both sexes should have evolved to enact situationally-contingent mating strategies and tactics. The fact that women also evolved to conceive during a small window of



their monthly reproductive cycle permits the derivation of some specific and nonobvious predictions about the kinds of men or male attributes that women should find appealing in short-term versus long-term mates, depending on whether or not women are ovulating. Indeed, research testing for specific patterns of mate preferences in different interpersonal contexts at different points of the female reproductive cycle provides some of the strongest evidence to date for evolved psychological adaptations in humans (see Gangestad, Thornhill, & Garver-Apgar, 2005; Simpson & Campbell, 2005). In this chapter, we first review some basic evolutionary concepts and principles associated with mating, including the twin concepts of sexual selection and adaptations, evolutionary functional analysis, and trade-offs in mating contexts. We then discuss mating strategies and tactics, the extensive within-sex variation that exists on various mating measures, the concept of good genes sexual selection, and the tenets of the Strategic Pluralism Model of mating (SPM; Gangestad & Simpson, 2000). SPM melds principles of “good provider” and “good genes” mate selection and specifies some of the personal and environmental conditions under which women may have evolved to preferentially value and “trade-off” evidence of a mate’s investment potential for evidence of his viability (e.g., good condition and perhaps immune system functioning). After outlining what is currently known about women’s mate preferences across the ovulatory cycle, we showcase two recent studies that have examined women’s mate preferences in response to men’s social behavior depending on where women are in their ovulatory cycle. Supporting SPM, both studies confirm that certain behavioral cues displayed by men are differentially attractive to women, depending on: (1) whether women are evaluating men as long-term or short-term mates, and (2) whether or not they are at peak fertility. These highly specific patterns of findings are difficult to derive or anticipate from nonevolutionary models of human mating. We conclude the chapter by discussing the wider theoretical implications of these and other recent findings.

BASIC EVOLUTIONARY CONCEPTS Sexual Selection and Adaptations Sexual selection refers to differences in reproduction in individuals due to differential advantages in mating, independent of advantages associated with differential survival. Sexual selection produces two types of adaptations (Andersson, 1994): intrasexual competitive abilities and intersexual attraction cues (e.g., signals that most opposite-sex people find attractive). In many species, the number of mates that males attract is closely tied to their ultimate reproductive success, whereas total mate number has appreciably less impact on the reproductive success of females given the prolonged gestation and lactation associated with each birth. Even in most mammals, females are a limited reproductive resource


over which males usually compete. Accordingly, sexual selection pressures should have operated more strongly on male intrasexual competitive abilities and male intersexual attraction cues than vice versa (Cronin, 1991; Trivers, 1972). This premise has been supported in many different species (see Andersson, 1994; Trivers, 1985). Theories of intersexual signaling have focused on two clusters of attributes: (1) signals conveying qualities of a good parent (or “good provider”), and (2) signals that honestly advertise “good genes” (Cronin, 1991; Gangestad & Thornhill, chapter 3, this volume). Historically, theories of good parenting have been fairly uncontroversial, whereas those invoking good genes have been hotly debated. Within the past decade, modeling studies have confirmed that goodgenes selection may have evolved in several species (Kirkpatrick, 1996; Møller, 1994), even those in which males provide substantial parental care and investment. This has led researchers to consider whether good-parenting and goodgenes selection processes might have generated “mixed” mating strategies in human males and females, as Trivers (1972) originally envisioned. Adaptations are traits or behaviors that conferred a gene-transmitting advantage to individuals who possessed them over those who had different variants of the traits/behaviors in evolutionary history. Adaptations are revealed by evidence of their special design features (Andrews, Gangestad, & Matthews, 2002; Williams, 1966). A trait/behavior has special design if it produced specific beneficial effects that should have enhanced inclusive fitness during evolutionary history with a high degree of specificity, precision, and efficiency. Tooby and Cosmides (1992) have further claimed that adaptive behavioral flexibility should have been facilitated by the evolution of domain-specific psychological mechanisms. These mechanisms are believed to operate via specific decision rules that are evoked by specific environmental cues and that produce efficient, stable, persistent, and finely tuned responses (e.g., specific perceptions or behavioral reactions) that solved recurrent adaptive problems during evolutionary history (e.g., choosing or attracting a desirable mate). These decision rules need not be conscious or premeditated, and many may operate outside of awareness.

Evolutionary Functional Analysis and Trade-Offs Individuals must invest considerable time, effort, and energy to accomplish the major life-tasks that most directly impact their inclusive fitness. Decisions about how to invest time, effort, and energy are important not only because they put individuals at risk, but also because individuals might have used or allocated their resources differently. Adaptations, therefore, have opportunity costs, including lost fitness benefits that might have been gained by using resources in a different manner. Accordingly, benefits must be evaluated in relation to costs to discern whether and how specific adaptations could have evolved (Parker & Maynard Smith, 1991). One objective of evolutionary functional analysis is to identify the cost–benefit “trade-offs” that motivated people to allocate their time, energy, and


effort toward activities in ways that typically would have improved their inclusive fitness. In many mating contexts, direct trade-offs must be made between the allocation of effort to parenting activities versus mating activities. Trivers (1972), in fact, defined parental investment as “any investment by the parent in an individual offspring that increases the offspring’s chance of surviving (and hence reproductive success) at the cost of the parent’s ability to invest in other [including future] offspring” (p. 139). Thus, although parental investment can increase the probability that a given offspring will survive and eventually reproduce, it also carries costs in terms of the lost benefits of alternate investments, including missed opportunities to invest in different or future offspring. Individuals who allocate high effort to parenting could have pursued other endeavors, such as devoting more time or effort to finding and attracting different mates (Gross & Sargent, 1985).

MATING STRATEGIES IN HUMANS Mating Strategies and Tactics Mating strategies are integrated sets of cognitive and behavioral adaptations that organize and guide an individual’s general reproductive efforts (see Lieberman, 2006). They are often conceptualized as implicit decision rules that motivate individuals to allocate their somatic and reproductive effort in functionally adaptive ways (Andersson, 1994). Mating tactics, in contrast, represent the behavioral “output” of individuals who are pursuing a general mating strategy. The enactment of a particular strategy usually involves the deployment of multiple behavioral tactics. For example, males in most biparentally investing species often invest quite heavily in their offspring, yet also remain open to low-cost mating opportunities with other females. According to Trivers (1972), this represents a mixed mating strategy, one that entails multiple behavioral tactics. If each tactic is evoked by specific environmental stimuli such as the prolonged absence of a mate, having short-term sex only with mates who have certain attributes, or seeking short-term sex only when such efforts have worked well in the past, the strategy is deemed a conditional one. Given the varied and changing environments in which reproduction occurred during evolutionary history, selection pressures should not have produced a single mating strategy or set of tactics for males and females in most species, especially in humans. Rather, selection should have shaped a small and flexible set of ecologically-contingent strategies and tactics (see Gross, 1996). One recent study on humans supports this premise. Li, Bailey, Kenrick, and Linsenmeier (2002) gave people “mate dollars” that they could spend to increase the quality of an ideal mate. If their budgets were tight, males invested in the “necessities” of a healthy, fertile mate, whereas females invested in “necessities”


associated with acquiring resources. Both sexes shifted to the “luxuries” of mates who were more creative and talented as budgets increased. This evidence suggests that individuals appear to utilize ecologically contingent strategies in response to relevant environmental factors.

Within-Sex Variation in Human Mating Strategies Until recently, most evolutionary theories focused on the use of different mating strategies by women and men. For example, Wilson (1978) originally proposed that, given how the sexes reproduce, human males should have evolved to be uniformly aggressive, hasty, and undiscriminating in mating contexts, whereas human females should be uniformly coy and should defer mating until the males who have the “best” genes are identified. Although there are between-sex differences in sociosexual attitudes and behaviors (Oliver & Hyde, 1996; Simpson & Gangestad, 1991), there is considerably more within-sex than between-sex variation. Roughly 30% of men, for instance, hold less favorable attitudes about casual sex than the median attitudes of women in North American samples (Gangestad & Simpson, 2000). What might explain this appreciable within-sex variation in mating-related attitudes and behaviors?

Genetic Fitness and Sexual Selection According to good genes selection models, females should have evolved to prefer males who possessed indicators of viability and good physical condition, including adaptive attributes that might be passed on to their offspring via genetic inheritance. Mildly deleterious alleles and lower pathogen resistance has to be inferred from phenotypic markers such as physical or social “advertisements” (Zahavi, 1975). In order to evolve, good genes sexual selection must be based on “honest” signaling, which explains why only certain features serve as valid cues of an individual’s physical condition and, potentially, his/her genetic fitness (Gangestad & Thornhill, chapter 3, this volume; Grafen, 1990). An attribute can remain an “honest” advertisement if individuals who have deleterious alleles or less pathogen resistance cannot develop or sustain the attribute without incurring considerable costs. One constellation of attributes that meet this criterion is those that conditionally “handicap” individuals who have deleterious mutations or are less pathogen-resistant (Gangestad & Thornhill, chapter 3, this volume). Mutations (Pomiankowski, Iwasa, & Nee, 1991) and pathogens (Hamilton & Zuk, 1982) tend to divert or compromise an individual’s energy and resources. Consequently, honestly advertised traits are energetically costly to develop and maintain because “handicapped” individuals cannot develop these traits without diverting valuable resources from other competing demands, such as sustaining their already burdened immune systems (Folstad & Karter, 1992). In birds that have extravagant features, costly handicaps include exaggerated sexual ornaments and colorful plumage (Zuk, Thornhill, Ligon, & Johnson,


1990). In mammals, they include large size and increased musculature that result in sexual dimorphism, particularly in polygynous species (Alexander, Hoogland, Howard, Noonan, & Sherman, 1979).

The Strategic Pluralism Model Informed by these ideas, Gangestad and Simpson (2000) developed a model of human mating that blends principles of “good-genes” and “good provider” models. Although the Strategic Pluralism Model (SPM) applies to both sexes, it focuses primarily on the mating strategies and tactics of women. According to SPM, both sexes should have evolved to engage in conditional or “mixed” mating strategies. Generally speaking, human females should have evolved to value and selectively “trade-off” evidence of a mate’s investment potential for evidence of his viability, contingent on various factors. For example, if a man is perceived as less investing than other available suitors, he must evince higher viability to be viewed as a commensurately attractive mate. If, however, he is seen as relatively more investing than other suitors, he need not be quite as high on viability to be perceived as attractive. Decisions about the relative weighting and value of these two sets of attributes should also be contingent on other factors, one of which ought to be where women are in their reproductive cycles when making mating decisions. According to SPM, women should have evolved to be more attracted to men who display attributes that “honestly” signal their greater viability. This should be especially true in short-term mating contexts when women are ovulating and, thus, could pass the “good genes” of these mates on to their offspring. The question then turns to which interpersonal behaviors might be good candidates to investigate. Trivers (1972) proposed that intrasexual competitive abilities could have evolved as valid cues of heritable fitness. Successful intrasexual competition requires developing costly attributes used in competition (e.g., muscularity, social dominance, willingness to directly compete with other men) along with the expenditure of energy during competitions. Because males who have greater viability should be more capable of withstanding the costs of “handicapping” traits than less viable males, they should be able to devote greater energy to developing the physical and interpersonal tools necessary to succeed in most intrasexual competitions. Females, in turn, should have evolved to attend to the outcomes of intrasexual competitions to evaluate male fitness.

STUDY 1 Several lines of research have revealed that the criteria women use to evaluate men’s attractiveness shift across the reproductive cycle. One major line of work has shown that women prefer the scent of men who have greater developmental


stability (measured by fluctuating asymmetry), especially during the fertile days of their cycles (Gangestad & Thornhill, 1998; Rikowski & Grammer, 1999; Thornhill et al., 2003). Another line of research has confirmed that women prefer masculine faces more on fertile than on nonfertile days of their cycles (Johnston, Hagel, Franklin, Fink, & Grammer, 2001; Penton-Voak & Perrett, 2000; Penton-Voak et al., 1999). And other evidence has revealed that women prefer creativity over wealth (Haselton & Miller, 2006) and deeper voices (Puts, 2005) when they are fertile. Importantly, these shifts in preferences for select male traits emerge only when women evaluate men as potential short-term partners. Collectively, these findings could reflect evolved adaptations for women to choose mates who are capable of providing genetic benefits to their offspring. Heightened attraction to men who possess putative indicators of genetic benefits (e.g., body symmetry, facial masculinity) could increase the likelihood that women will have sex with these men when fertile, even if such men are not their primary romantic partners (Jones et al., 2001). This interpretation is bolstered by the fact that women’s attraction to masculine facial features tends to be strongest during ovulation when women evaluate men as possible short-term sex partners, but not when they evaluate men as long-term, stable partners (PentonVoak et al., 1999). These preference shifts may also explain why women report increased sexual attraction to extrapair men, but only when they are fertile and romantically involved with men who have traits signaling lower genetic benefits (Gangestad, Thornhill, & Garver, 2002). Although scent and facial attractiveness affect women’s attraction to men in important ways, men’s behavior may be an even more critical determinant of attraction. In general, women prefer men who display self-assurance and who stand up for themselves vis-à-vis other men, but who also exhibit warmth and agreeableness (Cunningham, Druen, & Barbee, 1997; Graziano, JensenCampbell, Todd, & Finch, 1997). According to Trivers (1972), the former attributes, which constitute intrasexual competitiveness, may partly function as signals of genetic benefits (i.e., heritable condition) that are conveyed by facial masculinity and developmental stability. The latter attributes, on the other hand, may be more highly valued in long-term, stable mates. Indeed, men who are more symmetrical tend to use more direct intrasexual competitive tactics when interacting with attractive women than do less symmetrical men (Simpson, Gangestad, Christensen, & Leck, 1999). Moreover, Johnston et al. (2001) have found that men who have more masculine faces are perceived as more socially dominant and less investing as fathers than are men who have less masculine faces. The purpose of Study 1 was to test whether women’s preferences for men’s behavioral displays shift depending on women’s fertility status. In this study, Gangestad, Simpson, Cousins, Garver-Apgar, and Christensen (2004) had women view videotaped segments of men who had been interviewed for a possible lunch date. Each man answered a series of questions posed by one of


two attractive women whom they thought would be choosing them or another man for the date. Following the interview, each man was asked to tell his “competitor” why the female interviewer should choose him instead of the competitor. Women raters watched the interviews and evaluated each man’s attractiveness as both a short-term mate (i.e., a sex or “affair” partner) and a long-term mate. We then examined whether women’s ratings were associated with variation in men’s behavioral displays on two dimensions of men’s observer-rated interview behavior: (1) their degree of social presence, and (2) their degree of direct intrasexual competitiveness. Based on the premise that these behavioral displays should partially convey (signal) viability in men, we predicted that women would prefer behavioral displays signaling greater social presence and direct intrasexual competitiveness in short-term mates, but chiefly on days when they were ovulating (i.e., could potentially conceive).

Conception Risk by Male Behavior by Mating Context Interaction Effect The results supported the basic predictions. As shown in Figure 10.1, we found the predicted three-way interaction between women’s ovulatory status (estimated from actuarial data), observer ratings of men’s social presence and direct

FIGURE 10.1 Preference for social presence/direct intrasexual competitiveness as a function of day of the cycle (adjusted for cycle length; points are 3-day moving averages). Preference reflects the mean regression slope of individual women’s ratings regressed on men’s social presence and direct intrasexual competitiveness, with men’s physical attractiveness controlled. The figure illustrates the Conception Risk × Behavioral Display × Mating Context (short-term vs. long-term) interaction. High fertility days run from about Day 6 to Day 14, with fertility peaking at Day 12.


intrasexual competitiveness during the interview, and whether men were evaluated as short-term or long-term mates. Specifically, women perceived men who displayed greater social presence and direct competition as more attractive, but only if they were ovulating and evaluating men as short-term mates. This effect held when several potential confounds (e.g., independent ratings of each man’s level of physical attractiveness) were statistically controlled. Greater social presence and direct competitiveness were also more preferred in short-term than in long-term mates, which is understandable if these behavioral displays advertise traits that are “traded-off” against perceived investment in a committed relationship. Women also rated men as more attractive when they were ovulating, a finding that is consistent with past studies indicating that sexual desire tends to be slightly higher during ovulation (see Gangestad et al., 2005, for a review). These findings contribute to the growing literature demonstrating systematic shifts in mate preferences across the female reproductive cycle. Because women’s attraction to men may depend on men’s behavioral traits even more strongly than on men’s scent or facial masculinity, the preference shifts documented in Study 1 are consequential. The fact that these mate preference shifts are specific to women’s evaluations of short-term mates provides further support for the notion that these results may reflect an evolved female adaptation to garner genetic benefits via extrapair mating.

STUDY 2 If women evolved to enact conditional mating strategies, occasionally engaging in short-term or extrapair sex to obtain heritable fitness benefits at the risk of possibly losing or damaging primary relationships, selection could have shaped women’s mate preferences to be contingent on their fertility status. Indeed, Study 1 reveals that when evaluating men as short-term mates, women find male behavioral attributes that may “honestly” signal heritable fitness more appealing during the fertile phase of their reproductive cycles, but less so when they are not fertile. The logic underlying this prediction is that, because women cannot benefit from a short-term mate’s heritable fitness when they cannot conceive, women should value indicators of heritable fitness less when genetic benefits cannot be gained. When women evaluate long-term mates, however, shifts in these mate preferences should be minimal or nonexistent. One appealing feature of this hypothesis is that it makes predictions that are derived directly from a good genes theoretical position, but are difficult to derive from other theories. Confirmations of predictions believed to have a low probability of being correct if the theory that generated them is wrong provide particularly compelling support for that theory (see Salmon, 1966). Accordingly, evidence for systematic ovulatory cycle preference shifts constitutes strong support for the hypothesis that women have evolved preferences for good genes


indicators in men, particularly when evaluating them as short-term mates (Simpson & Campbell, 2005). There is, however, a possible alternate explanation. During evolutionary history, individuals had competing demands on their time and effort. In addition to finding and retaining mates, ancestral women had to perform many other complex and crucial tasks, such as securing food, caring for and protecting young, maintaining social alliances, and fending off aggressive males. Given that the relative importance of these tasks may have differed depending on specific social or environmental circumstances, selection should have shaped women’s allocation of effort or attention to vary in response to factors that affected the relative importance of each task, especially as it impacted their likely reproductive fitness. Selection, for example, may have favored an allocation strategy that motivated women to pay greater attention to the reproductive task of selecting a good mate when women were fertile and less attention when they were not. Consistent with this perspective, Fessler (2003) has found that the appetite of women tends to be lower during ovulation, despite the fact that women have the greatest caloric needs at this particular time. Fessler suggests that appetite motivates the search for food, which might be allotted lower priority at mid-cycle than immediate reproductive tasks. This “wiser mate selection” hypothesis suggests that women’s mate preferences and standards for attractiveness in men should change across their cycles because women may be more attuned to the general task of finding a good, compatible mate when they are fertile. According to this view, women’s stronger preference for developmental stability, facial masculinity, and displays of social presence and competitiveness could be due to the fact that women prefer male features that are generally valued in mates rather than because they place special emphasis on select male indicators of “good genes.” This alternate view implies that women should make wiser mate choices in general when they are fertile, but should not differentially weigh one type of preferred mate trait or attribute over others. In a second study, Gangestad, Garver-Apgar, Simpson, and Cousins (in press) had normally ovulating women view the same videotapes of men being interviewed for a possible lunch date that were used in Study 1. After viewing each man, women rated his attractiveness as both a long-term and a short-term mate. A different sample of women then rated each man’s perceived traits and characteristics on 10 global mate attribute dimensions that could be preferred in long-term and/or short-term mates—each man’s inferred intelligence, kindness/ warmth, social influence (socially respected), capacity to be a good father, sexual faithfulness, capacity for financial success, physical attractiveness, muscularity, confrontativeness with other men, and arrogance. We tested two competing models. The fertile-women-favor-good-genes hypothesis anticipates that women should value and prefer ostensible markers of genetic benefits most strongly when they are fertile and evaluating men as conceivable short-term mates. The fertile-women-possess-wiser-preferences


hypothesis, in contrast, anticipates that women should strongly yet equally prefer both long-term and short-term male attributes when they are ovulating versus when they are not ovulating.

Women’ s Attractiveness Ratings Male Attribute × Mating Context Interaction Effects.

As expected, men’s perceived arrogance, confrontativeness, muscularity, and physical attractiveness predicted their attractiveness better as short-term mates than as long-term mates. Conversely, men’s perceived faithfulness, warmth, intelligence, potential to be a good father, and potential for financial success predicted men’s attractiveness better as long-term than short-term mates.

Conception Risk × Male Attribute × Mating Context Interaction Effects. Across the 10 mate attributes, we found several predicted Conception

Risk × Male Attribute × Mating Context interactions. Relative to women low in conception risk, those high in conception risk preferred as short-term mates men whom they perceived as more confrontative, arrogant, muscular, socially respected, and physically attractive. Such women were also more attracted to men perceived as less faithful as short-term mates. These effects parallel those previously reported by Penton-Voak et al. (1999), Gangestad et al. (2004), and Haselton and Miller (2006) in that women’s mate preferences predictably shift across the reproductive cycle to favor attributes that may signal a man’s “good genes,” particularly in short-term mating contexts. Similar interactions did not emerge for attributes valued highly in long-term mates—warmth, intelligence, potential to be a good father, and potential for financial success.

Characterizing the Interaction Effects The results of Study 2 indicate that women’s standards of attractiveness do not change across the ovulatory cycle for all kinds of mate attributes. Only standards associated with particular male attributes systematically change. These results support the fertile-women-favor-good-genes hypothesis. However, the “good genes” hypothesis makes an even more specific prediction about which male attributes should be most appealing to fertile women. Fertile women should be especially drawn to men who have attributes typically valued in short-term sex partners. Figure 10.2 summarizes these results. It shows that the extent to which male attributes were preferred in short-term mating contexts strongly predicted the extent to which this was especially true of fertile women. Additional analyses revealed a similar pattern of results for the dimensions of good investing mate qualities and intrasexual competitiveness (see Gangestad et al., in press). This evidence provides further support for the fertile-women-favor-good-genes hypothesis.


FIGURE 10.2 Effects of the three-way interactions between trait factor, mating context, and conception risk as a function of the effects of mating context (short-term vs. long-term) on the trait preferences across the 10 perceived mate attributes. Values on the Y-axis are t-test values.

Short-Term and Long-Term Mate Preferences Gangestad et al. (in press) also examined women’s ratings of short-term mate attractiveness and long-term mate attractiveness separately. Men’s intrasexual competitiveness strongly and positively predicted their short-term attractiveness, whereas men’s good investing mate qualities predicted their short-term attractiveness negatively and much more weakly. As anticipated, intrasexual competitiveness interacted with conception risk to predict men’s short-term attractiveness ratings. In particular, with increasing probability of fertility, women were more attracted to men perceived to have attributes signaling their greater intrasexual competitiveness. By contrast, there was no evidence that women preferred good investing mate qualities in short-term mates as a function of their fertility status, nor was there evidence that women who varied in fertility status responded differently to combinations of the two factors. Intrasexual competitiveness, good investing mate qualities, and their interaction all predicted long-term mate attractiveness, but there was no evidence that women’s attraction to these attributes in long-term mates systematically changed across the ovulatory cycle.


Summary Study 2 provides further evidence—particularly nice discriminant validity evidence—for the fertile-women-favor-good-genes hypothesis. When fertile, women are more strongly drawn to attributes that tend to be valued in shortterm mates. No parallel preference shifts were found for attributes that typically are valued more highly in long-term mates. Thus, consistent with earlier research (e.g., Gangestad et al., 2004; Johnston et al., 2001; Penton-Voak et al., 1999), we did not find general shifts in mate preferences across the reproductive cycle. Instead, mating context interacted with fertility status to predict women’s specific mate preferences. Near ovulation, traits that are likely to be “honest” signals of greater intrasexual competitiveness, physicality, attractiveness, and arrogance are viewed as especially attractive in short-term sexual partners.1

BROADER THEORETICAL CONSIDERATIONS One might wonder how SPM differs for other major models of human mating. Perhaps the most significant alternate model is Sexual Strategies Theory (SST; Buss & Schmitt, 1993), which also posits that both sexes should have evolved to engage in long-term and short-term mating strategies. There are some critical differences between the two models, however. For instance, although SST acknowledges that women should have evolved to engage in short-term sexual relations, it focuses more directly on gender differences in mating strategies. SST also does not directly apply good genes sexual selection thinking to explain why women prefer certain attributes in short-term mates, such as physical attractiveness and sex appeal (see Buss & Schmitt, 1993). Moreover, according to SST, women frequently use short-term mating to attract and evaluate men as possible long-term mates. Women engage in short-term mating, in other words, to facilitate their long-term mating goals and objectives. SPM (Gangestad & Simpson, 2000), by comparison, focuses on the nature and trade-off dynamics of variation in mating strategies within each sex. Melding good genes and good provider principles, SPM proposes that women evolved to engage in short-term mating selectively to obtain the “good genes” of certain men, independent of women’s long-term mating goals. One issue that complicates the testing of mating theories is that men who have attributes suggestive of “good genes” might have been more capable of providing more or better benefits and resources to their partners, above and beyond the genetic benefits they may have conferred. If so, men who have “good genes” might also be better or more investing long-term mates (Gangestad & Simpson, 2000). Indeed, even though the intrasexual competitiveness factor did predict women’s ratings of men’s short-term attractiveness more strongly in Study 2, the good investing mate qualities factor also independently predicted women’s short-term attractiveness ratings, albeit more weakly.


What, then, constrains women from pursuing mates who possess both desired short-term and long-term attributes? Men who score high on both dimensions tend to be highly sought after by women and, accordingly, have many good mating options, making them especially difficult to attract and retain as mates (Simpson & Gangestad, 1992). Despite this fact, many women should still find such men highly attractive, which is exactly what emerged in Study 2. That is, men perceived as having higher standing on both good investing mate qualities and intrasexual competitiveness were unusually attractive to women, more than would be expected from men’s standing on each mate choice factor considered alone. Given that such paragons of virtue are very difficult to attract and retain, women should remain open to flexible, ecologically-contingent mating strategies, selectively engaging in short-term sex with men who display evidence of “good genes” (particularly during ovulating), and preferentially engaging in long-term mating with men who display strong paternal investment (Gangestad & Simpson, 2000). These findings support basic tenets of SPM. Although SPM assumes that female extrapair mating occurred in ancestral environments, it does not assume that women engaged in extrapair mating frequently or indiscriminately. Along with potential genetic benefits, there should have been immediate and severe costs associated with engaging in extrapair mating given that established mates may have harmed or abandoned unfaithful partners. Most men, in fact, are vigilant of their partners’ whereabouts, particularly when their partners are ovulating (see Gangestad et al., 2002). Hence, despite the fact that women may be more strongly attracted to men other than their primary partners during ovulation, most women are not likely to act on these attractions. One of the most novel components of SPM is the premise that environment conditions should influence how women weigh, evaluate, and make trade-offs between male viability and male investment. According to the model, ancestral women should have placed greater weight on investment when local environments were demanding and required sustained biparental care of children. On the other hand, women should have placed more weight on viability when local environments indicated the presence of heavy pathogen load. Provisional support for these hypotheses at a cross-cultural level has been marshaled by Gangestad and Buss (1993) and more recently by Schmitt (2005). Because women’s mating decisions should also be impacted by local environmental parameters, some of the ovulatory cycle findings reported above might be moderated by variation in the need for biparental care or pathogen presence. For example, the interaction depicted in Figure 10.1 of Study 1 could be qualified by the severity of pathogen load in the local environment. When pathogen load is particularly high, women might place even greater emphasis on male viability, amplifying the appeal of certain short-term behavioral cues especially when women are ovulating. When pathogen load is lower, however, women may put less weight on male viability, perhaps attenuating the effect reported in Figure 10.1. The crucial point is that the local ecology may amplify


or attenuate some of the “good genes” interaction effects described in this chapter. In conclusion, the studies reviewed in this chapter contribute to the rapidly growing literature on ovulatory cycle effects. One of the most unique aspects of the current set of findings is their specificity coupled with the fact they cannot be easily derived from theories or models other than those suggesting that good genes sexual selection operated on humans. It is important to reiterate that just because women might have benefited from selective extrapair mating with certain men, many women in evolutionary history probably did not do so on a regular basis. Such matings should have been fairly infrequent and opportunistic, occurring when assorted circumstances—the lower viability of one’s current partner, the stellar short-term attributes of select extrapair or short-term partners, the discreetness and confidentiality of the liaison—all coalesced. Nevertheless, current empirical evidence now suggests that both sexes probably evolved to engage in conditional mating strategies, contingent on both personal factors (e.g., one’s attractiveness, social status, resources, and ovulatory status in the case of women) and local environmental conditions (e.g., cues indicating the need for prolonged biparental care, cues signaling the need to mate with partners who have weathered pathogen-prevalent environments). Rather than adopting sex-linked or invariant mating strategies, men and women evolved to be strategic pluralists, adopting ecologically-contingent mating strategies that, on average, enhanced their inclusive fitness in response to varied, fluctuating, and sometimes uncertain physical and social environments.

NOTE 1. Although information processing was not the focus of this work, important outcomes of cognitive processing—including attribute inferences of men on different mating dimensions and evaluations of men’s appeal as short-term and long-term mates—were

examined. Future work needs to investigate how women reason when making trait inferences about certain men and how these judgments inform women’s evaluation of men’s attractiveness as short-term and long-term mates (see Lieberman, 2006).

REFERENCES Alexander, R. D., Hoogland, J. L., Howard, R. D., Noonan, K. M., & Sherman, P. W. (1979). Sexual dimorphism and breeding systems in pinnipeds, ungulates, primates and humans. In N. A. Chagnon & W. Irons (Eds.), Evolutionary biology and human social behavior: An anthropological perspective. North Scituate, MA: Duxbury.

Andersson, M. B. (1994). Sexual selection. Princeton, NJ: Princeton University Press. Andrews, P. W., Gangestad, S. W., & Matthews, D. (2002). Adaptationism: How to carry out an exaptationist program. Behavioral and Brain Sciences, 25, 489–553. Buss, D. M., & Schmitt, D. P. (1993). Sexual strategies theory: An evolutionary perspective

176 EVOLUTION AND THE SOCIAL MIND on human mating. Psychological Review, 100, 204–232. Cronin, H. (1991). The ant and the peacock. Cambridge, UK: Cambridge University Press. Cunningham, M. R., Druen, P. B., & Barbee, A. P. (1997). Angels, mentors, and friends: Trade-offs among evolutionary, social, individual variables in physical appearance. In J. A. Simpson & D. T. Kenrick (Eds.), Evolutionary social psychology (pp. 109–140). Mahwah, NJ: Lawrence Erlbaum Associates, Inc. Fessler, D. M. T. (2003). No time to eat: An adaptationist account of periovulatory behavioral changes. Quarterly Review of Biology, 78, 3–21 Folstad, I., & Karter, A. J. (1992). Parasites, bright males, and the immunocompetence handicap. The American Naturalist, 139, 603–622. Gangestad, S. W., & Buss, D. M. (1993). Pathogen prevalence and human mate preferences. Ethology and Sociobiology, 14, 89–96. Gangestad, S. W., Garver-Apgar, C. E., Simpson, J. A., & Cousins, A. J. (in press). Changes in women’s mate preferences across the ovulatory cycle. Journal of Personality and Social Psychology. Gangestad, S. W., & Simpson, J. A. (2000). The evolution of human mating: The role of trade-offs and strategic pluralism. Behavioral and Brain Sciences, 23, 573–587. Gangestad, S. W., Simpson, J. A., Cousins, A. J., Garver-Apgar, C. E., & Christensen, P. N. (2004). Women’s preferences for male behavioral displays change across the menstrual cycle. Psychological Science, 15, 203–207. Gangestad, S. W., & Thornhill, R. (1998). Menstrual cycle variation in women’s preference for the scent of symmetrical men. Proceedings of the Royal Society of London, Series B, 262, 727–733. Gangestad, S. W., Thornhill, R., & Garver, C. E. (2002). Changes in women’s sexual interests and their partners’ mate retention tactics across the menstrual cycle: Evidence for shifting conflicts of interest. Proceedings of the Royal Society of London, Series B, 269, 975–982. Gangestad, S. W., Thornhill, R., & GarverApgar, C. E. (2005). Adaptations to

ovulation. In D. M. Buss (Ed.), The handbook of evolutionary psychology (pp. 68–95). Hoboken, NJ: Wiley. Grafen, A. (1990). Biological signals as handicaps. Journal of Theoretical Biology, 144, 517–546. Graziano, W. G., Jensen-Campbell, L. A., Todd, M., & Finch, J. (1997). Interpersonal attraction from an evolutionary perspective: Women’s reactions to dominant and prosocial men. In J. A. Simpson & D. T. Kenrick (Eds.), Evolutionary social psychology (pp. 141–167). Mahwah, NJ: Lawrence Erlbaum Associates, Inc. Gross, M. R. (1996). Alternative reproductive strategies and tactics: Diversity within sexes. Trends in Ecology and Evolution, 11, 92–98. Gross, M. R., & Sargent, R. C. (1985). The evolution of male and female parental care in fishes. The American Zoologist, 25, 807–822. Hamilton, W. D., & Zuk, M. (1982). Heritable true fitness and bright birds: A role for parasites. Science, 218, 384–387. Haselton, M. G., & Miller, G. F. (2006). Women’s fertility across the cycle increases the short-term attractiveness of creative intelligence. Human Nature, 18, 50–73. Johnston, V. S., Hagel, R., Franklin, M., Fink, B., & Grammer, K. (2001). Male facial attractiveness: Evidence for hormone medicated adaptive design. Evolution and Human Behavior, 21, 251–267. Jones, B. C., Little, A. C., Penton-Voak, I. S., Tiddeman, B. P., Burt, D. M., & Perrett, D. I. (2001). Facial symmetry and judgments of apparent health: Support for a “good genes” explanation of the attractiveness-symmetry relationship. Evolution and Human Behavior, 22, 417–429. Kirkpatrick, M. (1996). Good genes and direct selection in the evolution of mating preferences. Evolution, 50, 2125–2140. Li, N. P., Bailey, J. M., Kenrick, D. T., & Linsenmeier, J. A. W. (2002). The necessities and luxuries of mate preferences: Testing the trade-offs. Journal of Personality and Social Psychology, 82, 947–955. Lieberman, D. (2006). Mate selection: Adaptive problems and evolved cognitive programs. In P. Noller & J. A. Feeney (Eds.), Close relationships: Functions, forms, and processes (pp. 245–266). New York: Psychology Press.


Møller, A. P. (1994). Sexual selection in the barn swallow. Cambridge, UK: Cambridge University Press. Oliver, M. B., & Hyde, J. S. (1996). Gender differences in sexuality: A meta-analysis. Psychological Bulletin, 114, 29–51. Parker, G. A., & Maynard Smith, J. (1991). Optimality theory in evolutionary biology. Nature, 348, 27–33. Penton-Voak, I. S., & Perrett, D. I. (2000). Female preference for male faces changes cyclically: Further evidence. Evolution and Human Behavior, 21, 39–48. Penton-Voak, I. S., Perrett, D. I., Castles, D., Burt, M., Koyabashi, T., & Murray, L. K. (1999). Female preference for male faces changes cyclically. Nature, 399, 741– 742. Pomiankowski, A., Iwasa, Y., & Nee, S. (1991). The evolution of costly mate preferences: I. Fisher and biased mutation. Evolution, 45, 1422–1430. Puts, D. (2005). Mating context and menstrual phase affect women’s preferences for male voice pitch. Evolution and Human Behavior, 26, 388–397. Rikowski, A., & Grammer, K. (1999). Human body odour, symmetry and attractiveness. Proceedings of the Royal Society of London, Series B, 266, 869–874. Salmon, W. C. (1966). The foundations of scientific inference. Pittsburgh, PA: University of Pittsburgh Press. Schmitt, D. P. (2005). Sociosexuality from Argentina to Zimbabwe: A 48-nation study of sex, culture, and strategies of human mating. Behavioral and Brain Sciences, 28, 247–311. Simpson, J. A., & Campbell, L. (2005). Methods of evolutionary sciences. In D. M. Buss (Ed.), The handbook of evolutionary psychology (pp. 119–144). Hoboken, NJ: Wiley.

Simpson, J. A., & Gangestad, S. (1991). Individual differences in sociosexuality: Evidence for convergent and discriminant validity. Journal of Personality and Social Psychology, 60, 870–883. Simpson, J. A., & Gangestad, S. W. (1992). Sociosexuality and romantic partner choice. Journal of Personality, 60, 31–51. Simpson, J. A., Gangestad, S. W., Christensen, P. N., & Leck, K. (1999). Fluctuating asymmetry, sociosexuality, and intrasexual competitive tactics. Journal of Personality and Social Psychology, 76, 159–172. Thornhill, R., Gangestad, S. W., Miller, R., Scheyd, G., McCollough, J. K., & Franklin, M. (2003). Major histocompatibility complex genes, symmetry, and body scent attractiveness in men and women. Behavioral Ecology, 14, 668–678. Tooby, J., & Cosmides, L. (1992). The psychological foundations of culture. In J. Barkow, L. Cosmides, & J. Tooby (Eds.), The adapted mind: Evolutionary psychology and the generation of culture (pp. 19–136). New York: Oxford University Press. Trivers, R. (1972). Parental investment and sexual selection. In B. Campbell (Eds.), Sexual selection and the descent of man, 1871–1971 (pp. 136–179). Chicago: Aldine. Trivers, R. (1985). Social evolution. Menlo Park, CA: Benjamin/Cummings. Williams, G. C. (1966). Adaptation and natural selection. Princeton, NJ: Princeton University Press. Wilson, E. O. (1978). On human nature. Cambridge, MA: Harvard University Press. Zahavi, A. (1975). Mate selection: A selection for a handicap. Journal of Theoretical Biology, 53, 205–214. Zuk, M., Thornhill, R., Ligon, J. D., & Johnson, K. (1990). Parasites and mate choice in red jungle fowl. The American Zoologist, 30, 235–244.


Aligning Evolutionary Psychology and Social Cognition

Inbreeding Avoidance as an Example of Investigations into Categorization, Decision Rules, and Emotions DEBRA LIEBERMAN

Introduction What is a Computational Theory of Mind and Why Should SocialCognitive Scientists Care? Selection Pressures Guiding the Evolution of Inbreeding Avoidance Mechanisms An Information-Processing View of Inbreeding Avoidance: What Would a Well-Engineered System for Inbreeding Avoidance Look Like? Empirical Investigation of Systems for Inbreeding Avoidance Conclusion

INTRODUCTION s an evolutionary psychologist employing theoretical tools from biology and cognitive science to study human social behavior, I often find I speak a completely different language from those trained in the field of social cognition, despite the similar goals of understanding human sociality. There are (at least) two reasons why translation between these two frameworks has been difficult. First, research in social cognition typically has not considered




important theoretical contributions from evolutionary biology (e.g., kin selection, parental investment), principles known to organize cognitive processes and behavior in nonhuman animals. For example, despite the fact that humans likely evolved in small kin-based groups and that much of our social interactions would have been with kin of varying degrees, kinship has been an area largely neglected within social cognition (Daly, Salmon, & Wilson, 1997). One area of research that has overlooked the importance of kinship is social categorization and person perception. Researchers interested in the dimensions into which we categorize social targets have focused on the big three—sex, age, and race—yet ignored kinship. Kinship, however, is an important social dimension organizing a variety of different individual and group-level behaviors. For example, tracking kin relations would have enabled inferences such as who is likely to come to the aid of another, who is not likely to be a competitor for a particular mate, or who is likely to band together and form a collective action, among other things. Indeed, recent empirical findings indicate that kinship, much like age and sex, is a dimension implicitly encoded by our psychological architecture (Lieberman, Oum, & Kurzban, 2006). A second reason for the language divide is that social-cognitive psychologists have not fully adopted a computational theory of mind—a tool that has proved powerful in other areas of cognitive research and, as I hope to demonstrate, in evolutionary approaches to understanding human behavior. I certainly do not believe that social cognition is the only field in need of updating. Evolutionary-minded researchers would do well to integrate the robust methodologies developed and debugged by social psychologists. This would allow hypotheses derived from an evolutionary-computational framework to be better tested and then more easily shared across academic boundaries (von Hippel, Haselton, & Forgas, chapter 1, this volume). Evolutionary psychologists should also take note of the phenomena social cognitive psychologists have documented and ask what clues they provide about our species’ evolution. Nevertheless, I do see the theoretical contributions flowing primarily from evolutionary science. If evolution can be used to understand the behavior of every other living organism, surely it can be used to understand aspects of human cognition, social or otherwise (Darwin, 1859/2003). This chapter aims to address these two communication roadblocks separating the fields of evolutionary psychology and social cognition. Specifically, my goal in this chapter is to (1) demonstrate how evolutionary principles can be used to study human social cognition and behavior using categorization and decisionmaking processes relating to inbreeding avoidance as an example and, in the process, (2) describe what an evolutionary-computational approach to social cognition might look like. In the next section, I briefly discuss what is meant by an evolutionary-computational framework and why it can be of value to the field of social cognition. Then, using my own research on inbreeding avoidance as an example, I walk through the logic of what an evolutionary-computational analysis entails. I start with a review of the evolutionary reasons (i.e., selection


pressures) why inbreeding avoidance mechanisms are hypothesized to exist. Following this, I outline the kinds of information processing procedures (i.e., computational components such as decision rules and categorization processes) required to achieve inbreeding avoidance. In particular, I focus on: (1) how we categorize others according to kinship using cues available in the social world (see Gangestad & Thornhill, chapter 3, this volume for a discussion of cues and signals), and (2) the emotion of disgust and its role in motivating the sexual avoidance of family members. Last, I share some recent empirical findings and make suggestions for the continued investigation of kinship cues and the programs regulating kin-directed behaviors.

WHAT IS A COMPUTATIONAL THEORY OF MIND AND WHY SHOULD SOCIAL-COGNITIVE SCIENTISTS CARE? A computational theory of mind (CTM) conceives of the mind as an information processing system (Fodor, 1983). According to Barrett (2005), “CTM views thought as computation: the use of algorithmic rules to systematically map inputs, i.e., information instantiated in neurochemical patterns, onto outputs, i.e. different patterns of information that have been systematically transformed. What makes the patterns in question information is that they ‘stand for’ something: that they, in turn, can be mapped onto something in the world or mind. The mapping operations are computations” (p. 260). This view of the mind allows for the generation of highly detailed models of the cognitive architecture underlying a specific behavior (e.g., the cues/information taken as input, the procedures that transform this information, and the outputs stored or behavioral sequences activated; see Barrett & Kurzban (2006) for a thorough discussion of modularity, a computational view of the mind, and evolution). An evolutionary approach to the CTM is important because computational systems are highly improbable orderings of matter and their existence requires explanation. The only known causal force likely to generate such complex functional design is natural selection (Tooby, Cosmides, & Barrett, 2003). Therefore, an understanding of the selective forces that shaped our computational architecture (e.g., why we take certain aspects of the environment as input, why certain outputs are generated over others, why given outputs effect behavior in the manner they do) is essential in the investigation of human social cognition. Without an evolutionary toolbox, computational analyses and neurological investigations can become haphazard and led astray by our folk intuitions, intuitions that might not map directly onto our cognitive architecture. Given their interests in social perception and decision making, it is surprising that socialcognitive psychologists have not taken advantage of these tools to discover the nature of our evolved perceptual systems and the kinds of decision-making processes evolution shaped to direct social behavior. To help bridge the gap between evolutionary approaches to psychology and



the field of social cognition, I use the topic of my own research, kin detection and inbreeding avoidance in humans, as an example of how social perception and decision making can be investigated using an evolutionary-computational framework. I start with an analysis of the function of inbreeding avoidance, that is, why evolution is hypothesized to have selected for information-processing systems motivating the sexual avoidance of close relatives. Then, I outline the computational architecture of what a well-designed system for inbreeding avoidance might look like. First: Why is inbreeding bad?

SELECTION PRESSURES GUIDING THE EVOLUTION OF INBREEDING AVOIDANCE MECHANISMS There are sound biological reasons why psychological mechanisms designed to avoid mating with a close genetic relative are expected to exist. Throughout our species’ evolutionary history, the selection pressures posed by deleterious recessive mutations (e.g., Bittles & Neel, 1994) and short-generation pathogens (e.g., Tooby, 1982) would have severely negatively impacted the health and viability of offspring of individuals who were close genetic relatives. As a result, individuals who avoided mating with close genetic relatives and, instead, mated with someone who did not share an immediate common ancestor would have enjoyed greater reproductive success. What follows is a brief discussion of each selection pressure.

Deleterious Recessive Mutations Because humans are a diploid species, it is possible for harmful mutations to “hide out” in the genome. This is because a fully functional gene inherited from one parent can mask the effects of a dysfunctional and potentially harmful gene inherited from the other. That is, when the functional gene is dominant an individual can display a normal phenotype despite the presence of a recessive deleterious version of the gene. Consequently, deleterious recessives can accumulate in the population until they reach relatively high frequencies. For example, a given deleterious allele, if dominant, might exist in a population at a frequency of 1 in 1,000,000. This same allele, if recessive, would approach a frequency of 1 in 1000—that is, 1000 times more frequent. The negative consequences of inbreeding occur when the same recessive damaged allele is supplied from both the mother and the father to the resulting offspring. In this case, the deleterious recessive genes are expressed decreasing the health and viability of the individual. According to Bittles and Neel (1994) “all of us are thought to carry in the heterozygous condition ‘several’ rare recessive genes which, if rendered homozygous, would result in a significant medical handicap, ranging from severe defects of vision and hearing to disorders incompatible with survival beyond childhood” (p. 17). The estimated number of


rare lethal genes in a genome is termed lethal equivalents (Cavalli-Sforza & Bodmer, 1971; Crow & Kimura, 1970). Data from a number of studies suggest that each of us possess, on average, somewhere between two (Bittles & Neel, 1994; Carter, 1967; May, 1979) to six (Kumar, Pai, & Swaminathan, 1967) lethal equivalents: alleles that, if homozygous, would cause death before an individual reached reproductive age (Burnham, 1975; Morton, Crow, & Muller, 1956). We are not dead many times over because at the great majority of these loci, we are heterozygous, and the damaging gene is masked by an intact gene. The effects of inbreeding should be apparent: If two close genetic relatives mate with one another, versus mating with an individual who doesn’t share an immediate common ancestor, there is a greatly increased chance that the resulting offspring will be homozygous for many deleterious recessives, leading to a decreased chance of survival and reproduction. The more closely related the parents, the greater the likelihood the offspring will suffer a decrease in health and viability, and the selection pressures become very intense whenever two parents are siblings, or parent and child. For this reason, deleterious recessive mutations posed a strong selection pressure against close-kin matings and would have led to the evolution of mechanisms reducing the probability such matings occurred.

Pathogens A second selection pressure that would have led to the evolution of incest avoidance mechanisms is pathogens (O’Brien, Roelke, & Marker, 1985; Tooby, 1982). The presence of disease-causing agents, such as viruses and bacteria, in and around an organism’s body was a constant feature of our evolutionary past. Due to their short generation time, pathogens have the ability to become finely tuned to the biochemistry of their host. The better adapted a pathogen is to its host’s microenvironment, the more efficient it becomes at acquiring the necessary resources, evading cells of the immune system, and replicating. As a consequence, they can become extremely detrimental to the health of the host. The recurrent presence of pathogens in our ancestral environments would have created intense selection pressures for genetic diversity between individuals in a population (Tooby, 1982). This is because the more genetically homogenous the sequence of hosts encountered by a parasitic lineage, the faster an infection is able to spread. Moreover, this selection pressure would have been especially severe the longer-lived the host species—and, compared to the average gardenvariety bacterium, humans are very long lived. As a result it is hypothesized that natural selection would have engineered a solution to maintain genetic diversity. From an evolutionary point of view, the function of sexual reproduction is to introduce genetic variability into offspring sets, and to make organisms genetically different from their neighbors (Ebert & Hamilton 1996; Hamilton, Axelrod, & Tanese, 1990; Tooby, 1982). During the process of reproduction, pathogens are transmitted from parent to offspring. The presence of a unique



internal environment in the offspring renders pathogens that were well adapted to a parental internal environment less suited to the offspring’s novel environment. Mating with a close relative, then, as opposed to a nongenetically related individual, maintains a more similar microenvironment for pathogens, a condition favoring the evolution and spread of more harmful pathogen strains. To the extent that incestuous matings led to an increased genetic uniformity in ancestral hunter-gatherer groups, then increased parasite load would have been a second, significant factor selecting against potentially fertile incestuous matings. In summary, given the selection pressures posed by deleterious recessive mutations and short-generation pathogens, evolution is hypothesized to have selected for reliably developing neural circuitry that was well-engineered for decreasing the probability of close-kin matings. The question is, what would a system designed for avoiding sexual contact (not necessarily all contact) with close genetic relatives look like? This topic is taken up in the next section.

AN INFORMATION-PROCESSING VIEW OF INBREEDING AVOIDANCE: WHAT WOULD A WELL-ENGINEERED SYSTEM FOR INBREEDING AVOIDANCE LOOK LIKE? A useful tool for exploring our cognitive architecture is to take the perspective of an engineer and describe the kinds of information processing programs required to perform a specific function. With respect to inbreeding avoidance, what kinds of programs would be required? This question amounts to a description of the computational procedures governing inbreeding avoidance in humans and at first glance at least two kinds of procedures would be needed: (1) procedures for categorizing individuals in the social environment according to genetic relatedness (i.e., kin detection), and (2) procedures that take as input information regarding the relatedness of another individual and regulate sexual attraction/ avoidance accordingly (see Lieberman, Tooby, & Cosmides, 2003, for a discussion of this model). These two procedures are discussed in turn.

Procedures for Categorizing Others by Genetic Relatedness For an inbreeding avoidance system to be functional there must exist mechanisms for discovering who is likely to be a close genetic relative. Categorization along the dimension of genetic relatedness requires the existence of cues that correlated with relatedness over our species’ evolutionary history. There are a number of possible cues kin detection systems might have been designed to take as input. One potential source of information regarding kinship is linguistic and cultural input (e.g., during development you are told who counts as a close genetic relative and how to feel about them). However, these prove problematic since: (1) Kin terms can be used across genetic boundaries blurring the


distinction between types of close genetic relatives and between kin and non-kin (e.g., the term “aunt” or “brother” in the US), (2) there exist asymmetries in relatedness and thus, individuals may not share common “interests” regarding, for example, whom to help and when (e.g., a woman with children of different paternity is motivated to treat each one equally, whereas a particular child is more likely to want to help their full sibling over their half sibling leading to conflicts of interest between parent and offspring; see Trivers, 1974), and (3) systems for categorizing others according to genetic relatedness exist in many other animal species and predate the evolution of language and culture (Hepper, 1991). For these reasons, it is unlikely that evolution used linguistic information as anchorpoints for assessing relatedness. Though kin terms do correlate with relatedness (Jones, 2004) and individuals do show greater altruism toward those who share common names (Oates & Wilson, 2002), it is more likely that evolved systems for categorizing kin pattern linguistic terms rather than vice versa.

Ecologically Valid Cues to Genetic Relatedness Natural selection, rather than relying on linguistic and cultural input, is hypothesized to have shaped kin categorization mechanisms to take advantage of cues that reliably correlated with genetic relatedness in the ancestral past. To the extent that different cues reliably correlated with an individual being a particular type of close genetic relative (e.g., mother, father, offspring, or sibling), different categorization mechanisms are expected to exist. For example, because ancestrally a female always gave birth to her own offspring, she could have relied on the process of birth and/or the visual and olfactory cues derived from a newborn to reliably and accurately categorize that child as a close genetic relative (e.g., Porter, Matochik, & Makin, 1983, 1984). However, due to the fact that males of our species could not be 100% certain of their paternity, seeing one’s mate give birth to an offspring would not have solved the problem of assessing degree of relatedness to that offspring. Rather, for males, assessments of paternity might rely on cues signaling the sexual fidelity of their mate. Therefore, there may not be a general kin detection mechanism that relies on the same set of information for detecting all types of close genetic relatives. Instead, the advantages of kin selection would accrue most strongly to individuals that possessed specialized detection systems capable of narrowing in on the small subset of states that correlated with an individual being a particular kind of kin. These states may not be signals in the sense that they function to communicate information regarding kinship (see Gangestad & Thornhill, chapter 3, this volume). Rather, they may be stable social arrangements that came into existence due to adaptations serving different functions (e.g., parental care) or as by-products of adaptations that can then be used as anchor-points for kinship categorization. The following discussion focuses on the cues used to detect a particular class of kin, siblings.



Cues to Siblingship: Coresidence Duration and Exposure to Early Maternal Care What cues could evolution have used for categorizing an individual as a sibling? The most likely candidates are cues that reliably carved siblings from, for example, cousins and other individuals that would not have posed as great a threat to reproductive success. Categorization errors would have been costly in two different ways (e.g., see Haselton & Buss, 2000). Using a cue that cast its net too widely and included not only siblings but other, more distantly related kin and non-kin would have excluded potential mating partners. On the other hand, restricting sibling categorization to those meeting too stringent criteria may have excluded actual genetic relatives leading to an increased chance of choosing a genetic relative as a mate and producing offspring that suffered from inbreeding depression. Two cues that appear to walk this fine line are childhood coresidence duration and exposure to one’s mother caring for an infant. In 1891, Edward Westermarck, a Finnish social scientist, made the commonplace observation that siblings rarely find one another sexually attractive. He proposed that the early childhood association, a pattern typical among siblings, serves as a cue to relatedness and leads to the development of a sexual aversion later during adulthood (Westermarck, 1891/1921). This has come to be known as the Westermarck Hypothesis. Specifically, Westermarck stated,

Coresidence Duration.

Generally speaking, there is a remarkable absence of erotic feelings between persons living very closely together from childhood. Nay more, in this, as in many other cases, sexual indifference is combined with the positive feeling of aversion when the act is thought of. . . . Persons who have been living together from childhood are as a rule near relatives. Hence their aversion to sexual relations with one another displays itself in custom and law as a prohibition of intercourse between near kin. (Westermarck, 1891/1921, p. 192)

The cue of coresidence duration is plausible considering our evolutionary history. The nutritional demands of breastfeeding along with the need for protection would have meant that children of the same mother were typically reared in close proximity during early childhood. Also, when hunter-gatherer bands fissioned into smaller units (e.g., due to size or difficult times), nuclear families (including siblings) would have stayed together as a unit (Chagnon, 1992; Lee & DeVore, 1968). This means that in ancestral environments, early childhood would have offered valuable information regarding the relatedness of individuals in prolonged close association. A number of researchers have tested the Westermarck Hypothesis (see, e.g., Bevc & Silverman, 1993, 2000; Fessler & Navarrete, 2004; Lieberman et al., 2003; Shepher, 1971, 1983; Williams & Finkelhor, 1995; Wolf, 1995). For some, the focus of research has been testing the Westermarck Hypothesis in


populations where genetically unrelated individuals were reared together as siblings. Most notable are the anthropological reports on Israeli Kibbutzim (Shepher, 1971, 1983; Spiro, 1958; Talmon, 1964) and on the Taiwanese minor form of marriage (Wolf, 1995). In these two cases cultural institutions inadvertently created a “natural experiment” where children who were not genetically related to one another were reared together from very early childhood. As the Westermarck Hypothesis predicts, lower rates of marriage and sexual interest were found (for the peer groups in Israeli Kibbutzim; Shepher, 1983) as well as greater rates of divorce and extramarital affairs and lower rates of fertility (for Taiwanese marriages where the bride was adopted into her husband’s family as a child; Wolf, 1995). These cross-cultural studies provide support for the hypothesis that coresidence duration serves as a cue to relatedness. Empirical investigations that have gone beyond sociological measures and sought responses from actual living individuals have also found that longer, uninterrupted periods of childhood coresidence are associated with greater disgust at sexual behavior (Lieberman, Tooby, & Cosmides, in press), a reduced probability of engaging in sexual behavior (Bevc & Silverman, 1993, 2000), and greater moral opposition to sibling incest (Fessler & Naverrete, 2004; Lieberman et al., 2003). However, coresidence duration may not be the best cue available for detecting siblings and, further, may not be used as cue for detecting other categories of genetic relatives (e.g., see Williams & Finkelhor, 1995). Exposure to Early Maternal Care. Though coresidence would have done a good job carving the social world into siblings versus other kin and non-kin, there may have been an even better cue. The stable association between mother and newborn that exists due to the demands of breastfeeding and care provides a reliable cue from which inferences regarding relatedness may be generated. Seeing one’s mother (i.e., the female from whom one breastfed) breastfeeding another infant would have meant, under ancestral conditions, that the infant in question was at least a half sibling (how cues to shared paternity are assessed is a good question and currently under exploration). Moreover, this cue would have been valid independent of one’s age. That is, no matter whether one is 3, 13, or 23, seeing one’s mother breastfeeding another child is a good cue to relatedness. However, this information would only have been accessible to older siblings. For younger siblings, the arrow of time prevents them from seeing their older sibling breastfed. Thus, for younger siblings, the best cue available for assessing relatedness might very well be coresidence duration. This suggests different decision rules might be used to assess siblingship. Indeed, in a recent study by Lieberman et al. (in press), coresidence duration appears to be used as a cue to kinship only when information regarding maternal care during infancy (e.g., breastfeeding information) is absent. The above discussion shows how an evolutionary analysis can help narrow the sets of cues or information evolution is likely to have used to solve the



recurring problem of inbreeding depression. Coresidence duration and exposure to early maternal care are two cues, among others, that are hypothesized to govern the categorization of individuals in the social environment into kin versus non-kin. However, this analysis represents the front end of the system. To effectively avoid inbreeding, kin detection systems need to hook into decision-making procedures regulating sexual motivation. This is the topic of the next section.

Procedures for Regulating Sexual Avoidance: The Emotion of Disgust In addition to procedures for categorizing individuals in the social environment by genetic relatedness, procedures for regulating sexual attraction/avoidance are required to prevent inbreeding. What would a well-designed system for motivating sexual avoidance look like? Such a system should have a number of properties. For example, it should: (i) be efficient at motivating sexual avoidance; (ii) associate sexual aversions with particular individuals in the social environment based on cues to kinship; and (iii) be able to output different intensities to match the different probabilities individuals have of being a close genetic relative and the different types of kin that exist (e.g., cousins and siblings). Each property is discussed in turn. The Motivation of Sexual Avoidance. Programs that were simply indifferent to sexual relations with close genetic relatives would not have solved the problem of inbreeding avoidance because family members can have strong sexual desires and motivations to mate that would not be strongly deterred by a disinterested disposition (e.g., I can be sexually disinterested in a chair, but the chair, unlike an animate being, does not have intentions of its own and thus is unlikely to pursue me sexually). The situation of family members finding one another sexual attractive may arise for at least two reasons: (1) Cues to kinship may not be similar for each individual within a dyad (e.g., the cues fathers use to assess paternity may be different from the cues a child uses to assess who their father is) leading to the possibility that only one person in a dyad has categorized the other as kin and developed a sexual aversion, and (2) depending on the pathogen load of the environment and available mates, incest may have paid as a mating strategy but asymmetrically so for the sexes (e.g., incest might have been beneficial under certain circumstances to the father but still prohibitively costly to the daughter; see Haig, 1999; Tooby, 1977). For these reasons, a program that actively motivated sexual avoidance would have out-competed one that was simply disinterested in sex with family members.

Kinship is a dimension that needs to be discovered anew for each individual generation after generation. For this reason, the program regulating sexual avoidance must be flexible and easily assigned based on the cues of kinship. That is, a sexual avoidance program

The Association to Specific Individuals.


should accept as input representations of any individual displaying cues to kinship, information that cannot be specified in advance. The intensity of sexual avoidance should be a function of genetic relatedness and the consequences of inbreeding. For example, the intensity of sexual avoidance toward a cousin should be less than the intensity toward a sibling and the intensity of avoidance toward a half sibling should be less than the intensity toward a full sibling. Therefore, rather than having a simple on/off switch, a well-designed program motivating sexual avoidance should output a graded response based on estimates of kinship. Is there a program that meets the above criteria and appears well suited to perform the function of inbreeding avoidance? Yes—disgust. This is hardly a new idea as Westermarck himself identified disgust as the response to incest (as would any thought experiment involving sex with a parent, for example). Further, Westermarck identifies disgust as governing not only incest but other costly sexual acts.

Variation in Intensity.

The objection will perhaps be made that the aversion to sexual intercourse between persons living very closely together from early youth is too complicated a mental phenomenon to be a true instinct, acquired through spontaneous variations intensified by natural selection. But there are instincts just as complicated as this feeling, which, in fact, only implies that disgust is associated with idea of sexual intercourse between persons who have lived in a long-continued, intimate relationship from a period of life at which the action of desire is naturally out of the question. This association is not matter of course, and certainly cannot be explained by the mere liking for novelty. It has all the characteristics of a real, powerful instinct, and bears evidently a close resemblance to the aversion to sexual intercourse with individuals belonging to another species. (Westermarck, 1891/1921, p. 353)

It has been hypothesized that the original function of disgust is to avoid the oral incorporation of various harmful substances (see, for example, Ekman & Davidson, 1994; Izard, 1993; Rozin & Fallon, 1987). Perhaps more specifically, the emotion of disgust evolved to inhibit the ingestion and contact with substances associated with disease-causing agents (e.g., feces, dead organisms, and spoiled food; Curtis & Biran, 2001; see also Schaller, chapter 18, this volume). Disgust could have been co-opted during human evolution to motivate the withdrawal from sexual relations with a close genetic relative (as well as other sexual partners imposing a cost on one’s reproductive success). The characteristic trait of disgust to motivate avoidance means that it can be mobilized to deter an unsolicited advance by a close family member. Moreover, it can also function to counteract any sexual desire that may arise due to the fact that one’s close genetic relatives may be an attractive member of the opposite sex and possess traits (including accessibility) that feed into sexual attraction systems. Furthermore, disgust varies in intensity and can be associated with novel stimuli (see Haidt & Bjorklund, in press) making it a good solution for an inbreeding avoidance system. Last,



disgust may have been relatively easy to coopt for this new function of sexual avoidance since it is already associated with sexual behavior: Disgust needs to be down-regulated before intimate contact (i.e., contact exposing one to the pathogens of another) can occur (Angyal, 1941; as an aside, the ratcheting down of disgust may explain the occurrence of some interesting sexual fetishes). If, instead of down-regulating disgust, a mutation caused its up-regulation in response to sexual behavior with specific individuals, inbreeding avoidance would have gained a foothold and slowly, over evolutionary time, become refined into a well-functioning inbreeding avoidance system. This section illustrated that inbreeding avoidance requires both categorization procedures that assess whether an individual has a probability of being a close genetic relative, and decision rules regulating sexual motivations. The emotion we term disgust is hypothesized to describe, in part, the procedures that take kinship information as input and adjust sexual motivations according to the costs (and benefits) such behavior had in ancestral environments. Disgust, then, is a type of cognitive (i.e., information processing) program (see Oum & Lieberman, in press, for a detailed discussion of emotions as cognitive programs). The model of inbreeding avoidance described above is a simplified version of what has been recently developed (see Lieberman et al., in press). But even this simple model has greatly aided empirical investigations of inbreeding avoidance systems in humans.

EMPIRICAL INVESTIGATION OF SYSTEMS FOR INBREEDING AVOIDANCE The model of a human inbreeding avoidance system proposed herein provides an empirical framework within which information hypothesized to serve as cues to relatedness can be tested. The magnitude of the sexual aversion (or attraction) associated with a particular individual should be a function of the exposure to cues correlating with genetic relatedness in our ancestral past. So, for example, longer durations of childhood coresidence should translate into greater sexual aversions toward a sibling. It is therefore possible to reverse engineer the kinds of cues used to detect each type of close genetic relative. This can be done by quantitatively matching individual variation in opposition to incest (i.e., sexual disgust) to individual variation in parameters that may have served as cues to relatedness (e.g., coresidence or maternal perinatal association). Recently, researchers have employed this method to investigate the nature of the cues our mind uses to identify siblings (e.g., DeBruine, 2002; Fessler & Navarrete, 2004; Lieberman, 2003; Lieberman et al., 2003; in press). Converging lines of evidence for the cues used to categorize individuals as different types of kin can be found by exploring the domain of kin-directed altruism (see Lieberman et al., in press). Kin selection, like inbreeding avoidance,


requires procedures for categorizing kin. To the extent that the same procedures for categorizing kin are used in both domains, cues signaling relatedness are hypothesized to regulate these disparate systems in parallel. Therefore, strong evidence that a particular cue is used to detect siblings would be if variations in this cue predict both sexual aversions and altruistic motivations toward that sibling. Using this logic, Lieberman et al. (in press) have found that coresidence duration and exposure to one’s mother caring for a newborn are two cues to estimate siblingship. That is, both cues were found to regulate sexual opposition to sibling incest and separately, motivations to help a sibling. Other cues also might be used to assess siblingship including physical resemblance (e.g., DeBruine, 2002; Park & Schaller, 2005) and olfactory cues derived from the catabolism of elements of our immune system (e.g., scents derived from the major histocompatibility complex; Wedekind, Seebeck, Bettens, & Paepke, 1995). In addition to the exploration of kinship cues, more research is needed on the structure of the emotion programs regulating sexual avoidance (for a discussion on evolution and emotions see Tooby & Cosmides, 1990, 2000). For example, does sexual disgust activate the same set of physiological and psychological features (e.g., components of the immune system and memory for substances ingested) as pathogen-related disgust? Are these systems neurally dissociable? Are there neurological conditions that impair disgust in one domain but not the other (e.g., work on Huntington’s disease suggests that individuals are impaired in their ability to detect facial expressions of disgust [Sprengelmeyer et al., 1996]—whether sexual disgust is impaired to the same extent as pathogenrelated disgust remains an open question). Just as an engineering perspective can provide a guide-rail for exploring the organization of systems for detecting kin, it can also aid the investigation of emotion programs that evolved to serve a particular function.

CONCLUSION The fields of evolutionary psychology and social cognition share the similar goal of understanding the cognitive processes (e.g., categorization and decisionmaking procedures) regulating social behavior. Greater progress can be made in both disciplines by employing an evolutionary theoretical framework—a framework successfully used to investigate the cognitive processes and behaviors in nonhuman animals. For example, kin selection and parental investment theories provide sturdy guide-rails for generating hypotheses about our cognitive architecture. However, as Daly et al. (1997) have noted, evolutionary concepts such as kinship have been surprisingly absent from the social psychological literature. Yet, as this chapter has demonstrated, kinship is a strong organizing force regulating sexual behavior as well as altruism, two large areas of research in social psychology.



Greater progress can also be made by adopting a computational view of the mind. As this chapter has demonstrated, the generation of a detailed model of what a well-designed system for avoiding inbreeding might look like has led to a research program for investigating the cues mediating kin detection and the programs regulating sexual motivations and kin-directed altruism. The same logic that led to the generation of this model can be used to investigate other areas of interest in social cognition and social psychology in general. The success of such investigations will be wide-spread. For example, not only will evolutionary-computational models in social cognition allow for more careful investigations into the neurological correlates of social behavior, they will also provide a stronger backbone for other fields, such as clinical and school psychology, that look to social cognition for models of human cognition. As scientists, psychologists share the goal of uncovering how the mind works. Therefore, we would do well to utilize the strengths each discipline in the behavioral and natural sciences has to offer. This would help clear the lines of communication between subdisciplines and result in a more unified science of the mind.

REFERENCES Angyal, A. (1941). Disgust and related aversions. Journal of Abnormal and Social Psychology, 36, 393–412. Barrett, H. C. (2005). Enzymatic computation and cognitive modularity. Mind and Language, 20, 259–287. Barrett, H. C., & Kurzban, R. (2006). Modularity in cognition: Framing the debate. Psychological Review, 113, 628–647. Bevc, I., & Silverman, I. (1993). Early proximity and intimacy between siblings and incestuous behavior: A test of the Westermarck theory. Ethology and Sociobiology, 14, 171–181. Bevc, I., & Silverman, I. (2000). Early separation and sibling incest: A test of the revised Westermarck theory. Evolution and Human Behavior, 21, 151–161. Bittles, A. H., & Neel, J. V. (1994). The costs of human inbreeding and their implications for variation at the DNA level. Nature Genetics, 8, 117–121. Burnham, J. T. (1975). Incest avoidance and social evolution. Mankind, 10, 93–98. Carter, C. O. (1967). Risk of offspring of incest. Lancet, 1, 436. Cavalli-Sforza, L. L., & Bodmer, W. F. (1971).

The genetics of human populations. San Francisco: W. H. Freeman. Chagnon, N. A. (1992). Yanomamo: The last days of Eden. San Diego, CA: Harcourt Brace & Company. Crow, J. F., & Kimura, M. (1970). An introduction to population genetics theory. New York: Harper & Row. Curtis, V., & Biran, A. (2001). Dirt, disgust, and disease: Is hygiene in our genes? Perspectives in Biology and Medicine, 44, 17–31. Daly, M., Salmon, C., & Wilson, M. I. (1997) Kinship: The conceptual hole in psychological studies of social cognition and close relationships. In J. A. Simpson & D. Kenrick (Eds.), Evolutionary social psychology (pp. 265–296). Mahwah, NJ: Lawrence Erlbaum Associates, Inc. Darwin, C. (2003). The origin of species. New York: Signet Classics. (Original work published 1859) DeBruine, L. M. (2002). Facial resemblance enhances trust. Proceedings: Biological Sciences, 269, 1307–1312. Ebert, D., & Hamilton, W. D. (1996). Sex against virulence: The coevolution of


parasitic diseases. Trends in Ecology and Evolution, 11, 79–82. Ekman, P., & Davidson, R. J. (1994). The nature of emotion: Fundamental questions. New York: Oxford University Press. Fessler, D. M. T., & Navarrete, C. D. (2004). Third-party attitudes toward sibling incest: Evidence for Westermarck’s hypotheses. Evolution and Human Behavior, 25, 277–294. Fodor, J. A. (1983). The modularity of mind. Cambridge, MA: MIT Press. Haidt, J., & Bjorklund, F. (in press). Social intuitionists answer six questions about moral psychology. In W. Sinnott-Armstrong (Ed.), Moral psychology. Haig, D. (1999). Asymmetric relations: Internal conflicts and the horror of incest. Evolution and Human Behavior, 20, 83–98. Hamilton, W. D., Axelrod, R., & Tanese, R. (1990). Sexual reproduction as an adaptation to resist parasites. Proceedings of the National Academy of Science, 87, 3566–3573. Haselton, M. G., & Buss, D. M. (2000). Error management theory: A new perspective on biases in cross-sex mind reading. Journal of Personality and Social Psychology, 78, 81–91. Hepper, P. G. (1991). Kin recognition. Cambridge, UK: Cambridge University Press. Izard, C. E. (1993). Organizational and motivational functions of discrete emotions. In M. Lewis & J. M. Haviland (Eds.), Handbook of emotions (pp. 631–665). New York: Guilford Press. Jones, D. (2004). The universal psychology of kinship: Evidence from language. Trends in Cognitive Sciences, 8, 211–215. Kumar, S., Pai, R. A., & Swaminathan, M. S. (1967). Consanguineous marriages and the genetic load due to lethal genes in Kerala. Annals of Human Genetics, 31, 141–145. Lee, R. B., & DeVore, I. (1968). Man the hunter. New York: Aldine de Gruyter. Lieberman, D. (2003). Mapping the cognitive architecture of systems for kin detection and inbreeding avoidance: The Westermarck hypothesis and the development of sexual aversions between siblings. UMI Digital Dissertation, AAT 3103441. Lieberman, D., Oum, R. E., & Kurzban, R. (2006). Does the family of fundamental social categories include kinship? Manuscript under review. Lieberman, D., Tooby, J., & Cosmides, L.

(2003). Does morality have a biological basis? An empirical test of the factors governing moral sentiments relating to incest. Proceedings of the Royal Society: Biological Sciences, 270, 819–826. Lieberman, D., Tooby, J., & Cosmides, L. (in press). The architecture of human kin detection. Nature. May, R. M. (1979). When to be incestuous. Nature, 279, 192–194. Morton, N. E., Crow, J. F., & Muller, H. J. (1956). An estimate of mutational damage in man from data on consanguineous marriages. Proceedings of the National Academy of Science, 42, 855–863. Oates, K., & Wilson M. (2002). Nominal kinship cues facilitate altruism. Proceedings of the Royal Society of London, Series B, 269, 105–109. O’Brien, S. J., Roelke, M. E., & Marker, L. (1985). Genetic evidence for species vulnerability in the cheetah. Science, 227, 1428– 1434. Oum, R. E., & Lieberman, D. (in press). Emotion as cognition: An information processing view of the mind. In K. D. Vohs, R. F. Baumeister, & G. Loewenstein (Eds.), Do emotions help or hurt decision making? Perspectives from psychology and economics. New York: Russell Sage. Park, J. H., & Schaller, M. (2005). Does attitude similarity serve as a heuristic cue for kinship? Evidence of an implicit cognitive association. Evolution and Human Behavior, 26, 158–170. Porter, R. H., Matochik, J. A., & Makin, J. W. (1983). Evidence for phenotype matching in spiny mice (Acomys cahirinus). Animal Behavior, 31, 978–984. Porter, R. H., Matochik, J. A., & Makin, J. W. (1984). The role of familiarity in the development of social preferences in spiny mice. Behavioral Proceedings, 9, 241–254. Rozin, P., & Fallon, A. E. (1987). A perspective on disgust. Psychological Review, 94, 23–41. Shepher, J. (1971). Mate selection among second-generation kibbutz adolescents: Incest avoidance and negative imprinting. Archives of Sexual Behavior, 1(1), 293–307. Shepher, J. (1983). Incest: A biosocial view. New York: Academic Press. Spiro, M. E. (1958). Children of the Kibbutz. Cambridge, MA: Harvard Press.


194 EVOLUTION AND THE SOCIAL MIND Sprengelmeyer, R., Young, A. W., Calder, A. J., Karnat, A., Lange, H., Hömberg, V., et al. (1996). Loss of disgust: Perception of faces and emotion in Huntington’s disease. Brain, 119, 1647–1665. Talmon, G. Y. (1964). Mate selection in collective settlements. American Sociological Review, 29, 408–491. Tooby, J. (1977). Factors governing optimal inbreeding. Proceedings of the Institute for Evolutionary Studies, 77(1), 1–54. Tooby, J. (1982). Pathogens, polymorphism, and the evolution of sex. Journal of Theoretical Biology, 97, 557–576. Tooby, J., & Cosmides, L. (1990). The past explains the present: Emotional adaptations and the structure of ancestral environment. Ethology and Sociobiology, 11, 375–424. Tooby, J., & Cosmides, L. (2000). Evolutionary psychology and the emotions. In M. Lewis & J. M. Haviland-Jones (Eds.), Handbook of emotions (2nd ed.). New York: Guilford Press. Tooby, J., Cosmides, L., & Barrett, H. C.

(2003). The second law of thermodynamics is the first law of psychology: Evolutionary developmental psychology and the theory of tandem, coordinated inheritances: Comment on Lickliter and Honeycutt. Psychological Bulletin, 129, 858–865. Trivers, R. E. (1974). Parent–offspring conflict. American Zoology, 14, 249–264. Wedekind, C., Seebeck, T., Bettens, F., & Paepke, A. (1995). MHC-dependent mate preferences in humans. Proceedings of the Royal Society of London: Biological Sciences, 260, 245–249. Westermarck, E. A. (1921). The history of human marriage. London: Macmillan. (Original work published 1891) Williams, L. M., & Finkelhor, D. (1995). Paternal caregiving and incest: Test of a biosocial model. American Journal of Orthopsychiatry, 65, 101–113. Wolf, A. P. (1995) Sexual attraction and childhood association: A Chinese brief for Edward Westermarck. Stanford, CA: Stanford University Press.


The Self in Intimate Relationships A Social Evolutionary Account


A Conceptual and Methodological Backdrop The Role of the Self in Mate Selection The Self Never Sleeps Conclusions ntimate sexual relationships are typically composed of two people—the self and one other individual. In human relationships both men and women choose their mates, as one would expect given that both men and women make substantial investments in long-term relationships (consistent with Triver’s, 1972, parental investment theory). However, in humans mate selection never sleeps and can only be fully understood in relation to mate deselection. Individuals continue to make decisions and evaluations about their partners and relationships, up to and including the decision to abandon specific relationships (and perhaps start again). Evolutionary psychologists and social psychologists deal with both the psychology of mate selection and the psychological processes involved in ongoing relationships. However, evolutionary psychologists concentrate on the first aspect of relationship development (mate selection), whereas social psychologists concentrate on the latter domain (the development and maintenance of intimate relationships). This chapter concerns the role of the self both in terms of initial mate selection choices and in the context of ongoing intimate relationships. However, the role of the self (vis-à-vis the partner) is more conceptually and methodologically complex than it appears at first sight. Thus, the first section of the chapter completes some preliminary theoretical and methodological




work as a springboard for the chapter. Next, we deal with the role of the self in mate selection, and consider various explanations that have been proposed for the phenomenon of assortative mating. In the subsequent section we consider several ways in which the self continues to play a role in ongoing relationships. Finally, we offer some conclusions and discuss some central themes in the chapter.

A CONCEPTUAL AND METHODOLOGICAL BACKDROP It is important to make some key distinctions (that are sometimes blurred) when referring to the self and partner in intimate relationships. Let’s consider Mary and Bill, who are involved in an ongoing sexual relationship. The self and partner are represented cognitively in the brains (or minds) of Mary and Bill, as lay theories, cognitions, and beliefs. Social psychologists have carried out much research using designs in which couples answer questions about one another in terms of their personality, attractiveness, and so forth. Sometimes they also measure behavior or obtain outside objective ratings of each partner (for example, on their facial attractiveness). Measuring these classes of variables can answer some intriguing questions. Imagine Mary believes she is only moderately attractive (she thinks she is too fat), but that Bill is very sensitive and wonderfully kind. Bill, in turn, believes that Mary is very attractive (she has the perfect body) but that he is inclined to be insensitive and only moderately kind. The example is not merely apocryphal, but reflects a partner-serving positively-biased pattern repeatedly found in couplebased research (see Fletcher, Simpson, & Boyes, 2006, for a review). Note that using the self-judgments as benchmarks in such analyses is a conservative measure of partner bias, given the evidence that such self-judgments themselves are likely to be on the rose-tinted side (see Boyes & Fletcher, in press). Collecting this kind of data (illustrated by Mary and Bill’s thoughts above) enables different kinds of questions to be answered. First, one can measure the extent to which individuals project or assume that self is similar to partner on the same traits. Second, the researcher can assess the extent to which couples are similar using either judgments of self or objective measures (assortative mating evidence). Third, the researcher can assess accuracy by correlating partner judgments with the relevant self-judgments (or objective measures of the trait). And, fourth, individual or couple differences in similarity, projection, positive bias, and accuracy can be used to predict mate choice, relationship stability, and satisfaction. Finally, it is important to note that bias and accuracy can happily coexist. To take an example adapted from Fletcher (2002), Mary rates herself (accurately) as very warm, moderately attractive, but as lacking ambition. Bill, her partner, rates Mary as incredibly warm, very attractive, and moderately ambitious. Bill is, thus, positively biased (he is consistently more positive than Mary’s self-perceptions)


but quite accurate (he is tracking her self-perceived qualities accurately across the three traits, albeit in a positively-inflated fashion). Indeed, researchers have often reported evidence consistent with this example. For example, Sprecher (1999) reported that those individuals who had flat trajectories of satisfaction over time tended to recall at a later time that they had steadily improved (a clear example of positive bias). Nevertheless, the sample overall quite accurately retrospectively tracked and reported relative increases or decreases in love and satisfaction over past periods in their relationships. With this brief analysis in mind, we move into dealing with the role of the self in mate selection contexts.

THE ROLE OF THE SELF IN MATE SELECTION A widely replicated finding is that partners in sexual relationships (dating or married) tend to be similar across many characteristics. However, the degree of similarity varies depending on the trait in question. Similarity (assortative) correlations are moderate to high for many traits (from .30 to .80) including age, height, attractiveness, values and attitudes, smoking behavior, intelligence, and so forth (see, for example, Luo & Klohnen, 2005; Watson et al., 2004). However, similarity on personality traits like kindness and conscientiousness is much lower (from zero to .20) (see, for example, Zentner, 2005). There is no evidence that such findings are simply a product of sub-groups of individuals who happen to be similar (e.g., university students) tending to meet, work, and live near each other (Watson et al., 2004). Evolutionary and social psychologists have suggested several substantive explanations, all involving self-perceptions, which we consider in turn.

Going for the Best Deal The first, and most parsimonious explanation, is that individuals simply attempt to get the best deal on offer in the mating market (see Todd, chapter 9, this volume). In a graduate class the first author (Garth Fletcher) teaches on the science of intimate relationships, he starts the year’s work with a demonstration developed by Ellis and Kelley (1999). The 25 or so students in the class are randomly given cards with numbers on them, ranging from 1 to 10, that represent their assigned fictional mating value. These cards are held to their foreheads in such a way that others can see them, but remain out of sight for the cardbearer (so that each individual does not know his or her own mate value). The aim of the game is to get together with the individual with the highest mate value possible (biological sex is ignored). Once a mate selection is made, the initiator indicates his selection by attempting to shake hands. If the individual approached spurns the handshake, then the initiator must look elsewhere. As the class members mill about, individuals pair off until a small and disconsolate



group is left standing in the middle of the room. Inevitably, this group represents the dregs of the mating market, but they too finally pair off in a crestfallen sort of way. The results typically reveal that the mating values of the paired-up partners are highly correlated (about .70 or so). This demonstration suggests that merely utilizing the heuristic—get the best deal possible—is enough to produce assortative mating in situations where both parties exercise choice. This is, of course, a very stripped down and closed version of real-life settings. In the real-world individuals believe that similarity is important for successful relationships (Hassebrauck & Fehr, 2002) and seem to prefer individuals they believe are similar to themselves (Klohnen & Luo, 2003). Moreover, it seems likely that from the onset of adolescence onwards individuals receive copious feedback (rejections and dating successes), along with a flood of benchmark images and information in the media, that enable them to accurately assess their own mate value. The development of self-perceptions of mate value should save time and effort, and reduce humiliating rejections to the minimum by encouraging the development of realistic standards (see Todd, chapter 9, this volume). Some preliminary support for this thesis comes from the class exercise previously described. One procedure sometimes added is that when a couple is formed, indicated by a handshake, then each individual first guesses his or her own mate value number, before taking a peek at the assigned value. Correlations of around .70 between the predicted self-mating values and the actual numbers assigned are produced. Even the limited information gained from their observations of others, and their own experiences, in this class exercise allow individuals to rapidly and accurately assess their own mate value. More substantial evidence for this thesis will be cited in due course.

Seeking Similarity The fact that people appear to prefer mates who are similar to themselves has led to the suggestion by Buss (1999) that such a heuristic helps solve adaptive problems of compatibility and promotes bonding, relationship satisfaction, and stability, working off the assumption that higher similarity in, for example, personality traits, will produce these kinds of payoffs. There is certainly a host of evidence that people who believe they are more similar to their partners in terms of their personalities, abilities, attitudes, and so forth, are happier and more committed to their partners (e.g., Acitelli, Douvan, & Veroff, 1993; Hammond & Fletcher, 1991). However, although consistent with conventional wisdom, the research evidence that more similarity confers such benefits is decidedly mixed, with many null findings being published (Acitelli. Kenny, & Weiner, 2001; Robins, Caspi, & Moffit, 2000; Watson et al., 2004). The fact is that similarity is not all it’s cracked up to be, which brings us back to the explanation that assortative mating effects are by-products of going for the best deal in town. But, what are people looking for in the mating market? In the next section we discuss


the standards people use in selecting and evaluating mates, before considering when and how the self comes back into the picture.

Relationship and Partner Standards Gangestad and Simpson’s (2000) Strategic Pluralism Model of human mating is based around three major postulates. First, the model argues that selection should have generated a small and flexible set of mating strategies and tactics that should be enacted conditionally according to cues in the social and physical environment. Second, without gainsaying the importance of gender differences in mate selection, its main focus is on explaining the considerable within-sex variability of sexual attitudes and behavior. Third, the model suggests that human mating strategies and tactics are a function of two kinds of goals—the search for the kind of mate who happens to provide good genes and the search for a mate who would be a good mate and provider (also see Simpson & LaPaglia, chapter 10, this volume). Evidence has been steadily building for this model (see Simpson & LaPaglia, chapter 10, this volume), but of course, it has its limitations. For example, like most evolutionary models, its central concern is with mate selection and not what happens after eyes meet across a crowded room or boy meets girl on the Internet. Nevertheless, social psychologists will immediately recognize that some key features of the model (principally its focus on the interaction between goals or tactics and cues in the social environment, and its concern with within-sex variability) provide an admirable platform on which to develop a social psychological theory dealing with the proximal-level processes involved in both mate selection and ongoing relationships. In this vein, then, Fletcher and Simpson developed a model that is centrally concerned with the standards that people utilize within the context of sexual relationships (termed the Ideal Standards Model) (see Simpson, Fletcher, & Campbell, 2001, for a review). This model was built on five main interlocking hypotheses. First, individuals should possess chronically accessible mate and relationship ideal standards that predate specific relationships and embody considerable within-sex variability. Second, the dimensions on which partners will be evaluated will be derived from the two main goals specified by Gangestad and Simpson (2000), and will not simply represent global mate evaluations. Third, relationship and partner judgments should be driven by the perceived consistency between ideal standards and accompanying perceptions. Fourth, the functions of the resultant perceived discrepancies revolve around helping individuals to assess and meet three main goals—relationship evaluation, explanation, and regulation (in both initial mate selection and ongoing relationship contexts). Fifth, the levels of accuracy and/or bias in relevant judgments are likely to be motivated by two independent overarching goals: the drive for accuracy and truth versus the desire to maintain positively-biased judgments in ongoing relationships.



The broad tenants of the Ideal Standards Model have been increasingly supported by research evidence. In the first published study, Fletcher, Simpson, Thomas, and Giles (1999) showed in a series of factor analytic studies (both exploratory and confirmatory) that there exist three major dimensions that individuals consider when evaluating both prospective and current romantic partners: warmth/trustworthiness, attractiveness/vitality, and status/resources. In addition, there is substantial within-sex variability in the importance attached to each trait or dimension. Table 12.1 shows the items for each scale derived from this work that we have since used successfully in further research. These findings are consistent with the argument from the Strategic Pluralism Model that evaluating mates could have promoted the reproductive success of our ancestors via two distinct routes—either good investment and/or good genes. The possession of warmth and trustworthiness signals the motivation to invest in the mate and resulting children, the possession of status and resources (or the drive to obtain them) signals that the individual has the means to provide good investment, and the possession of attractiveness and vitality is often held to indicate good genes, signaling higher fertility and perhaps better long-term health, although this latter claim remains controversial (for recent reviews see Rhodes, 2006, and Gangestad & Scheyd, 2005). In addition, there is good evidence across several studies (both crosssectional and longitudinal) that greater perceived consistency between standards and partner perceptions (for both men and women) is related to (1) higher levels of relationship satisfaction (Fletcher et al., 1999; Fletcher, Simpson, & Thomas, 2000), (2) lower rates of relationship dissolution (Fletcher et al., 2000), and (3) weaker desires and attempts to regulate or change the partner (e.g., make them more attractive or ambitious) (Overall, Fletcher, & Simpson, 2006). Moreover, in all of these studies, the links between standards/perceptions consistency and evaluation or regulation operate within each dimension independently; they are not simply artifacts of halo effects or global evaluations of the relationship. Finally, to add some further predictive validity, using the scales shown in Table 12.1, we have consistently found the same sex differences reported in other research; namely, women give greater importance than men (in long-term relationships) to warmth/trustworthiness and status/resources, and less importance to attractiveness/vitality (see, for example, Fletcher, Tither, O’Loughlin, Friesen, & Overall, 2004). Table 12.1 Partner Ideal Standard Scales (from Fletcher et al., 1999) Ideal Dimensions

Short Scale Items

Partner Warmth/Trustworthiness Partner Attractiveness/Vitality Partner Status/Resources

understanding, supportive, considerate, kind, a good listener, sensitive adventurous, nice body, outgoing, sexy, attractive, good lover good job, financially secure, nice house or apartment, successful, dresses well


What is the proximal origin of individual within-sex differences in the importance attached to mate standards? The answer suggested by our prior discussion is that the prime determinant is likely to be self-perceptions of mate value. However, the research and theorizing on the Ideal Standards Model suggests that such self-perceptions are not likely to operate simply in a global fashion, but according to the dimensions already found to exist for evaluating potential or current mates; namely, warmth/trustworthiness, attractiveness/ vitality, and status/resources. Provisional evidence is consistent with this thesis. Fletcher and Boyes (2004) had 200 individuals rate their own mate value using the scales developed by Fletcher et al. (1999) to assess the importance attached to qualities in prospective mates (see Table 12.1). A confirmatory factor analysis revealed the same tripartite structure for self-mate evaluations as found in prior research for potential partners. Moreover, correlating the two sets of ratings showed the expected pattern of discriminant and convergent correlations (see Table 12.2). More positive self-evaluations on given dimensions were related to higher importance attached to ideal standards on the same dimensions, but typically not for the off-diagonal correlations. Although this evidence is suggestive it is hardly conclusive, given that the scales used to assess self-perceived mate value were derived from research specifically designed to assess individual differences in the extent to which people attach importance to different characteristics of potential or actual partners. What comes out of a factor analysis depends on what items go in, it could be reasonably argued. Thus, Fletcher, Boyes, Overall, and Kavanagh (2006) recently completed a series of studies designed to assess self-perceptions of mate value starting from scratch. In the first study, samples of university students and older individuals (both men and women) wrote down their strengths and weaknesses in terms of what they could offer in finding either a desirable mate or maintaining a successful intimate relationship. These items were then coded, sticking close to the wording used by respondents, to produce 60 individual items. These items were then turned into a scale, which could be rated in terms of how accurately each item describes the self. The next two studies involved different samples of Table 12.2 Convergent/Discriminant Correlations Among Self-Rated Mate Value and Partner Ideal Standards (from Fletcher, 2002) Partner Ideal Standards Self-rated Mate Value

Warmth/Trustworthiness Attractiveness/Vitality Status/Resources * p < .05, n = 200.

Warmth/ Trustworthiness



.53* .04 .10

.26* .54* .29*

.11* .23* .37*



200 individuals. Initially using an exploratory factor analysis followed by a confirmatory factor analysis, the same factorial structure was produced across samples, with primarily the same items loading on each factor. Moreover, this factor structure replicated across both gender and relationship status. Table 12.3 shows the three best loading items for each factor, their factor loadings, and the suggested names for each factor. At face value, five out of the six self-perception categories seem to overlap with the scales previously derived to assess the perceived merits of traits in potential or actual partners in relationships: caring, open, sexy, outgoing, and status. The odd one out is secure. To establish the case empirically, and to further assess the convergent and discriminant validity of the scale, we administered the self-perception of mate value scale to a sample of 200 individuals (100 men and 100 women) along with scales measuring the importance of ideals standards in partners, self-esteem, the Big Five personality ratings, attachment, and relationship quality (completed by a subsample of 100 currently involved in sexual relationships). The results were very close to what was expected. Of the 66 correlations, all 14 of the predicted convergent correlations were significant, ranging from .20 to .64. Of the remaining 52 discriminant correlations, only five exceeded .20 and none exceeded .23. Most importantly (controlling for self-esteem), (1) superior self-perceptions of caring and openness were associated with more weight given to warmth/trustworthiness in a potential partner, (2) higher self-perceptions of being sexy and outgoing were associated with more importance attached to attractiveness/vitality, and (3) more positive perceptions of status in the self were associated with more weight given to status/resources. The correlations with the Big Five were also revealing, showing that the Big Five caught three of the six mate self-perception categories (caring, outgoing, and stable) but missed three distinctive categories of mate evaluation (open, status, and sexy). Not surprisingly, global self-esteem was positively and significantly related to all six self-perception categories. However, when we regressed self-esteem on all six selfperception categories, only two self-perception categories remained significant positive predictors (for both men and women): sexy and stable (βs from .30 to .41). Table 12.3 Factors and Loadings for Self-Perceptions of Mate Value (n = 200) (from Fletcher, Boyes, Overall, & Kavanagh, 2006) Dimensions

Items (and Factor Loadings)

Caring Open Sexy Outgoing Status Stable

caring (.85), kind (.84), considerate (.81) talks openly (.79), open (.79), ability to commit (.64) sexy (.87), attractive (.84), nice body (.77) outgoing (.80), good social skills (.77), funny (.64) good job (.80), wealthy (.72), ambitious (.54) emotionally stable (.87), secure (.84), good self esteem (.75)


To summarize, these results revealed independent evidence that selfperceptions of mate value largely mirror the three important categories previously found for evaluating potential or actual relationship partners, and that self-perceptions of mate value may undergird the way on which individuals set their standards for evaluating their mates. In addition, the measurement of selfperceptions of mate value does not seem to be redundant with existing measures of the self developed by personality and social psychologists, but includes specific categories suggested by the Ideal Standards Model and its evolutionary cousins.

Caveats and Objections Thus far, we have argued that (mirroring the research on ideal standards) selfperceptions of mate value operate along independent dimensions. However, this is somewhat misleading. In all the CFA analyses mentioned, the results actually show that the separate categories (for both partner ideal standards and selfperceptions of mate value) are quasi-independent and that the best-fitting models include second-order factors representing the possession of more global demanding standards or more positive perceptions of mate value respectively. These results suggest that the relevant social cognitive modules are stored in both a simple global form and in terms of more differentiated structures. Depending on the demands of the context, and the importance and nature of the decision, either cognitive model may be accessed and used. For example, rating a potential date may invoke a quick initial judgment based on the overall impression of the individual, whereas deciding whether to marry an individual will be likely to involve a more considered evaluation, analyzing each domain in turn. Finally, consistent again with Gangestad and Simpson’s (2000) model, there is no doubt that people frequently engage in trade-offs, taking into account many factors including their own mate value and that of their potential partner. However, yet another prediction of the Ideal Standards Model is that such tradeoffs will typically occur across the three dimensions. Li and his colleagues have argued, with research support, that some features are likely to be necessities and others may be luxuries (Li, Bailey, Kenrick, & Linsenmeier, 2002). For example, in long-term relationships they propose that for men partner attractiveness is a necessity, whereas for women status/resources is a necessity. For both genders, in contrast, warmth/trustworthiness is a necessity. We do not find their arguments wholly convincing. In Fletcher et al. (2004) we manipulated the extent to which the familiar three mate selection categories were present in potential mates across both short-term and long-term relationship contexts. The results suggested that such choices were the product of a complex interaction between gender, context, and the traits being traded off. For example, in long-term relationships most individuals chose a warm, homely person over a cold, attractive alternative, whereas for a short-term fling the trade-offs went in the opposite direction. However significant gender differences were also apparent in both short-term and long-term settings, with women more consistently choosing a warm, kind



individual over a sexy, attractive person. In short, although this pattern of findings was consistent with Li et al.’s argument, “necessities” were commonly traded off depending on the context. Two crucial empirical questions remain that potentially sink virtually all evolutionary and social psychological models of mate selection, including those presented here; namely, are individuals’ judgments of their own (relevant) traits and those of potential mates accurate? One cannot assume in advance that the answers will be in the affirmative. As previously noted, for example, individuals believe that more similarity between partners produces more happiness, and happier couples believe they have more similar personalities. However, the research evidence indicates individuals are largely mistaken on both counts. Most relevant research has been conducted in relation to two out of the three pivotal mate selection categories: characteristics such as warmth/trustworthiness, extraversion, and attractiveness. The evidence is clear-cut. First, after minimal observation or interaction, ratings of strangers in terms of both physical attractiveness and extraversion are reasonably accurate (see, for example, Albright, Kenny, & Malloy, 1988; Langlois et al., 2000). These conclusions are based on findings that examine the extent to which different people achieve consensus about a target and the extent to which self-perceptions of the target are consistent with either some objective benchmark or the views of observers. To take one particularly compelling research example, high in ecological validity, Marcus and Miller (2003) had participants rate their own physical attractiveness and that of other men and women who were sitting together in small groups. There was good consensus on the level of attractiveness for specific targets, and targets’ self-perceptions generally matched well with how they were perceived (correlations ranging from .28 to .53). Moreover, individuals’ metaperceptions of how they were perceived generally by others were accurate (correlations ranging from .26 to .49). As the authors conclude, “we know who is handsome or pretty, and those who are attractive know it as well” (p. 344). From an evolutionary standpoint, one would also expect men to produce particularly accurate perceptions of women’s attractiveness and women should be on the money when it comes to judging how they are rated by men. Both predictions were confirmed. The highest level of consensus was reached by different men rating the same women (41% of the target variance), and the most accurate meta-awareness was achieved by women rating how they were generally perceived by men (r = .49). In contrast to physical attractiveness and extraversion, the accuracy in rating strangers in terms of traits, such as warmth and kindness, is typically abysmal, but does climb to quite respectable levels as a function of increased closeness and knowledge of the target (Funder & Colvin, 1988; Letzring, Wells, & Funder, 2006; Thomas, 1999). For example, Thomas (1999) had individuals observe men and women currently involved in sexual relationships having a 5-minute discussion of capital punishment, and then rate each partner on the Big Five traits. Self–other agreement was low for all five categories when strangers carried


out the task (from zero to .20; mean r = .10), considerably higher when friends carried out the same task (from .20 to .45, mean r = .34), and better still when partners rated each other (from .29 to .48, mean r = .41). Consensus across raters told the same story, with good agreement across partners and friends when rating the same targets (from .20 to .42, mean r = .34) and weak consensus across strangers and either partners and friends (zero to .21; mean r = .09). The difference between the accuracy in rating personality traits like emotional stability or conscientiousness versus more immediately observable qualities (such as attractiveness), when levels of acquaintanceship are low, may explain why the assortative correlations for the latter traits are much higher than the former traits (as described previously). People may seek similarity on personality traits, but their personality judgments are generally initially inaccurate. By the time individuals get to know their partners better, the inexorable processes of attraction, falling in love, and bonding (driven by neuropeptides such as oxytocin) are too far advanced to claw back, and the actual similarity between the partners may become largely irrelevant. The pivotal and reassuring conclusion to be drawn from this work is that, although the evidence is partial, both individuals’ self-judgments and their judgments of potential or existing partners seem to be reasonably accurate for both men and women, specifically for traits that are pivotal in mate selection contexts.

THE SELF NEVER SLEEPS Having discussed the pivotal role of the self in mate selection contexts, we move to briefly consider its fate and fortune in ongoing relationships. An examination of the research which has directly compared the impact of self to partner judgments gives initial pause to an overenthusiastic endorsement of the role of the self in ongoing relationships. In brief, the standard research finding reveals that partner judgments play a much more powerful role in predicting relationship satisfaction than self-judgments (e.g., Fletcher & Fincham, 1991; Fletcher & Thomas, 2000; Friesen, Fletcher, & Overall, 2005; Overall et al., 2006; Sümer & Cozzarelli, 2004). Even so, it is clear that the self remains directing traffic behind the relationship scenes after relationships are in full swing, and is itself buffeted by the powerful psychological forces at work in intimate relationships. As already described, the perceived consistency between ideal standards and perceptions is a fundamental driver of important evaluations and decisions in the development of intimate relationships, up to and including decisions to leave. And selfjudgments continue to influence the importance attached to specific standards. However, self-judgments are particularly vulnerable to change in intimate relationships, readily becoming the dependent variable instead of the independent variable. A recent study by Overall et al. (2006) is a good illustration of this process.



Overall et al. (2006) reported (in two studies) that greater regulation attempts of the partner were associated with lower consistency between ideal standards and perceptions, which in turn were related to lower relationship satisfaction. A longitudinal cross-lagged study confirmed that lower perceived consistency between standards and perceptions of the partner motivated more regulation over time, and that more regulation also produced lower levels of standard/perception consistency (and associated relationship satisfaction) over time. The latter finding may seem paradoxical—the aim of regulation is presumably to make relationships better and lower the gap between perceptions and expectations, yet more regulation seems to make things worse. In further analyses of their data, Overall et al. (2006) discovered that the key to these latter findings was related to what regulation attempts communicate to the partner. As this (and other research) has suggested, a powerful determinant of relationship satisfaction is how individuals believe they are viewed by their partners. If Bill is unhappy with Mary’s level of attractiveness, and suggests she lose weight and join a gym, then, according to the findings from Overall et al., Mary is likely to do one or more of the following: (1) develop more negative perceptions of her own attractiveness, (2) realize that Bill does not fulsomely accept her the way she is, and (3) start regulating herself on this dimension. Mary’s relationship satisfaction is also likely to become more negative. Interestingly, consistent with evolutionary models of mate selection, women tended to regulate their own levels of attractiveness specifically in response to their male partners’ regulatory attempts, whereas men reported trying to change themselves principally in response to their female partners’ attempts to improve their ambition and status. The fact that regulation efforts tend to commonly backfire, and people become even unhappier with the relationship, raises questions about the functions of the relationship monitoring and regulation system. From a distal evolutionary approach the functions of an adaptation or behavior are defined in terms of the costs and benefits vis-à-vis reproductive fitness, and do not necessarily equate to increased happiness. For example, one reason why humans may have evolved the relationship monitoring and regulation system was to loosen the powerful bonds of love and attachment when standards were not being met, thus enabling individuals to look elsewhere for a new partner and relationship. Alternately, perhaps a principal function of our ancestral relationship monitoring and regulation system was indeed to improve relationships, but it fails because the contemporary social and cultural environment has changed so that it no longer matches the ancestral environment within which the relevant adaptations developed. For example, perhaps contemporary Western cultures, with thousands of accessible partners apparently a mouse-click away, barrages of self-help books and TV shows about relationships and how to make them better, constant images of attractive alternatives, and people apparently having great sex everywhere, have heightened people’s expectations and standards to the extent that the relationship monitoring and regulation system has been put into overdrive. Thus, the monitoring


and regulation system has become relatively dysfunctional in the modern environment. A critic may argue that the regulation system in intimate relationships is a by-product of other adaptations, such as the need and means to regulate kin or perhaps even the physical environment. This is certainly a possibility, but we would argue that intimate relationships comprise a key fulcrum around which powerful evolutionary forces and adaptations are likely to be marshaled. Having the ability and motivation to play the piano or read a book are certainly byproducts of adaptations that were not “designed” by evolution in furtherance of these particular activities. Possessing the ability and motivation to monitor and regulate relationships or partners, in contrast, are likely to be evolutionary adaptations.

CONCLUSIONS The mate-selection decisions people make seem to be based (in part) on realistic assessments of their own mate value, and, thus, guide who they are likely to be able to attract or who to realistically settle for in a long-term (or even short-term) relationship. Hence, in our view, the major driver of assortative mating correlations is probably the motivation for individuals to obtain the best deal going, rather than the desire to find a similar soul-mate. However, humans have unusual mating patterns compared to other species, continuing to monitor and regulate their sexual relationships (often ending and replacing them) over long periods of time (see Fletcher, 2002). Accordingly, the role of the self does not cease after mates have been selected but continues throughout the course of specific relationships. We trust this chapter shows at least a glimpse of the theoretical and research gains that are obtainable when social cognitive and evolutionary approaches join forces to detail and explain the workings of the intimate relationship mind, and its links to interpersonal behavior.

REFERENCES Acitelli, L. K., Douvan, E., & Veroff, J. (1993). Perceptions of conflict in the first year of marriage: How important are similarity and understanding? Journal of Social and Personal Relationships, 10, 5–19. Acitelli, L. K., Kenny, D. A., & Weiner, D. (2001). The importance of the similarity and understanding of partners’ marital ideals to relationship satisfaction. Personal Relationships, 8, 167–186. Albright, L., Kenny, D., & Malloy, T. (1988).

Consensus in personality judgments at zero acquaintance. Journal of Personality and Social Psychology, 55, 387–395. Boyes, A. D., & Fletcher, G. J. O. (in press). Meta-awareness of bias in intimate relationships. Journal of Personality and Social Psychology. Buss, D. M. (1999). Evolutionary psychology: The new science of the mind. Boston: Allyn & Bacon. Ellis, B. J., & Kelley, H. H. (1999). The pairing


208 EVOLUTION AND THE SOCIAL MIND game: A classroom demonstration of the matching phenomenon. Teaching of Psychology, 26, 118–121. Fletcher, G. J. O. (2002). The new science of intimate relationships. Malden, MA: Blackwell Publishers. Fletcher, G. J. O., & Boyes. A. D. (2004). Bias and accuracy in perceptions of mate value. Paper presented at the Society for Personality and Social Psychology, Austin, TX. Fletcher, G. J. O., Boyes, A. D., Overall, N. C., & Kavanagh, P. (2006). Self-perceptions of mate value. Unpublished manuscript. University of Canterbury, Christchurch, New Zealand. Fletcher, G. J. O., & Fincham, F. D. (1991). Attributional processes in close relationships. In G. J. O. Fletcher & F. D. Fincham (Eds.), Cognition in close relationships (pp. 6–34). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. Fletcher, G. J. O., Simpson, J. A., & Boyes, A. D. (2006). Accuracy and bias in romantic relationships: An evolutionary and social psychological analysis. In M. Schaller, J. A. Simpson, & D. T. Kenrick (Eds.), Evolution and social psychology (pp. 189–210). New York: Psychology Press. Fletcher, G. J. O., Simpson, J. A., & Thomas, G. (2000). Ideals, perceptions, and evaluations in early relationship development. Journal of Personality and Social Psychology, 79, 933–940. Fletcher, G. J. O., Simpson, J. A., Thomas, G., & Giles, T. (1999). Ideals in intimate relationships. Journal of Personality and Social Psychology, 76, 72–89. Fletcher, G. J. O., Tither, J. M., O’Loughlin, C., Friesen, M. D., & Overall, N. C. (2004). Warm and homely or cold and beautiful? Sex differences in trading off traits in mate selection. Personality and Social Psychology Bulletin, 30, 659–672. Fletcher, G. J. O., & Thomas, G. (2000). Behavior and on-line cognition in marital interaction. Personal Relationships, 7, 111– 130. Friesen, M. D., Fletcher, G. J. O., & Overall, N. C. (2005). A dyadic assessment of forgiveness in intimate relationships. Personal Relationships, 12, 61–77. Funder, D. C., & Colvin, C. R. (1988). Friends and strangers: Acquaintanceship, agreement,

and the accuracy of personality judgment. Journal of Personality and Social Psychology, 55, 149–158. Gangestad, S. W., & Scheyd, G. J. (2005). The evolution of human physical attractiveness. Annual Review of Anthropology, 34, 523–548. Gangestad, S. W., & Simpson, J. A. (2000). The evolution of human mating: Trade-offs and strategic pluralism. Behavioral and Brain Sciences, 23, 573–644. Hammond, J. R., & Fletcher, G. J. O. (1991). Attachment styles and relationship satisfaction in the development of close relationships. New Zealand Journal of Psychology, 20, 56–62. Hassebrauck, M., & Fehr, B. (2002). Dimensions of relationship quality. Personal Relationships, 9, 253–270. Klohnen, E. C., & Luo, S. (2003). Interpersonal attraction and personality: What is attractive—self similarity, ideal similarity, complementarity, or attachment security? Journal of Personality and Social Psychology, 85, 709–722. Langlois, J. H., Kalakanis, L., Rubenstein, A. J., Larson, A., Hallam, M., & Smoot, M. (2000). Maxims or myths of beauty? A metaanalytic and theoretical review. Psychological Bulletin, 126, 390–423. Letzring, T. D., Wells, S. A., & Funder, D. C. (2006). Information quantity and quality affect the realistic accuracy of personality judgment. Journal of Personality and Social Psychology, 91, 111–123. Li, N. P., Bailey, J. M., Kenrick, D. T., & Linsenmeier, J. A. W. (2002). The necessities and luxuries of mate preferences: Testing the tradeoffs. Journal of Personality and Social Psychology, 82, 947–955. Luo, S., & Klohnen, E. C. (2005). Assortative mating and marital quality in newlyweds: A couple-centered approach. Journal of Personality and Social Psychology, 88, 304–326. Marcus, D. K., & Miller R. S. (2003). Sex differences in judgments of physical attractiveness: A social relations analysis. Personality and Social Psychology Bulletin, 29, 325–335. Overall, N, C., Fletcher, G. J. O., & Simpson J. A. (2006). Regulation processes in intimate relationships: The role of ideal standards. Journal of Personality and Social Psychology, 91, 662–685.


Rhodes, G. (2006). The evolutionary psychology of facial beauty. Annual Review of Psychology, 57, 199–226. Robins, R. W., Caspi, A., & Moffitt, T. E. (2000). Two personalities, one relationship: Both partners’ personality traits shape the quality of their relationship. Journal of Personality and Social Psychology, 79, 251– 259. Simpson, J. A., Fletcher, G. J. O., & Campbell, L. (2001). The structure and function of ideal standards in close relationships. In G. J. O. Fletcher & M. Clark (Eds.), Blackwell handbook of social psychology: Interpersonal processes (pp. 86–106). Oxford, UK: Blackwell. Sprecher, S. (1999). “I love you more today than yesterday”: Romantic partners’ perceptions of changes in love and related affect over time. Journal of Personality and Social Psychology, 76, 46–53. Sümer, N., & Cozzarelli, C. (2004). The impact of adult attachment on partner and

self-attributions and relationship quality. Personal Relationships, 11, 355–371. Thomas, G. (1999). Accuracy in empathic and trait judgments: The quest for the good judge, good target, good trait, and good relationship. Unpublished doctoral dissertation, University of Canterbury, Christchurch, New Zealand. Trivers, R. (1972). Parental investment and sexual selection. In B. Campbell (Ed.), Sexual selection and the descent of man 1871–1971 (pp. 136–179). Chicago: Aldine. Watson, D., Klohnen, E. C., Casillas, A., Simms, E. N., Haig, J., & Berry, D. S. (2004). Match makers and deal breakers: Analyses of assortative mating in newlywed couples. Journal of Personality, 72, 1029– 1068. Zentner, M. R. (2005). Ideal mate personality concepts and compatibility in close relationships: A longitudinal analysis. Journal of Personality and Social Psychology, 89, 242– 256.






A Social Cognitive Evolutionary Approach to Jealousy The Automatic Evaluation of One’s Romantic Rivals ABRAHAM P. BUUNK KARLIJN MASSAR PIETERNEL DIJKSTRA

The Importance of Jealousy Inventory of Relevant Rival Characteristics Experimentally Manipulating Rival Characteristics Body Build Sexual Versus Emotional Infidelity Conclusion

THE IMPORTANCE OF JEALOUSY he defining feature of a jealousy-evoking situation is that it involves a rival who is interested in one’s partner, or in whom one’s partner is interested. Individuals do not become jealous when their mate ends the relationship for other reasons, such as when the partner is killed in an automobile accident, moves to a far away city for work (Mathes, Adams, & Davies, 1985), or ends the relationship without getting involved with someone else (Parrott, 1991). An illustration of the centrality of a rival for the occurrence of jealousy was provided by Hupka, Otto, Tarabrina, and Reidl (1993) who found that individuals in three cultures (Russia, the US, and Germany) agreed that the




words “rival” and “sex” were associated strongly with jealousy, but not with emotions such as anger, envy, and fear. In addition, jealousy should not be viewed as a basic or specific emotion (cf. Ellsworth, chapter 5, this volume), but rather as an evaluative-motivational complex aimed at assessing the threat that a rival may impose to one’s reproductive interests (e.g., Buunk & Dijkstra, 2000; Parrott, 1991). Among human males, the inclination to assess the potential threat of a rival has a long evolutionary history that is rooted in fights over the access to females, behaviors that are found in many species (cf. Barash & Lipton, 2001; Buss, 1994; Buunk, 1986). Moreover, although in many primates males direct special attention to offspring likely to be their own (Hrdy, 1981), human males invest much more heavily in their offspring than males of other primate species, and will therefore be particularly alert to rivals who interfere in the relationship. As females, more than males, value dominance and status in a mate, supposedly because these features are related to a man’s ability to provide protection and resources, a basic assumption in our research is that jealousy in human males is likely to be influenced particularly by the rival’s dominance and status. While intrasexual competition among human males has a long evolutionary history, intrasexual competition among human females seems an evolutionarily more novel phenomenon that arose with the advent of pair bonding, due to which females began to compete over males who would be willing and able to invest and to protect (cf. Campbell, 2002; see also Simpson & LaPaglia, chapter 10, this volume). In the case of a rival vying for her mate’s attention, a major threat to a woman is that she may have to share her partner’s resources with another woman, and, even more threatening, that her partner will direct all of his support to another partner. Because men, more than women, value physical attractiveness in a partner, supposedly because this signals women’s reproductive value, in our research we assumed that women will have evolved a tendency to compete with other females in this domain, and jealousy in women is likely to be driven particularly by a rival’s physical attractiveness (e.g., Buss, 1989; Gangestad & Thornhill, chapter 3, this volume).

INVENTORY OF RELEVANT RIVAL CHARACTERISTICS Our research program began with examining what individuals spontaneously mention when asked about those characteristics that would most strongly evoke feelings of jealousy (Dijkstra & Buunk, 2002). To evoke these characteristics, we presented participants with the following scenario: [Y]ou are at a party with your girlfriend and you are talking with some of your friends. You notice your girlfriend across the room talking to a man you do not know. You can see from his face that he is very interested in your girlfriend. He is listening closely to what she is saying and you notice that he casually touches her


hand. You notice that he is flirting with her. After a minute, your girlfriend also begins to act flirtatiously. You can tell from the way she is looking at him that she likes him a great deal. They seem completely absorbed in each other.

With regard to the person their partner was flirting with, participants were asked what kind of person would make them feel most jealous if this situation would happen to them. In total, participants mentioned over 600 rival characteristics. Men more often than women mentioned a rival’s physical dominance, “smoothness,” and social status as characteristics that would make them jealous. In contrast, women more often than men mentioned a rival’s sexy appearance and slenderness as characteristics that would make them jealous. On the basis of these spontaneously mentioned rival characteristics, a questionnaire was constructed that included 56 characteristics. The same scenario as in the previous study was used with the additional question “When my partner and a different man would flirt with each other, I would feel particularly jealous when that other man. . . .” In a study among 240 college students a factor analysis on these characteristics showed five factors: social dominance, physical attractiveness, seductive behavior, physical dominance, and social status (Dijkstra & Buunk, 2002). It may be noted that the first, second, and fifth factors are in part similar to those Fletcher and Overall (chapter 12, this volume) found for self-perceived mate value (i.e., open and outgoing, sexy, and status), underlining the assumption that rivals are primarily evaluated in terms of their mate value. Consistent with our expectations, in the student sample men experienced more jealousy than women when their rival was more socially or physically dominant or had a higher status than themselves, whereas women experienced more jealousy than men when their rival was more physically attractive. Men and women did not differ in the extent to which the seductive behavior of their rival evoked feelings of jealousy. Next, in a community sample of 144 individuals, these findings were replicated, demonstrating that the sex differences were not restricted to college students. In the same vein, studies in other cultures (United States and Korea) have found that women reported more distress to a rival who surpassed them on facial and bodily attractiveness, whereas men reported more distress to a rival who had better financial and job prospects (Buss, Shackelford, Choe, Dijkstra, & Buunk, 2000). Finally, we like to mention that, underlining the importance of social comparison in jealousy, in the community sample, social comparison orientation (i.e., the dispositional tendency to engage in social comparisons; Buunk & Gibbons, 2005) was positively correlated with jealousy in response to a rival’s social dominance, social status, and physical attractiveness, and, in men, also with jealousy in response to a rival’s physical dominance (Dijkstra & Buunk, 2002).



EXPERIMENTALLY MANIPULATING RIVAL CHARACTERISTICS As a next step in our research program, we conducted a series of studies in which these rival characteristics were experimentally manipulated. Only a handful of studies on jealousy have employed a similar method (e.g., Nadler & Dotan, 1992; Shettel-Neuber, Bryson, & Young, 1978). In our experiments, participants were presented with the scenario mentioned previously in which the participant’s current (real or imagined) partner was flirting with an opposite-sex individual. Next, participants received one of four profiles of the individual flirting with their partner, consisting of a picture and a personality description. The picture showed an individual of either high or low physical attractiveness, and the personality description depicted someone who was either high or low in dominance in terms of characteristics such as being a good judge of character, taking initiative, influencing people, and livening things up at parties. These are precisely those characteristics that are, according to van Vugt and Kurzban (chapter 14, this volume), typical of individuals who emerge as leaders in groups. After they had read the scenario and the profile, participants were asked how they would respond to this situation. A first study among college students showed that the hypothesized sex difference clearly emerged: jealousy in men was in particular influenced by the rival’s dominance (especially when the rival was unattractive), whereas jealousy in women was in particular influenced by the rival’s physical attractiveness (Dijkstra & Buunk, 1998).

Nature of the Rival Evaluation Mechanism Although these last findings are in line with evolutionary predictions, they may reflect at least two different types of adaptive mechanisms. One possibility is that there are sex-specific rival oriented mechanisms; that is, as a result of intrasexual competition, males and females have evolved to be sensitive to sex-specific rival characteristics. A second possibility is that males and females possess a general partner-oriented mechanism, i.e., a sensitivity to what one’s partner and one’s potential partners may find attractive in a mate. A homosexual sample offers the opportunity to examine the validity of both interpretations, because the two perspectives lead to different predictions. A sex-specific rival oriented mechanism would be expected on the basis of the reasoning put forward by Symons (1979) that homosexuals have the same set of sexual mental mechanisms as do heterosexuals, except for the sex of their sex mates. The notion of modularity (Kenrick, Keefe, Bryan, Barr, & Brown, 1995) elaborates on Symons’s assumption by stating that different psychological processes involved in reproduction, such as sexual orientation, mate preferences, and jealousy, are controlled by a number of independent mechanisms. Therefore, a change in sexual orientation would not necessarily alter the rival characteristics that evoke jealousy, suggesting that the same sex differences in rival characteristics—social dominance being more salient among men and physical attractiveness being more salient among women—


would emerge in homosexuals as in heterosexuals. In contrast, a general partneroriented mechanism would imply that such sex differences would depend on whether individuals are heterosexual or homosexual. That is, differences between gay men and lesbian women in the rival characteristics that evoke jealousy would reflect differences in the characteristics that gay men and lesbian women, respectively, value in a mate. A number of studies have found that gay men’s mate preferences are in general rather similar to those of heterosexual men (e.g., Bailey, Gaulin, Agyei, & Gladue, 1994; Kenrick et al., 1995). That is, like heterosexual men, gay men show little interest in a potential partner’s status but show high interest in a potential partner’s physical attractiveness (see also Symons, 1979). However, lesbian women seem to have, compared with heterosexual women, a more masculine pattern of mating psychology. For instance, compared to heterosexual women, lesbian women seem more interested in younger partners (Jankowiak, Hill, & Donovan, 1992), and less concerned with their partner’s status (Bailey et al., 1994). Given these mate preferences among homosexual men and women, the existence of a general partner-oriented mechanism would be supported when jealousy in homosexual men would be evoked more by a rival’s physical attractiveness than a rival’s dominance, whereas jealousy in lesbian women would be evoked more by both a rival’s physical attractiveness and a rival’s dominance. Using the same paradigm as in the Dijkstra and Buunk (1998) study, a study showed clear support for the existence of a sex-specific rival mechanism. That is, lesbian women, but not gay men, reported more jealousy when they were exposed to a physically attractive rival as compared to a physically unattractive rival. Gay men, but not lesbian women, reported more jealousy when they were exposed to a rival high in dominance as compared to a rival low in dominance, and, like among heterosexual men, especially when exposed to a physically unattractive rival. Thus, these findings strongly suggest that males and females possess an evolved mechanism through which they respond more or less automatically to those rival characteristics that have been important in intrasexual competition in our evolutionary past.

Subliminal Perception of Rival Characteristics One potential problem with both the correlational and experimental studies we described here is demand characteristics. That is, participants may have had theories about research hypotheses and may have responded accordingly. In order to circumvent this problem, we have conducted a new series of experiments in which we presented participants subliminally with rival characteristics. Given the importance of rival evaluation for reproductive success, it seems plausible that sensitivity to rival characteristics has evolved in such a way that these characteristics may be perceived even outside of conscious awareness. The social cognition literature suggests that unobtrusively presenting participants with certain cues may nonconsciously influence their evaluations of others (e.g., Wegner



& Bargh, 1998; see also Kenrick, Delton, Robertson, Vaughn Becker, & Neuberg, chapter 4, this volume, for examples of social cognitive approaches of mating and related phenomena). Directly relevant to the present issue, recent research suggests that people may make social comparisons with these targets that are presented subliminally either in the form of photographs of well-known people (e.g., Stapel & Blanton, 2004), or in the form of names of well-known people (e.g., Mussweiler, Ruter, & Epstude, 2004). Applying such findings to rival evaluation, we hypothesized that the mere exposure to rival characteristics through subliminal priming would induce comparison between oneself and the rival literally in the blink of an eye, and that the degree of jealousy would be based on the outcome of this comparison. In the first study using subliminal priming (Massar & Buunk, 2005a), participants were asked to indicate as quickly as possible whether two neutral words presented on the screen were related to each other by pressing one of two colored keys on the keyboard. The words in this “association task” had no relation to rival characteristics, but were neutral words like “house” and “garden”. In between these two neutral words, participants were subliminally exposed to rival characteristics. To ensure that participants would relate the rival characteristics to another individual, and not to themselves, each word was preceded by an implicitly presented personal pronoun, “he” for the men and “she” for the women. The rival characteristics were those that were in a preliminary study most often mentioned when men and women were asked to generate words relating to attractiveness and social dominance. For the attractiveness condition, these were “pretty,” “beautiful,” “slender,” and “sexy” (these are imperfect translations of Dutch words that apply equally to men and women), and for the social dominance condition, “tough,” “money,” “power,” and “success.” Each word was presented five times, making a total of 20 trials. A trial would consist of a neutral word (presented for 1.5 s), a personal pronoun (17 ms), a rival characteristic (17 ms), and then another neutral word (1.5 s). After completing the association task, participants read a shortened version of the jealousy scenario used in the studies described above, and were asked how jealous they would be in such a situation. The results showed that subliminal priming in this context clearly had the hypothesized effects, albeit only for individuals with a high mate value. These participants obviously differentiated between the rival characteristics: Women reported more jealousy following exposure to attractiveness words than to social dominance words, and men reported more jealousy following exposure to social dominance words than to attractiveness words. Participants with a low mate value reported more overall jealousy, independently of the characteristics of the rival. From an evolutionary point of view, this seems adaptive, because individuals with a low mate value have fewer options on the mating market, and they may easily be surpassed by even a relatively unattractive rival. In contrast, people with a high mate value will not feel threatened by an undesirable rival, but only by a rival who is better than they are (e.g., more physically attractive for women


and more socially dominant for men). Thus, this study established for the first time that it is possible to induce jealousy in participants through subliminal presentation of sex-specific rival characteristics. Our findings suggest that unconsciously linking certain features to a third person may lead to “projecting” these characteristics onto a rival that is described without any characteristics in a scenario.

Influence of the Ovulatory Cycle In a subsequent experiment, using the same paradigm, we examined the extent to which the jealousy evoking effects of rival characteristics depended on the ovulatory cycle of the woman. During the fertile period of a woman’s cycle, the presence of a physically attractive rival might be an especially large threat to the relationship for women as in this period they would desire the exclusive sexual and emotional attention of their mate. Previous research has indeed established that during the time of high fertility risk, women tend to be more prone to feelings of jealousy, and are especially sensitive to cues of emotional infidelity (e.g., Gaulin, Silverman, Phillips, & Reiber, 1997; see also Simpson & LaPaglia, chapter 10, this volume). Moreover, for men the cycle of their mates matters as well: They tend to be more attentive and proprietary during the fertile phases of their girlfriends’ ovulatory cycle (Gangestad, Thornhill, & Garver, 2002). Therefore, it can be expected that, for men, the presence of a socially dominant rival would be especially threatening during the fertile phases of their partner’s ovulatory cycle. The results from this study, which employed the same paradigm as the previous study (e.g., participants were subliminally primed with words relating to rival characteristics), showed that women in the fertile phase of their cycle did indeed report more jealousy than women in the nonfertile phase of their cycle, and did report especially more jealousy after exposure to a physically attractive rival than after exposure to a socially dominant rival. We also found some preliminary evidence that men whose girlfriends were in the fertile phase of their menstrual cycle at the time of the experiment reported more jealousy after exposure to a socially dominant rival than after exposure to a physically attractive rival (Massar & Buunk, 2005b). Thus, these results show that ovulatory cycle apparently has a strong effect on the sensitivity to intrasexual competition as it affects how men and women respond to subliminally presented rival characteristics.

BODY BUILD Not only facial attractiveness, but also body attractiveness may be an important cue individuals pay attention to when confronted with a rival. Indeed, Dijkstra and Buunk (2002) found that features like more beautiful legs, a better figure, a



more attractive body, more beautiful hips, and a lighter and more slender body build were spontaneously mentioned by participants as quite important rival characteristics. Many studies have shown that the body is at least as an important determinant of physical attractiveness as the face, particularly when individuals are observed from a distance (e.g., Alicke, Smith, & Klotz, 1986). A low waistto-hip ratio (WHR) is a particularly important feature of female attractiveness that is independent of weight (e.g., Singh, 1993). The ultimate reason that a low WHR is perceived as attractive is that it is actually associated with health and fertility (but see, for example, Wetsman & Marlowe, 1999, for contrary evidence). Given the importance of WHR for female attractiveness, we expected that, especially among women, rivals with a favorable WHR would evoke more jealousy than rivals with an unfavorable WHR. We used the stimuli developed by Singh (1993) that manipulate the rival’s WHR by varying the size of the waist. However, with this procedure, one unintentionally manipulates a rival’s degree of body taper as well: As the waist narrows, not only does the WHR decrease, but also the body taper seems to increase. This is quite relevant as there is considerable evidence that body taper is a more important determinant of male than of female physical attractiveness (e.g., Franzoi & Herzog, 1987), presumably because it reflects a man’s level of physical dominance, a feature highly valued by women, but not by men in a mate. There is indeed evidence that the pelvic-toshoulder ratio correlates positively with beta-lipoproteins, hormones that are related to testosterone levels and muscle development in men (e.g., Evans, 1972). We manipulated body taper by varying the shoulder-to-hip ratio (SHR). While we expected rivals with lower WHRs to evoke relatively more jealousy in females than in males, we expected rivals with higher SHRs to evoke relatively more jealousy in males than in females. We also asked participants which body parts they had paid attention to while evaluating the rivals. In a first study with this paradigm (Dijkstra & Buunk, 2001), we presented a sample of students with line drawings (derived from the work by Singh, 1993) of individuals of the same sex as themselves. The drawings had identical facial and bodily features and only differed in the size of their WHR and SHR. The results showed that rivals with a low as opposed to a high WHR evoked indeed more jealousy in women than in men. In contrast, rivals with a high as opposed to low SHR evoked more jealousy in men than in women, particularly when the rival had a high WHR. In evaluating the rivals, women indicated that they had paid more attention to the rivals’ waist, hips, and legs, and men indicated that they had paid more attention to the rivals’ shoulders, chest, and belly.

Role of Life History According to life history theory, men may follow two important strategies to achieve reproductive success (e.g., Hill & Hurtado, 1996): a strategy of physical dominance—an elevated social rank achieved by physical competition—or a


strategy of eminence—an elevated rank achieved through socially approved accomplishments (Kemper, 1990). Physical dominance contributes especially to the mate value of young men who are at their peak with regard to health and fitness, whereas eminence will peak as men get older. Because men most often will be confronted with rivals of approximately the same age (due to women’s preference for males who are only slightly older than themselves; Kenrick & Keefe, 1992), it was expected that as men get older, a rival’s SHR will play a less important role in evoking jealousy. In contrast, it was expected that a rival’s WHR will continue to evoke jealousy among women as they get older, particularly because, regardless of age, men tend to prefer women who signal health, youth, and fertility (e.g., Buunk, Dijkstra, Kenrick, & Warntjes, 2001; Kenrick & Keefe, 1992). In a study in a community sample in which we used the same method as in the previous study among students (Buunk & Dijkstra, 2005), we found that, as predicted, as men were older, the SHR of the rival was a less important factor in evoking jealousy, whereas among women jealousy in response to the rival’s WHR was not affected by age. Moreover, the rival’s SHR was also a more important determinant of perceptions of social and physical dominance as men were younger. Remarkably, among men, the low WHR–low SHR rival, that is, the rival with a linear and slender body build, evoked the highest level of jealousy and was perceived as the most attractive and the most socially dominant of all rivals, probably because this type of rival is perceived by adult men as having the highest level of eminence. Indeed, there is some evidence that men with a linear and slender body build attain a higher occupational level (Deabler, Hartl, & Willis, 1975), and that an athletic and muscular body build is associated with lower impulse control and higher competitive aggressiveness, which may not be conducive to attaining a high position in the societal hierarchy (for a review see Kemper, 1990).

Subliminal Perception of Body Build The previous studies used a within-subjects design in which participants viewed the various figures simultaneously, which may have evoked demand characteristics as participants may have had theories about the research hypotheses and may respond accordingly. Therefore, in a subsequent experiment (Massar & Buunk, 2005b), we examined whether individuals may perceive the figure of the rival automatically and unconsciously. In this experiment, a parafoveal priming procedure was used, whereby the primes were presented in the periphery of the attended region (Bargh & Chartrand, 2000). The primes were for male participants a figure with either a high shoulder-to-hip ratio (e.g., an attractive body shape, indicating social dominance), or a low shoulder-to-hip ratio (an unattractive body shape), and for female participants a figure with either a low WHR (a physically attractive body shape), or a high WHR (unattractive body shape). They were told to focus on the “X” in the centre of the screen and to indicate as fast as possible on which side of the “X” they saw a flash by pressing a



key on the keyboard. The prime (the line drawing) was randomly presented for 60 ms in one of the four parafoveal regions, and was then immediately masked. A random delay between the primes was inserted to avoid an anticipated response by participants. Moreover, the primes were alternated with neutral pictures of geometrical shapes. All in all, participants were exposed to 64 trials, 16 of which consisted of the primes. After the priming procedure, the shortened version of the jealousy evoking scenario was presented to the participants and jealousy was measured with a slider on the computer screen. The results from this experiment were in line with the results from the Dijkstra and Buunk (2001) study: Males reported significantly more jealousy after subliminal exposure to the figure with the attractive body shape (high SHR) than after exposure to the figure with the unattractive body shape (low SHR). Apparently, the men in this study were able to detect another man’s body shape without being aware of it. We did not find an effect among females. These findings suggest that assessing the body shape of one’s rival is a more automatic mechanism for males than for females, which may be due to the evolutionarily more ancient, and therefore more automatic nature of male jealousy.

SEXUAL VERSUS EMOTIONAL INFIDELITY One might argue that even automatic gender differences in the importance attached to the dominance and attractiveness of rivals do not necessarily reflect evolved differences, but may simply be due to cultural learning. The validity of an evolutionary perspective would be particularly strengthened if we could define on the basis of such a perspective conditions under which the opposite sex difference would occur. An evolutionary approach may make very specific predictions concerning the way in which sex differences may depend on the context, in the present case the type of threat implied by the infidelity. Because men, but not women, have faced in the course of evolution the problem of uncertainty with regard to their genetic relation to their offspring, male jealousy may have evolved primarily as a mechanism to prevent one’s partner’s sexual involvement with another man (Buss, Larsen, Westen, & Semmelroth, 1992; Wilson & Daly, 1992). In contrast, because for women, more than for men, their partner’s infidelity might imply the risk of losing the partner’s investment and resources, female jealousy may have evolved primarily as a mechanism to prevent one’s partner emotional involvement with another woman (Buss et al., 1992). A series of studies in the United States, the Netherlands, China, Germany, Korea, Sweden, and Japan has found that when asked to choose what they find most upsetting, more men than women do indeed find sexual infidelity of their partner more upsetting, whereas more women than men find emotional infidelity of their partner more upsetting (e.g., Bailey et al., 1994; Buss et al., 1992; Buunk, Angleitner, Oubaid, & Buss, 1996; Harris & Christenfeld, 1996). In addition, participants are also more physiologically upset, as measured by heart


rate, electrodermal response, and corrugator supercilii contraction, in line with the predicted gender difference (see also Pietrzak, Laird, Stevens, & Thompson, 2002), although these physiological data could not later be replicated by either Grice and Seely (2000) or by Harris (2000). It must be noted, however, that the gender difference may not occur when rating scales instead of a forced-choice paradigm are used, when personal experiences with a partner’s actual infidelity are recalled, when individuals are under cognitive constraint, or when individuals do have experience with infidelity (e.g., DeSteno, Bartlett, Braverman, & Salovey, 2002; Sagarin, Becker, & Guadagno, 2003). Furthermore, men are mostly equally split when it comes to choosing which type of infidelity they find the most upsetting.

Type of Infidelity and Rival Characteristics Despite these empirical inconsistencies, we reasoned that sexual and emotional infidelity will evoke qualitatively different types of affective responses (e.g., Parrot & Smith, 1993). That is, in the case of a partner’s emotional infidelity, anxiety and insecurity due to a threat to the primary relationship will become salient (Buunk, 1997), whereas in the case that extradyadic sex has already occurred, individuals will tend to respond with anger and betrayal to an extradyadic affair of one’s partner, in particular when the infidelity is perceived as undeserved or unfair (Buunk, 1995; Parrot, 1991). More importantly, we assumed that the type of infidelity may determine which rival characteristics will most strongly affect these emotional responses. Under conditions of “pure” emotional infidelity, rivals will be evaluated more as potential threats to the relationship, and the jealousy-evoking effect of rival characteristics may strongly reflect the importance of long-term partner preferences. However, when confronted with unequivocal sexual infidelity without the potential of the development of an emotional attachment, gender differences in the characteristics that evoke jealousy may be quite different, and even the opposite. Ultimate motivations for why women engage in extradyadic sex include the acquisition of “good genes” that would increase offspring quality, and the acquisition of “sexy sons” genes that would increase a son’s chance of reproductive success (e.g., Barash & Lipton, 2001; Gangestad, Simpson, Cousins, Garver-Apgar, & Christensen, 2004). All of these potential benefits would be served by having sex with physically attractive men because a man’s physical appearance is the only quick way to assess the quality of his genes (e.g., Buss, 1994). Therefore, in the case of “pure” sexual infidelity, men will be particularly attentive to the attractiveness of the rival rather than to his social dominance or status. Because for women sexual infidelity as such does from an evolutionary point of view not pose a threat to a woman’s reproductive success, for them it is relatively unimportant who the rival is in a purely sexual fling of their partners. In our study (Buunk & Dijkstra, 2004), we exposed men and women to the scenario used in our previous studies. However, the scenario was expanded in



that it indicated that one lost track of one’s partner, and that the next day the partner indicated to have had a very intense and special sexual experience (sexual infidelity condition), or unique and special communication and connection (emotional infidelity). The results showed indeed that jealousy evoked by emotional infidelity was primarily characterized by feelings of threat, and jealousy evoked by sexual infidelity was primarily characterized by feelings of betrayal and anger (cf. Ellsworth, chapter 5, this volume). More importantly, and as predicted, following emotional infidelity, in men a rival’s dominance, and in women a rival’s physical attractiveness, evoked feelings of threat (but not of angerbetrayal). In contrast, after sexual infidelity, in men, but not in women, a rival’s physical attractiveness evoked feelings of betrayal-anger (but not of anxiety or suspicion). Thus, this study showed that the gender differences found in many of our studies are confined to “pure” emotional infidelity, and that in the case of “pure” sexual infidelity, the sex difference is in part reversed, with men, and not women, responding with more jealousy to physically attractive rivals. This latter finding reflects the importance of physical attractiveness as an attribute for women in the context of casual sexual affairs (Barash & Lipton, 2001; Gangestad et al., 2004).

Subliminal Activation of Context In a related experiment, we examined whether subliminally presenting sexrelated versus intimacy-related words would affect responses to different types of rivals, and whether this would depend on one’s sex drive. We reasoned that for individuals with a high sex drive, activating sex-related constructs will make intrasexual competition particularly salient, whereas for individuals with a low sex drive, activating intimacy-related constructs will make intrasexual competition particularly salient. That is, someone with a high sex drive will be more oriented to having various sex partners, whereas someone with a low sex drive will be more oriented toward developing a committed intimate relationship with a single partner. In this experiment, participants were subliminally primed with words related either to sex (sex, passion, making out, and aroused ) or with words relating to intimacy (warmth, intimate, attached, and committed ). After the priming procedure, they were told to imagine their partner coming home one day and telling them “I found someone else.” Next, they indicated, among others, how upset they would be if this “other person” had better career prospects than they had and was more attractive than they were. The results showed that males and females with a high sex drive reported feeling more upset than men with a low sex drive over the rival, but only when they had been primed with sex. These results suggest that individuals with a high sex drive are more prone to engage in intrasexual competition, especially when they are confronted with the sexual infidelity of their partner (Massar & Buunk, 2006).


CONCLUSION Although some people like to believe that, despite their biological differences, men and women are essentially the same, according to evolutionary psychology men and women do not only have different bodies, they also have different minds. This is in particular due to the fact that, during their life, women produce only a limited amount of eggs, whereas men produce billions of sperms. This has led men and women to make essentially different investments in their offspring, producing different adaptive problems for men and women (Buss, 1994; Miller, 2000). In this chapter we presented a series of results from our program that examined the consequences from these different adaptive problems for intrasexual competition. Using both descriptive and experimental methods as well as various kinds of stimuli material, our findings show that there are consistent gender differences in the jealousy-evoking effect of particular rival characteristics. Most of our findings sing a single song; that is, whereas jealousy in women is evoked more than in men by a rival’s physical attractiveness, jealousy in men more than in women is evoked by a rival’s status and dominance related features. This type of gender difference was found when physical attractiveness was defined as general attractiveness, facial attractiveness, and waist-to-hip ratio, and when status- and dominance-related features were defined as social dominance, physical dominance, shoulder-to-hip ratio, and social status. The finding that women are particularly sensitive to the physical attractiveness of the rival may also help explain why Kenrick et al. (chapter 4, this volume) found that primarily women who were involved in committed relationships overestimated the number of attractive women. Of course, one might argue that such sex differences mainly reflect culturally learned norms concerning what is appropriate for each sex. However, this explanation appears to fall short in explaining an increasing and diverse number of findings. First, among lesbian women and gay men the same sex differences were found, suggesting that, overall, males and females possess an evolved mechanism through which they respond more or less automatically to those rival characteristics that have been important in sexual selection in our evolutionary past, even when, as in the case of homosexuals, this mechanism does not parallel those characteristics that, given the mate preferences of their partners, constitute the largest threat. Second, we found in line with evolutionary reasoning, for males a reversal of the importance of dominance versus physical attractiveness as an important rival characteristic in the case of sexual infidelity. That is, when confronted with an intense single sexual contact of one’s partner with a rival, males responded with more jealousy to a physically attractive rather than to a socially dominant rival, paralleling precisely what women find important in short-term mating. Third, we are now obtaining increasing evidence that men and women respond differently to subliminal cues of rival characteristics, and do so in a similar way as to explicit descriptions of rival characteristics. Our findings seem to suggest that the evaluation of rivals in a romantic jealousy situation is a basic mechanism that may function



unconsciously and automatically, and is affected by factors that are relevant from an evolutionary perspective, such as mate value and fertility of the female. In conclusion, our research program is unraveling in more and more detail how the male and female minds are made up to pay attention in different ways to specific rival characteristics.

REFERENCES Alicke, M. D., Smith, R. H., & Klotz, M. L. (1986). Judgements of physical attractiveness: The role of faces and bodies. Personality and Social Psychology Bulletin, 12, 381–389. Bailey, J. M., Gaulin, S., Agyei, Y., & Gladue, B. A. (1994). Effects of gender and sexual orientation on evolutionary relevant aspects of human mating. Journal of Personality and Social Psychology, 66, 1081–1093. Barash, D. P., & Lipton, J. E. (2001). The myth of monogamy: Fidelity and infidelity in animals and people. New York: Freeman. Bargh, J. A., & Chartrand, T. L. (2000). The mind in the middle. A practical guide to priming and automaticity research. In H. T. Reis & C. M. Judd (Eds.), Handbook of research methods in social and personality psychology (pp. 253–285). New York: Cambridge University Press. Buss, D. M. (1989). Sex differences in human mate preferences: Evolutionary hypotheses tested in 37 cultures. Behavioral and Brain Sciences, 12, 1–49. Buss, D. M. (1994). The evolution of desire: Strategies of human mating. New York: Basic Books. Buss, D. M., Larsen, R. J., Westen, D., & Semmelroth, J. (1992). Sex differences in jealousy: Evolution, physiology, and psychology. Psychological Science, 3, 251–255. Buss, D. M., Shackelford, T. K., Choe, J., Dijkstra, P., & Buunk, B. P. (2000). Distress about mating rivals. Personal Relationships, 7, 235–243. Buunk, B. P. (1986). Husband’s jealousy. In R. A. Lewis & R. E. Salt (Eds.), Men in families (pp. 97–114). Beverly Hills, CA: Sage. Buunk, B. P. (1995). Sex, self-esteem, dependency and extradyadic sexual experiences as related to jealousy responses. Journal of Social and Personal Relationships, 12, 147–153. Buunk, B. P. (1997). Personality, birth order

and attachment styles as related to various types of jealousy. Personality and Individual Differences, 23, 997–1006. Buunk, B. P., Angleitner, A., Oubaid, V., & Buss, D. M. (1996). Sex differences in jealousy in evolutionary and cultural perspective: Test from the Netherlands, Germany, and the United States. Psychological Science, 7, 359– 363. Buunk, B. P., & Dijkstra, P. (2000). Extradyadic relationships and jealousy. In C. Hendrick & S. S. Hendrick (Eds.), Close relationships: A sourcebook (pp. 317–329). Thousand Oaks, CA: Sage. Buunk, B. P., & Dijkstra, P. (2001). Evidence from a homosexual sample for a sex-specific rival-oriented mechanism: Jealousy as a function of a rival’s physical attractiveness and dominance. Personal Relationships, 8, 391–406. Buunk, B. P., & Dijkstra, P. (2004). Gender differences in rival characteristics that evoke jealousy in response to emotional versus sexual infidelity. Personal Relationships, 11, 395– 408. Buunk, B. P., & Dijkstra, P. (2005). A narrow waist versus broad shoulders: Sex and age differences in the jealousy-evoking characteristics of a rival’s body build. Personality and Individual Differences, 39, 379–389. Buunk, B. P., Dijkstra, P., Kenrick, D. T., & Warntjes, A. (2001). Age preferences for mates as related to gender, own age and involvement level. Evolution and Human Behavior, 22, 241–250. Buunk, A. P. & Gibbons, F. X. (2005). Social comparison orientation: A new perspective on those who do, and those who don’t compare with others. In S. Guimond (Ed.), Social comparison and social psychology: Understanding cognition, intergroup relations and culture (pp. 15–32). Cambridge, UK: Cambridge University Press.


Campbell, A. (2002). A mind of her own: The evolutionary psychology of women. Oxford, UK: Oxford University Press. Deabler, H. L., Hartl, E. M., & Willis, C. A. (1975). Physique and personality: Somatotype and vocational interest. Perceptual and Motor Skills, 41, 382. DeSteno, D. A., Bartlett, M. Y., Braverman, J., & Salovey, P. (2002). Sex differences in jealousy: Evolutionary mechanism or artifact of measurement? Journal of Personality and Social Psychology, 83, 1103–1116. Dijkstra, P., & Buunk, B. P. (1998). Jealousy as a function of rival characteristics: An evolutionary perspective. Personality and Social Psychology Bulletin, 24, 1158–1166. Dijkstra, P., & Buunk, B. P. (2001). Gender differences in the jealousy-evoking nature of a rival’s body build. Evolution and Human Behavior, 22, 335–341. Dijkstra, P., & Buunk, B. P. (2002). Sex differences in the jealousy-evoking effect of rival characteristics. European Journal of Social Psychology, 32, 829–852. Evans, R. B. (1972). Physical and biochemical characteristics of homosexual men. Journal of Consulting and Clinical Psychology, 39, 140– 147. Franzoi, S. L., & Herzog, M. E. (1987). Judging physical attractiveness: What body aspects do we use? Personality and Social Psychology Bulletin, 13, 19–33. Gangestad, S. W., Simpson, J. A., Cousins, A. J., Garver-Apgar, C. E., & Christensen, P. N. (2004). Women’s preferences for male behavioral displays change across the menstrual cycle. Psychological Science, 15, 203–207 Gangestad, S. W., Thornhill, R., & Garver, C. E. (2002). Changes in women’s sexual interests and their partners’ mate retention tactics across the menstrual cycle: Evidence for shifting conflicts of interest. Proceedings of the Royal Society of London, Series B, 269, 975– 982. Gaulin, S. J. C., Silverman, I., Phillips, K., & Reiber, C. (1997). Activational hormone influences on abilities and attitudes: Implications for evolutionary theory. Evolution and Cognition, 3, 191–199. Grice, J. W., & Seely, E. (2000). The evolution of sex differences in jealousy: Failure to replicate previous results. Journal of Research in Personality, 34(3), 348–356.

Harris, C. R. (2002). Sexual and romantic jealousy in heterosexual and homosexual adults. Psychological Science, 13, 7–12. Harris, C. R., & Christenfeld, N. (1996). Jealousy and rational responses to infidelity across gender and culture. Psychological Science, 7, 378–379. Hill, K., & Hurtado, A. M. (1996). Ache life history. New York: Aldine de Gruyter. Hrdy, S. B. (1981). The woman that never evolved. Cambridge, MA: Harvard University Press. Hupka, R. B., Otto, J., Tarabrina, N. V., & Reidl, L. (1993). Cross-cultural comparisons of nouns associated with jealousy and the related emotions of envy, anger, and fear. Cross Cultural Research: The Journal of Comparative Social Science, 27, 181–211. Jankowiak, W. R., Hill, E. M., & Donovan, J. M. (1992). The effect of sex and sexual orientation on attractiveness judgements: An evolutionary interpretation. Ethology and Sociobiology, 13, 73–85. Kemper, D. (1990). Social structure and testosterone. London: Rutger University Press. Kenrick, D. T., & Keefe, R. C. (1992). Age preferences in mates reflect sex differences in human reproductive strategies. Behavioral and Brain Sciences, 15, 75–133. Kenrick, D. T., Keefe, R. C., Bryan, A., Barr, A., & Brown, S. (1995). Age preferences and mate choice among homosexuals and heterosexuals. Journal of Personality and Social Psychology, 69, 1166–1172. Massar, K., & Buunk, B. P. (2005a). Jaloezie vanuit een evolutionair perspectief: onbewuste waarneming van rivaalkenmerken [Jealousy from an evolutionary perspective: Unconscious perception of rival characteristics]. Tijdschrift voor Criminologie, 47, 188– 192. Massar, K., & Buunk, A. P. (2005b, June). Subliminal rivals: Unconscious processes in jealousy and rival evaluation. Paper presented at the annual meeting of the Human Behavior and Evolution Society (HBES), Austin, TX. Massar, K., & Buunk, A. P. (2006, June). Sex drive, infidelity type and intrasexual competition: A subliminal priming paradigm. Poster presented at the annual meeting of the Human Behavior and Evolution Society (HBES), Philadelphia, PA.


228 EVOLUTION AND THE SOCIAL MIND Mathes, E. W., Adams, H. E., & Davies, R. M. (1985). Jealousy: Loss of relationship rewards, loss of self-esteem, depression, anxiety, and anger. Journal of Personality and Social Psychology, 48, 1552–1561. Miller, G. (2000). The mating mind: How sexual choice shaped the evolution of human nature. London: Heinemann. Mussweiler, T., Ruter, K., & Epstude, K. (2004). The man who wasn’t there: Subliminal social comparison standards influence self-evaluation. Journal of Experimental Social Psychology, 40, 689–696. Nadler, A., & Dotan, I. (1992). Commitment and rival attractiveness: Their effects on male and female reactions to jealousy arousing situations. Sex Roles, 26, 293–310. Parrot, W. G. (1991). The emotional experience of envy and jealousy. In P. Salovey (Ed.), The psychology of jealousy and envy (pp. 3–30). New York: Guilford Press. Parrot, W. G., & Smith, R. H. (1993). Distinguishing the experiences of envy and jealousy. Journal of Personality and Social Psychology, 64, 906–920. Pietrzak, R. H., Laird, J. D., Stevens, D. A., & Thompson, N. S. (2002). Sex differences in human jealousy: A coordinated study of forced-choice, continuous rating-scale, and physiological responses on the same subjects. Evolution and Human Behavior, 23, 83–94. Sagarin, B. J., Becker, D. V., & Guadagno, R. E. (2003). Sex differences (and similarities) in jealousy: The moderating influence of infidel-

ity experience and sexual orientation of the infidelity. Evolution and Human Behavior, 24, 17–23. Shettel-Neuber, J., Bryson, J. B., & Young, L. E. (1978). Physical attractiveness of the “other person” and jealousy. Personality and Social Psychology Bulletin, 44, 612–615. Singh, D. (1993). Adaptive significance of female physical attractiveness: Role of waistto-hip-ratio. Journal of Personality and Social Psychology, 65, 293–307. Stapel, D. A., & Blanton, H. (2004). From seeing to being: Subliminal social comparisons affect implicit and explicit self-evaluations. Journal of Personality and Social Psychology, 87, 468–481. Symons, D. (1979). The evolution of human sexuality. Oxford, UK: Oxford University Press. Wegner, D., & Bargh, J. A. (1998). Control and automaticity in social life. In D. T. Gilbert, S. T. Fiske, & G. Lindzey (Eds.), The handbook of social psychology (Vol. 1, pp. 446–496). Boston: McGraw-Hill. Wetsman, A., & Marlowe, F. (1999). How universal are preferences for female waist-to-hip ratios? Evidence from the Hadza of Tanzania. Evolution and Human Behavior, 20, 219– 228. Wilson, M., & Daly, M. (1992). The man who mistook his wife for chattel. In J. Barkow, L. Cosmides, & J. Tooby (Eds.), The adapted mind: Evolutionary psychology and the generation of culture (pp. 289–322). New York: Oxford University Press.


Cognitive and Social Adaptations for Leadership and Followership Evolutionary Game Theory and Group Dynamics


Leadership and Followership in an Evolutionary Framework An Evolutionary Game Analysis of Leadership Non-human Evidence for Leadership Leadership in Humans Discussion nthropologists have identified leadership as a human universal (Boehm, 1999; Brown, 1991). In his analysis of the anthropological evidence for a range of universal human social behaviours, Brown (1991)—using his fictitious Universal People (UP) as a vehicle—says that:


[t]he UP have leaders, though they may be ephemeral or situational. The UP admire, or profess to admire, generosity and this is particularly desired in a leader. No leader of the UP ever has complete power lodged in himself alone. UP leaders go beyond the limits of UP reason and morality. Since the UP never have complete democracy, and never have complete autocracy, they always have a de facto oligarchy. (p. 138)

This description resonates with the literature in social psychology, which suggests that whenever individuals come together to form a group, a leadership 229


structure quickly emerges (Bass, 1954; van Vugt, 2006). However, despite leadership’s crucial role in social groups and the enormity of the research directed at understanding it, there exists no overarching theoretical structure that organizes the wealth of thought and empirical evidence gathered on this topic (Chemers, 2000; Yukl, 1989). As Hogan and Kaiser (2005) note: The academic tradition [in the study of leadership] is a collection of dependable empirical nuggets, but it is also a collection of decontextualized facts that do not add up to a persuasive account of leadership. (p. 171)

Here, we present a set of ideas grounded in the theory of evolution (Darwin, 1859) in a tentative attempt to integrate the massive body of empirical data into a coherent conceptual framework. (For a set of related ideas and a more thorough review, see van Vugt, 2006.)

LEADERSHIP AND FOLLOWERSHIP IN AN EVOLUTIONARY FRAMEWORK Leadership and followership have been defined in a great many ways in the literature (Bass, 1990). Two common construals of leadership are, first, as an individual difference variable (Stogdill, 1974) and, second, the outcome of strategic interactions among rational actors (Hollander, 1985). With deference to the thought that has gone into the research traditions from which these definitions emerge, we introduce here a set of definitions that diverge from these and are, unlike their predecessors, decidedly adaptationist in nature. In particular, we take leadership and follower behaviour to be the product of cognitive adaptations designed to solve adaptive problems humans faced throughout evolutionary history. In this case, these problems are associated with particular features of humans’ social and physical environment (Dunbar, chapter 2, this volume; Kenrick, Delton, Robertson, Vaughn Becker & Neuberg, chapter 4, this volume; Tooby & Cosmides, 1992). Leadership is defined here as design for inducing others to coordinate their actions or goals with those of the individual, the leader, to foster the leader’s proximate goals. Followership is defined as design to coordinate one’s actions or goals with those of another individual—the leader—in order to foster the leader’s proximate goals. There are a few important aspects to these definitions. First, because we define leadership and followership in terms of design, cases in which individuals accidentally coordinate their actions with one another are not included in the definition. For example, a broken-down car blocking the road forces other drivers to take a different route, but this is not leadership. The focus on design rather than behaviour correctly excludes “accidental” coordination and emphasizes that evidence for leadership and followership will be in the form of evidence of special design of the associated cognitive adaptations (cf. Tooby & Cosmides, 1992).


Second, rather than being a generic source of social influence like status, prestige or social dominance (Buunk, Massar & Dijkstra, chapter 13, this volume; Henrich & Gil-White, 2001; Simpson & LaPaglia, chapter 10, this volume), leadership involves specifically the solution to a coordination problem (Cartwright & Zander, 1968; Schelling, 1960; van Vugt, 2006). To illustrate, by virtue of his contributions, Charles Darwin is a person of great esteem, yet it would stretch our definition past its limits to suggest that his scientific endeavours were driven by adaptations designed to coordinate others’ behaviour. Finally, the definition of followership suggests that there are adaptations designed to cause followers to adopt the goals of the leader. This can be as simple as, for example, following the leader to his or her preferred location. This does not, of course, mean that a follower is not executing a strategy that furthers that organism’s ultimate or proximate goals as well. Although many definitions from the psychological literature embody the assumption that the goals of leaders and followers necessarily converge into a single group goal (Chemers, 2000; Hogg, 2001), we include the possibility that organisms can simultaneously pursue multiple proximate goals. By furthering a leader’s proximate goals, followers can be fostering their own as well. This idea is clarified in the discussion of leadership using simple two-player coordination games, below. In certain respects, what is puzzling from an evolutionary perspective is followership. It has been argued that adaptations for striving to lead—to cause others to coordinate their actions with one’s goals—can evolve because there may be clear advantages associated with leading (van Vugt, 2006). Yet, given what is known about the process of evolution through natural selection, adaptations designed to adopt another organism’s goals stand in need of special explanation. Understanding followership adaptations is of as much, if not greater, theoretical interest than those surrounding leadership. It is, then, in some sense surprising that questions about the origins of followership are not normally posed in the psychological literature.

AN EVOLUTIONARY GAME ANALYSIS OF LEADERSHIP If leadership and followership adaptations evolved for the purpose of solving coordination problems amongst organisms, we should be able to model the evolution of these traits. Evolutionary game theory provides a useful tool (Maynard-Smith, 1982; see also Gangestad & Thornhill, chapter 3, this volume). Evolutionary game theory models social interactions as games in which strategies compete in a Darwinian fashion. In evolutionary game theory, the agents are genes which embody strategies that over the course of evolution are tested against alternative strategies and copies of themselves. Strategies (genes) spread through a population by virtue of the effect they have on their own replication rate as a result of their strategic interactions with other agents



(Dawkins, 1976). If we can model leadership and followership as strategies in a coordination game, then we should be able to examine how well they fare in terms of relative fitness. The simplest strategic interaction in this context is a coordination game involving two organisms, 1 and 2, who must decide between two alternatives. To make this concrete, consider the case in which the organisms must decide whether to go to waterhole A or B. Each organism is indifferent between the two waterholes, but both prefer to go to the same waterhole as the other. This coordination problem is characteristic of many social species that need to stay close to each other for safety (Dunbar, chapter 2, this volume). The pay-off structure of this game is depicted in Table 14.1. There are four cells in this game, each with two pay-offs representing the fitness outcomes of Players 1 and 2, respectively. If 1 and 2 go to Hole A or B together, they receive a pay-off of one unit. If they end up at different holes, they receive a zero pay-off. Selection will favour adaptations designed to execute equilibrium strategies (DeScioli & Kurzban, in press; Maynard Smith, 1982), and this game illustrates how followership adaptations might emerge. If the game is played sequentially, the first player chooses randomly and the second player’s best response is to choose the hole that the first player selected.1 The game is, in effect, sequential if organisms can signal their intentions to each other (see Gangestad & Thornhill, chapter 3, this volume); in this case, adaptations can be designed to reach the equilibrium outcomes. This situation selects for followership because it is in the interest of the organism to follow whoever moves first, adopting the goal of the first mover (i.e., which hole to go to), regardless of the hole that individual chooses. Note that this model makes no commitment regarding certain details of the leader/follower adaptations. In this example, it is easy to imagine that the leader/ follower role is adopted facultatively. That is, in some cases, an organism might find itself in a position to be the first mover in such a coordination game, and in other cases, might find itself better off being the follower in such a game. This is consistent with conditional strategies models (Simpson & LaPaglia, chapter 10, this volume; West-Eberhard, 2003), which assume that leader and follower roles are adopted flexibly by the same organisms. Table 14.1 Coordinating Leadership. A simple coordination game in which pay-offs (in reproductive success) are for Players 1 and 2 respectively. Equilibria are indicated with asterisks Player 2

Player 1

Hole A Hole B

Hole A

Hole B

1,1* 0,0

0,0 1,1*


It is, by the same token, not implausible that there might be adaptations in which pure strategies are coded in relatively static fashion, with the population reaching an equilibrium of varying frequencies of individuals, each of whom plays a pure strategy, sustained through frequency-dependent selection (Maynard-Smith, 1982; Wilson, Near & Miller, 1996). This distinction between conditional and pure strategies (the latter maintained via frequency-dependent selection) is analogous to the distinction in the psychological literature between state and trait accounts of leadership (van Vugt, 2006). A slightly more complex coordination game is depicted in Table 14.2. The pay-off structure resembles the game of Leader, one of four classic 2 × 2 games identified by Rapoport (1967). In this game, Players 1 and 2 benefit from going to the same waterhole, but Player 1 benefits more than Player 2 if they go to Hole A. Such an asymmetry in the coordination benefits can derive from any number of causes. For example, Player 1 could know the way to Hole A better than Hole B (Couzin, Krause, Franks & Levin, 2005). In any case, as in the pure coordination game, both players prefer to go to the same hole as the other player. However, each has a preference for a different hole. Unlike the first game, there is now an advantage to being the first mover. By taking the initiative, a player creates the incentive for the other player to adopt the first mover’s preferred hole. Once the first player has committed to one hole, the follower’s best response is to coordinate. The second mover profits from coordination, but not as much as the leader does. Moving first induces an iterated dominance game in which the second mover chooses the best outcome possible given the choice of the first (DeScioli & Kurzban, in press). As circumstances change, the ability of organisms to take the initiative and move first might vary. However, there might be stable differences between players that might make it more likely that pairs end up at Hole A in a series of interactions over time. For example, there might be individual differences in activity or energy levels, knowledge, size, power or dominance that might make one individual be more likely to emerge as leader in a given context (Couzin et al., 2005; see also Buunk et al., chapter 13, this volume). There is another possibility for the emergence of followership in this situation, suggested by multi-level selection theory (Sober & Wilson, 1998).

Table 14.2 Strategic Leadership. The Leader Game, in which pay-offs (in reproductive success) are for Players 1 and 2 respectively. Equilibria are indicated with asterisks Player 2

Player 1

Hole A Hole B

Hole A

Hole B

3,1* 0,0

0,0 1,3*



Although followers do less well than leaders in the Leader Game, it is also clear that aggregate pay-offs are higher in the Leader Game when there is coordination. Groups with a leader–follower structure (i.e., where the pair ends up coordinating at the same hole) have higher aggregate fitness (i.e., where the pair ends up at a different hole). Thus, there could, plausibly, be a between-group selection pressure. Under the right conditions—discussed at length elsewhere (Sober & Wilson, 1998)—it is plausible that between group fitness differentials would constitute pressure against which natural selection could act. We take no strong position on this issue, but merely point out that gains from coordination lead to potentially interesting multi-level selection dynamics (see Kurzban & Aktipis, 2006, for a brief discussion of finding evidence for the action of multi-level selection). Finally, consider the game depicted in Table 14.3. For Player 1, choosing Hole A is a dominant strategy because the pay-offs associated with Hole A are always higher, independent of Player 2’s choice. Because of this pay-off structure, Player 2 cannot change Player 1’s incentive such that Player 1’s best response will be to choose Hole B. Player 2’s best response, therefore, is to “make the best of a bad situation” (Dawkins, 1976; Simpson & LaPaglia, chapter 10, this volume), and choose Hole A. In this game, moving first or communicating one’s intentions does not matter. The dominant individual (Player 1) always emerges as leader and the non-dominant as follower. This kind of leadership matches our definition in the sense that the dominant individual induces the subordinates to adopt the goals of the leader by following them wherever they go. These games are obviously simplified versions of any real-world situation, and are intended to give a sense of plausible incentive structures that shaped human cognitive adaptations surrounding leadership and followership. In humans, because of the ability to coordinate actions in groups of larger than size two, leadership can involve multi-party coalitions. In such groups, one individual can lead a group of people. The sketch here is intended as a heuristic tool for understanding leadership in dyads. We look forward to additional theoretical work generalizing from groups of size two to groups of size N.

Table 14.3 Leadership and Dominance. The Dominance Game, in which pay-offs (in reproductive success) are for Players 1 and 2 respectively. A/A is an equilibrium Player 2

Player 1

Hole A Hole B

Hole A

Hole B

3,1* 0,0

2,0 1,3


NON-HUMAN EVIDENCE FOR LEADERSHIP Evolutionary biologists have historically reserved the term leadership for behaviours that determine the type, timing and duration of group activity (Wilson, 1975). In any species an important set of adaptive problems revolve around deciding what to do, when and where. For animals living in social groups, a further complication is the presence of conspecifics (see, in this volume, Dunbar, chapter 2; Kenrick et al., chapter 4; Ybarra et al., chapter 16). As mentioned above, it is often safer to move together as a unit, forage as part of a group and sleep at a communal site. This favours some coordination of activity (Krause & Ruxton, 2002). This problem could be solved by one or several individuals taking the initiative and others in the group acquiescing and following. These sorts of problems are likely to have paved the way for the emergence of leadership and followership in many social species, including humans. There are many examples of putative leadership in the animal behaviour literature. The waggle-dance of the honey bee that recruits hive members to visit food resources has been construed as a kind of leadership; the aerial formations of certain bird species and the swimming patterns of schools of fish are also often cited as examples of leader–follower patterns (Couzin et al., 2005; Krause & Ruxton, 2002; Lamprecht, 1996). Because individuals hold different preferences about the type and timing of group activities, some individuals stand to benefit more than others from group coordination, and therefore have an incentive to get others to adopt their preferences. Further, individuals’ preferences are likely to differ systematically. For example, some organisms might simply consume more energy and digest their food more quickly. Because they get hungry sooner, they decide the timing of the group movement, inducing others to follow them (Couzin et al., 2005). Leading by taking the initiative is not automatically going to be effective, as the game being played is not always as simple as the ones sketched above. An interesting example is found among the nomadic Hamadryas baboons (Dunbar, chapter 2, this volume; Kummer, 1968). When they decide upon which sleeping site to move to on a given night, one individual might make a move in a particular direction. Sometimes they are followed by the rest, but sometimes they are not, in which case the individual is forced to return to the group and the decision process is repeated. Individual recognition makes it possible that some individuals are more likely to be followed, based on such factors as age and knowledge. Dominant individuals also sometimes take on leadership functions in groups, being supported in this role by the rest of the group. De Waal (1996) observed an example when the dominant (alpha) male in a troop of chimpanzees that he studied at Arnhem Zoo intervened in a fight: “A quarrel between Mama and Spin got out of hand and ended in fighting and biting. Numerous apes rushed up to the two warring females and joined in the fray. A huge knot of fighting, screaming apes rolled around in the sand, until Luit [the alpha male]



leapt in and literally beat them apart. He did not choose sides in the conflict, like others; instead anyone who continued to act received a blow from him” (p. 129). De Waal argued that this control role is only effective if it is endorsed by the majority of the group, thus constituting an example of leadership. Boehm (1999) observed an instance of leadership displayed by the alpha male in the chimpanzee colony at Gombe. When members of this colony encountered the members of a different troop, the alpha charged towards them and the rest followed his example, until the enemy slowly retreated into their home range. Finally, there is evidence for dominant leadership among other social mammals (lions and wolves), suggesting that dominant individuals emerge as leaders more frequently in chasing prey or chasing away intruders (Heinsohn & Packer, 1995; Wilson, 1975). In these instances, dominant individuals make the first move in initiating group action. Once a leader commits to a course of action, the best move on the part of others is often to follow (see Table 14.3).

LEADERSHIP IN HUMANS Obviously, selection pressures that created leadership in other social species might not be the same as those observed in humans (cf. Kurzban & Leary, 2001). The above analysis suggests that in humans, the concept of “leadership” might be best understood as a constellation of adaptations designed to solve two qualitatively different types of group problems. Note that these adaptations are flexible in the sense that behaviours producing leadership and followership are only elicited in specific environments that resemble the ones in which these roles were originally displayed and selected for (Cosmides & Tooby, 1992). Which are these situations?

Coordinating Leadership In some cases, individuals are playing relatively simple coordination games. That is, people care less about which specific action is taken than about coordinating on one action. Such cases require coordination and, hence, a coordinating leader. Real-world examples have been discussed at length by Schelling (1960). This type of leadership ties well with the literature on task-oriented leadership, which is the most common form of leadership displayed in human groups (Cartwright & Zander, 1968; Hemphill, 1950). Task leaders display activities to promote task completion, such as the coordination of group activities, assignment of sub-tasks and performance monitoring (Fiedler, 1967). Leadership as coordination is also seen in highly cohesive groups in which members have very similar preferences. Cohesive groups actually do better with a randomly changing leader than a permanent leader, presumably because structural leadership creates power differences between members, which undermines group cohesion


(Haslam et al., 1998; Hogg, 2001). Finally, in pure coordination problems, groups might choose one individual to make a decision on behalf of the entire group, paving the way for a highly directive, autocratic leader (Peterson, 1999). Interestingly, in pure coordination situations, leadership does not have to be granted to a person. Groups might converge on a coordination point by all following the same social norm (e.g., driving on the left; wearing a tie) or adhering to the same vision (e.g., religion, political belief, corporate strategy, etc.; see Dunbar, chapter 2, this volume). Again, Schelling (1960) is an excellent source for discussions of problems of this type.

Strategic Leadership Strategic leadership emerges when one individual can change the pay-offs of others’ actions so that they are induced to assume the leader’s proximate goals, even if these do not, globally, maximize the followers’ outcomes. Table 14.2 illustrates such a case: In such situations, there is an incentive for individuals to move first and seize the initiative. Real-world examples are not always so clean. Clearly, there are cases in which potential followers must be persuaded that coordinating on the leader’s goal is more beneficial than not doing so (Bass, 1990). This idea touches the broad literature on relational-style leadership. Relational leaders’ primary concern is to build up good relations with followers (Cartwright & Zander, 1968; Hemphill, 1950). One form of persuasion is to be fair and generous to followers so that they expect to get what they have been promised (Boehm, 1999; de Cremer & van Vugt, 2002; Hardy & van Vugt, 2006; Tyler & Lind, 1992). Another strategy is to develop a unique skill or competency that attracts followers. Apparently, members of task groups are very good at recognizing the strengths and weaknesses of each other (Littlepage, Robinson & Reddington, 1997). In these cases, potential followers are persuaded about the benefits of relevant choices; this stands in contrast to purely coordinating leadership, in which a leader need not worry about persuasion as it is to others’ advantage to coordinate on the selected act. The concept of strategic leadership is of potential value in understanding interesting correlations between leadership and traits like ambition, intelligence, self-esteem, extraversion, sociability, empathy and Machiavellianism (Lord, DeVader & Alliger, 1986; van Vugt, 2006). To emerge as leaders, individuals must think and behave strategically to influence potential followers. Having the ability to put oneself in other people’s shoes (empathy) as well as having excellent communication skills fosters leadership. Research suggests that leadership in ad hoc groups is granted to the most talkative group member, regardless of what they have to say (babble hypothesis; Sorentino & Boutillier, 1975). Also, coming across as intelligent—Machiavellianists are particularly good at this (Wilson et al., 1996)—might persuade others to give up their preferred option, following the leader instead.



Of course, not all forms of strategic interactions are equal. One way to change potential followers’ pay-offs is to use coercion or the threat of coercion. It is not clear the extent to which pure coercion is used as strategic leadership in humans. To achieve important adaptive goals, such as food gathering and selfprotection, humans depend upon other humans (Kenrick et al., chapter 4, this volume; Kenrick, Li & Butner, 2003). Human leaders often cannot achieve their goals without the voluntary support of followers. Further, comparative research suggests that human hierarchies—particularly in non-industrialized societies— are often flatter than the non-human primates, where examples of dominant leadership have been obtained (Boehm, 1999; de Waal, 1996; Dunbar, 2004). In humans, if followers disagree with their leaders, they have strategies to deal with them (Boehm, 1999). Humans also form unusually large coalitions (Dunbar, chapter 2, this volume; Kurzban, Tooby & Cosmides, 2001; Navarete, Kurzban, Fessler & Kirkpatrick, 2004), which makes it difficult for any one individual to dominate an entire group. The psychological literature suggests that there is a weak correlation between dominance and leadership in humans (van Vugt, 2006). It seems that people do not want to be dominated by others. If they are, they either form a coalition to turn against their leader (Boehm, 1999) or simply leave the group (van Vugt, Jepson, Hart & de Cremer, 2004). The importance of cooperation, combined with the availability of exit and voice strategies, often prevents leaders from turning into dictators (cf. Vehrencamp, 1983). This is not to say that leaders are not tempted to dominate followers. One might expect that there would be a tendency in leaders to try to dominate followers (van Vugt, Hogan & Kaiser, 2006). However, this is likely to be counteracted by subversive tactics among followers through gossip, disobedience and open rebellion (Boehm, 1999; Dunbar, 2004; van Vugt et al., 2006). In short, while there is no doubt that human history is replete with examples of coercive strategic leadership, it is not clear the extent to which this has been typical of human social groupings.

DISCUSSION In this chapter, we have offered an evolutionary perspective on leadership. To understand leadership, it is important to ask the previous question surrounding why individuals might have come to be designed to follow a leader. In particular, we suggested that followership adaptations emerged in order to reap the benefits of coordination in response to problems that many group-living species face (see the other chapters in this volume).

Evolutionary Game Analysis Our analysis of leadership heavily relies on evolutionary game theory. The evolutionary game analysis enabled us to model the evolution of leader and follower


traits through identifying problems (games) that could be solved through leadership. Without attempting to be exhaustive, our analysis identified two basic forms of leadership, coordinating and strategic leadership. We first reviewed the animal literature and found some support for this distinction. Subsequently, we reviewed the human (psychological) literature to illuminate some of the conditions under which each of these leader types might develop.

Leadership and Social Cognition We can offer a few speculations to make concrete our ideas surrounding the cognitive adaptations designed for leadership and followership. First, it seems plausible that there are specialized mechanisms designed to identify situations as coordination problems. It is not clear, as yet, what these adaptations would look like, but certainly such skills might look like facets of what has come to be understood under the broad umbrella of intelligence (Dunbar, 2004). Not surprisingly, psychological research shows that leadership in humans is consistently linked with intelligence (van Vugt, 2006). Second, cognitive mechanisms associated with timing would have been instrumental for the development of leadership. Our game theoretical analysis suggests that any trait that enabled an individual to pre-empt a move would have increased the chances of them emerging as leaders. Not surprisingly, the literature shows a strong correlation between leadership and traits associated with initiative taking such as boldness, extraversion, impatience, risk taking and self-esteem (van Vugt, 2006). Third and related, the emergence of leadership is tightly bound to the ability to predict others’ actions and intentions (Ybarra et al., chapter 16, this volume). This idea has close connections with belief/desire-psychology, or what Dennett (1987) has termed the Intentional Stance. The broad literatures on theory of mind and, obviously, language, are relevant for understanding how humans came to coordinate actions (Dunbar, 2004). Mind-reading abilities, supported by language, would have enabled individuals to coordinate their actions in larger, dispersed groups, which opened up new opportunities for leadership. Indeed, leadership is correlated with superior mind-reading and communication skills such as empathy and verbal giftedness (van Vugt, 2006). Finally, leader–follower relationships would be smoothed if individuals were able to recognize those in the group with a particular skill, ability, or piece of information that could help them achieve a desired goal such as solving a coordination problem. Perhaps this is the underlying reason that task ability is considered to be highly important for qualifying as leader (Palmer, 1962). Furthermore, given enough experience working together, people can easily rank one another in terms of skills and knowledge (Littlepage et al., 1997). Thus, cognitive traits associated with problem recognition, timing, mindreading, language and social competence recognition would have all contributed to the emergence of leadership in humans. However, whether these modules



were specifically designed to solve group coordination problems remains an empirical issue.

Implications One implication of the approach sketched here is that leadership is tightly bound to the need to coordinate (Kurzban, 2001). To the extent that individuals are able to determine when they are in such a context, one would expect leader/ follower adaptations to be differentially activated. Second, the distinction we draw in terms of types of leadership implies that different types of leadership might be more appropriate under different conditions. Pure coordination games should be solved best by coordinating leaders, whereas other types of game might be better suited to strategic leadership, as illustrated by the example of chimpanzees. As a corollary to this, our analysis suggests that an individual’s style of leadership—whether oriented towards coordinating or strategic—might be a crucial determinant of the success of groups and their members (Sober & Wilson, 1998). Our analysis suggests various new directions for research on leadership. One straightforward prediction is that leadership emerges more quickly when groups are under threat and there is an urgent need for coordination. Studies could also investigate the emergence of different forms of leadership in response to different kinds of threats such as an intra-group conflict or inter-group aggression (de Cremer & van Vugt, 2002; Kenrick et al., chapter 4, this volume; Schaller & Duncan, chapter 18, this volume; van Vugt & de Cremer, 1999). More research is needed to study the benefits of leadership for groups. Multi-level selection theory suggests that leadership might have evolved as group-level adaptation (Boehm, 1999). One way to test this idea would be to show that groups with a functioning leader–follower structure fare better than groups without this structure, thus creating variations between groups on which selection could have operated in the past. Perhaps more generally, the leadership field could benefit from developing a more integrated research agenda involving efforts from many behavioural science disciplines. In addition, this analysis suggests that the light of inquiry should shine more brightly on followership. Though this topic has important historical roots (e.g., Freud, 1922; Fromm, 1941), it has received far less attention than leadership. Examining the conditions under which people choose to coordinate their actions to help leaders achieve their—or the group’s—goals holds the promise of balancing the vast research on leadership with the other side of the equation, the mechanisms that underpin a potentially well-elaborated psychology of followership.


NOTE 1. In the one-shot, simultaneous version of this game, there are equilibria in pure strategies in which each chooses the same hole and an additional equilibrium in mixed strategies in which each player chooses each watering hole with probability. 5. In the third equilibrium, they reach the non-zero pay-offs

only half the time. For standard game theoretical discussions, which typically do not include sequential play, see Rapoport (1967). We use these games, and include the possibility of sequential play, simply to illustrate our broader points about leadership and followership.

REFERENCES Bass, B. M. (1954). The leaderless group discussion. Psychological Bulletin, 51, 465–492. Bass, B. M. (1990). Bass and Stogdill’s handbook of leadership: Theory, research, and managerial applications (3rd ed.). New York: Free Press. Boehm, C. (1999). Hierarchy in the forest. London: Harvard University Press. Brown, D. (1991). Human universals. Boston: McGraw-Hill. Cartwright, D., & Zander, A. (1968). Group dynamics: Research and theory. Evanston, IL: Row Peterson. Chemers, M. M. (2000). Leadership research and theory: A functional integration. Group Dynamics, 4, 37–43. Cosmides, L., & Tooby, J. (1992). Cognitive adaptations for social exchange. In J. Barkow et al. (Eds.), The adapted mind: Evolutionary psychology and the generation of culture (pp. 163–128). New York: Oxford University Press. Couzin, I. D., Krause, J., Franks, N. R., & Levin, S. A. (2005). Effective leadership and decision-making in animal groups on the move. Nature, 434, 513–516. Darwin, C. (1859). The origins of species. London: Murray. Dawkins, R. (1976). The selfish gene. Oxford, UK: Oxford University Press. De Cremer, D., & van Vugt, M. (2002). Intra and intergroup dynamics of leadership in social dilemmas: A relational model of cooperation. Journal of Experimental Social Psychology, 38, 126–136. Dennett, D. C. (1987). The intentional stance. Cambridge, MA: MIT Press/Bradford Books. DeScioli, P., & Kurzban, R. (in press). The game people play. In S. Gangestad & J. Simpson (Eds.), Evolution of mind: Fundamental

questions and controversies. New York: Guilford Press. De Waal, F. (1996). Good natured: The origins of right and wrong in humans and other animals. Cambridge, MA: Harvard University Press. Dunbar, R. M. (2004). Gossip, grooming, and the evolution of language. London: Faber & Faber. Fiedler, F. E. (1967). A theory of leadership effectiveness. New York: McGraw-Hill. Freud, S. (1922). Group psychology and the analysis of the ego. New York: Liveright. Fromm, E. (1941). Escape from freedom. New York: Holt. Hardy, C. L., & van Vugt, M. (2006). Nice guys finish first: The competitive altruism hypothesis. Personality and Social Psychology Bulletin, 32, 1402–1413. Haslam, A., McGarthy, C., Brown, P., Eggins, R., Morrison, B., & Reynolds, K. (1998). Inspecting the emperor’s clothes: Evidence that random selection of leaders can enhance group performance. Group Dynamics, 2, 168–184. Heinsohn, R., & Packer, C. (1995). Complex cooperative strategies in group-territorial African lions. Science, 269, 1260–1262. Hemphill, J. K. (1950). Leader behaviour description. Columbus, OH: Ohio State University Henrich, J., & Gil-White, F. (2001). The evolution of prestige: Freely conferred deference as a mechanism for enhancing the benefits of cultural transmission. Evolution and Human Behavior, 22, 165–196. Hogan, R., & Kaiser, R. (2005). What we know about leadership. Review of General Psychology, 9, 169–180. Hogg, M. A. (2001). A social identity theory of


242 EVOLUTION AND THE SOCIAL MIND leadership. Personality and Social Psychology Review, 5, 184–200. Hollander, E. P. (1985). Leadership and power. In G. Lindzey & E. Aronson (Eds.), The handbook of social psychology (pp. 485–537). New York: Random House. Kenrick, D., Li, N. P., & Butner, J. (2003). Dynamical evolutionary psychology: Individual decision rules and emergent social norms. Psychological Review, 110, 3–28. Krause, J., & Ruxton, G. D. (2002). Living in groups. Oxford, UK: Oxford University Press. Kummer, H. (1968). Social organization of hamadryas baboons. Chicago: University of Chicago Press. Kurzban, R. (2001). The social psychophysics of cooperation: Nonverbal communication in a public goods game. Journal of Nonverbal Behavior, 25(4), 241–259. Kurzban, R., & Aktipis, C. A. (2006). Modular minds, multiple motives. In M. Schaller, J. Simpson, & D. Kenrick (Eds.), Evolution and social psychology (pp. 39–54). New York: Psychology Press Kurzban, R., Tooby, J., & Cosmides, L. (2001). Can race be erased? Coalitional computation and social categorization. Proceedings of the National Academy of Sciences, 98(26), 15387– 15392. Kurzban, R. K., & Leary, M. (2001). Evolutionary origins of stigmatization: The functions of social exclusion. Psychological Bulletin, 127, 187–208. Lamprecht, J. (1996). What makes an individual the leader of its group? An evolutionary concept of distance regulation and leadership. Social Sciences Information, 35, 595–617. Littlepage, G. E., Robinson, W., & Reddington, K. (1997). Effects of task experience and group experience on group performance, member ability, and recognition of expertise. Organizational Behaviour and Human Decision Processes, 69, 133–147. Lord, R. G., DeVader, C. L., & Alliger, G. M. (1986). A meta-analysis of the relation between personality traits and leadership perceptions: An application of validity generalization procedures. Journal of Applied Psychology, 71, 402–410. Maynard-Smith, J. (1982). Evolution and the theory of games. Cambridge, UK: Cambridge University Press.

Navarete, C., Kurzban, R., Fessler, D., & Kirkpatrick, L. (2004). Anxiety and worldview defense: Terror-management or coalitional psychology? Group Processes and Intergroup Relations, 7(4), 370–397. Palmer, G. J. (1962). Task ability and effective leadership. Psychological Reports, 10, 863– 866. Peterson, R. S. (1999). Can you have too much of a good thing? The limits of voice for improving satisfaction with leaders. Personality and Social Psychology Bulletin, 25, 313– 324. Rapoport, A. (1967). Exploiter, leader, hero, and martyr: The four archetypes of the 2×2 game. Behavioral Science, 12, 81–84. Schelling, T. (1960). The strategy of conflict. Cambridge, MA: Harvard University Press. Sober, E., & Wilson, D. S. (1998). Unto others: The evolution and psychology of unselfish behaviour. Cambridge, MA: Harvard University Press. Sorrentino, R. M., & Boutillier, R. G. (1975). The effect of quantity and quality of verbal interaction on ratings of leadership ability. Journal of Experimental Social Psychology, 11, 403–411. Stogdill, R. M. (1974). Handbook of leadership. New York: Free Press. Tooby, J., & Cosmides, L. (1992). Psychological foundations of culture. In J. Barkow et al. (Eds.), The adapted mind: Evolutionary psychology and the generation of culture (pp. 19–136). New York: Oxford University Press. Tyler, T. R., & Lind, E. A. (1992). A relational model of authority in groups. In M. Zanna (Ed.), Advances in experimental social psychology (Vol. 25, pp. 115–191). New York: Academic Press. Van Vugt., M. (2006). Evolutionary origins of leadership and followership. Personality and Social Psychology Review. Van Vugt, M., & de Cremer, D. (1999). Leadership in social dilemmas: The effects of group identification on collective actions to provide public goods. Journal of Personality and Social Psychology, 76, 587–599. Van Vugt, M., Hogan, R., & Kaiser, R. (2006). What evolution teaches us about leadership. Unpublished manuscript. Van Vugt, M., Jepson, S. F., Hart, C. M., & de Cremer, D. (2004). Autocratic leadership in


social dilemmas: A threat to group stability. Journal of Experimental Social Psychology, 40, 1–13. Vehrencamp, S. L. (1983). A model for the evolution of despotic versus egalitarian societies. Animal Behaviour, 31, 667–682. West-Eberhard, M. J. (2003). Developmental plasticity and evolution. Oxford, UK: Oxford University Press.

Wilson, D. S., Near, D., & Miller, R. R. (1996). Machiavellianism: A synthesis of the evolutionary and psychological literatures. Psychological Bulletin, 119, 285–299. Wilson, E. O. (1975). Sociobiology: The new synthesis. Harvard, MA: Harvard University Press. Yukl, G. A. (1989). Leadership in organizations. Englewood Cliffs, NJ: Prentice Hall.



Proximate and Ultimate Origins of a Bias for Prototypical Faces

An Evolutionary Social Cognitive Account JAMIN HALBERSTADT

The Prototypicality Bias as an Adaptation Domain Specificity of the Prototypicality Bias The Prototypicality Bias as a Side-Effect Conclusion he use of evolutionary reasoning has been sometimes fruitful, and equally often controversial. Skeptics rightly criticize sometimes-sloppy or seemingly unfalsifiable adaptive stories that are all-too-easy to muster as support for otherwise weak theoretical positions. Meanwhile evolutionary psychologists tout the field as the critical link between psychology and the natural sciences, sometimes condemning the unconvinced as unscientific, or even intellectually backward (Tooby & Cosmides, 1992). Ketelaar and Ellis (2000) have argued that evolutionary psychology itself is not a theory subject to falsification, but a “metatheory” to be judged on its success in organizing empirical findings. This perspective echoes similar arguments for the primacy of ultimate explanations based on their ability to constrain research: evolutionary theory defines the space of plausible proximate psychological mechanisms, and therefore must be considered before exploring those mechanisms. Evolutionary psychologists ask, in effect: How can we understand how a psychological process works without knowing what it is designed to do? Thus, there exists a tension between a belief, fostered by evolutionary psychologists, that an evolutionary approach is necessary




to scientific psychological progress, and the often legitimate sense that particular evolutionary accounts of psychological phenomena are untested and untestable. In this chapter I try to relieve this tension by examining a specific and potentially important attractiveness phenomenon as a test case for a hybrid approach to social psychological explanation. The phenomenon is this: Faces that have been digitally averaged, made more similar to a prototypical face, or simply independently rated as highly “face-like”, are generally judged as more attractive than mathematically or subjectively distinctive faces. The effect, first reported in the modern literature by Langois and Roggman (1990), and illustrated in Figure 15.1, has been replicated many times (e.g., Langlois, Roggman, & Musselman, 1994; Rhodes, Sumich, & Byatt, 1999; Rhodes & Tremewan, 1996), in same- and other-race faces (Jones & Hill, 1993; Light, Hollander, & Kayra-Stuart, 1981; Rhodes & Tremewan, 1996; Rhodes et al., 1999), and in both adult and child raters (Rubenstein, Kalakanis, & Langlois, 1999). In the process, a number of correlates and confounds of the averaging process, such as blurriness, symmetry, and youthfulness (attractive in and of themselves), have been ruled out (Langois et al., 1994; Rhodes et al., 1999; Rhodes et al., 2004; Rubenstein, Langois, & Roggman, 2002). Over and above these factors, faces are generally more attractive as a positive function of their prototypicality. Herein, I refer to this phenomenon as the “prototypicality bias.” In this chapter I use evolutionary reasoning to generate functional explanations of the prototypicality bias, whose plausibility are then evaluated with social cognitive data and research paradigms. Thus, unlike evolutionary research that looks to archival, historical, or even prehistorical evidence (or speculation) to reconstruct the adaptive pressures in our evolutionary past, this approach examines proximate causes and correlates of attractiveness to deduce likely origins of the prototypicality bias. As will be seen, the results not only provide insight into the nature of facial attractiveness, but also expose more general, and arguably more important, associations between category structure and affect, as well as interesting distinctions between natural and artificial categories. Hopefully, they also provide some insight into the potential for a successful integration of evolutionary and social cognitive explanation.

FIGURE 15.1 Illustration of the prototypicality bias. A face becomes increasingly attractive

when it is part of a 2-face, 4-face, 8-face, 16-face, and 32-face composite (from left to right).


THE PROTOTYPICALITY BIAS AS AN ADAPTATION The first thing to acknowledge when searching for the adaptive function of any psychological phenomenon is that there might not be one. Adaptations are specific mechanisms that evolved to solve specific and recurring reproductive problems. However, there are other, arguably more frequent products of natural selection: side-effects, which are features associated with adaptations that may or may not be functional themselves; and noise, which is random genetic, environmental, or developmental variability with no adaptive significance. Thus, the first substantive question to ask in any evolutionary psychological inquiry is not whether a particular mechanism is functional, but whether it evolved because it was functional. This is not a trivial or merely technical distinction. Identifying the ultimate origin of a psychological mechanism is, according to evolutionary psychologists at least, a necessary first step in studying it, because it constrains the possible research space by identifying plausible moderators and mediators. Whether or not this strong view of the primacy of evolutionary psychology is correct, it seems clear that if, for example, we assume prototypicality to be an evolved cue to mate choice when it has nothing whatsoever to do with mate choice (because it is a coincident side-effect of some more general cognitive mechanism perhaps), this error is likely to impede research progress. So, are humans attracted to prototypical faces because that attraction conferred a reproductive advantage in the past? Possibly, if we make the controversial assumption that mates with prototypical faces are more genetically fit (as well as the somewhat tautological assumption that we tend to mate with individuals to whom we are attracted). From this perspective, prototypicality is simply a signal of reproductive fitness (see Gangestad & Thornhill, chapter 3, this volume, for discussion of the evolution of signaling systems). The goal of the researcher would then be to verify that people with prototypical faces are indeed genetically fit. Indeed, evolutionary psychologists studying distinctive (i.e., atypical) attractive features generally take this approach (e.g., Cunningham, 1986; Cunningham, Barbee, & Pike, 1990; Penton-Voak & Perrett, 2001; Rhodes, Hickford, & Jeffrey, 2000; Thornhill & Gangestad, 1999; Zebrowitz, 1997). For example Cunningham (1986; Cunningham, Roberts, Wu, Barbee, & Druen, 1995) used men’s ratings of beauty pageant contestants to identify both neotonous features such as large eyes and pupils (see Zebrowitz, 1997, for a review of the literature on “babyfacedness”) and sexually mature features, such as prominent cheekbones and narrow cheeks, associated with attractiveness in women. Perrett et al. (1998) increased attractiveness in both sexes by enhancing their “feminine” features, which the researchers found to be associated with positive personality characteristics such as warmth and honesty. More controversially, some studies have also found extreme “masculine” traits to be attractive in men (Cunningham et al., 1990; but see Penton-Voak et al., 1999; Perrett et al., 1998; Rhodes et al., 2000). Intriguingly, a woman’s attraction to such traits depends on



the stage in her ovulatory cycle and the type of mate (short vs. long term) she is rating (Simpson & LaPaglia, chapter 10, this volume). Although this research associates distinctive features with presumably reproductively beneficial traits, evolutionary psychologists have explained attraction to prototypicality—effectively the absence of distinctive features—using the same logic. Such hypotheses are generally of two types: So-called “good genes” hypotheses link prototypicality to features that are positively predictive of reproductive fitness, such as current health or fertility, developmental stability (Thornhill & Møller, 1997), or resistance to parasites (Gangestad & Buss, 1993; Thornhill & Gangestad, 1993). “Bad genes” hypotheses, on the other hand, propose relationships between deviations from prototypicality and fitness impairments, such as some genetic disorders. The two accounts of course are not incompatible, and in fact both rely on the same logic, namely that prototypicality is a signal of an individual’s reproductive fitness, and attraction a mechanism for identifying such individuals. Though controversial and limited, there is some recent evidence for the relationship between prototypicality and reproductive fitness. Rhodes et al. (2000) found evidence for the good genes hypothesis in correlations between facial prototypicality and current (for women) and childhood (for men) health (in contrast to a number of prior studies reporting weak or absent relationships; cf. Kalick, Zebrowitz, Langois, & Johnson, 1998; see Rubenstein et al., 2002, for review). Thornhill and Møller (1997) provide some evidence for the “bad genes” hypothesis by identifying facial distortions associated with some conditions, such as Down’s syndrome, autism, and learning disabilities, which are presumably associated with decreased reproductive fitness.

DOMAIN SPECIFICITY OF THE PROTOTYPICALITY BIAS Of course, even a strong relationship between prototypicality and reproductive fitness is no guarantee that humans find prototypical faces attractive because of this relationship—in other words, that the prototypicality bias is an adaptation. How else might the adaptation hypothesis be evaluated? Well, one characteristic of adaptations, touted by many (but not all) evolutionary psychologists, is “domain specificity.” Psychological mechanisms, like biological ones, evolved to solve very specific problems, and therefore to operate in the very narrow range of input defined by those problems. For example, humans do not have a single organ for “perception,” but rather very specific organs for processing visual auditory, olfactory, tactile, and gustatory stimuli, each itself tuned only to a narrow range of input (presumably the range most useful for our survival and reproduction). Similarly, we (arguably) do not have an evolved mechanism for “logic,” but rather specific logical abilities in domains that matter, such as the detection of cheaters or the evaluation of social contracts (Tooby & Cosmides, 1992). Is the prototypicality bias domain-specific, limited to the particular class


of input (i.e., human faces, and perhaps bodies) relevant to mate selection? Or do humans simply prefer prototypes of any category, reproductively relevant or not? Surprisingly, despite its seemingly central importance in the study of cognition and affect, this question has rarely been asked (for isolated exceptions see Martindale & Moore, 1988; Repp, 1997; Smith & Melara, 1990; Whitfield & Slatter, 1979), and never systematically studied. Therefore in a series of correlational and experimental studies Gillian Rhodes and I examined the relationship between prototypicality and attractiveness in a variety of natural and artificial categories (Halberstadt, 2006; Halberstadt & Rhodes, 2000, 2003). In one study, for example (Halberstadt & Rhodes, 2000), participants rated the attractiveness of 50 dogs of different breeds, while an independent group of participants rated how prototypical each dog was (specifically, how similar each dog was to their image of a “typical” dog). The ratings, averaged across participants (interrater reliability was high in almost all cases reported in this chapter), revealed that the most attractive dog (an American Eskimo) was judged the ninth most prototypical, whereas the least attractive dog (a Bedlington Terrier) was also judged least prototypical. Overall, attractiveness and prototypicality correlated .69, approximately as strongly as in faces, e.g., r (24) = .73 for female European faces in our research. In fact, replications on all manner of animal and artifact stimuli (see Table 15.1 for examples), from monkeys and horses, to wristwatches and handguns, to Chinese ideographs and random dot patterns, reveal correlations between prototypicality and attractiveness, almost always strong and positive (the strongest so far was found, curiously, for eyeglasses, r[50] = .98). The relationship also holds when prototypicality is manipulated rather than Table 15.1 The Prototypicality Bias in Various Natural and Artificial Categories (Data from Halberstadt, 2006; Halberstadt & Rhodes, 2000, 2003) Category

Birds Butterflies Dogs Eyeglasses Handguns Rings Roses Spiders Watches Passerines (N = 98) Fish (N = 63) Cars (N = 63)

r (Prototypicality, Attractiveness)

r (Familiarity, Attractiveness)

Partial r (Prototypicality, Attractiveness)

.50 .34* .69 .98 .87 .30* .32* .20 .65 .94 .75 .54

.20 .23 .47 .98 .89 .25 .33 .12 .69 .93 .46 .74

.47 .31* .58 .08 .24 .19 .00 .22 .13 .46 .69 .01

Note: Partial correlations between prototypicality and attractiveness are controlling for familiarity. All correlations in bold are significant at p < .001, except those marked by an asterisk, which are significant at p < .05. N = 50 for each category unless otherwise noted.



FIGURE 15.2 Average prototypicality and attractiveness ratings of original, caricatured, and anticaricatured New Zealand passerines (from Halberstadt & Rhodes, 2003). The example stimulus set shown is based on the native Tui.

measured, using methods modeled on morphing techniques used to manipulate prototypicality in faces (e.g., Langlois & Roggman, 1990; Rhodes & Tremewan, 1996). For example, the attractiveness of New Zealand resident passerines (the largest family of birds) varies linearly with their manipulated distance from a prototype (a morphed average of all stimuli used in the study; Halberstadt & Rhodes, 2003; see Figure 15.2).

THE PROTOTYPICALITY BIAS AS A SIDE-EFFECT Thus, far from being domain specific, people’s preference for prototypes appears very general: in nonhuman, natural and artificial, positive and negative, manipulated and naturally varying categories, people generally prefer good category exemplars over poor ones. This finding does not preclude the possibility that a preference for prototypical faces is an evolutionary adaptation for identifying reproductively fit mates—nonfaces could be attractive for some other reason(s), perhaps unrelated to the bias in faces. However the prospect of multiple, independent adaptations, though possible, is theoretically awkward: Parsimony demands that we first assume that very similar looking effects are the result of a single cognitive mechanism, and only abandon this null hypothesis in favor of a more complex account if the evidence requires it. So, if people’s attraction to prototypical faces is a product of evolution (as it


must be); and if the only two substantive products of evolution are adaptations and their side-effects; and if, due to its generality, the prototypicality bias is unlikely to be an adaptation (i.e., a domain-specific mate selection mechanism), then the most likely explanation is that it is a side-effect (or a side-effect of a side-effect) of another adaptation. Next, I consider some plausibly functional cognitive mechanisms of which an attraction to average faces could be a side-effect.

Generalization of Category-Level Affect The categorization process itself is often claimed to be critical to cognitive functioning. Kunda’s (1999) portrayal of the situation in terms of social cognition is typical: Without concepts, our world would make little sense. We would be unable to extract meaning from the huge amount of information that surrounds us, unable to generalize from one experience to another, and unable to communicate effectively with each other. (p. 17)

Extracting meaning, generalizing, communicating: These are critical social tasks if ever there were ones, and at least plausible adaptive problems (i.e., impacting on reproductive fitness). The formation and use of categories, if they indeed solve these problems, may be a genuine adaptation, and the affect associated with category members merely an incidental (though perhaps functional) consequence of the fact that some categories are liked better than others. However, theories that assume that affect is a concomitant of categorization can explain why individuals might respond positively toward a face once it is categorized as such (on the assumption that faces are a generally positive stimulus class), but not why their response should vary with the prototypicality of the face. In fact, social cognitive theories of person perception and stereotyping generally do not allow for affective responses proportional to category fit (Fiske’s, 1982 theory of “schema-triggered affect” is a notable exception): Theoretically (if somewhat implausibly) someone who likes liberals (for example) likes any individual they categorize as “liberal,” regardless of how liberal the individual is (i.e., how well they fit the perceiver’s prototype). However, even allowing for graded affective responses to category members (and such a parameter would be relatively easy to incorporate into theories of stereotyping), such theories cannot explain the positive response to prototypes of negatively valenced categories. If affect is merely a side-effect of categorization, then good examples of positive categories should be especially liked, and good examples of negative categories should be especially disliked. However, there is little evidence in our data that category valence moderates the prototypicality bias. In fact, one of the strongest correlations in our data exists for handguns (r = .87), which are judged by participants living in handgun-free New Zealand as a



very negative stimulus. Thus, mere generalization of category-level affect cannot provide a general answer to the question of why category prototypes are attractive.

Fluency Although positive responses to prototypes are thus unlikely to be simply extensions of category-level affect to category members, the process of categorization itself produces affective responses. Cognitive fluency—the ease with which a stimulus can be identified as a member of its category, as quantified by speeded classification time—varies with the prototypicality of the stimulus: the closer to the prototype, the more quickly a stimulus can be processed (Posner & Keele, 1968). More important, the more fluent a stimulus, the more positively it is judged (e.g., Reber, Winkielman, & Schwartz, 1998), either because the experience of fluent processing is itself positive, or because the experience is interpreted as a signal of something positive about the stimulus. In evolutionary psychological terms, the attractiveness of prototypical faces, like the attractiveness of other prototypes, could be simply a side-effect of the ease with which they can be processed. My colleagues and I recently reported results that lend initial plausibility to this account (Winkielman, Halberstadt, Fazendeiro, & Catty, 2006). In two studies, either meaningful (squares and diamonds) or meaningless (random dot patterns) prototypes were distorted to different degrees using a probabilistic algorithm (Posner, Goldsmith, & Welton, 1967). Participants rated the attractiveness of the stimuli and, in a separate experimental phase, categorized them according to the prototype from which they were distorted. For both types of stimuli, prototypicality (i.e., stimulus distortion) predicted both attractiveness and categorization speed, which were themselves correlated. More importantly, in both studies, partialling out the effect of fluency significantly reduced the prototypicality–attractiveness relationship, although prototypicality accounted for additional, unique variance in attractiveness. A third study used facial electromyography to verify that participants’ attractiveness ratings reflected genuine positive affective reactions (as opposed to “cold” judgments, such as distance from the prototype, for which “attractiveness” might have served as a judgment proxy). Results confirmed that “prepared” prototypes (i.e., novel stimuli, distortions of which had been previously seen) elicited more activity in the zygomaticus major muscle region (the “cheek” muscle associated with smiling), and higher explicit liking ratings, than unprepared prototypes. Together, the results of this research suggest that prototypical stimuli are attractive at least in part because the ease with which they are processed elicits positive affective reactions. However, fluency cannot provide a complete account of the appeal of prototypes, and the role of fluency in the appeal of average faces in particular is still unknown, the subject of current research in our laboratory.


Prior Exposure Another source of affect associated with prototypes comes from their relationship to previously seen stimuli (Zajonc, 1968). Stimuli to which individuals have been exposed are judged more positively than novel stimuli on a variety of dimensions, including attractiveness, likeability, and semantic meaning (the “mere exposure effect”; see Bornstein, 1989, for review). Arguably, a bias toward the previously encountered could be functional, perhaps even an adaptation for identifying “safe” stimuli (Garcia-Marquez & Mackie, 2000; Smith, 2000), since anything a living organism has previously encountered has by definition been nonlethal in the past. Such a mechanism would be particularly useful in the absence of conscious stimulus recognition (Zajonc, 1980), and indeed the mere exposure effect appears to be more powerful when stimulus exposure is subliminal (Bornstein, 1989). However, the role of prior exposure on the attractiveness of prototypical faces has not been born out in our research: Although participants in some cases found facial composites more likeable when they had been exposed to the individual faces from which the composites were created, they usually failed to generalize attractiveness to these composites (Halberstadt, Rhodes, & Catty, 2003; Rhodes, Halberstadt, & Brajkovich, 2001; Rhodes, Halberstadt, Jeffrey, & Palermo, 2005). However, theoretically independent of whether a face has been in fact seen before is the feeling that the face has been seen before, what Halberstadt et al. (2003) call subjective familiarity (in contrast to objective familiarity, one’s actual prior exposure to a stimulus). Subjective familiarity typically covaries with objective familiarity, but not always, as evidenced by manipulations, such enhancing stimulus clarity, that increase subjective familiarity independent of actual exposure (e.g., Whittlesea, Jacoby, & Girard, 1990). More important for the present purposes, prototypes are subjectively familiar, even when they are objectively novel (Gordon & Holyoak, 1983) and subjective familiarity predicts positive affect (Reber et al., 1998). Therefore, in evolutionary psychological terms, the appeal of prototypical faces could be a side-effect of their familiarity, a specific case of a more general cognitive mechanism linking subjective familiarity and positive affect. To examine this hypothesis we partialled out subjective familiarity, averaged across new sets of independent participants, from the prototypicality– attractiveness relationship for each stimulus category in Table 15.1. The results provide some unexpected insights: As expected, familiarity was highly correlated with attractiveness in all categories, but unexpectedly it fully mediated the prototypicality–attractiveness relationship only in artifact categories. An informal meta-analysis confirmed a case of moderated mediation of the prototypicality bias by familiarity, depending on category type (natural vs. artificial): the attractiveness of artifact categories such as eyeglasses and handguns can be entirely explained by their subjective familiarity (overall partial r = .08, ns), but



prototypes of natural categories are attractive above and beyond their familiarity (overall partial r = .43, p < .001).

Prototypicality as a Cue to Quality Thus, the evidence so far is most consistent with a dual-origin account of the prototypicality bias. The attractiveness of prototypical artifacts, and possibly natural but nonanimal kinds, can be explained entirely by their subjective familiarity: prototypes feel more familiar than unusual exemplars, and perceivers prefer the subjectively familiar, either for specific functional reasons or as a side-effect of yet another psychological adaptation. In animals, however, prototypicality makes an independent contribution to attractiveness beyond subjective familiarity, and therefore a different or supplementary mechanism is required. One possibility is that prototypicality is associated with something important to know about animals generally, and the affect associated with prototypicality is a mechanism for bringing organisms closer to things that are fitness-promoting (see, for other examples of emotions’ functionality, contributions in this volume by Badcock & Allen, chapter 8; Buck, chapter 6; Forgas, chapter 7; Lieberman, chapter 11; and Schaller & Duncan, chapter 18). For example, prototypicality might be associated with the health of an organism, and accurate assessment of health would be valuable across species, though for different reasons. In the context of human face perception, this information would be useful for evaluating the reproductive fitness of potential mates. In other species, however, health information might be useful for establishing an organism’s value as food or prey or its risk as a predator. As noted above, prototypicality has tentatively been associated with both current and childhood health (Rhodes et al., 2000) in humans. Is there evidence that prototypicality could be a useful heuristic more generally? As an initial examination of this question, independent participants rated the prototypicality and “dangerousness” of a number of plant and animal categories (berries, fish, frogs, mushrooms, snakes, and spiders; N = 50 in each case) whose consumption, contact, or attack was either harmless, injurious, or lethal. The results, which appear in Table 15.2, do not support a general link between prototypicality and objective danger, which significantly correlated in the predicted direction only in two cases (prototypical berries and frogs tend to be safer), and in fact correlated in the opposite direction in one case (prototypical spiders are more dangerous). Even in these two cases, prototypicality failed to mediate the accuracy of danger judgments (i.e., the correlation between judged and actual dangerousness of the individual members of a category). Interestingly, prototypicality correlated strongly with perceived danger in all categories, except for spiders, where the relationship was reversed (i.e., prototypical spiders were judged as more dangerous, r = .36). Thus, at least in this limited sample of animal categories, prototypicality does not consistently predict the dangerousness of particular category members, although participants may believe it does,


Table 15.2 Correlations Between Prototypicality and Actual and Perceived Dangerousness, for Six Natural Categories Category

Berries Fish Frogs Mushrooms Snakes Spiders

r (Prototypicality, Dangerousness)

r (Prototypicality, Perceived Dangerousness)

−.49 −.15 −.50 .27 .21 .44**

−.72 −.72 −.90 −.77 −.60 .36*

Note: All correlations in bold are significant at p < .001, except those marked by an asterisk, which are significant at p < .05, or two asterisks, which are significant at p < .01. N = 50 for each category.

which may partially account for the generally low accuracy of their danger judgments. Therefore, the hypothesis that a general prototypicality bias evolved as a heuristic tool for inferring reproductively-relevant information about animals (including conspecifics) is tentatively rejected, although additional research examining other functional correlates of prototypicality is obviously needed to evaluate the hypothesis fully.

Stimulus Generalization Even if an attraction to prototypical animals is not ultimately functional in every case, it is possible that it represents an adaptation in some cases but not in others, or even that the latter are a consequence of the former. Give the theoretical and empirical relationships between prototypicality and reproductive fitness in human faces, a particularly intriguing possibility is that the prototypicality bias evolved as a functional mechanism for identifying quality mates, and has since generalized to the evaluation of other animals. Research on stimulus generalization is nearly as old as psychology itself (see Ghirlanda & Enquist, 2003, for a recent review), and the idea that responses to novel stimuli vary as some function of their similarity to previously encountered stimuli is a well-established principle among behavioral, cognitive, developmental, social, clinical, and ethological researchers, among others (e.g., Buss, Murray, & Buss, 1968; Guttman & Kalish, 1954; Nosofsky, 1987; Rescorla, 1976; Shepard, 1987; Siegel, Hearst, George, & O’Neal, 1968). Although theory and research are more limited, there is also precedent for explaining social perceptual phenomena as generalizations or overgeneralizations of otherwise functional evolutionary mechanisms (Zebrowitz, Fellous, Mignault, & Andreoletti, 2003). For example, individuals’ avoidance of, and negative emotions toward, disabled others have been interpreted as an overgeneralization of an otherwise functional “disease-avoidance” mechanism (Schaller & Duncan, chapter 18, this volume). Conversely, and more relevant to the current chapter,



Keating, Randall, Kendrick, and Gutshall (2003) found that adults with “babyish” features received more help than adults with more mature features, a phenomenon the researchers attribute to overgeneralization of help-eliciting facial cues that are otherwise functional in actual babies (Berry & MacArthur, 1986; Zebrowitz, 1997). If, analogously, the prototypicality bias represents an overgeneralization of an otherwise functional mechanism to identify reproductively fit mates, then theories of stimulus generalization predict a generalization gradient, such that the correlation between prototypicality and attractiveness in any given animal category is a positive monotonic function of that animal’s actual or perceived similarity to humans. Indeed, although there is no evidence of such a relationship when considering all of the categories in Table 15.1, when only nonmanipulated animal data are considered, the magnitude of the prototypicality bias for a particular animal appears to map well onto that animal’s phylogenetic similarity to humans: Dogs show the strongest association, followed by birds, butterflies, and spiders. To examine the generalization hypothesis more formally, a new group of participants judged natural stimuli representing five qualitative levels of similarity to humans: monkeys; dogs and horses; birds and fish; butterflies, beetles, spiders, and jellyfish; and mushrooms. Male and female, Caucasian and Japanese faces were also included for comparison, for a total of 14 categories (N = 24 exemplars per category). Participants rated all stimuli in random order, blocked by category, on either attractiveness or prototypicality, as well as (for nonhuman categories) their similarity to human beings. Results, shown in Table 15.3, Table 15.3 Ordinal Phylogenetic Similarity to Humans, Judged Similarity to Humans, and Prototypicality−Attractiveness Correlations for 14 Stimulus Categories Category (Ns = 24)

Asian females Asian males European females European males Chimpanzees Horses Dogs Birds Fish Butterflies Beetles Spiders Jellyfish Mushrooms

Phylogenetic Similarity to Humans

Judged Similarity to Humans

r (Prototypicality− Attractiveness)

1 1 1 1 2 3 3 4 4 5 5 5 5 6

— — — — 8.18 5.22 5.65 3.90 2.90 2.23 2.12 2.10 1.75 1.50

.65 .49 .72 .69 .70 .79 .51 .29 .22 .26 .16 .05 .16 .10

Note: Smaller numbers indicate a closer phylogenetic similarity in column 2, and less judged similarity in column 3, to humans. Correlations in bold are significant at p < .05 or less.


confirm that the more similar a category is to humans, the better the prototypicality of its members predicts their attractiveness. While the correlation for faces was approximately .70 and .57 for Caucasian and Japanese faces, respectively, the correlations in the five groups of nonhuman stimuli averaged .70, .65, .26, .16, and .10. Overall the relationship between perceived similarity to humans of the 10 animal categories (which correlated .96 with their level of phylogenetic similarity) and the prototypicality bias was .89. The size of this relationship is surprising—almost 80% of the variance in the prototypicality bias can be accounted for by a species’ perceived (and actual) similarity to human beings—and lends some support to the hypothesis that the use of prototypicality was “designed” for humans but is used to evaluate other animals when preference judgments are required. Nevertheless, even such strong data are ultimately limited by the fact that similarity and species are effectively confounded. That is, it is not clear that perceived similarity per se moderates the prototypicality bias, rather than another variable associated with similarity, such as the variability, formal structure, or familiarity of a category. Ideally, the prototypicality bias should be evaluated in identical stimuli that vary only in their perceived similarity to human beings. As an initial study utilizing this experimental logic, a new set of participants rated 50 digitally constructed faces that differed in their “human-ness.” Specifically, the faces, which were created using feature-based face composite software (“Faces”, InterQuest, Inc., 1988; see Figure 15.3 for examples), were described as police sketches of suspects in either “criminal” or “alien” abductions. All participants judged the prototypicality, attractiveness, and familiarity, of all 50 faces, which differed (between participants) only in their purported genetic relationship to human beings. The results showed that humans were more attractive (Ms = 3.83 vs. 3.41), as well as more familiar (5.10 vs. 4.21), than aliens, and that the prototypicality bias was stronger for humans than aliens, r (50) = .81 versus .55. Most intriguingly, when familiarity (which correlated strongly with both prototypicality and attractiveness in both conditions) was partialled out of the prototypicality bias, the correlation remained highly significant in the human face condition (partial r = .63), but dropped to zero in the alien face condition (partial r = .07). Thus, as was observed in comparisons between animal and

FIGURE 15.3 Examples of faces described as “human” or “alien”.



nonanimal categories, prototypicality makes an independent contribution to attractiveness (i.e., over and above subjective familiarity) in “human” faces, but not when the very same faces are believed to be nonhuman. In this set of stimuli, then, differences in the relationships among the variables cannot be due to differences in their formal features or category structure (which were identical between conditions), but are more consistent with the strategic use of prototypicality to assess mate quality, a domain-specific psychological mechanism elicited most strongly by human faces.

CONCLUSION The goal of this volume is to explore both the theoretical and working relationships between social cognition and evolutionary psychology—two extremely influential, nonmutually exclusive, yet largely incommunicado approaches to social psychological inquiry. The goal of the current chapter in particular is to use the prototypicality bias as a test case for a hybridized investigative strategy. After all, the bias, discovered entirely independent of evolutionary considerations, is nonetheless a product of evolution, and as such is either an adaptation itself—a mechanism evolved specifically to solve a recurring reproductive problem, most likely related to the identification of quality mates—or a side-effect of an adaptation. Yet the plausibility of various functional mechanisms, it is argued, can be best evaluated with proximate social cognitive data. Taking this intellectual approach, I initially discounted an account of the prototypicality bias as a domain-specific mate selection mechanism, on the basis that, as it turns out, people find prototypical exemplars of almost everything attractive. Instead, I considered a number of more general cognitive mechanisms related to categorization as potential primary adaptations of which the prototypicality bias could be a side-effect. Of these, the link between prototypicality subjective familiarity was most theoretically and empirically promising, but with a critical and unexpected qualification. In categories such as eyeglasses, wristwatches, and handguns, defined largely by a common function, category prototypes may be preferred simply because they feel more familiar (even if they are objectively novel), and familiar things are preferred. However, in animal categories, organized around a common and often strong formal structure, prototypes are attractive above and beyond their familiarity. I explored two plausible hypotheses accounting for the generality of the prototypicality bias in animals: (1) Prototypicality is a cue to a characteristic, such as health, that is relevant to a number of different evaluations of different animal categories; or (2) prototypicality is a mate-assessment mechanism that generalized to evaluations of other animals in other contexts. The data, preliminary though they are, support the generalization hypothesis: Prototypicality failed to predict at least one useful fact about animals—whether or not they can kill you—while the magnitude of the prototypicality bias varied linearly with the similarity of an animal category to


humans. Perhaps most interestingly, the bias appears tuned to human faces in particular and is moderated by perceivers’ beliefs about how human a given stimulus is. Thus, our investigations have come full circle. Although the generality of the prototypicality bias was initially seen as cause for rejection of a simple mate choice account, when the nature and magnitude of the bias in different categories are considered, the most logical conclusion is that human faces are indeed the origin of the bias in animals, and that prototypical faces themselves are preferred because they signal reproductive fitness in a potential mate. It is noteworthy that this conclusion is the result of a genuine and bi-directional interaction of social cognitive and evolutionary approaches, a series of empirical and falsifiable studies on proximate mechanisms that were informed by evolutionary principles. It is also noteworthy, however, that contrary to some bolder evolutionary psychological claims, the current research was not obviously constrained by these principles. In part this is because not enough is known, or perhaps will ever be known, about the relationship between cognitive processes and reproductive fitness. While a good adaptive case can be made for some psychologically and neurologically specific processes, such as language and face recognition, in many others, including categorization and its associated effects, the arguments are more speculative. Indeed, one could argue that it is proximate cognitive data such as those reported here that constrain the plausibility of adaptive problems in our evolutionary past, not the other way around. At a minimum, it seems that any insights gleaned from the current research were the result of a vigorous interaction between proximate and ultimate considerations. Such a symbiotic approach, rather than a hierarchical one, may prove to be more productive for both evolutionary and nonevolutionary psychologists.

REFERENCES Berry, D. S., & McArthur, L. A. (1986). Perceiving character in faces: The impact of age-related craniofacial changes on social perception. Psychological Bulletin, 100, 3–18. Bornstein, R. F. (1989). Exposure and affect: Overview and meta-analysis of research. Psychological Bulletin, 106, 265–289. Buss, A. H., Murray, E. N., & Buss, E. (1968). Stimulus generalization and fear of snakes. Journal of Personality and Social Psychology, 10, 134–141. Cunningham, M. R. (1986). Measuring the physical in physical attractiveness: Quasiexperiments on the sociobiology of female facial beauty. Journal of Personality and Social Psychology, 50, 925–935.

Cunningham, M. R., Barbee, A. P., & Pike, C. L. (1990). What do women want? Facial metric assessment of multiple motives in the perception of male facial physical attractiveness. Journal of Personality and Social Psychology, 59, 61–72. Cunningham, M. R., Roberts, A. R., Wu, C., Barbee, A. P., & Druen, P. B. (1995). “Their ideas of beauty are, on the whole, the same as ours”: Consistency and variability in the cross-cultural perception of female physical attractiveness. Journal of Personality and Social Psychology, 68, 261–279. Fiske, S. T. (1982). Schema-triggered affect: Applications to social perception. In M. S. Clark & S. T. Fiske (Eds.), Affect and cognition: The seventeenth annual Carnegie


260 EVOLUTION AND THE SOCIAL MIND Symposium on Cognition (p. 66). Mahwah, NJ: Lawrence Erlbaum Associates, Inc. Gangestad, S. W., & Buss, D. M. (1993). Pathogen prevalence and human mate preferences. Ethology and Sociobiology, 14, 89–96. Garcia-Marquez, T., & Mackie, D. M. (2000). The positive feeling of familiarity: Mood as an information processing regulation mechanism. In H. Bless & J. P. Forgas (Eds.), The message within: The role of subjective experience in social cognition and behavior (pp. 240– 261). Philadelphia: Psychology Press. Ghirlanda, S., & Enquist, M. (2003). A century of generalization. Animal Behaviour, 66, 15–36. Gordon, P. C., & Holyoak, K. J. (1983). Implicit learning and generalization of the “mere exposure” effect. Journal of Personality and Social Psychology, 45(3), 492–500. Guttman, N., & Kalish, H. I. (1956). Discriminability and stimulus generalization. Journal of Experimental Psychology, 51, 79–88. Halberstadt, J., & Rhodes, G. (2003). It’s not just average faces that are attractive: Computer-manipulated averageness makes birds, fish and automobiles attractive. Psychonomic Bulletin and Review, 10(1), 149–156. Halberstadt, J., Rhodes, G., & Catty, S. R. (2003). Subjective and objective familiarity as explanations for the attraction to average faces. In F. Columbus (Ed.), Advances in psychology research (pp. 91–106). New York: Nova Science Publishers. Halberstadt, J. B. (2006). The generality and ultimate origins of the attractiveness of prototypes. Personality and Social Psychology Review, 10, 166–183. Halberstadt, J. B., & Rhodes, G. (2000). The attractiveness of non-face averages: Implications for an evolutionary explanation of the attractiveness of average faces. Psychological Science, 11, 289–293. InterQuest, Inc. (1988). Faces: The Ultimate Composite Picture [Computer software]. Quebec, Canada: Author. Jones, D., & Hill, K. (1993). Criteria of attractiveness in five populations. Human Nature, 4, 271–296. Kalick, S. M., Zebrowitz, L. A., Langlois, J. H., & Johnson, R. M. (1998). Does human facial attractiveness honestly advertise health?

Longitudinal data on an evolutionary question. Psychological Science, 9, 8–13. Keating, C. F., Randall, D. W., Kendrick, T., & Gutshall, K. A. (2003). Do babyfaced adults receive more help? The (cross-cultural) case of the lost resume. Journal of Nonverbal Behavior, 27, 89–109. Ketelaar, T., & Ellis, B. J. (2000). Are evolutionary explanations unfalsifiable? Evolutionary psychology and the Lakatosian philosophy of science. Psychological Inquiry, 11, 1–21. Kunda, Z. (1999). Social cognition. Cambridge, MA: MIT Press. Langois, J. H., & Roggman, L. A. (1990). Attractive faces are only average. Psychological Science, 1, 115–121. Langois, J. H., Roggman, L. A., & Musselman, L. (1994). What is average and what is not average about attractive faces? Psychological Science, 5, 214–220. Light, L. L., Hollander, S., & Kayra-Stuart, F. (1981). Why attractive people are harder to remember. Personality and Social Psychology Bulletin, 7, 269–276. Martindale, C., & Moore, K. (1988). Priming, prototypicality, and preference. Journal of Experimental Psychology: Human Perception and Performance, 14, 661–670. Nosofsky, R. M. (1987). Attention and learning processes in the identification and categorization of integral stimuli. Journal of Experimental Psychology: Learning, Memory, and Cognition, 13, 87–108. Penton-Voak, I. S., & Perrett, D. I. (2001). Male facial attractiveness: Perceived personality and shifting female preferences for male traits across the menstrual cycle. In P. J. Slater, & J. S. Rosenblatt (Eds.), Advances in the study of behavior (pp. 219–259). San Diego, CA: Academic Press. Penton-Voak, I. S., Perrett, D., Castles, D., Burt, M., Koyabashi, T., & Murray, L. K. (1999). Female preferences for male faces change cyclically. Nature, 399, 741–742. Perrett, D. I., Lee, K. J., Penton-Voak, I., Rowland, D., Yoshikawa, S., Burt, D. M., et al. (1998). Effects of sexual dimorphism on facial attractiveness. Nature, 394, 884–887. Posner, M. I., Goldsmith, R., & Welton, K. E., Jr. (1967). Perceived distance and the classification of distorted patterns. Journal of Experimental Psychology, 73(1), 28–38.


Posner, M. I., & Keele, S. W. (1968). On the genesis of abstract ideas. Journal of Experimental Psychology, 77, 155–158. Reber, R., Winkielman, P., & Schwarz, N. (1998). Effects of perceptual fluency on affective judgments. Psychological Science, 9, 45–48. Repp, B. H. (1997). The aesthetic quality of a quantitatively average music performance: Two preliminary experiments. Music Perception, 14, 419–444. Rescorla, R. A. (1976). Stimulus generalization: Some predictions from a model of Pavlovian conditioning. Journal of Experimental Psychology, 2, 88–96. Rhodes, G., Halberstadt, J., & Brajkovich, G. (2001). Generalization of mere exposure effects to averaged composite faces. Social Cognition, 19, 57–70. Rhodes, G., Halberstadt, J. B., Jeffrey, L., & Palermo, R. (2005). The attractiveness of average faces is not a generalized mere exposure effect. Social Cognition, 23, 205–217. Rhodes, G., Hickford, C., & Jeffrey, L. (2000). Sex-typicality and attractiveness: Are supermale and superfemale faces super-attractive? British Journal of Psychology, 91, 125–140. Rhodes, G., Sumich, A., & Byatt, G. (1999). Are average facial configurations attractive only because of their symmetry? Psychological Science, 10, 52–58. Rhodes, G., & Tremewan, T. (1996). Averageness, exaggeration and facial attractiveness. Psychological Science, 7, 105–110. Rhodes, G., Yoshikawa, S., Palermo, R., Simmons, L. W., Peters, M., Lee, K., et al. (2004). The contribution of familiarity and perceived health to the appeal of average faces. Unpublished data. Rubenstein, A. J., Kalakanis, L., & Langlois, J. H. (1999). Infant preferences for attractive faces: A cognitive explanation. Developmental Psychology, 15, 848–855. Rubenstein, A. J., Langois, J. H., & Roggman, L. A. (2002). What makes a face attractive and why: The role of averageness in defining facial beauty. In G. Rhodes & L. A. Zebrowitz (Eds.), Advances in visual cognition: Vol. 1. Facial attractiveness: Evolutionary, cognitive, and social perspectives (pp. 1–33). Westport, CT: Ablex Publishing. Shepard, R. N. (1987). Toward a universal law

of generalization for psychological science. Science, 237, 1317–1323. Siegel, S., Hearst, E., George, N., & O’Neal, E. (1968). Generalization gradients obtained from individual subjects following classical conditioning. Journal of Experimental Psychology, 78, 171–174. Smith, D. J., & Melara, R. J. (1990). Aesthetic preference and syntactic prototypicality in music: ’Tis the gift to be simple. Cognition, 34, 279–298. Smith, E. R. (2000). Subjective experience of familiarity: Functional basis in connectionist memory. In H. Bless & J. P. Forgas (Eds.), The message within: The role of subjective experience in social cognition and behavior (pp. 109–124). Philadelphia: Psychology Press. Thornhill, R., & Gangestad, S. W. (1993).Human facial beauty: Averageness, symmetry, and parasite resistance. Human Nature, 4, 237–269. Thornhill, R., & Gangestad, S. W. (1999). Facial attractiveness. Trends in Cognitive Science, 3, 452–460. Thornhill, R., & Møller, A. P. (1997). Developmental stability, disease and medicine. Biological Reviews, 72, 497–548. Tooby, J., & Cosmides, L. (1992). The psychological foundations of culture. In J. H. Barkow & L. Cosmides (Eds.), The adapted mind: Evolutionary psychology and the generation of culture (pp. 19–136). London: Oxford University Press. Whitfield, T. W., & Slatter, P. E. (1979). The effects of categorization and prototypicality on aesthetic choice in a furniture selection task. British Journal of Psychology, 70, 65–75. Whittlesea, B. W., Jacoby, L. L., & Girard, K. (1990). Illusions of immediate memory: Evidence of an attributional basis for feelings of familiarity and perceptual quality. Journal of Memory and Language, 29, 716–732. Winkielman, P., Halberstadt, J. B., Fazendeiro, T., & Catty, S. (2006). Faces are attractive because they are easy on the mind. Psychological Science, 17, 799–806. Zajonc, R. B. (1968). Attitudinal effects of mere exposure. Journal of Personality and Social Psychology, 9(Suppl. 2, Pt. 2). Zajonc, R. B. (1980). Feeling and thinking: Preferences need no inferences. The American Psychologist, 35, 151–175.


262 EVOLUTION AND THE SOCIAL MIND Zebrowitz, L. A. (1997). Reading faces: Window to the soul? Boulder, CO: Westview Press. Zebrowitz, L. A., Fellous, J, Mignault, A., & Andreoletti, C. (2003). Trait impressions as

overgeneralized responses to adaptively significant facial qualities: Evidence from connectionist modelling. Personality and Social Psychology Review, 7, 194–215.



Social Pressures on Cognitive Evolution On Social Scientists Trying to Predict People The Averseness of Being the Target of Prediction The Social Prediction Dynamic: Delineating the Theoretical Framework Situational Factors Moderate the Need to be Unpredictable Conclusion hat does evolutionary psychology have to do with social cognition? A lot. Following in the footsteps of other scholars, we argue that cognition should be calibrated to the tasks and challenges that confront a species on a recurring basis (e.g., in this volume, Kenrick, Delton, Robertson, Becker, & Neuberg, chapter 4; Lieberman, chapter 11; and also Tooby & Cosmides, 1992). Although some researchers have proposed that for many species, including humans, such challenges have been ecological in nature (e.g., Clutton-Brock & Harvey, 1980; Gibson, 1986), for example, needing to




exploit the environment, a recurring set of challenges for humans is intrinsically tied to group life (cf. Dunbar, 1992; Dunbar, chapter 2, this volume; Humphrey, 1976; Tooby & Cosmides, 1992). Thus, despite the various ecological problems humans and their ancestors have faced over millennia, which by the way have likely varied as a function of niche and geography, the perpetual challenge for all humans, regardless of what niche they have inhabited, has been having to navigate the social sphere and to live in groups. One consequence of this is that cognition, shaped by evolutionary processes, is at its core attuned to the social world.

SOCIAL PRESSURES ON COGNITIVE EVOLUTION Being attuned to the social world can mean various things, but a basic social cognitive process (call it a fundamental social motive) for all people is the need to understand and predict other people’s behavior; this is what we all do, not just social scientists. By knowing others’ mental states, such as their intentions and desires, we can anticipate others’ actions. Thus, “good enough” guesses of others’ mental states should facilitate social interaction (e.g., Asch, 1952; Heider, 1958; Tagiuri, 1958). However, smooth and coordinated social interaction is an outcome people should want from their interactions with kin, friends, and other trusted individuals. But the mental transparency that facilitates coordinated interaction is not something people should want from their interactions with unknown or contentious parties who could potentially predict a person’s behaviors to the detriment of the predicted. The concern people should have with predicting others, and, as we will argue, with having others predict them, may be an outcome of primate brain evolution. As noted already, in contrast to the idea that primate intelligence and primate brains evolved in response to ecological pressures, other research indicates that many brain characteristics can be better explained in terms of features of the social environment (e.g., Dunbar, 1992). This should be expected given the potential for benefits but also costs associated with group living (Humphrey, 1976). Consistent with this perspective, studies have shown that across various primate species the typical group size for a species correlates with neocortex size, such that larger brains are associated with living in larger groups (Barton & Dunbar, 1997; Dunbar, 1992; Sawaguchi & Kudo, 1990). This suggests that social complexity (e.g., tracking others, triadic awareness, etc.) was an important force in brain evolution (Barton & Dunbar, 1997; de Waal, 1998). Moreover, recent research has shown that the degree of deception displayed by various primate species is highly related to neocortex size (Byrne & Corp, 2004). The above findings suggest that although theory of mind and our social perception apparatus allows for some ability to predict others, which should prove advantageous in social domains, the findings on deception in primates


suggest that with social prediction also comes the ability to manipulate and to keep from being predicted (cf. Humphrey, 1976), which should in itself be advantageous under specific circumstances. It does not make sense that everyone under all circumstances would have the ability to predict and thus manipulate everyone else’s behaviors. Under some circumstances, individual advantage would accrue not only to those good at social prediction (e.g., able to read minds), but also those better able to keep their own minds from being known and their behavior from being predicted (Miller, 1997; also see Krebs & Dawkins, 1984). We propose that such easily activated reactions in people, especially when confronted with unfamiliar, uncertain, or threatening circumstances, contribute to the difficulty social psychologists and social scientists more generally have in predicting and chasing this social animal. Before delineating our theoretical framework further and describing some initial studies from our lab, we will review approaches taken by social scientists over the last few decades to predict their reticent and less than cooperative human participants. In addition, we will review various findings in a subsequent section that detail some reactions people experience at the prospect of being predicted and figured out.

ON SOCIAL SCIENTISTS TRYING TO PREDICT PEOPLE Social scientists often lament that human behavior is less predictable than the topics studied by the “harder” sciences, ostensibly because many factors go into determining why people do the things they do. Normally, this unpredictability is treated as the “noise” or measurement error of social science studies. However, as some investigators have pointed out, such noise may reflect meaningful behavior in complex systems (Gilden, 2001) and has actually been modeled and used to inform theory in the physical and biological sciences (Miller, 1997). Applying our perspective to this issue, we argue that the trouble with predicting human behavior, the noise, may be due in part to the fact that many social science studies trigger people’s need to be opaque, unpredictable, and to remain unknown.

What People Say and Do Social psychologists have long regarded people’s attitudes and beliefs and their reports on them as central to predicting social behavior. However, it was not long before they discovered that the road from such reports to corresponding behavior was not straightforward. The extent to which attitudes predict behavior is regarded as one of the most important controversies in attitude research (cf. Fishbein & Ajzen, 1975; Kraus, 1995; Wicker, 1969). Early findings by LaPiere (1934) showing a lack of correspondence between prejudiced attitudes and prejudiced behavior served as the impetus for much



research aimed at determining the factors that augment the attitude/behavior correlation. Some of these factors are psychometric in nature and deal with equivalence in measurement (Ajzen & Fishbein, 1977). Others include the stability of the attitude over time (Kraus, 1995), the strength and importance of the attitude (e.g., Krosnick, Boninger, Chuang, Berent, & Carnot, 1993), attitude accessibility (e.g., Fazio, Chen, McDonel, & Sherman, 1982), and whether or not the attitude is formed through direct experience (e.g., Regan & Fazio, 1977) (for a meta-analysis see Kraus, 1995). In addition, factors other than the attitude itself, such as the norms surrounding the performance of certain behaviors, also impact the relationship (e.g., Ajzen, 1991; Ybarra & Trafimow, 1998). As the catalog of factors above suggests, researchers have made great strides in uncovering the many conditions that affect an attitude’s influence on behavior. However, given the diversity of approaches to this issue, with no one approach applying broadly and robustly, we are left with the feeling that predicting people’s behavior from their attitudes requires considerable knowledge of the nature of attitudes, knowledge of various situational factors and the person’s state of mind, and so on. To be cognizant of all these factors should amount to an overwhelming task for the most expert researchers, let alone the lay person. Predicting people’s behavior is difficult not only because behavior is determined by many factors and situational cues, but also because, as we argue, at a fundamental level people do not want to be sized up and do not want their behavior predicted. Thus, people either do not provide accurate information on their attitudes and intentions or they intentionally behave in ways that are inconsistent with their attitudes. Given the trouble with predicting people’s behavior from their expressed beliefs and attitudes, it makes much sense that researchers have attempted to assess the content of people’s minds in ways that preclude people’s tendency not to be predicted.

The Bogus Pipeline One first attempt to assess people’s attitudes indirectly was carried out by Jones and Sigall (1971). These researchers created the bogus pipeline, an impressive contraption made up of a pile of electronic hardware with dials the experimenter could secretly manipulate. The machine was presented to participants as a type of lie detector, whereas participants in the control condition were simply asked to indicate their attitudes on a paper and pencil measure. In one study, for example, Sigall and Page (1971) found that students readily expressed more racial prejudice in the presence of the bogus pipeline than the control condition (for a review of similar findings, see Roese & Jamieson, 1993). Research efforts have moved beyond the bogus pipeline to trying to assess people’s mental contents in a manner that minimizes participant control, such as using measures of implicit cognition and techniques that allow researchers to assess brain activity. We review these trends next.


Implicit Measures of Cognition and Brain Activity There has been an explosion of research on implicit attitudes and cognition and their measurement (e.g., Dovidio, Kawakami, Johnson, Johnson, & Howard, 1997; Fazio & Olson, 2003; Greenwald & Banaji, 1995; Wilson, Lindsey, & Schooler, 2000). Implicit attitude assessment usually rests on people’s response latencies to a stimulus of interest. For example, after activating a stereotype of a social group in memory outside conscious awareness (e.g., by subliminally presenting race-relevant words such as white or black), a question of interest might be whether people are subsequently faster at identifying words that are positive or negative in valence, the assumption being that faster recognition for the former represents an implicit positive White bias (e.g., Wittenbrink, Judd, & Park, 1997). According to research on implicit attitudes, attitudes may be implicit for various reasons (for a review see Wilson & Dunn, 2004), and there are important theoretical reasons for undertaking such approaches to their measurement. For example, as both philosophers and psychologists have mused, human behavior is many times prompted and directed by cues in the environment of which the person has little knowledge or awareness (Bargh, 1996; Dennett, 1991). But, regardless of the assessment technique used or the theoretical approach taken, we believe there are potentially other reasons for these research efforts: Researchers may believe that by doing so they are better able to excavate and make their way to the true nature of people’s minds. An interest in implicit cognition may be deemed necessary because the people being studied many times do not want to be known and predicted. Research in social neuroscience and neuroscience more generally, which many times relies on brain imaging techniques (e.g., PET, fMRI), may similarly serve the unacknowledged purpose of measuring people’s minds more directly by bypassing people’s social and communicative processes. These different techniques have become of great interest to psychologists because they provide a sense that it is possible to assess participants’ responses in a manner that precludes the various problems associated with gaining valid information from people. By examining people’s implicit responses or peering straight into the brain structures themselves (assuming there is good consensus as to what those structures do), researchers may gain confidence that they are assessing “true” responses. We offer our proposals as informed speculation. Researchers may go to great lengths to study cognition in the ways described above because it cannot be studied by asking the participants themselves. And one reason they cannot rely on participants’ responses is that people many times do not want to be known and predicted (whether consciously or not).



THE AVERSENESS OF BEING THE TARGET OF PREDICTION We argue that in general people should find the prospect of being the object of prediction aversive. Thus, at times, they may want to avoid these unpleasant states or circumstances by being unpredictable, a hypothesis we test in studies to be reviewed subsequently. At this point, our aim is to review several findings from the literature to show across various paradigms that people are prone to feel scrutinized and as if they are being watched by others. These various reactions make sense in the context of our framework, in that such aversive feelings may signal to people that they should exit such situations or enact responses that will foil others’ social prediction attempts.

Day-to-Day Paranoia One such phenomenon is referred to as paranoid cognition (e.g., Kramer, 1994), which relates to a person’s beliefs, for example, that they are being persecuted (cf. Colby, 1981). More recently researchers have explored the social and situational aspects of paranoid cognition instead of focusing on individual pathology (e.g., Fenigstein & Vanable, 1992; Zimbardo, Andersen, & Kabat, 1981). Such research suggests that people in many social situations, especially novel ones, may be prone to feelings of uncertainty about what others think of them (cf. Festinger, 1954). Hence, people are apt to be sensitive to various social cues and to ruminate about them (Colby, 1981). Although this can be regarded as a normal process, certain situational factors and reactions can transform these cognitions into more dysfunctional forms, such as greater self-consciousness and the feeling that one is being scrutinized (Lord & Saenz, 1985). Social uncertainty can also increase people’s misinterpretation of events (and elaboration of such misinterpretations), which can result in feelings of discomfort in the presence of others. Social uncertainty can also lead to judgments that others are not open to them and can give rise to distrust (as reported in Kramer & Wei, 1999). Uncertain social situations can prompt people to feel as if they are being scrutinized. But it is also interesting to consider the possibility that people, regardless of the circumstances, are generally inclined to such thinking. For example, in one sample of 324 college students, researchers found that 47% of those participants had had some experience of paranoia (Ellet, Lopes, & Chadwick, 2003). In a different study, Fenigstein and Vanable (1992, Study 1) constructed and validated a measure of paranoia. In this study paranoid responses were highly and negatively correlated with social desirability, meaning people seemed aware of the potential for social disapproval from endorsing such items. Nevertheless, many respondents felt that some of the paranoid experiences described in the items applied to them. The above findings suggest that even people from nonclinical samples may be apt to feel on occasion that they are being watched, that there is a need to be suspicious, and to potentially assume ill will on the part of others. Other


phenomena to be reviewed presently further corroborate the idea that people often do feel scrutinized, and as we suggest, feel that they are being subjected to other’s prediction efforts. Below we discuss these reactions, namely the “spotlight” effect and the “illusion of transparency.” Spotlight Effect. The “spotlight” refers to an effect in which people overestimate the degree to which their actions and appearance are noticed and evaluated by others (Gilovich, Medvec, & Savitsky, 2000; Gilovich & Savitsky, 1999). Interestingly, the feelings that others are taking special notice and that one is being watched are thought of as classical indicators of paranoid ideology (e.g., Magaro, 1980; Millon, 1981). Fenigstein and Vanable (1992, Study 2) showed, for example, that the more people were predisposed toward having paranoid thoughts, the more they felt they were being observed. In two studies on the “spotlight” effect, participants were asked to wear a t-shirt with a potentially embarrassing or flattering image (Gilovich et al., 2000, Studies 1 and 2). In these studies participants overestimated the number of people who would be able to remember the images on the shirts. Thus, participants felt that they stood out and that others were scrutinizing them much more so than was actually the case. Feeling as if one is in the spotlight thus represents one way in which most people can feel as if others are taking notice and scrutinizing the self.

The feelings of transparency studied by Gilovich and colleagues (e.g., Gilovich & Savitsky, 1999; Vorauer & Ross, 1999) may be slightly different manifestations of the same reactions underlying the spotlight effect and paranoid cognition more generally. According to Gilovich and colleagues, to feel transparent is for a person to feel as if their thoughts, feelings, and reactions are leaking out for all to see. Evidence for the transparency effect comes from various sources. For example, Barr and Kleck (1995) had participants watch a video and then judge how expressive they had been during the viewing. The results showed that the participants rated themselves as more expressive than a group of observers judged them to be. The researchers then had the participants watch the covertly recorded videos. The participants actually expressed surprise as to how inexpressive they seemed in the video recordings. This study nicely illustrates that people are prone to think that they are being explicitly analyzed, and that their internal states are readily available to others. Due to different factors, the above review suggests that people are predisposed to assume some form of paranoid thinking, which usually coincides with overgeneralization of scrutiny and the unpleasant feelings of selfconsciousness (Fenigstein & Vanable, 1992). We suggest that such states should be unpleasant to people because they serve as a signal that the situation they are in is one in which others are potentially keen on figuring them out and predicting their behavior.

Feeling Transparent.



THE SOCIAL PREDICTION DYNAMIC: DELINEATING THE THEORETICAL FRAMEWORK In describing the efforts of social scientists to find better ways to ascertain valid information from their participants we have aimed to highlight the person perception or prediction side of the social prediction dynamic. Our discussion of people’s readiness to feel scrutinized speaks to the responses of the person as the object of prediction. Up to this point, we have thus focused on the form the social prediction dynamic takes when people find themselves in uncertain, contentious, or unfamiliar social circumstances. But social prediction and people’s responses to prediction are dynamic in that people should be sensitive not only to the negative and threatening aspects of social situations but also their positive and propitious characteristics as well. Thus, it is important to consider the social interaction factors that may highlight people’s need to be predicted in some situations but not others.

Cooperation, Competition, and the Benefits and Costs of Group Living Group living creates benefits but also potential costs for individuals (Barash, 1977; Humphrey, 1976). For example, group living brings together potential mates but also increases competition for them. Social living is beneficial in the acquisition of food, but competition surfaces when resources are scarce. Living with others aids in the defense against predators or other groups, but costs include the problem of free-riders. Thus, group living is and was likely to be a complex and ever shifting mix of both competitive and cooperative scenarios. At the same time, while being able to predict another’s behavior was probably adaptive regardless of the situation, the adaptive benefit of being predicted was probably contextual, depending on whether the situation was cooperative or competitive. We propose that reactions to others’ attempts to predict the self have been shaped by natural selection to be sensitive to the cooperative (safe, favorable) or competitive (threatening, unfavorable) nature of the immediate situation. Cooperative and safe contexts should shift one toward lessened discomfort at the prospect of being known and predicted. In such situations, benefits should result from mutual prediction by known and trusted others. Many activities throughout human and primate evolution have required coordinated responses and rest on the individual’s ability to build and maintain coalitions for mutual protection against enemies or individuals who abuse power within the group (Boehm, 1997; de Waal, 1998). Hunting and food-gathering activities can also require high levels of coordination. The benefits of mutual “mind reading” between members of such cooperative groups should not require elaboration.


In addition, members of a group must be transparent enough to members of the opposite sex for courting and reproduction to take place. Although there is much uncertainty in the courting of a mate, there will be no offspring and protection of those offspring if some transparency of intentions, desires, and needs fails to be established between the courting parties and, eventually, their offspring. In essence, we predict transparency and predictability to increase to the degree that fates are shared between predictors and the targets of prediction. Importantly, such a shift should have at least as much to do with the situation as it does the people involved; a desire to be transparent to someone in one context (e.g., hunting) does not necessarily transfer to a desire to be transparent to the same person in a different context (e.g., coalition building). Competitive contexts should do the opposite: people should strive to be opaque and unpredictable to unfamiliar or potentially contentious parties (cf. Miller, 1997). The adaptive response to many situations throughout human and primate evolution, involving both “ingroup” and “outgroup” members, should have depended on the individual’s ability to keep oneself from being known, sized-up, and predicted by others when this information could be used to one’s detriment. Considerations of signal detection theory as applied to human cognitive evolution (Haselton & Buss, 2003; Nesse, 2001) would predict a cognitive system that commits fewer of the costliest errors at the expense of committing more of the less costly errors. Thus, we might expect that the occasional event of being predicted by a competitor would be more costly than the occasional event of being unpredictable to a friend, which should incline humans, as a default, to be unpredictable, opaque, and difficult to “figure out.” Some of the benefits (or reduction in costs) of being unknown and unpredictable in competitive situations would include not being harmed, betrayed, or taken advantage of. Avoiding such negative outcomes should have proven useful in winning competitions, conflict resolution, securing resources and perhaps mates, and maintaining a given role or rank within a group. Some of the costs of being unknown and unpredictable when the situation was truly cooperative might be that one would be distrusted, disliked, or ostracized by potential cooperative partners. Thus, for uncertain or contentious social circumstances, we predict that people should want to be opaque and unpredictable. In the next section we test some implications of our model, which suggest that people’s reactions to being sized up and predicted should be responsive to the contentious or cooperative (threatening or safe) nature of the social interaction. Although people may generally be biased toward being unpredictable, our perspective holds that people should relax these tendencies when circumstances prove to be cooperative or more generally safe and favorable.



SITUATIONAL FACTORS MODERATE THE NEED TO BE UNPREDICTABLE Evidence for Changes in Self-Reported Behavioral Orientations In two pretests (n = 366) (Ybarra, Keller, Baron, Chan, Garcia, & SanchezBurks, 2006), we created and validated measures that assessed three behavioral orientations reflecting unpredictability and not wanting to be known. These included: Unpredictability, which assessed the degree to which participants felt they were complex, difficult to understand, and inconsistent from situation to situation; unwillingness to self-disclose, which measured reticence and an unwillingness to open up to others; and deceptiveness, which reflected a willingness to use deception in dealing with people. In validating the dependent measures we treated social distrust as a measure of competition and as the predictor variable and then ran multivariate regressions on the three subscales. We predicted that people who viewed their social world as more competitive would report a greater degree of these behavioral orientations. The regression analyses were consistent with expectations. In both surveys, the more people tended to see the world as competitive, the more they judged themselves as unpredictable, less willing to self-disclose, and as more deceptive. Having validated the dependent measures above, we conducted an experiment in which we led participants to believe they would interact with another participant in a competitive or cooperative situation (Ybarra et al., 2006, Study 1). We also included a control group of participants who did not expect to be involved in any interaction. Regardless of condition the participants blindly selected a colored badge from a bag at the beginning of the session and pinned the badge to their shirts. Participants were then asked to find their “opponent” (competition condition) or “partner” (cooperation condition) with the same badge color and sit across from them. The participants were told they and the person sitting across from them would go to another room to play a game called “Matching Wits.” We manipulated the participants’ expectations of competition or cooperation by describing this game differently depending on condition. Prior to going forward with the game, we asked participants to fill out some personality questionnaires for a second, unrelated study. Embedded in the battery were the three subscales validated in the pretests above. We expected that participants who anticipated to compete in the upcoming game would report being more unpredictable, less willing to disclose information about themselves, and more deceptive than participants who anticipated to cooperate in the upcoming interaction. Based on our analysis, the default for people is to express tendencies toward unpredictability. So, we did not expect the control condition to differ significantly from the two experimental conditions. This was indeed the case, although it did produce results that descriptively lay between the two experimental conditions (Figure 16.1). The main focus in this study was on the two experimental


FIGURE 16.1 The three measures (+/− 1 standard error) as a function of experimental


conditions, which we expected to differ from each other, with competition participants reporting being more unpredictable, less willing to disclose, and more deceptive than participants who expected to cooperate. In testing our specific prediction, the contrast comparing the competition versus cooperation condition was significant (p < .01). The univariate contrasts conducted on each dependent measure separately were significant for unwillingness to self-disclose (p < .01) and unpredictability (p < .01), and marginally so for deceptiveness (p = .06). Thus, consistent with expectations, participants’ behavioral tendencies were modified depending upon whether they thought they were going to compete or cooperate.

Evidence that Behavior Changes Toward Greater or Lesser Predictability In the above studies participants reported on behavioral tendencies that had to do with being unpredictable, not wanting to be known, and being deceptive. But the limitation of these data is that they only inform us about how people judged or defined themselves, not how they actually behaved as a function of their interpersonal expectations. In the next experiment (Ybarra et al., 2006, Study 2), we wanted to examine people’s behavior more directly. This study, although supposedly measuring how people describe themselves, actually provides a behavioral measure of unpredictability, as determined by people’s use of the survey instrument itself. How can we infer a shift in behavior toward more or less predictability? Well, surveys can be created to show this. On a standard semantic differential scale, participants choose between two bipolar or opposing trait adjectives, such as “interesting” on one end and “boring” at the other end, and the intervening intervals are delineated so that the midpoint represents no endorsement of



“interesting” or “boring.” However, what we did in this experiment was give participants a modified semantic differential scale in which they had to choose one of two bipolar traits or they had the option of choosing “both” or “neither” if they felt both traits captured what they were like or if they did not feel either of the traits was self-descriptive, respectively. We argue that the use of the “both” response serves as a marker of behavioral unpredictability, as the person is ambiguating what they are like by indicating they can assume either of the opposing dispositional qualities, in addition to being cagey. The critical question we examined in this study was whether a subtle experimental intervention aimed at shifting people’s interpersonal expectations could create changes in the use of these survey responses. We expected that people with negative interpersonal expectations (to be described presently) would choose the “both” option to a greater extent than people with positive interpersonal expectations. In this study (Ybarra et al., 2006, Study 2) we induced interpersonal expectations by having people in the first part of two, ostensibly unrelated studies bring to mind and write about a person they knew. In the negative interpersonal expectancies condition the participants were asked to bring to mind the image of a person they disliked. In the positive interpersonal expectations condition participants were asked to bring to mind the image of a person they liked. After completing this part of the experiment, the participants were presented with the second, unrelated study in which they were asked to endorse from a list those traits that described them. The traits were presented as pairs (n pairs = 21) on a horizontal line. Participants had to choose one of the traits or if they wanted they could choose “both” or “neither.” The main results for the study are presented in Table 16.1. “Neither” responses were infrequent and thus were not analyzed. The results are presented as an interaction of the two factors for ease of comprehension. However, given the correlated nature of the scale responses (i.e., if participants select a trait they will not select “both” or vice versa), we conducted the analysis on participants’ difference scores (no. of “both” responses—no. of selected traits) as a function of condition, which yields the exact same result (p < .05). As indicated in Table 16.1, participants who brought to mind an image of a person they did not like Table 16.1 The Mean Number of Times Specific Traits and the “Both” Option were Selected as a Function of Experimental Condition Condition

No. of Traits No. of “Both”

Recall Liked Other (n = 18)

Recall Disliked Other (n = 17)

11.28 (3.49) 9.28 (3.61)

8.47 (4.23) 11.65 (3.99)


chose “both” more often than participants who recalled a liked other, who in turn chose more specific traits than participants who recalled the disliked other. The findings from this experiment thus show that by simply having participants bring to mind the image of a liked or disliked other, we impacted their survey responses. More specifically, the use of the “both” option in the survey provided people with the opportunity to be unpredictable and hide the self. As predicted by our model and consistent with the results from the previous experiment, this tendency toward not wanting to be known and to be unpredictable was more likely when the interpersonal context was negative or threatening than positive or safe.

CONCLUSION Our general purpose in this chapter has been to put forth a theoretical framework that emphasizes not only the prediction side of social perception but also how the object of prediction, many times a moving target, responds to the prospect of being predicted. We argue that our behavioral tendencies carry the imprint of our ancestral social past, with a sensitivity to the threatening or safe nature of interpersonal situations. We believe the analysis has broad implications for other variables that affect the tenor of social interactions or relationships, for example, the relative power between parties, whether the interaction is one-shot or long term, or other factors that influence trust or approachability more generally. But, given the frequency with which people are likely to find themselves in uncertain social situations in modern times, it may be that we have inherited mechanisms that predispose us too easily and too often toward not wanting to be known and toward unpredictability. It is such reactions that may many times lay the foundation for the trouble with predicting human behavior and the lament of the social scientist. Hence, the present analysis reveals the value in predicting human behavior in unpredictable ways.

REFERENCES Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50, 179–211. Ajzen, I., & Fishbein, M. (1977). Attitude– behavior relations: A theoretical analysis and review of empirical research. Psychological Bulletin, 84, 888–918. Asch, S. E. (1952). Social psychology. New York: Prentice Hall. Barash, D. P. (1977). Sociobiology and behavior. New York: Elsevier. Bargh, J. A. (1996). Automaticity in social

psychology. In E. T. Higgins & A. W. Kruglanski (Eds.), Social psychology: Handbook of basic principles (pp. 169–183). New York: Guilford Press. Barr, C. L., & Kleck, R. E. (1995). Self–other perception of the intensity of facial expressions of emotion: Do we know what we show? Journal of Personality and Social Psychology, 68, 608–618. Barton, R. A., & Dunbar, R. I. M. (1997). Evolution of the social brain. In A. Whiten & R. E. Byrne (Eds.), Machiavellian intelligence


276 EVOLUTION AND THE SOCIAL MIND II: Extensions and evaluations (pp. 240–263). Cambridge, UK: Cambridge University Press. Boehm, C. (1997). Egalitarian behaviour and the evolution of political intelligence. In A. Whiten & R. E. Byrne (Eds.), Machiavellian intelligence II: Extensions and evaluations (pp. 341–364). Cambridge, UK: Cambridge University Press. Byrne, R. W., & Corp, N. (2004). Neocortex size predicts deception rate in primates. Proceedings of the Royal Society, London, 271, 1693–1699. Clutton-Brock, T. H., & Harvey, P. H. (1980). Primates, brains and ecology. Journal of Zoology, 207, 151–169. Colby, K. M. (1981). Modeling a paranoid mind. Behavioral and Brain Sciences, 4, 515–560. Dennet, D. C. (1991). Consciousness explained. Boston: Little, Brown. De Waal, F. (1998). Chimpanzee politics. Baltimore: Johns Hopkins University Press. Dovidio, J. F., Kawakami, K., Johnson, C., Johnson, B., & Howard, A. (1997). On the nature of prejudice: Automatic and controlled processes. Journal of Experimental Social Psychology, 33, 510–540. Dunbar, R. I. M. (1992). Neocortex size as a constraint on group size in primates. Journal of Human Evolution, 20, 469–493. Ellet, L., Lopes, B., & Chadwick, P. (2003). Paranoia in a nonclinical population of college students. Journal of Nervous and Mental Disease, 191, 425–430. Fazio, R. H., Chen, J., McDonel, E. C., & Sherman, S. J. (1982). Attitude accessibility, attitude–behavior consistency, and the strength of the object–evaluation association. Journal of Personality and Social Psychology, 18, 339–357. Fazio, R. H., & Olson, M. A. (2003). Implicit measures in social cognition research: Their meaning and uses. Annual Review of Psychology, 54, 297–327. Fenigstein, A., & Vanable, P. A. (1992). Persecutory ideation and self-consciousness. Journal of Personality and Social Psychology, 62, 129–138. Festinger, L. (1954). A theory of social comparison processes, Human Relations, 7, 117–140. Fishbein, M., & Ajzen, I. (1975). Belief, atti-

tude, intention and behaviour: An introduction to theory and research. Reading, UK: AddisonWesley. Gibson, K. R. (1986). Cognition, brain size, and the extraction of embedded food resources. In J. G. Else & P. C. Lee (Eds.), Primate ontogeny, cognitive and social behavior (pp. 93–105). Cambridge, UK: Cambridge University Press. Gilden, D. L. (2001). Cognitive emissions of 1/f noise. Psychological Review, 108, 33–56. Gilovich, T., Medvec, V. H., & Savitsky, K. (2000). The spotlight effect in social judgment: An egocentric bias in estimates of the salience of one’s own actions and appearance. Journal of Personality and Social Psychology, 78, 211–222. Gilovich, T., & Savitsky, K. (1999). The spotlight effect and the illusion of transparency: Egocentric assessments of how we are seen by others. Current Directions in Psychological Science, 8, 165–168. Greenwald, A. G., & Banaji, M. R. (1995). Implicit social cognition: Attitudes, selfesteem, and stereotypes. Journal of Personality and Social Psychology, 102, 4–27. Haselton, M. G., & Buss, D. M. (2003). Biases in social judgment: Design flaws or design features? In J. Forgas, K. Williams, & B. von Hippel (Eds.), Responding to the social world: Implicit and explicit processes in social judgments and decisions (pp. 21–43). New York: Cambridge University Press. Heider, F. (1958). The psychology of interpersonal relations. New York: Wiley. Humphrey, N. K. (1976). The social function of intellect. In P. P. G. Bateson & R. A. Hinde (Eds.), Growing points in ethology (pp. 303– 317). Cambridge, UK: Cambridge University Press. Jones, E. E., & Sigall, H. (1971). The bogus pipeline: A new paradigm for measuring affect and attitude. Psychological Bulletin, 76, 349–364. Kramer, R. M. (1994). The sinister attribution error: Paranoid cognition and distrust in organizations. Motivation and Emotion, 18, 199–230. Kramer, R. M., & Wei, J. (1999). Social uncertainty and the problem of trust in social groups: The social self in doubt. In T. R. Tyler, O. P. John, & R. M. Kramer (Eds.), The psychology of the social self (pp. 145–170).


Mahwah, NJ: Lawrence Erlbaum Associates, Inc. Kraus, S. J. (1995). Attitudes and the prediction of behavior: A meta-analysis of the empirical literature. Personality and Social Psychology Bulletin, 21, 58–75. Krebs, J. R., & Dawkins, R. (1984). Animal signals: Mindreading and manipulation. In J. R. Krebs & N. B. Davies (Eds.), Behavioral ecology: An evolutionary approach (pp. 380– 402). Oxford, UK: Blackwell Scientific. Krosnick, J. A., Boninger, C. S., Chuang, Y. C., Berent, M. K., & Carnot, C. G. (1993). Attitude strength: One construct or many related constructs. Journal of Personality and Social Psychology, 65, 1132–1151. LaPiere, R. T. (1934). Attitudes versus actions. Social Forces, 13, 230–237. Lord, C. G., & Saenz, D. S. (1985). Memory deficits and memory surfeits: Differential cognitive consequences of tokenism for tokens and observer. Journal of Personality and Social Psychology, 49, 918–926. Magaro, P. A. (1980). Cognition in schizophrenia and paranoia: The interpretation of cognitive processes. Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. Miller, G. F. (1997). Protean primates: The evolution of adaptive unpredictability in competition and courtship. In A. Whiten & R. W. Byrne (Eds.), Machiavellian intelligence II: Extensions and evaluations (pp. 312–340). Cambridge, UK: Cambridge University Press. Millon, T. H. (1981). Disorders of personality. New York: Wiley. Nesse, R. M. (2001). The smoke detector principle: Natural selection and the regulation of defenses. Annals of the New York Academy of Sciences, 935, 75–85. Regan, D. T., & Fazio, R. (1977). On the consistency between attitudes and behavior: Look to the method of attitude formation. Journal of Experimental Social Psychology, 13, 28–45. Roese, N. J., & Jamieson, D. W. (1993). Twenty years of bogus pipeline research: A critical review and meta-analysis. Psychological Bulletin, 114, 363–375. Sawaguchi, T., & Kudo, H. (1990). Neocortical

development and social structure in primates. Primates, 31, 283–290. Sigall, H., & Page, R. (1971). Current stereotypes: A little fading, a little faking. Journal of Personality and Social Psychology, 18, 247–255. Tagiuri, R. (1958). Introduction. In R. Tagiuri & L. Petrullo (Eds.), Person perception and interpersonal behavior (pp. 9–16). Stanford, CA: Stanford University Press. Tooby, J., & Cosmides, L. (1992). The psychological foundations of culture. In J. Barkow, L. Cosmides, & J. Tooby (Eds.), The adapted mind: Evolutionary psychology and the generation of culture (pp. 19–136). New York: Oxford University Press. Vorauer, J. D., & Ross, M. (1999). Selfawareness and feeling transparent: Failing to suppress one’s self. Journal of Experimental Social Psychology, 35, 415–440. Wicker, A. W. (1969). Attitudes versus actions: The relationship of verbal and overt behavioral responses to attitude objects. Journal of Social Issues, 25, 41–78. Wilson, T. D., & Dunn, E. W. (2004). Selfknowledge: its limits, value, and potential for improvement. Annual Review of Psychology, 55, 493–518. Wilson, T. D., Lindsey, S., & Schooler, T. Y. (2000). A model of dual attitudes. Psychological Review, 107, 101–126. Wittenbrink, B., Judd, C., & Park, B. (1997). Evidence for racial prejudice at the implicit level and its relationship with questionnaire measures. Journal of Personality and Social Psychology, 72, 262–274. Ybarra, O., Keller, M. C., Baron, A. S., Chan, E., Garcia, S. M., & Sanchez-Burks, J. (2006). Predicting human unpredictability: The effect of cooperative and competitive social circumstances. Manuscript under review. Ybarra, O., & Trafimow, D. (1998). How priming the private self or collective self affects the relative weights of attitudes and subjective norms. Personality and Social Psychology Bulletin, 24, 362–370. Zimbardo, P. G., Andersen, S. M., & Kabat, L. G. (1981). Induced hearing deficit generates experimental paranoia. Science, 212, 1529–1531.



The Evolution of an Ostracism Detection System


Introduction The Evolutionary Importance of Inclusion in Groups Model of Ostracism Detection Experimental Tests of the Ostracism Detection System The Indiscriminate Ostracism Detection System Implications and Conclusions

INTRODUCTION For animals, ostracism from the group is often the first step toward starvation and early death (Goodall, 1986). Ostracism, exclusion and rejection by others (Williams, 2001), is a ubiquitous phenomenon among humans, occurring around the world and throughout history. In this chapter, we suggest that evolutionary pressures favoring group and cooperative living shaped an ostracism detection system that is sensitive to rejection from groups and close others. As such, humans are acutely sensitive to actual and threatened rejection. After ostracism has occurred, the system works to cope with the rejection and, in many situations, obtain reacceptance into the group. Drawing on Williams’s (1997, 2001, 2007) model of ostracism detection, we elaborate some of the evolutionary bases of the model. We first outline some general principles in evolutionary psychology that provide a framework for understanding the adaptive significance of an ostracism detection system. We then review aspects of Williams’s model and discuss illustrative research where appropriate. We conclude with a discussion of the implications of an 279


evolutionary perspective for research on ostracism, an illustrative study, and some directions for future research.

THE EVOLUTIONARY IMPORTANCE OF INCLUSION IN GROUPS It is clear that humans are social animals (Aronson, 1999; Caporael, 2001; see all chapters in this volume). In early evolutionary development, individuals depended upon each other and group interaction to complete necessary survival functions (Buss & Kenrick, 1998; Dunbar, chapter 2, this volume). Thus, early humans’ survival and propagation of their genetic line was directly tied to their group’s survival, and early humans faced a recurring adaptive problem of monitoring and maintaining their inclusion in groups. Theoretically, individuals who quickly detected and responded to threats to inclusion (i.e., actual or threatened ostracism) would have had a fitness advantage over those who did not. Evidence that an ostracism detection system responds to such threats with relative efficiency and economy provides evidence that the system is a domain-specific adaptation (Andrews, Gangestad, & Matthews, 2002; see also Simpson & LaPaglia, chapter 10, this volume) that may meet the requirements for special design (Tooby & Cosmides, 1992). Another example of a domain-specific psychological mechanism tied to group living is reasoning about social exchange. Cosmides and Tooby (Cosmides, 1989; Cosmides & Tooby, 2005) argue that as early group interactions expanded to involve social exchange with individuals beyond immediate kin, a new problem arose: Some individuals may opt to cheat and fail to reciprocate their obligation. Thus, they argue that humans developed a cognitive module to reason about social exchange and cheater detection. Supportive evidence is provided in studies using a modified version of the Wason selection task (e.g., Cosmides, 1989; see also Badcock & Allen, chapter 8, this volume). Participants are asked to test whether a given rule has been violated, and performance on the original version is abysmal: Fewer than 30% of participants select the logically correct responses. However, when modified so that participants test whether a target has paid the cost for taking a benefit, performance improves to around 70% correct responses. Improved performance is also found on unfamiliar, abstract versions of the social contract, and performance on an unfamiliar switched version of the task is in a direction predicted by the adaptive logic of detecting cheaters. Thus, although humans’ logical reasoning may be flawed in many domains, humans reason about cooperation and social dilemmas in an efficient, effective, and adaptive manner. The ostracism detection system may be adaptive for selfish individuals, but it may also facilitate the coordination and behaviors of groups, reinforcing the system’s development and maintenance. Brewer and Caporael (Brewer, 1997; Caporael, 1997; Caporael & Brewer, 1991) argue that because early humans’ interactions with the physical environment were mediated by the group


environment, a primary adaptive problem was coordinating group activities to maintain safe and effective functioning. Several strategies may have developed to improve coordination of group activity, including affective influences in groups (Spoor & Kelly, 2004), mimicry (Lakin, Jefferis, Cheng, & Chartrand, 2003), and leadership (van Vugt & Kurzban, chapter 14, this volume). Ostracism and ostracism detection may also improve group coordination and predictability. Ostracism allows the group to exert control and maintain cooperation within the group (Ouwerkerk, Kerr, Gallucci, & van Lange, 2005), and may protect the group against targets with disease and poor fitness potential (Schaller & Duncan, chapter 18, this volume). Ostracism detection, in contrast, signals threats to inclusion and may prompt affiliation with the group. Because ostracism threatens group membership, which was fundamental for survival, individuals may be more sensitive to ostracism than other types of aversive social interaction. In the following sections, we review evidence suggesting that the ostracism detection system is an adaptive response to a recurring evolutionary problem. We also review evidence that the system helps to coordinate group activities and maintain smoother interactions

MODEL OF OSTRACISM DETECTION We are guided by Williams’s (Williams, 1997, 2001, 2007; Williams & Zadro, 2005) model of how targets respond to social ostracism, and extensive reviews of the model can be found elsewhere (e.g., Williams, 1997, 2001, 2007). For the present purposes, we focus on how targets immediately respond to ostracism, reflecting an adaptive response to evolutionary pressures. While also affected by evolutionary pressures, more intermediate responses are more likely to be mediated by social cognitive processes.

Immediate Reactions to Ostracism As discussed above, we assume that ostracism, or other threats to inclusion in groups, was a recurring adaptive problem, such that humans (and other social animals) would benefit from quickly and efficiently detecting such threats. Indeed, research in which ostracism is induced in the lab suggests that targets immediately react to perceived ostracism in an automatic, reflexive, and nearly universal way. Ostracized targets report pain, hurt feelings, and depressed mood after only brief (e.g., less than 5 minutes) exposure to ostracism (see Williams, 2007; Williams & Zadro, 2005). The immediate feelings of pain may be crucial in signaling ostracism threats. In humans, physical pain signals threats or problems in the environment (MacDonald & Leary, 2005), whereas social pain refers to the emotional reaction resulting from the realization that one is being excluded from important


relationships or groups. Social pain may also be manifested via mild depression (Badcock & Allen, chapter 8, this volume). Ostracism produces social pain, and pain affect in particular. Pain affect (Price, 2000) draws attention to the injury or experience as unpleasant, and also elicits emotional reactions that motivate actions to end the negative experience. Ostracism also triggers both the neurological and physiological mechanisms for responding to social pain. Using fMRI technology, Eisenberger, Lieberman, and Williams (2003) found that individuals who had been excluded showed increased activity in the area of the brain associated with both physical and social pain. Ostracism also triggers a generalized threat response system (MacDonald, Kingsbury, & Shaw, 2005), producing an analgesic effect of increased pain tolerance that allows one to temporarily ignore the pain and effectively react. McDonald et al. review research that this analgesic effect occurs in both animals and humans who experience isolation and ostracism. Simply remembering instances of being ostracized also results in presently-experienced pain levels exceeding dental pain, and on par with back pain and labor pain (Williams, Fitness, Newton, & Cheng, 2006). As discussed in more detail below (see “The Indiscriminate Ostracism Detection System” section), numerous studies have attempted to delineate minimal conditions in which exclusion will not affect targets. However, the cumulative evidence suggests that the ostracism detection system reacts indiscriminately to ostracism of all sorts. This system may at times overreact, mistakenly labeling a benign situation as ostracism. Ostracism detection may be a case in which the costs of errors are asymmetrical. Drawing on error management theory (Haselton & Buss, 2000; Haselton & Nettle, 2006), failing to detect ostracism may be incredibly costly in terms of fitness, while mistakenly perceiving ostracism that has not actually occurred may have relatively minimal costs.

Intermediate Responses to Ostracism After the immediate response to ostracism, targets move into an intermediate, reflective stage, in which the target’s responses are mediated by cognitive appraisal of how ostracism has threatened various needs. According to Williams (2001), ostracism potentially threatens four core needs: • • • •

belonging self-esteem control meaningful existence.

Each of these core needs is tied to inclusion in groups, reinforcing the development of the system over time. Ostracism may threaten only some or all of the core needs simultaneously, and the specific needs that are threatened may be critical in elucidating the target’s cognitive responses to ostracism, their


short-term coping, and the behaviors enacted. We discuss these differential reactions in the section on “Short-Term Coping with Ostracism”, and only briefly outline the general importance of each need next. There is clear evidence that interpersonal rejection, including ostracism, negatively impacts the need to belong (e.g., Leary, 2005; Williams & Zadro, 2005). The need to belong (Baumeister & Leary, 1995) contributes to individuals’ physical and mental well-being and functioning (see also Buck, chapter 6, this volume). Although belonging needs may be met by acceptance by a few important others (Baumeister & Leary, 1995), the importance of group living in evolutionary history suggests that even ostracism by a group of strangers will threaten belonging needs. There is also evidence that ostracism threatens belonging needs more strongly than other types of unpleasant interaction, such as verbal disagreement (Zadro, Williams, & Richardson, 2005). Ostracism may also threaten self-esteem, particularly social self-esteem, or how we perceive that others view our goodness and worth (Leary & Baumeister, 2000). Self-esteem serves as a proxy for belonging because it gauges one’s relative inclusion and worth within a group. Ostracism may be especially threatening to self-esteem because targets of ostracism are rarely given an explicit reason for the ostracism (Williams & Zadro, 2005). Instead, they are left to imagine the perhaps myriad reasons why their own actions may have incited the exclusion. Ostracism may also threaten the need for control over the environment (Seligman, 1975). In the ancestral environment, living in groups necessarily required a level of control and predictability (Brewer, 1997), thus ostracism threatens both adaptive inclusion in the group, as well as the sense of control and predictability offered by the group. After being ostracized, one has little control or influence over the course of an interaction or how the group members will react. Particularly if it is not obvious what offense was committed, the ostracized target may not even have a sense of how to go about regaining control. Finally, ostracism may threaten the need for meaningful existence. According to terror management theory (Greenberg, Pyszczynski, & Solomon, 1986), humans are strongly motivated to buffer the terror and fear of their own death and insignificance. In many tribes, social ostracism is one of the most severe forms of punishment (Case & Williams, 2004), and many terms for ostracism translate to a form of social death. Ostracism may increase mortality salience, prompting ostracized targets to imagine what the world would be like if they did not exist.

EXPERIMENTAL TESTS OF THE OSTRACISM DETECTION SYSTEM In this section we elaborate on some ways that ostracism has been studied using experimental methods, elucidating how targets respond in both the immediate and intermediate stages. We briefly review two commonly used methods of manipulating ostracism, the ball-toss paradigm and Cyberball, and then review


research examining how people cope with ostracism in both the short- and longterm.

Ball-Toss Paradigm and Cyberball Experimental studies of ostracism have demonstrated that targets respond quickly and powerfully to relatively minor instances of ostracism. For example, in early studies (e.g., Williams, 1997; Williams & Sommer, 1997) one real participant engaged in a ball-toss game with two confederates. Participants were randomly assigned to either be included (i.e., tossed the ball about one-third of the time), or excluded (i.e., tossed the ball a few times during the beginning of the game). The manipulation lasted for only 5 minutes and involved no other interaction among the players. Ostracized participants were visibly and negatively affected by only brief ostracism by two strangers. More recently, Williams, Cheung, and Choi (2000) developed Cyberball, an Internet version of the ball-toss game, which was designed to add more experimental control and further minimize any social cues present in the balltoss game. Participants logged onto an Internet website ostensibly to participate in a study of mental visualization, including a virtual game of ball toss with players at other locations. The other “players” were computer generated, and participants were randomly assigned to be included or ostracized. Even though participants had no direct contact with the other players, no shared history, nor any prospect of future interaction, the experience of ostracism was quite aversive. More recently, Cyberball has been used in laboratory studies with experiment participants (Williams & Jarvis, 2006), and being ostracized in the lab version results in similarly negative reactions as the ball-toss and Internet version of Cyberball. It should be noted that the negative effects of ostracism occur because of being excluded and not from knowing that others are being included. Smith and Williams (2004) conducted a study using cell phone text messages, in which excluded participants initially received messages and then did not receive any additional messages. Thus, participants could only imagine that they were being ostracized and could not determine whether the other participants were communicating with each other. Ostracized participants again reported depressed mood and reductions in the four needs consistent with other ostracism manipulations. The above examples demonstrate that the ostracism detection system is sensitive to subtle and relatively minor instances of ostracism. Although the actual ostracism episodes described above were typically brief and perpetrated by strangers, it was clear that the experience had powerful negative consequences. The impact of these relatively stripped-down manipulations also supports our contention that the ostracism detection system is acutely sensitive to any threat to inclusion.


Short-Term Coping with Ostracism According to the model, individuals respond to the immediate pain of ostracism by attempting to fortify the core needs. From an evolutionary perspective, the most adaptive response may often be to attempt to gain readmittance into a group, and there is some evidence that some individuals do respond to ostracism by attempting to affiliate. For example, Williams and Sommer (1997) had participants complete the ball-toss game and then work either collectively or coactively on a brainstorming task. Women who previously had been included tended to loaf when working collectively. However, women who previously had been excluded actually contributed more on the collective task. Presumably, these women were attempting to gain approval from the group by demonstrating that they were indeed “good” group members. Similar results were obtained in a study in which targets believed that they were playing Cyberball with two members of an ingroup, two members of an outgroup, or a mixed group (Williams et al., 2000). On a perceptual judgment task, ostracized participants were more likely to conform to an obviously incorrect response, especially after being ostracized by at least one ingroup member. Thus, ostracized targets often respond in ways that will help them regain acceptance and inclusion in a group. There is also some evidence that participants’ attempts to regain inclusion may occur automatically and without conscious awareness (Lakin & Chartrand, 2005). Lakin and Chartrand found that participants who were ostracized during Cyberball were more likely to nonconsciously mimic an interaction partner than those who had been included. Because mimicry is associated with a desire to affiliate with an interaction partner and also increases liking and rapport during the interaction (Lakin & Chartrand, 2003), they argued that the ostracized participants’ increased mimicry served the goal of reducing their own negative affective state, but also increased affiliation with the interaction partner. Targets of ostracism do not always respond with positive and affiliative behaviors, and some targets of ostracism have reacted with negative and aggressive behaviors (e.g., Twenge, Baumeister, Tice, & Stucke, 2001). Targets may be more likely to use aggression when control and meaningful existence needs are threatened (Warburton & Williams, 2005; Warburton, Williams, & Cairns, 2006; Williams, 2001). For example, Warburton et al. (2006) excluded or included participants using a variation of the ball-toss paradigm. Half of the participants were given the opportunity to restore control; the other half received a manipulation that further diminished their control. Later in the experiment, only ostracized participants in the diminished control condition were aggressive toward another participant. The intermediate response to ostracism may require more elaborate cognitive processing of the situation. Such processing may require a cost–benefit analysis of each reaction, as well as perspective taking at higher orders of intentionality (see Dunbar, chapter 2, this volume) to determine what the perpetrators intended to achieve by using ostracism. From an evolutionary perspective, the


substantial survival benefits of remaining in a group should increase the likelihood of prosocial behaviors aimed at affiliation and inclusion. There is some evidence that affiliation behaviors are more likely when the behavior is public and observable (Warburton & Williams, 2005), again suggesting that targets are sensitive to how their behaviors will be viewed by others. However, it is plausible that in some situations, aggression and antisocial behaviors may be a predictable and adaptive response. For example, cognitive appraisal of the costs and benefits of an aggressive versus a prosocial response may lead to aggression aimed at increasing control and predictability in the situation. Targets may also perceive that aggression will serve as retaliation, especially if the ostracism is perceived as unjustified. Aggressive responses may also deter the perpetrators from using ostracism in the future (i.e., the potential costs of aggression toward the group outweigh the costs of the target’s offense).

THE INDISCRIMINATE OSTRACISM DETECTION SYSTEM A number of studies have attempted to find moderators of the effects of ostracism, but these studies suggest that the ostracism detection system appears to react immediately and indiscriminately to ostracism of all sorts, even in those (perhaps rare) situations in which being ostracized may actually be beneficial for the target. For example, Gonsalkorale and Williams (in press) led participants to believe that they were playing Cyberball with members of either an ingroup, a rival outgroup, or a despised group (the Ku Klux Klan) and found that participants reacted negatively even when ostracized by members of a despised outgroup. Recent studies have also explored potential moderators by varying the meaning behind the ball tossing. In one variation on Cyberball, participants virtually tossed a bomb and responded negatively even when excluded from being tossed the bomb (van Beest & Williams, 2006). Although only symbolic, it suggests that individuals may actually feel bad being excluded from a game of Russian Roulette. Additionally, van Beest and Williams found that, when receiving the ball meant that money was deducted from one’s account, participants responded negatively. Thus, being ostracized meant that while the other players ended up with no money, the ostracized participant ended up with the maximum amount of money. Despite the relative monetary gain of being ostracized, ostracized participants felt bad. Finally, participants respond negatively to ostracism in the Cyberball paradigm, even when explicitly informed that they are playing the game with two computer-generated players or that the inclusion/exclusion is a scripted part of the experiment (Zadro, Williams, & Richardson, 2004). It may be that ostracism effects occur to the extent that participants construe the situation in social terms, thinking about other people as volitional agents (Law & Williams, 2006). Law and Williams have found preliminary evidence that using geometric forms


(squares and spheres) rather than traditional Cyberball icons and ball, combined with giving participants no instructions to mentally visualize the animated screen, results in no impact on the four needs. However, simply instructing participants to create a story around what they are viewing on the screen is sufficient to reduce the four needs and to increase sadness and anger. From an evolutionary perspective, it is not surprising that the ostracism detection system immediately responds to even mild (or mislabeled) instances of ostracism. The accumulation of evidence to date indicates that the system reacts similarly to any environmental cue of exclusion from a social/group situation. This finely tuned response is stable, persistent, and efficient, suggesting that it is a domain-specific adaptation (Andrews et al., 2002) resulting in a domainspecific psychological mechanism (Tooby & Cosmides, 1992). Additionally, the asymmetric costs of failing to detect true ostracism versus falsely detecting ostracism (Haselton & Buss, 2000; Haselton & Nettle, 2006) reinforce the use of the system across time.

Ostracism and Biological Fitness Humans are acutely sensitive to ostracism, which we argue derives from the fitness benefits of avoiding the dire consequences of exclusion from groups. However, other factors influence one’s fitness potential, including the fitness of one’s partner (or group). In other words, some relationships are highly desirable and have high fitness potential (e.g., physically attractive partners), while other relationships may actually be detrimental to fitness (e.g., diseased partners). It is possible that the ostracism detection system will distinguish and react differently to ostracism depending upon the relative fitness potential of the lost relationship. Sher, Vujic, Locke, and Williams (2006) tested this notion in two studies that examined immediate reactions to Cyberball-induced ostracism. In Study 1, participants were either ostracized or included by two other individuals who were either normally or highly physically attractive, where physical attractiveness served as a proxy for biological fitness. Consistent with previous studies, however, only a main effect for ostracism emerged. In Study 2, participants were ostracized or included by two individuals with or without a facial deformity, where the facial deformity served as a proxy for nonfitness (see Schaller & Duncan, chapter 18, this volume). Again, ostracism was distressing, regardless of the source. However, the pattern of means suggests that ostracism by individuals with the facial deformity was somewhat less distressing than ostracism by individuals without the facial deformity, especially in terms of meaningful existence. There was also some evidence that participants were sensitive to biological fitness when they had been included. Specifically, participants reported lower satisfaction levels of their needs when included by facially deformed individuals (but still higher than when ostracized). These studies suggest that the ostracism detection system functions to immediately detect and respond to any threat of ostracism, even in the face of other evolutionarily adaptive responses. Inclusion in a


group was so fundamental to early survival (Dunbar, chapter 2, this volume), that it may be more important to be included in any group, even one whose members have minimal fitness potential. The results of Study 2 are in line with this possibility, suggesting that reactions to factors related to biological fitness may reside in relatively safe inclusion circumstances rather than danger-signaling ostracism circumstances.

IMPLICATIONS AND CONCLUSIONS Reviewing the literature on ostracism and social exclusion, we conclude that there is good evidence that an ostracism detection system is an evolutionarily adaptive system that responds to protect organisms from threats to inclusion. Because ostracism signals potential death, and lost opportunities to reproduce, it would be evolutionarily adaptive for humans (or any social animal) to readily detect signs of ostracism. Thus, it is best to feel pain first, and ask questions later. This is demonstrated in the numerous studies that found that the immediate, automatic, and nearly universal response to ostracism is pain and hurt feelings. Additionally, as discussed in the previous section, ostracism may signal a more immediate danger, trumping signals of other forms of biological fitness. In the short term, the pain of ostracism can direct coping responses that can help individuals rehabilitate their behaviors to either gain readmittance into the ostracizing group, or become attractive for new groups. Thus, following ostracism, individuals have been shown to become more socially attentive (see Pickett & Gardner, 2005), hard-working (Williams & Sommer, 1997), conforming (Williams et al., 2000), and even gullible (Carter & Williams, 2005; Wheaton, 2001). They also engage in more nonconscious (Lakin & Chartrand, 2005) and strategic mimicry (Ouwerkerk et al., 2005). However, ostracism can also prompt seemingly dysfunctional reactions that may actually increase the probability of more ostracism. Under certain circumstances, ostracized individuals lose the ability to self-regulate (Baumeister, DeWall, Ciarocco, & Twenge, 2006), become cognitively impaired (Baumeister, Twenge, & Nuss, 2002), and lash out at the ostracizers and even naive others who had nothing to do with the ostracism (Twenge et al., 2001). From an evolutionary perspective, the prosocial response is probably more adaptive for most of the population. However, some researchers have argued that certain forms of aggression may have persisted because they were adaptive for a few (e.g., Wilson & Daly, 1996); thus, a few individuals may have benefited from responding aggressively to ostracism. Our hunch is that antisocial reactions occur only when ostracism (or social exclusion) is paired with a severe loss of control, when rehabilitative responses would appear to be fruitless. Thus, if the target is offered a genuine alternative to aggression (e.g., inclusion in another group or an alternative form of self-affirmation), antisocial responses may be avoided altogether.


Future research should be directed toward understanding more long-term coping with ostracism. Drawing on an evolutionary perspective, these more reflective and cognitively-mediated responses may be influenced by other fitness concerns. For example, individuals may react differently if the inclusion scenarios are clearly related to mating. Ostracism in a mating scenario may threaten different core needs than in a nonmating scenario because some needs, such as self-esteem and belonging, may be fulfilled via other group memberships. Rejection during mating, however, may threaten meaningful existence and control because, biologically speaking, organisms exist to propagate their genetic line. Thus, few hits at eHarmony (an online dating service) and no responses after speed dating may elicit different reactions than being rejected from a social club or at a party. Intermediate reactions to ostracism may also be affected by the target’s own fecundity. For example, women may be more sensitive to ostracism by males other than their partner while they are ovulating, but more sensitive to ostracism from their partner during other times of the cycle. We may also observe gender and developmental patterns in responses to ostracism by romantic partners. While women’s fertility decreases with age, men potentially can reproduce well into old age. Thus, men and women may react differently to ostracism at different points in their lives as they can seek out different routes to fortify their core needs. An evolutionary framework helps elucidate why ostracism continues to be such a ubiquitous phenomenon, and why individuals have quick and universal responses to even minor incidents of ostracism. The framework also prompts interesting questions and directions for future research. While individuals face a myriad of challenges each day, monitoring inclusion appears to be a fundamental and constant task, with powerful implications for short- and longterm coping.

ACKNOWLEDGMENT This chapter was supported by National Science Foundation grant 0519209-BCS and an Australian Research Council Discovery Grant awarded to K. Williams.

REFERENCES Andrews, P. W., Gangestad, S. W., & Matthews, D. (2002). Adaptationism: How to carry out an exaptationist program. Behavioral and Brain Sciences, 25, 489–553. Aronson, E. (1999). The social animal (8th ed.). New York: Freeman. Baumeister, R. F., DeWall, C. N., Ciarocco, N. L., & Twenge, J. M. (2006). Social exclusion

impairs self-regulation. Journal of Personality and Social Psychology, 88, 589–604. Baumeister, R. F., & Leary, M. R. (1995). The need to belong: Desire for interpersonal attachments as a fundamental human motivation. Psychological Bulletin, 117, 497– 529. Baumeister, R. F., Twenge, J. M., & Nuss, C. K.

290 EVOLUTION AND THE SOCIAL MIND (2002). Effects of social exclusion on cognitive processes: Anticipated aloneness reduces intelligent thought. Journal of Personality and Social Psychology, 83, 817–827. Brewer, M. B. (1997). On the social origins of human nature. In C. McGarty & S. A. Haslam (Eds.), The message of social psychology (pp. 54–62). Oxford, UK: Blackwell. Buss, D. M., & Kenrick, D. T. (1998). Evolutionary social psychology. In D. T. Gilbert, S. T. Fiske, & G. Lindzey (Eds.), The handbook of social psychology (4th ed., pp. 982–1026). New York: Oxford University Press. Caporael, L. R. (1997). The evolution of truly social cognition: The core configurations model. Personality and Social Psychology Review, 1, 276–298. Caporael, L. R. (2001). Parts and wholes: The evolutionary importance of groups. In C. Sedikides & M. B. Brewer (Eds.), Individual self, relational self, collective self (pp. 241– 258). New York: Psychology Press. Caporael, L. R., & Brewer, M. B. (1991). Reviving evolutionary psychology: Biology meets society. Journal of Social Issues, 47, 187–195. Carter, A. R., & Williams, K. D. (2005, May). Effects of ostracism on social susceptibility. Paper presented at the annual meeting of the Midwestern Psychological Association, Chicago. Case, T. I., & Williams, K. D. (2004). Ostracism: A metaphor for death. In S. L. Greenberg, S. L. Koole, & T. Pyszczynski (Eds.), Handbook of experimental existential psychology (pp. 336–351). New York: Guilford Press. Cosmides, L. (1989). The logic of social exchange: Has natural selection shaped how humans reason? Studies with the Wason selection task. Cognition, 31, 187–276. Cosmides, L., & Tooby, J. (2005). Neurocognitive adaptations designed for social exchange. In D. M. Buss (Ed.), Evolutionary psychology handbook (pp. 584–627). New York: Wiley. Eisenberger, N. I., Lieberman, M. D., & Williams, K. D. (2003). Does rejection hurt? An fMRI study of social exclusion. Science, 302, 290–292. Gonsalkorale, K., & Williams, K. D. (in press). The KKK won’t let me play: Ostracism by a

despised outgroup hurts. European Journal of Social Psychology. Goodall, J. (1986). Social rejection, exclusion, and shunning among the Gombe chimpanzees. Ethology and Sociobiology, 7, 227–236. Greenberg, J., Pyszczynski, T., & Solomon, S. (1986). The causes and consequences of the need for self-esteem: A terror management theory. In R. F. Baumeister (Ed.), Public self and private self (pp. 189–212). New York: Springer-Verlag. Haselton M. G., & Buss, D. M. (2000). Error management theory: A new perspective on biases in cross-sex mind reading. Journal of Personality and Social Psychology, 78, 81– 91. Haselton, M. G., & Nettle, D. (2006). The paranoid optimist: An integrative evolutionary model of cognitive biases. Personality and Social Psychology Review, 10, 47–66. Lakin, J. L., & Chartrand, T. L. (2003). Using nonconscious behavioral mimicry to create affiliation and rapport. Psychological Science, 14, 334–339. Lakin, J. L., & Chartrand, T. L. (2005). Exclusion and nonconscious behavioral mimicry. In K. D. Williams, J. P. Forgas, & W. von Hippel (Eds.), The social outcast: Ostracism, social exclusion, rejection, and bullying (pp. 279–296). New York: Psychology Press. Lakin, J. L., Jefferis, V. E., Cheng, C. M., & Chartrand, T. L. (2003). The chameleon effect as social glue: Evidence for the evolutionary significance of nonconscious mimicry. Journal of Nonverbal Behavior, 27, 145–162. Law, A. T., & Williams, K. D. (2006). The minimal conditions necessary to perceive ostracism. Working paper, Purdue University, West Lafayette, IN. Leary, M. R. (2005). Varieties of interpersonal rejection. In K. D. Williams, J. P. Forgas, & W. von Hippel (Eds.), The social outcast: Ostracism, social exclusion, rejection, and bullying (pp. 35–52). New York: Psychology Press. Leary, M. R., & Baumeister, R. F. (2000). The nature and function of self-esteem: Sociometer theory. In M. R. Leary (Ed.), Interpersonal rejection (pp. 3–20). New York: Oxford University Press. MacDonald, G., Kingsbury, R., & Shaw, S. (2005). Adding insult to injury: Social pain theory and response to social exclusion. In


K. D. Williams, J. P. Forgas, & W. von Hippel (Eds.), The social outcast: Ostracism, social exclusion, rejection, and bullying (pp. 77–90). New York: Psychology Press. MacDonald, G., & Leary, M. R. (2005). Why does social exclusion hurt? The relationship between social and physical pain. Psychological Bulletin, 131, 202–223. Ouwerkerk, J. W., Kerr, N. L., Gallucci, M., & van Lange, P. A. M. (2005). Avoiding the social death penalty: Ostracism and cooperation in social dilemmas. In K. D. Williams, J. P. Forgas, & W. von Hippel (Eds.), The social outcast: Ostracism, social exclusion, rejection, and bullying (pp. 321–332). New York: Psychology Press. Pickett, C. L., & Gardner, W. L. (2005). The social monitoring system: Enhanced sensitivity to social clues as an adaptive response to social exclusion. In K. D. Williams, J. P. Forgas, & W. von Hippel, The social outcast: Ostracism, social exclusion, rejection, and bullying (pp. 213–226). New York: Psychology Press. Price, D. D. (2000). Psychological and neural mechanisms of the affective dimension of pain. Science, 288, 1769–1772. Seligman, M. E. P. (1975). Helplessness: On depression, development, and death. San Francisco: Freeman. Sher, T., Vujic, T., Locke, V., & Williams, K. D. (2006). The effects of signals of biological fitness on reactions to ostracism and inclusion. Unpublished manuscript, University of Western Australia, Perth. Smith, A., & Williams, K. D. (2004). R U there? Ostracism by cell phone text messages. Group Dynamics: Theory, Research and Practice, 8, 291–301. Spoor, J. R., & Kelly, J. R. (2004). The evolutionary significance of affect in groups: Communication and group bonding. Group Processes and Intergroup Relations, 7, 398– 416. Tooby, J., & Cosmides, L. (1992). The psychological foundations of culture. In J. Barkow, L. Cosmides, & J. Tooby (Eds.), The adapted mind: Evolutionary psychology and the generation of culture (pp. 19–136). New York: Oxford University Press. Twenge, J. M., Baumeister, R. F., Tice, D. M., & Stucke, T. S. (2001). If you can’t join them, beat them: Effects of social exclusion

on aggressive behavior. Journal of Personality and Social Psychology, 81, 1058–1069. Van Beest, I., & Williams, K. D. (2006). When inclusion costs and ostracism pays, ostracism still hurts. Journal of Personality and Social Psychology, 91, 218–228. Warburton, W. A., & Williams, K. D. (2005). Ostracism: When competing motivations collide. In J. P. Forgas, K. D. Williams, & S. M. Laham (Eds.), Social motivation: Conscious and unconscious processes (pp. 294– 313). New York: Cambridge University Press. Warburton, W. A., Williams, K. D., & Cairns, D. R. (2006). When ostracism leads to aggression: The moderating effects of control deprivation. Journal of Experimental Social Psychology, 42, 213–220. Wheaton, A. (2001). Ostracism and susceptibility to the overtures of socially deviant groups and individuals. Unpublished Honours thesis, Macquarie University, Sydney, Australia. Williams, K. D. (1997). Social ostracism. In R. M. Kowalski (Ed.), Aversive interpersonal behaviors (pp. 133–170). New York: Plenum. Williams, K. D. (2001). Ostracism: The power of silence. New York: Guilford Press. Williams, K. D. (2007). Ostracism. In Annual Review of Psychology (Vol. 58). New York: Annual Reviews, Inc. Williams, K. D., Cheung, C. K. T., & Choi, W. (2000). CyberOstracism: Effects of being ignored over the Internet. Journal of Personality and Social Psychology, 79, 748–762. Williams, K. D., Fitness, J., Newton, N., & Cheng, Z. (2006). When hurt won’t heal: Exploring the capacity to relive social and physical pain. Manuscript under review. Williams, K. D., & Jarvis, B. (2006). Cyberball: A program for use in research on ostracism and interpersonal acceptance. Behavior Research Methods, Instruments, and Computers, 38, 174–180. Williams, K. D., & Sommer, K. (1997). Social ostracism by one’s coworkers: Does rejection lead to loafing or compensation. Personality and Social Psychology Bulletin, 23, 693–706. Williams, K. D., & Zadro, L. (2005). Ostracism: The indiscriminate early detection system. In K. D. Williams, J. P. Forgas, & W. von Hippel (Eds.), The social outcast: Ostracism, social exclusion, rejection, and bullying (pp. 19–34). New York: Psychology Press.

292 EVOLUTION AND THE SOCIAL MIND Wilson, M. I., & Daly, M. (1996). Male sexual proprietariness and violence against wives. Current Directions in Psychological Science, 5, 2–7. Zadro, L., Williams, K. D., & Richardson, R. (2004). How low can you go? Ostracism by a computer is sufficient to lower self-reported levels of belonging, control, self-esteem, and

meaningful existence. Journal of Experimental Social Psychology, 40, 560–567. Zadro, L., Williams, K. D., & Richardson, R. (2005). Riding the “O” train: Comparing the effects of ostracism and verbal dispute on targets and sources. Group Processes and Intergroup Relations, 8, 125–143.


The Behavioral Immune System Its Evolution and Social Psychological Implications MARK SCHALLER LESLEY A. DUNCAN

The Past The Present The Future ovember 27, 1966. For one of the authors (Mark Schaller) that was a bad day. It was his fourth birthday and his parents threw him a party, with balloons and ice cream and birthday party games. This wasn’t easy because the family was living on the Serengeti plains of Tanzania, where neither balloons nor ice cream (nor very many other young children for that matter) were readily available. Despite his parents’ intrepid efforts, the day was a disaster. Mark’s balloon popped. He dropped his ice cream into the dirt. He cried and cried and cried. Meanwhile, elsewhere in Tanzania, on exactly the same day, a fellow named McGregor was also having a bad day—a really bad day. Indeed, the minor setbacks of Mark’s birthday party are trivial in comparison to the truly tragic events that befell McGregor that day. McGregor lived on the eastern edge of Lake Tanganyika, in an area that is now Gombe National Park. McGregor was a chimpanzee. Chimpanzees are, like people, a highly social species. Chimps typically stay in very close contact with their fellow group members. Their health and reproductive success depend on it. They may spend several hours each day grooming each other—removing fleas, mites, and other ectoparasites from each other’s fur. For most of McGregor’s life,




his social experiences were not unusual. But, in 1966, a polio epidemic struck the Gombe chimpanzees, and McGregor fell victim. The consequences are described in detail by Goodall (1986). McGregor lost the use of his legs, forcing him to drag his body backwards with his arms, or to attempt a series of bizarre somersaults as a crude means of locomotion. He lost control of his bladder, and so his awkward movements were accompanied constantly by a buzzing swarm of flies. And, as if that wasn’t painful enough, McGregor’s physical privations precipitated near-complete social rejection—as indicated by observations recorded on November 27, 1966: Of the total number of 32 adult and adolescent chimpanzees who visited camp at the time, 17 approached the crippled male. . . . Only nine adults approached closely . . . and of these only four actually touched him (two aggressively). . . . Humphrey [possibly his biological nephew] was the only chimpanzee who sometimes slept within 20 meters of the stricken male. . . . Perhaps the most striking aspect was the fact that not once in the 24 hours was [he] involved in a session of social grooming. (Goodall, 1986, pp. 233–234)

Behavioral avoidance and social rejection of diseased individuals are observed not only in chimpanzees, but in many other species too. Mice avoid mating with mice that are infected with parasitic viruses, protozoa, and larval nematodes (e.g., Kavaliers, Colwell, Braun, & Choleris, 2003). Bullfrog tadpoles avoid swimming in proximity to tadpoles infected with debilitating intestinal parasites (Kiesecker, Skelly, Beard, & Preisser, 1999). Closer to home, human beings prefer to maintain distance from others who are described as diseased, especially if the alleged disease is perceived to be contagious (Crandall & Moriarty, 1995). Why is this? To some extent, the explanation is straightforward. Goodall (1986, p. 234) suggested that “avoidance of conspecifics showing abnormal behavior may be highly adaptive since it reduces the risk of spreading contagious disease.” More broadly, it has been suggested that many animals, including humans, have an evolved capacity to detect symptoms of parasitic infection in others, and to respond with behaviors—such as behavioral avoidance and social rejection—that reduce the likelihood of contracting that infection oneself (Eibl-Eibesfeldt, 1979; Kurzban & Leary, 2001; Schaller, Park, & Faulkner, 2003). Although that explanation is straightforward and perhaps even obvious, it has many additional implications that are more complex and a lot less obvious. A deeper consideration of how such an evolved process might operate yields a large number of novel implications for human social cognition and interpersonal behavior more generally. Evolved mechanisms designed to inhibit contact with disease-carrying conspecifics are likely to promote specific kinds of aversive reactions toward many specific kinds of people who are, in fact, perfectly healthy. The purpose of this chapter is to discuss some of these implications. But first, we need to consider carefully the selection pressures that presumably led to


the evolution of these psychological mechanisms in the first place. We must spend a bit of time in the past.

THE PAST Evolution of the Behavioral Immune System Parasites are an enduring part of human history. Infectious bacteria and viruses have existed on the planet far longer than people and other primates have; and as long as animals have had guts, those guts have been infected with helminths and worms (Brothwell & Sandison, 1967; Ewald, 1993; van Blerkom, 2003). Not all parasites are harmful, but many are. The European plague outbreak in the Middle Ages killed millions of people (Lippi & Conti, 2002). Bacterial diseases wiped out up to 90% of the native populations in the Americas (Guerra, 1993). These are relatively recent events, but the dangers posed by parasites are hardly recent phenomena. The very fact that humans and other animals have evolved extraordinarily sophisticated immune systems attests to the antiquity of parasitic infections, to the enormous selection pressures that parasitic infections have exerted on animal populations, and to the fitness-conferring benefits associated with any adaptation that contributes to an antiparasite defense system. The immune system is just one kind of antiparasite defense system, and, although effective in many ways, it has its downsides too. The mobilization of immunological defenses often consumes substantial metabolic resources, robbing individuals of energy that might be devoted to other fitness-enhancing tasks (Brown, 2003; Klein & Nelson, 1999). Specific features of immunological defense, such as fever, may be even further debilitating. Organisms are wellserved by the existence of an immune system, but they are best served when that immune system is engaged as infrequently as possible. In addition, the immune system is limited by the boundaries of physical anatomy. The immune system is designed to combat infectious agents only after they are detected upon contact with the individual’s body. It cannot prevent contact in the first place. Therefore, it is entirely plausible that selection pressures posed by parasites led to the evolution of an additional antiparasite defense system as well, one designed to inhibit contact with infectious agents in the first place. This system is comprised of a set of mechanisms that allow individuals to detect the potential presence of parasites in the objects and individuals around them, and to engage in behaviors that prevent contact with those objects and individuals. This has been called the behavioral immune system (Schaller, 2006). The operation of some sort of evolved behavioral immune system is implicated by abundant evidence—across many species—pertaining to foraging and feeding behavior. Sheep selectively avoid grazing on grasses contaminated with



their own fecal waste (Cooper, Gordon, & Pike, 2000). People too show a disgust reaction and behavioral rejection toward foods that are potentially contaminated by parasites (Rozin, Millman, & Nemeroff, 1986). Of course, it’s not just foods and other inanimate objects that host potentially-dangerous parasites. Other individuals do as well. So it’s no surprise that the behavioral immune system also compels avoidance, and even outright rejection, of conspecifics (like poor McGregor) that demonstrate symptoms of physical illness.

How It Works: Cue Detection and Response Any effective defense system requires the coordinated operation of at least two kinds of underlying mechanism: mechanisms designed to detect cues signaling threat, and other mechanisms that respond to those cues by mobilizing some sort of defensive response. This is the case for the “real” immune system. Specific mechanisms distinguish the difference between organic entities that belong in the body and those—like viruses—that don’t. When pathogenic intruders are detected, other mechanisms within the system are triggered that attempt to repel those pathogens through a variety of physiological means. In an analogous fashion, the behavioral immune system also is comprised of mechanisms designed for detection and response. The detection mechanisms employ the organism’s ordinary sensory organs as a means of recognizing parasite-connoting cues at a distance. Frogs use specific kinds of chemical signals for this purpose (Kiesecker et al., 1999). Many mammals use olfactory cues of some sort (Kavaliers et al., 2003), and surely people do too. In addition, given our highly-developed visual systems—which allow us to detect many different kinds of fitness-connoting signals from a distance—our parasite-detection mechanisms are sensitized to detect visual cues signaling possible parasitic infection. This makes sense, of course, given that the symptoms of many parasitic infections are manifest in individual’s superficial appearance or behavior (e.g., skin lesions, rashes, coughing spasms). The behavioral immune system also includes mechanisms designed to respond in functionally-useful (i.e., fitness-enhancing) ways once a parasiteconnoting cue has been detected. Behavioral avoidance is the functionallyrelevant “goal” for which these mechanisms are designed. But behavior doesn’t just happen; it is the product of underlying psychological activity. In humans, this activity involves both affective and cognitive mechanisms. Emotions are instrumental in motivating immediate behavioral reactions (see, in this volume, Buck, chapter 6; Ellsworth, chapter 5; Forgas, chapter 7; Lieberman, chapter 11). Both fear and disgust motivate behavioral avoidance. Disgust in particular seems likely to be an important part of the behavioral immune system. The capacity for disgust may have arisen originally to protect individuals from the ingestion of toxins and other food-based contaminants (Rozin & Fallon, 1987), but the mechanisms involved in the disgust experience appear to have evolved to serve a parasite-defense function as well. Disgust is


triggered by the visual perception of skin lesions, runny noses, and other obvious symptoms of parasitic infection (Curtis, Aunger, & Rabie, 2004; Curtis & Biran, 2001). Disgust may motivate an immediate and impulsive avoidant response, but that’s it. The emotional experience alone cannot compel wariness about future interactions, nor can disgust alone precipitate more planful actions (such as coordinated efforts at quarantine and social exclusion) that help to eliminate the long-term threat posed by possibly parasitized individuals. To facilitate these kinds of fitness-relevant behaviors, various cognitive processes must be engaged as well. In humans the detection of any parasite-connoting cue may have immediate implications on higher-order cognitive processes involved in inference and memory, which may then influence the specific nature of attitudes and other enduring social knowledge structures. These, in turn, are likely to have consequent effects on social decision making and behavior. If the behavioral immune system influenced reactions only to truly diseased individuals, it would still constitute a worthwhile topic of scientific inquiry, but would perhaps be of limited relevance to the broader range of social psychological phenomena. In fact, however, the behavioral immune system appears to operate in such a way that it often precipitates aversive reactions to individuals who are perfectly healthy. Consequently, it has direct implications for many phenomena that lie squarely in the center of the social psychological literature— including interpersonal attraction, intergroup prejudice, and social stigmas of various kinds. To understand why, it is useful to apply the logic of signal detection.

The Signal Detection Problem and its Solution: Oversensitivity and Overgeneralization The behavioral immune system is designed not to respond to the presence of parasites, per se, but rather to the perceived presence of parasites as indicated by superficial sensory signals. Many of these cues, presumably, are probabilistically predictive of the presence of parasites. But even the most diagnostic of symptoms is highly imperfect. (Some healthy people cough, and some sick people don’t.) The result is a classic signal-detection problem, with the potential to make both false-positive errors (a healthy person is erroneously perceived to be sick) and false-negative errors (a sick person is erroneously perceived to be healthy). Any general tendency toward avoiding false positives leads to an increase in the rate of false negatives, and vice versa. Evolutionary logic indicates that this dilemma will be resolved in favor of minimizing the error that poses the greatest costs to an individual’s fitness, even if that results in an increased rate of making the other kind of error (Haselton & Nettle, 2006; Nesse, 2005; for a broader discussion of signaling and signal-detection systems see Gangestad & Thornhill, chapter 3, this volume). In this case, as with most evolved systems designed for selfprotection, the fitness costs associated with false negatives are considerably



greater than those associated with false positives. The adaptive resolution is clear: The behavioral immune system errs on the side of false positives (Kurzban & Leary, 2001). Thus, we are hypervigilant for signs of sickness, and any such signal (whether it’s a tubercular cough or merely some innocuous guttural tic, whether it’s a rash of infectious pox or merely some superficial allergic inflammation) is liable to trigger aversive emotional, cognitive, and behavioral reactions. It is unlikely that there was a finite and stable set of symptoms associated with parasitic infections in ancestral environments. Different kinds of parasite would have produced different infectious symptoms. (The rash diseases—such as measles, mumps, and scarlet fever—are all evolutionarily ancient, as is tuberculosis, and all are associated with somewhat different specific symptoms.) Different individuals are likely to have responded differently to the same kind of parasitic infection (the rhinovirus may manifest in a cough, or in a runny nose, or both, or neither). And parasitic species themselves—especially bacteria and viruses—evolve at an exceptionally rapid pace, an evolution that is reflected in the highly variable nature of infectious symptoms over time (Ewald, 1993). A behavioral immune system that was calibrated too tightly to specific perceptual cues would have resulted, over time, in many costly false-negative errors. More adaptive would be a system that responded to a broader, more crudely-defined range of cues. This suggests that the behavioral immune system errs not merely on the side of oversensitivity, but also on the side of overgeneralization: Any gross deviation from the species-typical norms in morphology and motor behavior may be implicitly interpreted as symptomatic of a parasitic infection, and so may trigger the behavioral immune response (Kurzban & Leary, 2001; Zebrowitz & Montepare, 2006). Thus, the behavioral immune system operates in a manner analogous to the real immune system. Just as the antipathogen defense system provided by the real immune system is hypersensitive to intrusion, and may be mobilized in response to organic matter that is entirely benign (or even beneficial, as in the case of organ transplants), the behavioral immune system too responds in a hypersensitive and overgeneral way to the perceived presence of parasites in the sensory environment. This has far-reaching implications for social perception and behavior: Simply because people may display some superficial form of nonnormality, we may respond to them—even if they are perfectly healthy—as though they are carriers of some contagious disease.

The Cost–Benefit Problem and Its Solution: Functional Flexibility Antiparasite defense systems confer adaptive benefits, but they also incur costs whenever they are triggered. We have already mentioned the physiological costs associated with the mobilization of the real immune system. There are analogous costs associated with the operation of the behavioral immune system. The emotional, cognitive, and behavioral responses triggered by the behavioral immune


system all consume metabolic resources. And because of the finite resources available to an individual at any moment, the activation of the behavioral immune system limits the extent to which other adaptive behaviors might be engaged (e.g., disgust and behavioral avoidance are typically incommensurate with mating motives; for additional discussions of evolutionary cost–benefit analyses and their implications, see Simpson & LaPaglia, chapter 10, and Todd, chapter 9, both in this volume). Therefore, like many adaptive psychological systems, the operation of the behavioral immune system is likely to have evolved so as to be functionally flexible and responsive to regulatory cues (Schaller, Park, & Kenrick, 2007; see also Kenrick et al., chapter 4, this volume). Aversive responses to potentially-parasitized others are most likely to be triggered when additional cues in the immediate environment indicate that the functional benefits of these responses are especially likely to outweigh the functional costs. Some regulatory cues lie in chronic individual differences in attitudes, traits, and temperament. People differ in the extent to which they are vulnerable (or, perhaps more importantly, perceive themselves to be vulnerable) to the transmission of contagious diseases. Other regulatory cues lie in temporary features of the immediate situation. Information present in any specific context may make germs and their potential transmission especially salient for a short period of time. Still other cues lie in chronic features of the local ecology. In some geographical contexts, parasitic diseases have posed an especially strong threat to individual fitness, with persistent consequences on local rituals and norms pertaining to hygiene, food preparation, and so on. Regardless of the locus of these regulatory cues—whether chronic or temporary, and whether rooted in the external environment or a perceiver’s own idiosyncratic knowledge structures— the information they provide is likely to moderate the activation of the behavioral immune system. If one is unaware of (or feels invulnerable to) the threat of disease, the activation of the system is likely to be muted. On the other hand, if the threat of disease is highly salient (or if one feels highly vulnerable), the reactivity of the system is likely to be more pronounced.

THE PRESENT These speculations about the evolution of the behavioral immune system make sense within the adaptive framework of evolutionary psychology. Ideally, this kind of conjecture should not simply be sensible; it should also be useful—even to scholars who care nothing about the evolutionary past. In fact, the principles of adaptive overgeneralization and functional flexibility imply a broad range of effects on contemporary social cognition and behavior. Some of these implications have been empirically tested in recent years. The results highlight the operation of the behavioral immune system in a wide range of contemporary social psychological phenomena.



Aversive Responses to Superficial Disfigurements and Disabilities There is now a substantial body of work documenting aversive responses to people displaying non-normative morphological cues of various kinds, including superficial facial anomalies and physically disabling conditions. These aversive responses may result from a variety of conceptually distinct psychological processes, many of which have nothing to do with parasite avoidance at all (for reviews, see Heatherton, Kleck, Hebl, & Hull, 2000). Is there reason to suppose that, in addition to these other processes, the specific mechanisms implicated in the behavioral immune system also play a substantial role? Yes. Evidence in favor of that assertion emerges from studies that do at least one of two things. They take measures that assess the specific kinds of semantic information that are cognitively associated with morphologically anomalous individuals. Or they test the extent to which aversive responses are facilitated under circumstances in which perceivers feel more vulnerable to the potential spread of contagious disease. Or they do both. These studies not only implicate the role of the behavioral immune system in reactions toward a variety of objectively noncontagious peoples, they also document novel phenomena whereby these reactions vary under predictable circumstances. Park, Faulkner, and Schaller (2003) reported a pair of studies that implicate the behavioral immune system in aversive responses to individuals who are physically disabled. There is a large literature documenting the fact that people are uncomfortable around others who are disabled, and often attempt to behaviorally avoid close contact with these others (e.g., Snyder, Kleck, Strenta, & Mentzer, 1979). If this prejudice results in part from the heuristic operation of the behavioral immune system, it follows that behavioral avoidance might be especially strong among individuals who are chronically concerned about the spread of contagious diseases. Consistent with this hypothesis, Park et al. (2003) found that individuals who score highly on measure of “perceived vulnerability to disease” (PVD) were less likely to report having friends or acquaintances with disabilities. In addition, Park et al. employed reaction time methods to assess the extent to which disabled individuals (compared to morphologically normal individuals) were implicitly linked to semantic information connoting disease. Results revealed that, not only were disabled individuals more likely than nondisabled individuals to be associated with disease, this effect was stronger among perceivers who scored more highly on either the PVD measure or on a measure assessing sensitivity to disgust. In this implicit association study, the disabled target individuals were described in such a way that, by any objective standard, they posed no realistic disease threat whatsoever. The results are therefore consistent with the conjecture that the behavioral immune system responds automatically to visual cues of morphological anomaly, even when rational appraisal indicates the absence of any realistic threat. Duncan (2005) conducted a strong test of the alleged automaticity of this response. Participants were provided with brief biographical


sketches of two men, and each biographical sketch was accompanied by a facial photograph. One man had a very noticeable “port wine stain” birthmark on his face, but this birthmark was explicitly described as superficial and the man himself was described as strong and healthy. The other man looked just fine, but was described as suffering from a strain of drug-resistant tuberculosis. Participants then responded to a computer-based reaction time task, designed to assess which of the two men was more strongly associated with the semantic concept “disease.” Results showed that, across all participants, there was a general tendency to associate disease with the facially-disfigured man (who was known to be healthy) more strongly than the man who was actually known to suffer from a contagious disease (but who looked normal). In short, even when processes of rational appraisal explicitly indicate otherwise, facial disfigurements may implicitly connote the threat of contagious disease.

Antifat Attitudes Previous research has suggested that negative attitudes toward fat people are rooted, in part, in personal ideologies and cultural value systems that prescribe hard work, self-denial, and willpower (Crandall, 1994; Crandall & Martinez, 1996). Consistent with this perspective, fat people are commonly stereotyped as lazy, and are more strongly stigmatized when their obesity is attributed to personally-controllable causes (e.g., Teachman, Gapinski, Brownell, Rawlins, & Jeyaram, 2003). But fat people are also commonly stereotyped as dirty or smelly, and images of fat people tend to arouse disgust (Harvey, Troop, Treasure, & Murphy, 2002)—observations hinting at the possibility that antifat attitudes may also be rooted in the operation of the behavioral immune system. This possibility is entirely plausible, given our speculations about how the behavioral immune system operates. If the system is sensitive to any gross deviation from morphological norms, then it’s likely to react aversively to individuals with bodies that are either skeletally thin or hugely obese. There has been very little research examining aversive reactions to super-skinny people, but some recent studies explicitly examined whether the heuristic operation of the behavioral immune system might contribute to antifat attitudes. Park, Schaller, and Crandall (2006) examined whether antifat attitudes were predicted by individual differences in perceived vulnerability to disease— focusing specifically on a subscale that assesses wariness of germs and their transmission. Results indicated that these individual differences did indeed predict antifat attitudes: People who were chronically more concerned about germs also expressed a stronger dislike of fat people. This effect was especially strong when antifat attitudes were measured immediately after the visual perception of specific obese individuals—a result consistent with the idea that the behavioral immune system is hypersensitive to visual cues. It’s worth noting also that the effect on antifat attitudes was statistically independent of the predictive effect of separate measures assessing attributions about willpower. This suggests that



ideological processes and parasite-defense processes both contribute to antifat attitudes, but in different ways. This last conclusion is further substantiated by another study reported by Park et al. (2006). This experiment assessed cognitions implicitly associated with obese individuals, and examined the impact of a manipulation designed to make specific concerns temporarily salient. Results revealed that the implicit association linking fat people (compared to nonfat people) with disease was amplified following a manipulation that made infectious pathogens especially salient. The amplifying effect of the pathogen-salience manipulation emerged only on implicit associations linking fat people to disease; it did not increase associations linking fat people with unpleasant concepts in general. In contrast, a manipulation that made ideological concerns salient led to an increased implicit association between fat people and unpleasantness, but had minimal impact on the fat–disease association. These results not only have implications for understanding contemporary prejudices toward obese individuals, they also have unique implications for understanding the operation of the behavioral immune system itself. It might be logical to perceive dramatically underweight individuals as potential parasite carriers (given that many parasitic infections do result in substantial weight loss), but there is little logical basis to associate obesity with contagious parasites. Nor is there much reason to assume that truly obese individuals were evident in the ancestral environments during which the behavioral immune system presumably evolved. The results of Park et al. (2006) therefore highlight the heuristic (nonrational) operation of the behavioral immune system, and they highlight its adaptive overgeneralization. The behavioral immune system responds not merely to specific cues that were evident in ancestral environments; it appears to have evolved so as to respond to any kind of apparent morphological deviation from population norms.

Responses to Physically Attractive and Unattractive Others The behavioral immune system may be sensitive not only to gross deviations from morphological norms, but may also be attuned to some relatively subtle deviations—at least in the realm of facial physiognomy. Human visual systems are highly attuned to facial features. We have specialized neurological equipment dedicated to the visual perception of faces (Kanwisher, 2000). Our subjective impressions of another’s attractiveness are influenced by specific aspects of facial physiognomy (such as bilateral symmetry, and the extent to which the size of specific facial features match population prototypes) that we appear to process implicitly and without conscious awareness (e.g., Langlois & Roggman, 1990; see also Halberstadt, chapter 15, this volume). It has been argued that these sorts of subtle morphological variables are predictive of an individual’s health status and future health outcomes (Fink & PentonVoak, 2002; Thornhill & Gangestad, 1999). Consistent with this argument,


evidence reveals that not only are substantially anomalous faces judged to be less healthy, but so too are faces that are simply perceived to be subjectively less attractive (Zebrowitz, Fellous, Mignault, & Andreoletti, 2003; Zebrowitz & Rhodes, 2004). Gangestad and Buss (1993; see also Gangestad, Haselton, & Buss, 2006) report a particularly interesting finding bearing on the link between physical attractiveness and the presence of parasites. Employing a cross-cultural methodology to test a hypothesis about functional flexibility, they found physical attractiveness was an especially prized attribute in a mate within societies that historically had a high prevalence of infectious parasites. This evidence has been interpreted as indicating that a subjective assessment of another’s facial attractiveness serves as an indicator of that individual’s genetic fitness. However, the same evidence is consistent with a process whereby individuals use facial attractiveness (or rather, unattractiveness) as an heuristic indicating the actual presence of potentially-contagious parasites. Is there any special empirical reason to suppose that unattractiveness really does trigger the behavioral immune system? Possibly. If attractiveness was simply a clue to genetic fitness, one might expect the impact of physical (un)attractiveness to be rather constrained in scope—exerting effects primarily in the domain of mating relations, but of limited impact in other domains of social life. In fact, however, physical attractiveness is valued—and physical unattractiveness compels aversive responses—across a broad range of social inferences and interactions (Biddle & Hamermesh, 1998; Eagly, Ashmore, Makhijani, & Longo, 1991; Matter & Matter, 1989). In addition, if attractiveness was merely a cue to genetic fitness, then one might expect the effects of pathogen prevalence, described above, to be especially strong among female perceivers (because women are especially attentive to indicators of genetic fitness). In fact, however, the results of Gangestad et al. (2006) show the opposite effect: The moderating impact of pathogen prevalence was stronger among men than among women. These results don’t argue against the hypothesis that attractiveness serves as an heuristic cue for genetic fitness, but they do suggest that something else might be going on as well. That something else may be the operation of the behavioral immune system.

Xenophobia and Ethnocentrism In human populations, the behavioral immune system may be responsive not only to morphological cues, but also to a broader set of cues indicating that another individual is foreign to the local population. There are at least two plausible reasons why. First, contact with individuals from previouslyunencountered populations is associated with an increased risk of contracting contagious diseases to which one has no acquired immunity. Second, foreign peoples are likely to be unaware of, and more likely to violate, local customs (such as those pertaining to food preparation and personal hygiene) that serve as



barriers to the transmission of disease. Thus, in contemporary social ecologies, the mechanisms that define the behavioral immune system may generalize beyond the tendency to respond to cues signaling morphological anomaly; they may respond to cues signaling cultural foreignness as well. Regardless of their local social environment, individuals may be especially adept at learning to detect a wide range of inferential cues that discriminate between familiar and foreign peoples. And when those cues are detected, they may promote the familiar emotional, cognitive, and behavioral responses associated with the behavioral immune system. Consistent with this reasoning, Schiefenhövel (1997) observed that people often display disgust reactions when speaking about ethnic outgroups, and Rozin, Haidt, McCauley, and Imada (1997, p. 73) suggested that “disgust in humans serves as an ethnic or outgroup marker.” To more rigorously test this conjecture, Faulkner, Schaller, Park, and Duncan (2004) conducted a series of studies that exploited the logic of functional flexibility. In one set of studies, Faulkner et al. (2004) tested whether chronic concerns of vulnerability to parasitic infections—as measured by the “perceived vulnerability to disease” (PVD) scale—predicted attitudes towards immigrants from various geographical regions. Results revealed that higher levels of PVD predicted stronger anti-immigrant attitudes—but only toward immigrants from subjectively foreign locations. There was no such effect on attitudes toward culturally familiar immigrant populations. The contribution of the behavioral immune system to xenophobic attitudes was also implicated in a pair of experiments reported by Faulkner et al. (2004). In both experiments, participants were first exposed to a brief slide show that either made salient the potential dangers posed by germs and germ transmission, or (in a control condition) made salient other dangers that were irrelevant to disease (e.g., electrocution). Results from both experiments revealed more strongly xenophobic attitudes after germs (rather than disease-irrelevant threats) were made salient. For instance, in one of these experiments, participants in Vancouver were told about a government program designed to recruit new immigrants to Canada, and were asked to indicate how much money should be spent to recruit immigrants from a variety of different countries that had been prerated as either culturally familiar (e.g., Taiwan, Poland) or unfamiliar (e.g., Mongolia, Brazil). Participants who had been exposed to the control slide show allocated roughly equal amounts of money to recruit immigrants from both familiar and unfamiliar places, but those for whom germ transmission had been made salient were much more likely to allocate money to recruit immigrants from familiar rather than unfamiliar places. These findings are complemented by more recent work by Navarrete and Fessler (2006). In one study they observed that not only does perceived vulnerability to disease predict more negative attitudes toward foreign peoples (xenophobia), it also predicts more positive attitudes toward one’s own cultural ingroup (ethnocentrism). In another study, they found that another


disease-relevant individual difference variable—sensitivity to disgust—also predicts both xenophobia and ethnocentrism. These results do not diminish the importance of the many other psychological processes that contribute to xenophobia and ethnocentrism. There is no doubt that these phenomena are multiply-determined; they are influenced also by processes pertaining to fear, mistrust, conflict, social identity, and mere categorization. But the fact of those well-known processes should not blind us to the apparent role of a less obvious process that also contributes to xenophobia and ethnocentrism: The hypersensitive and overgeneralized operation of a psychological system designed to protect our bodies from contact with parasites.

THE FUTURE The results reviewed above suggest that the behavioral immune system has implications for a broad range of psychological responses to people who, in fact, may be completely healthy. In future research, it will be worthwhile to examine additional implications, perhaps particularly in the realm of actual interpersonal behavior. The subtle operation of the behavioral immune system may contribute, for instance, to many specific acts of aggression and social ostracism (e.g., see Spoor & Williams, chapter 17, this volume). It will also be worthwhile to consider implications that exist not merely at the individual level of analysis, but at the societal level of analysis. The behavioral immune system may play an important role in shaping the collective belief systems that define a culture (Schaller, 2006). One route is through interpersonal communication. Cultural norms are sculpted, often unintentionally, through communication processes (Schaller, 2001). People may be especially likely to communicate about things that seem relevant to disease and disease transmission—as indicated perhaps by the finding that disgust-arousing stories are especially likely to be communicated, and to become culturally popular (Heath, Bell, & Sternberg, 2001). Moreover, disease-relevant arguments and rhetorical devices may be especially persuasive in sculpting popular opinion and public policy. (The abundance of parasite-relevant imagery in Nazi propaganda offers one sobering historical example.) The intriguing upshot, still largely unexplored, is that the evolution of the behavioral immune system may not only exert a pervasive influence on human social cognition; it may also, as a consequence, influence human culture. REFERENCES Biddle, J. E., & Hamermesh, D. S. (1998). Beauty, productivity, and discrimination: Lawyers’ looks and lucre. Journal of Labor Economics, 16, 172–201.

Brothwell, D., & Sandison, A. T. (1967). Diseases in antiquity. Springfield, IL: Thomas. Brown, J. K. M. (2003). A cost of disease-


306 EVOLUTION AND THE SOCIAL MIND resistance: Paradigm or peculiarity. Trends in Genetics, 19, 667–671. Cooper, J., Gordon, I. J., & Pike, A. W. (2000). Strategies for the avoidance of faeces by grazing sheep. Applied Animal Behaviour Science, 69, 15–33. Crandall, C. S. (1994). Prejudice against fat people: Ideology and self-interest. Journal of Personality and Social Psychology, 66, 882– 894. Crandall, C. S., & Martinez, R. (1996). Culture, ideology, and antifat attitudes. Personality and Social Psychology Bulletin, 22, 1165–1176. Crandall, C. S., & Moriarty, D. (1995). Physical illness stigma and social rejection. British Journal of Social Psychology, 34, 67–83. Curtis, V., Aunger, R., & Rabie, T. (2004). Evidence that disgust evolved to protect from risk of disease. Proceedings of the Royal Society of London, Series B, 271, S131–S133. Curtis, V., & Biran, A. (2001). Dirt, disgust, and disease: Is hygiene in our genes? Perspectives in Biology and Medicine, 44, 17–31. Duncan, L. A. (2005). Heuristic cues automatically activate disease cognitions despite rational knowledge to the contrary. Unpublished master’s thesis, University of British Columbia, Vancouver, BC, Canada. Eagly, A. H., Ashmore, R. D., Makhijani, M. G., & Longo, L. C. (1991). What is beautiful is good, but . . .: A meta-analytic review of research on the physical attractiveness stereotype. Psychological Bulletin, 110, 109– 128. Eibl-Eibesfeldt, I. (1979). The biology of peace and war. New York: Viking. Ewald, P. W. (1993). Evolution of infectious disease. New York: Oxford University Press. Faulkner, J., Schaller, M., Park, J. H., & Duncan, L. A. (2004). Evolved disease-avoidance mechanisms and contemporary xenophobic attitudes. Group Processes and Intergroup Behavior, 7, 333–353. Fink, B., & Penton-Voak, I. (2002). Evolutionary psychology of facial attractiveness. Current Directions in Psychological Science, 11, 154–158. Gangestad, S. W., & Buss, D. M. (1993). Pathogen prevalence and human mate preferences. Ethology and Sociobiology, 14, 89–96. Gangestad, S. W., Haselton, M. G., & Buss, D. M. (2006). Evolutionary foundations of

cultural variation: Evoked culture and mate preferences. Psychological Inquiry, 17, 75–95. Goodall, J. (1986). Social rejection, exclusion, and shunning among the Gombe chimpanzees. Ethology and Sociobiology, 7, 227–239. Guerra, F. (1993). The European-American exchange. History and Philosophy of the Life Sciences, 15, 313–327. Harvey, T., Troop, N. A., Treasure, J. L., & Murphy, T. (2002). Fear, disgust, and abnormal eating attitudes: A preliminary study. International Journal of Eating Disorders, 32, 213–218. Haselton, M. G., & Nettle, D. (2006). The paranoid optimist: An integrative evolutionary model of cognitive biases. Personality and Social Psychology Review, 10, 47–66. Heath, C., Bell, C., & Sternberg, E. (2001). Emotional selection in memes: The case of urban legends. Journal of Personality and Social Psychology, 81, 1028–1041. Heatherton, T. F., Kleck, R. E., Hebl, M. R., & Hull, J. G. (2000). The social psychology of stigma. New York: Guilford Press. Kanwisher, N. (2000). Domain specificity in face perception. Nature Neuroscience, 3, 759– 763. Kavaliers, M., Colwell, D. D., Braun, W. J., & Choleris, E. (2003). Brief exposure to the odour of a parasitized male alters the subsequent mate odour responses of female mice. Animal Behaviour, 65, 59–68. Kiesecker, J. M., Skelly, D. K., Beard, K. H., & Preisser, E. (1999). Behavioral reduction of infection risk. Proceeding of the National Academy of Sciences, 96, 9165–9168. Klein, S. L., & Nelson, R. J. (1999). Influence of social factors on immune function and reproduction. Reviews of Reproduction, 4, 168–178. Kurzban, R., & Leary, M. R. (2001). Evolutionary origins of stigmatization: The functions of social exclusion. Psychological Bulletin, 127, 187–208. Langlois, J. H., & Roggman, L. A. (1990). Attractive faces are only average. Psychological Science, 1, 115–121. Lippi, D., & Conti, A. A. (2002). Plague, policy, saints and terrorists: A historical survey. Journal of Infection, 44, 226–228. Matter, D. E., & Matter, R. M. (1989). If beautiful is good, then ugly must be . . .: Confronting discrimination against the physically


unattractive child. Elementary School Guidance and Counseling, 24, 146–152. Navarrete, C. D., & Fessler, D. M. T. (2006). Disease avoidance and ethnocentrism: The effects of disease vulnerability and disgust sensitivity on intergroup attitudes. Evolution and Human Behavior, 27, 270–282. Nesse, R. M. (2005). Natural selection and the regulation of defenses: A signal detection analysis of the smoke detector principle. Evolution and Human Behavior, 26, 88–105. Park, J. H., Faulkner, J., & Schaller, M. (2003). Evolved disease-avoidance processes and contemporary anti-social behavior: Prejudicial attitudes and avoidance of people with physical disabilities. Journal of Nonverbal Behavior, 27, 65–87. Park, J. H., Schaller, M., & Crandall, C. S. (2006). Psychological disease-avoidance mechanisms and stigmatization of fat people. Unpublished manuscript, University of Groningen, The Netherlands. Rozin, P., & Fallon, A. E. (1987). A perspective on disgust. Psychological Review, 94, 23–41. Rozin, P., Haidt, J., McCauley, C., & Imada, S. (1997). Disgust: Preadaptation and the cultural evolution of a food-based emotion. In H. MacBeth (Ed.), Food preferences and taste (pp. 65–82). Providence, RI: Berghahn. Rozin, P., Millman, L., & Nemeroff, C. (1986). Operation of the laws of sympathetic magic in disgust and other domains. Journal of Personality and Social Psychology, 50, 703–712. Schaller, M. (2001). Unintended influence: Social-evolutionary processes in the construction and change of culturally-shared beliefs. In J. P. Forgas & K. D. Williams (Eds.), Social influence: Direct and indirect processes (pp. 79–93). Philadelphia: Psychology Press. Schaller, M. (2006). Parasites, behavioral defenses, and the social psychological mechanisms through which cultures are evoked. Psychological Inquiry, 17, 96–101. Schaller, M., Park, J. H., & Faulkner, J. (2003). Prehistoric dangers and contemporary prejudices. European Review of Social Psychology, 14, 105–137.

Schaller, M., Park, J. H., & Kenrick, D. T. (2007). Human evolution and social cognition. In R. I. M. Dunbar & L. Barrett (Eds.), Oxford handbook of evolutionary psychology (pp. 491–504). Oxford, UK: Oxford University Press. Schiefenhövel, W. (1997). Good tastes and bad tastes: Preferences and aversions as biological principles. In H. MacBeth (Ed.), Food preferences and taste (pp. 55–64). Providence, RI: Berghahn. Snyder, M. L., Kleck, R. E., Strenta, A., & Mentzer, S. J. (1979). Avoidance of the handicapped: An attributional ambiguity analysis. Journal of Personality and Social Psychology, 37, 2297–2306. Teachman, B. A., Gapinski, K. D., Brownell, K. D., Rawlins, M., & Jeyaram, S. (2003). Demonstrations of implicit anti-fat bias: The impact of providing causal information and evoking empathy. Health Psychology, 22, 68– 78. Thornhill, R., & Gangestad, S. W. (1999). Facial attractiveness. Trends in Cognitive Sciences, 3, 452–460. Van Blerkom, L. M. (2003). Role of viruses in human evolution. Yearbook of Physical Anthropology, 46, 14–46. Zebrowitz, L. A., Fellous, J. M., Mignault, A., & Andreoletti, C. (2003). Trait impressions as overgeneralized responses to adaptively significant facial qualities: Evidence from connectionist modeling. Personality and Social Psychology Review, 7, 194–215. Zebrowitz, L. A., & Montepare, J. (2006). The ecological approach to person perception: Evolutionary roots and contemporary offshoots. In M. Schaller, J. A. Simpson, & D. T. Kenrick (Eds.), Evolution and social psychology (pp. 81–113). New York: Psychology Press. Zebrowitz, L. A., & Rhodes, G. (2004). Sensitivity to “bad genes” and the anomalous face overgeneralization effect: Cue validity, cue utilization, and accuracy in judging intelligence and health. Journal of Nonverbal Behavior, 28, 167–185.


Author Index Acitelli, L. K., 198, 207 Ackerman, J. M., 58, 60, 66 Adams, H. E., 213, 227 Adolphs, R., 109, 122, 127, 139 Agyei, Y., 217, 226 Ainsworth, M. D. S., 129, 139 Ajzen, I., 265, 266, 275, 276 Aktipis, C. A., 234, 242 Albright, L., 204, 207 Alexander, M. G., 7, 16 Alexander, R. D., 40, 47, 166, 175 Alicke, M. D., 220, 226 Allan, S., 128, 130, 141 Allee, W. C., 95, 103 Allen, N. B., 3, 14, 65, 108, 125, 126, 130–135, 137–140, 254, 280, 282 Alliger, G. M., 237, 242 Alter, A., 121, 122 Ambady, N., 111, 122 Andersen, S. M., 268, 277 Anderson, J. L., 40, 46 Andersson, M. B., 162–164, 175 Andreoletti, C., 255, 262, 303, 307, 308 Andrews, P. W., 39, 46, 163, 175, 280, 287, 289 Angleitner, A., 222, 226 Angyal, A., 190, 192 Arendt, H., 100, 103 Aristotle, 118, 123 Arnold, M., 71, 77, 86 Aronson, E., 242, 280, 289 Asch, S. E., 4, 16, 264, 275 Ashmore, R. D., 303, 307 Astington, J. W., 23, 30 Atkinson, R. C., 50, 66 Aunger, R., 297, 307 308

Averill, J. R., 130, 137, 139 Axelrod, R., 183, 193 Badcock, P. B. T., 3, 14, 65, 108, 125, 126, 130, 131–135, 137–140, 254, 280, 282 Bailey, J. M., 61, 68, 164, 176, 203, 208, 217, 222, 226 Baird, J., 40, 48 Banaji, M. R., 267, 276 Barash, D. P., 214, 223, 224, 226, 270, 275 Barbee, A. P., 167, 176, 247, 259 Bargh, J. A., 218, 221, 226, 228, 267, 275 Barkow, J. H., 2, 16, 31, 67, 68, 124, 128, 140, 177, 228, 241, 242, 261, 277, 291 Baron, A. S., 263, 272, 277 Barr, A., 216, 227 Barr, C. L., 269, 275 Barrett, H. C., 72, 79, 83, 85, 86, 181, 192, 194 Barrett, L., 23, 26, 29, 30, 73, 76, 86, 307 Bartlett, M. Y., 223, 226 Barton, R. A., 25, 30, 264, 275 Bass, B. M., 230, 237, 241 Bassler, B. L., 96, 97, 105 Baumeister, R. F., 91, 103, 129, 132, 140, 193, 283, 285, 288–291 Beard, H. K., 294, 307 Beauchamp, G. K., 40, 48 Beck, A. T., 126, 140 Becker, D. V., 3, 17, 49, 51, 55, 56, 58, 59, 66, 68, 218, 223, 228, 230, 263 Becker, I., 95, 103 Bell, C., 305, 307 Ben-Jacob, E., 95, 103 Bentall, R., 23, 24, 27, 31


Berent, M. K., 266, 277 Berry, D. S., 209, 256, 259 Bettens, F., 40, 48, 191, 194 Bevc, I., 186, 187, 192 Biddle, J. E., 303, 305 Billari, F. C., 155, 159 Biran, A., 189, 192, 297, 306, 307 Bittles, A. H., 182, 183, 192 Bjorklund, F., 189, 193 Blanton, H., 218, 228 Blascovich, J., 109, 122 Bleske, A. L., 5, 16 Bless, H., 110, 111, 112, 118, 119, 122, 123, 260, 261 Boden, J., 129, 140 Bodmer, W. F., 183, 192 Boehm, C., 229, 236, 237, 238, 240, 241, 270, 276 Boesch, C., 41, 46 Bonanno, G. A., 76, 86 Boninger, C. S., 266, 277 Boone, R. T., 103 Bootzin, R. R., 141 Bornstein, R. F., 253, 259 Boutillier, R. G., 237, 242 Bower, G. H., 58, 67, 109, 110, 114, 122, 123 Bowlby, J., 99, 103, 129, 140 Boyd, R., 3, 16 Boyes, A. D., 50, 67, 196, 201, 202, 207, 208 Boyse, E. A., 40, 48 Brajkovich, G., 253, 261 Braun, W. J., 294, 306, 307 Braunstein, Y. M., 149, 159 Braverman, J., 223, 226 Brennan, P., 47 Brewer, M. B., 7, 16, 280, 283, 290 Bronstad, P. M., 42, 47 Brooks, R., 36, 37, 46, 47 Brothwell, D., 295, 305 Brown J. K. M., 295, 305 Brown, D., 229, 241 Brown, G. W., 130, 140 Brown, P., 241 Brown, R., 58, 66 Brown, S., 216, 227 Brownell, K., 301, 308

Brozinsky, C., 50, 68 Bryan, A., 216, 227 Bryne, R. E., 275, 276 Bryson, J. B., 216, 228 Buck, R., 3, 14, 89–93, 98–100, 102, 103, 108, 254, 283, 296 Buckling, A., 95, 104 Burnham, J. T., 183, 192 Burt, D. M., 38, 47, 176, 177, 260 Buss, A. H., 255, 259 Buss, D. M., 2, 4–6, 8, 16, 44, 46, 51, 67, 128, 129, 131, 140, 157, 158, 173–177, 185, 186, 193, 198, 207, 214, 215, 222, 223, 225, 226, 248, 260, 271, 276, 280, 282, 287, 290, 303, 306 Buston, P. M., 157, 158 Butner, J., 51, 67, 238, 242 Buunk, A. P., 3, 6, 12, 15, 61, 66, 100, 102, 104, 213–224, 226, 227, 231, 233 Byatt, G., 246, 261 Byrne, A., 141 Byrne, R. W., 25, 30, 130, 133, 140, 264, 276, 277 Cacioppo, J., 109, 123 Cairns, D. R., 285, 291 Calder, A. J., 194 Campbell, A., 214, 226 Campbell, B., 17, 177 Campbell, L., 162, 170, 199, 207 Caporael, L. R., 280, 290 Carnot, C. G., 266, 277 Caro, T. M., 40, 46 Carrasco, M., 64, 67 Carruthers, P., 159 Carter, A. R., 290 Carter, C. O., 183, 192 Cartwright, D., 231, 236, 237, 241 Case, T. I., 283, 290 Casillas, A., 209 Caspi, A., 198, 209 Castles, D., 177, 260 Catt, K. J., 99, 105 Catty, S. R., 252, 253, 260, 261 Cavalli-Sforza, L. L., 183, 192 Chadwick, P., 268, 276 Chagnon, N. A., 175, 186, 192


Champion, L. A., 127, 140 Chan, E., 263, 272, 277 Chan, N., xix Chandramani, S. P., 47 Chartrand, T. L., 221, 226, 281, 285, 288, 290 Chemers, M. M., 230, 231, 241 Chen, J., 266, 276 Cheng, Z., 281, 282, 290, 291 Cheung, C. K. T., 291, 284 Chiappe, D., 83, 86 Choe, J., 215, 226 Choi, W., 284, 291 Choleris, E., 294, 307 Christenfeld, N., 222, 227 Christensen, P. N., 167, 176, 177, 223, 227 Chuang, Y. C., 266, 277 Cialdini, R., 51, 67 Ciarocco, N. L., 288, 289 Ciarrochi, J. V., 108, 122 Clark, D. A., 126, 140 Clark, L. A., 134–136, 140 Clark, M. S., 103, 108, 110, 122, 159, 209, 259 Clark, R. D., 61, 67 Clarke, C. H., 97, 104 Clore, G. L., 109, 124 Clutton-Brock, T. H., 263, 276 Coale, A. J., 155, 158 Cohen, M. X., 50, 68 Colby, K. M., 268, 276 Cole, D. A., 133, 140 Collins, A., 109, 124 Colvin, C. R., 204, 209 Colwell, D. D., 294, 307 Conti, A. A., 295, 307 Conway, L. G., III, 5, 16 Cooper, J., 294, 307 Cooper, L., 118, 120, 123 Corp, N., 25, 30, 264, 276 Cosmides, L., 2, 4, 8, 16, 23, 31, 62, 63, 66–68, 71–73, 75–79, 82, 83, 85, 86, 88, 110, 122, 124, 126, 134, 140, 147, 158, 163, 177, 181, 184, 187, 191, 193, 194, 228, 230, 236, 238, 241, 242, 245, 248, 261, 263, 264, 277, 280, 290, 291

Cousins, A., 167, 170, 176, 223, 227 Couzin, I. D., 233, 235, 241 Cowan, N., 50, 67 Coyan, J. C., 131, 141 Cozzarelli, C., 205, 209 Craik, F. I., 50, 67 Crandall, C. S., 294, 300, 307, 308 Cronin, H., 163, 176 Crow, J. F., 183, 192, 193 Cunningham, M. R., 167, 176, 247, 259 Curran, M., 40, 48 Curtis, V., 189, 192, 297, 307 Daly, M., 7, 16, 180, 191, 192, 222, 228, 288, 292 Damasio, A. R., 71, 86, 109, 122, 123 Darwin, C., 2, 16, 33, 46, 73, 87, 93, 94, 104, 109, 123, 126, 128, 180, 192, 230, 231, 241 Davidson, J. R., 87, 193 Davies, N, B., 104, 277 Davies, R. M., 115, 213, 227 Dawkins, R., 43, 46, 94, 95, 97, 98, 104, 232, 234, 241, 265, 277 De Waal, F. M. B., 128, 140, 235, 236, 238, 241, 264, 270 De Wall, C. N., 276, 288, 289 Deabler, H. L., 221, 226 Deaner, R. O., 129, 140 DeBruine, L. M., 190–192 DeCremer, D., 237, 238, 240–242 Delton, A. W., 3, 49, 68, 218, 230, 263 Dennett, D. C., 239, 241, 267, 276 DePaulo, B. M., 114, 224 Deschner, T., 41 DeScioli, P., 232, 233, 241 DeSteno, D., 111, 124, 223, 226 DeVader, C. L., 237, 242 DeVore, I., 186, 193 Dijkstra, P., 3, 61, 66, 102, 213–217, 219–223, 226, 227, 231 Dimberg, U., 126, 141 Dobkin, D. S., 62, 67 Domb, L. G., 41, 46 Donovan, J. M., 217, 227 Dotan, I., 216, 228 Douvan, E., 198, 207 Dovidio, J. F., 267, 276


Downs, D. L., 139, 141 Dragoin, W. B., 62, 68 Druen, P. B., 167, 176, 247, 259 Dunbar, R. I. M., xvii, 3, 4, 6, 7, 14, 21, 23–31, 99, 104, 127, 230, 232, 235, 237–239, 241, 263, 275, 276, 280, 285, 288, 308 Duncan, L. A., 3, 4, 6, 7, 8, 11, 15, 16, 65, 67, 240, 254, 255, 281, 287, 293, 300, 304, 307 Duncan, N., 28, 30 Dunn, E. N., 267, 277 Dupré, J., 158 Eagly A. H., 303, 307 Eals, M., 66, 68 East, R., xix, 112, 114, 123 Eberhard, A., 96, 104 Eberhard, C., 104 Ebert, D., 183, 192 Ehrlich, P. R., 62, 67 Eibl-Eibesfeldt, I., 97, 104, 294, 307 Eisenberger, N. I., 4, 16, 65, 67, 91, 104, 282, 290 Ekman, P., 73–75, 87, 113, 123, 140, 189, 193 Ellet, L., 276 Ellis, B. J., 122, 124, 152, 155, 158, 159, 197, 207, 245, 260 Ellison, P. T., 38, 46 Ellsworth, P. C., 3, 14, 71, 72, 76–78, 80, 87, 214, 224, 296 Emlen, S. T., 36, 46, 157, 158 Engebrecht, J., 96, 104 Engelhardt, A., 42, 46 Enquist, M., 255, 260 Epstude, K., 218, 227 Eriksson, C. J. P., 47 Evans, R. B., 220, 227 Ewald, P. W., 298, 307 Fallon, A. E., 72, 87, 189, 193, 296, 308 Fanselow, C., 158, 159 Fasolo, B., 155, 157, 159 Faulkner, J., 7, 16, 17, 65, 67, 294, 300, 304, 307, 308 Fazendeiro, T., 261 Fazio, R. H., 266, 267, 276, 277

Feeney, J. A., 176 Fehr, B., 198, 208 Fell, C., xix Fellous, J., 255, 262, 302, 308 Fenigstein, A., 268, 269, 276 Ferguson, T. S., 149, 150, 158 Ferrer, R., 103 Ferstl, R., 47 Feshbach, S., 109, 123 Fessler, D. M. T., 170, 176, 186, 187, 190, 193, 238, 242, 304, 308 Festinger, L., 276 Fetchenhauer, D., 61, 66 Fiedler, F. E., 236, 241 Fiedler, K., 110, 111, 112, 118, 119, 122, 123 Finch, J., 167, 176 Fincham, F. D., 205, 208 Fink, B., 35, 46, 167, 176, 302, 307 Finkelhor, D., 186, 187, 194 Fishbein, M., 265, 266, 275, 276 Fiske, S. T., 251, 228, 259, 290 Fitness, J., 282, 291 Fletcher, G. J. O., 3, 15, 50, 67, 152, 157, 159, 195–202, 203, 205, 207–209, 215 Fodor, J. A., 72, 81, 85, 181, 193 Folstad, I., 165, 176 Forgas, J. P., xiii, xx, 1, 3, 14, 16, 50, 58, 67, 104–112, 114–124, 133, 134, 180, 254, 260, 261, 276, 290, 291, 296, 307 Frank, R. H., 39, 46 Franklin, M., 167, 176, 177 Franks, N. R., 233, 241 Franzoi, S. L., 220, 227 Freud, S., 108, 240, 241 Friesen, M. D., 200, 205, 208 Frijda, N. H., 76, 77, 87, 109, 123, 130, 140 Fromm, E., 240, 241 Funder, D. C., 204, 208 Fuqua, C., 96, 104 Gallio, M., 96, 104 Gallucci, M., 281, 291 Galperin, A., 46, 49 Gangestad, S. W., 3, 8, 9, 14, 33, 35, 41–43, 53, 61, 93, 95, 148, 161–163, 165–167, 169–174, 181, 185, 199,


200, 203, 214, 219, 223, 224, 231, 232, 247, 248, 297, 302, 303, 307, 308 Gapinski, K. D., 301, 308 Garcia, S., 263, 272, 277 Garcia-Marquez, T., 253, 260 Gardner, R., 128, 142 Gardner, W. L., 288, 291 Garver, C. E., 43, 46, 61, 67, 167, 176, 219, 227 Garver-Apgar, C. E., 162, 167, 170, 176, 223, 227 Gaulin, S. J. C., 39, 47, 217, 219, 226, 227 George, J. M., 108, 111, 120, 121, 123 George, N., 255, 261 Getty, T., 36, 46 Ghirlanda, S., 255, 260 Ghiselin, M. T., 2, 16 Gibbons, F. X., 100, 104, 215, 226 Gibson, K. R., 263, 276 Gigerenzer, G., 147, 159 Gilbert, D. T., 114, 115, 123, 228, 290 Gilbert, P., 125, 127–130, 140–142 Gilden, D. L., 265, 276 Giles, T., 200, 208 Gillath, O., 99, 105 Gilovich, T., 269, 276 Gil-White, F. E., 231, 241 Ginsburg, B., 98, 102, 103 Girard, K., 253, 261 Gladue, B. A., 217, 226 Gleicher, F., 133, 141 Goldberg, D., 137, 141 Goldsmith, R., 252, 260 Goldstein, D. G., 147, 159 Gonsalkorale, K., 65, 67, 286, 290 Gonzaga, G., 39, 40, 46 Goodall, J., 279, 290, 294, 307 Gordon, I. J., 294, 307 Gordon, P. C., 253, 260 Gottleib, G., 93, 104 Grafen, A., 36, 46, 165, 176 Grammer, K., 35, 38, 46, 48, 167, 176, 177 Gray, H. M., 111, 122 Graziano, W. G., 167, 176 Greenberg, E. P., 96, 104 Greenberg, J., 283

Greenberg, S. L., 290 Greenwald, A. G., 267, 276 Grice, J. W., 223, 227 Griffin, A. S., 95, 104 Gross, M. R., 164, 176 Groth, G. E., 155, 158, 159 Guadagno, R. E., 223, 228 Guerin, S., 51, 66 Guerra, F., 295, 307 Guimond, S., 104 Gürerk, O., 21, 31 Gutshall, K. A., 256, 260 Guttman, N, 255, 260 Hagel, R., 167, 176 Hagen, E. H., 9, 16 Haidt, J., 189, 193, 304, 308 Haig, D., 188, 193 Haig, J., 209 Halberstadt, J., 3, 4, 13, 14, 15, 109, 121, 124, 245, 249, 250, 252, 253, 260, 261, 302 Hallam, M., 208 Hamermesh, D. S., 303, 305 Hamilton, T. E., 152, 159 Hamilton, W. D., 165, 176, 183, 192, 193 Hammond, J. R., 198, 208 Hardy, C. L., 237, 241 Harrer, H., 64, 67 Harris, C. R., 222, 223, 227 Harris, T. O., 130, 140 Harris, V. A., 115, 123 Harris-Warrick, R. M., 98, 105 Hart, C. M., 238, 242 Hartl, E. M., 221, 226 Harvey P. H., 263, 276 Harvey, T., 301, 306, 307 Haselton, M. G., xx, 1, 5, 6, 8, 16, 39, 40, 43, 44, 46, 47, 51, 67, 110, 167, 171, 176, 180, 185, 186, 193, 271, 276, 282, 287, 290, 297, 303, 306 Haslam, A., 237, 241, 290 Haslam, N., 130, 138, 141 Hassebrauck, M., 198, 208 Hastorf, A. H., 123 Hatfield, E., 67 Hauser, M. D., 94, 104 Haviland-Jones, J., 86, 87, 194


Hearst, E., 255, 261 Heath, C., 305, 307 Heatherton, T. F., 300, 307 Hebl, M. R., 290, 307 Heider, F., 114, 123, 264, 276 Heinsohn, R., 236, 251 Heistermann, M., 41, 42, 46 Hellingwerf, K. J., 95, 104 Hemphill, J. K., 236, 237, 241 Hendrick, C., 226 Hendrick, S. S., 226 Henrick, J., 231, 241 Henzi, S. P., 23, 30 Hepper, P. G., 185, 193 Hepworth, C., 130, 140 Herrmann, R. K., 7, 16 Herzog, M. E., 220, 226 Hesse, F. W., 137, 142 Heuvelink, A., 148, 159 Hickford, C., 247, 261 Higgins, E. T., 124, 275 Hilgard, E. R., 107, 108, 123 Hill, E. M., 217, 220, 227 Hill, K., 227, 246, 260 Hill, R. A., 26, 27, 31 Hinde, R. A., 141 Hirschfeld, L. A., 63, 67 Hodges, K., 41, 46 Hofer, B., 16, 68 Hogan, R., 230, 238, 241, 242 Hogg, M. A., 231, 237, 241 Holland, B., 42, 47 Hollander, E. P., 230, 242 Hollander, S., 246, 260 Holyoak, K. J., 253, 260 Hömberg, V., 194 Hoogland, J. L., 166, 175 Houle, D., 35, 47 Houston, A. I., 36, 47 Howard, A., 267, 276 Howard, R. D., 166, 175 Hrdy, S. B., 214, 227 Hull, J. G., 300, 307 Humphrey, N. K., 127, 140, 264, 265, 270, 276 Hupka, R. B., 213, 227 Hurtado, A. M., 220, 227 Hutchinson, G. E., 84, 87

Hutchinson, J., 158, 159 Hutsler, J., 263, 272 Huxley, P., 137, 141 Hyde, J. S., 165, 177 Imada, S., 304, 308 Ingram, R. E., 129, 130, 141 Irlenbusch, B., 21, 31 Irons, W., 175 Isen, A. M., 108, 110, 122, 123 Ito, T., 109, 123 Iwasa, Y., 165, 177 Izard, C. E., 73, 87, 189, 193 Jacoby, L. L., 253, 261 James, W., 4, 16, 86, 87 Jamieson, D. W., 266, 277 Jarvis, B., 284, 291 Jasienska, G., 46 Jeffrey, L., 247, 253, 260 Jennions, M. D., 37, 46 Jensen-Campbell, L. A., 167, 176 Jepson, S. F., 238, 242 Jerison, H. J., 25, 31 Jeyaram, S., 301, 308 Joffe, T. H., 31 John, O. P., 276 Johnson, B., 267, 276 Johnson, C., 267, 276 Johnson, K., 165, 177, 248 Johnston, V. S., 167, 173, 176 Joiner, T. E., 131, 141 Joireman, J., 137, 142 Jones, B. C., 38, 47, 167, 176 Jones, D., 185, 193, 246, 260 Jones, E. E., 115, 123, 266, 276 Judd, C. M., 226, 267, 277 Jurgens, U., 93, 104 Kabat, L. G., 277 Kaiser, D., 97, 105 Kaiser, R., 230, 238, 241, 242 Kalakanis, L., 208, 246, 261 Kalick, S. M., 152, 159, 248, 260 Kalish, H. I., 255, 260 Kanwisher, N., 302, 306 Kappeler, P., 30, 41, 48 Karnat, A., 194


Karter, A. J., 165, 176 Kavaliers, M., 294, 296, 306 Kavanagh, P., 201, 202, 208 Kawakami, K., 267, 276 Kayabashi, T., 177, 260 Kayra-Stuart, F., 246, 260 Keating, C. F., 256, 260 Keefe, R. C., 62, 67, 216, 221, 227 Keele, S. W., 252, 261 Keller, M. C., xi, xv, 263, 272, 277 Kelley, H. H., 152, 158, 197, 207 Kelly, J. R., iii, 281, 291, 294, 306 Keltner, D., iii, 74, 76, 86, 87, 110, 124 Kendall, P. C., 140 Kendrick, T., 256, 260 Kenny, D. A., 198, 204 Kenrick, D. T., iii, viii, xv, 2, 3, 4, 6, 7, 8, 11, 12, 14, 16, 17, 49, 51, 61, 62, 66, 67, 68, 155, 159, 164, 176, 192, 203, 208, 216, 217, 218, 221, 225, 226, 227, 230, 235, 238, 240, 242, 263, 280, 290, 299, 307 Kenyon, C. L., 104 Kerr, N. L., ii, iii, 27, 31, 281, 291 Ketelaar, T., iii, 122, 124, 245, 260 Khera, A. V., 129, 140 Kiesecker, J. M., 294, 296, 306 Kimura, M., 183, 192 Kinderman, P., 23, 24, 31 Kirby, L. D., ii, iii, 109, 124 Kirkpatrick, L. A., 155, 159, 238, 242 Kirkpatrick, M., 37, 46, 163, 176 Kleck, R. E., 269, 275, 300, 306, 307 Klein, S. L., 295, 306 Kleinginna, A. M., 90, 104 Kleinginna, P. R., 90, 104 Klinger, E., 127, 141 Klohnen, E. C., 197, 198, 208, 209 Klotz, M. L., 220, 226 Kokko, H., 36, 37, 44, 46, 47 Koskela, E., 47 Kraft, P., 137, 142 Kral, P. A., 62, 68 Kramer, R. M., 268, 276 Krauel, K., 47 Kraus, S. J., 265, 266, 277 Krause, J., 233, 235, 241, 242 Kravitz, E. A., 99, 105

Krebs, D. L., 67 Krebs, J. R., 43, 46, 94, 95, 104, 265, 277 Kreft, J. U., 95, 97, 105 Krosnick, J. A., 266, 277 Kudo, H., 25, 31, 264, 277 Kuhlman, D. M., 137, 142 Kulik, J., 58, 66 Kumar, S., 183, 193 Kummer, H., 235, 242 Kunda, Z., 251, 260 Kurzban, R., x, xv, 3, 6, 7, 15, 17, 65, 72, 79, 82, 83, 85, 86, 155, 159, 180, 181, 192, 193, 216, 229, 232, 233, 234, 236, 238, 240, 241, 242, 281, 294, 298, 306 Kuukasjarvi, S., 42, 47 Kylsten, P., 96, 104 Lacey, J. J., 76, 87 Lackie, J. M., 97, 104, 105 Laham, S. M., iii, 65, 67, 116, 123, 291 Laird, J. D., 223, 228 Laird, K., 40, 46 Lakin, J. L., iii, 281, 285, 288, 290 Lamprecht, J., 235, 242 Lane, J. D., 114, 124 Lange, H., 194 Langlois, J. H., 204, 208, 246, 250, 260, 261, 302, 306 LaPaglia, J., ix, 3, 5, 15, 43, 161, 199, 214, 219, 231, 232, 234, 248, 280, 299 LaPiere, R. T., 265, 277 Larsen, R. J., 222, 226 Law, A. T., 286, 290 Lazarus, R., 71, 79, 87 Leahy, R. L., 127, 141 Leary, M. R., ii, iii, 7, 17, 91, 103, 12, 139, 140, 141, 236, 242, 281, 283, 289, 290, 291, 294, 298, 306 LeDoux, J. E., 71, 87 Lee, K. J., 260, 261 Lee, P. C., 276 Lee, R. B., 186, 193 Leinders-Zufall, T., 40, 47 Lenton, A. P., 155, 157, 159 Lerner, J. S., 110, 124 Letzring, T. D., 204, 208 Levenson, R. W., 73, 87


Levick, W. R., 63, 67 Levin, S. A., 233, 241 Levine, H., 95, 103 Levinson, S., 16 Lewinsohn, P. M., 130, 138, 141 Lewis, K., 25, 31 Lewis, M., 86, 87, 193, 194 Lewis, R. A., 226 Li, N. P., 51, 61, 67, 68, 164, 176, 203, 204, 208, 242 Libert, J., 138, 141 Lieberman, D., ii, ix, xv, 3, 4, 6, 16, 63, 65, 67, 68, 72, 91, 104, 164, 175, 176, 179, 180, 184, 186, 187, 190, 191, 193, 254, 263, 282, 290, 296 Light, L. L., 246, 260 Ligon, J. D., 165, 177 Lind, E. A., 237, 242 Lindsey, S., 267, 277 Lindzey, G., 228, 242, 290 Linsenmeier, J. A., 61, 68, 164, 176, 203, 208 Lippi, D., 295, 306 Lipson, S. F., 38, 46 Lipton, J. E., 214, 223, 224, 226 Little, A. C., 38, 47 Littlepage, G. E., 237, 242 Livingstone, M. S., 98, 105 Locke, V., 287, 291 Loewenstein, G., 193 Longo, L. C., 303, 306 Lopes, B., 268, 276 Lord, C. G., 268, 277 Lord, R. G., 237, 242 Losick, R., 97, 105 Loumaye, E., 99, 105 Low, B. S., 40, 47, Lowen, C. B., 25, 31 Luo, S., 197, 198, 208 Lycett, J. E., 26, 30 MacBeth, H., 307 MacDonald, G., iii, 83, 281, 282, 290, 291 Mack, A., 64, 68 Mackie, D. M., ii, 253, 260 MacLoed, A. K., 133, 141 MacQueen, J., 157, 159

Magaro, P. A., 269, 277 Makhijani, M. G., 303, 306 Makin, J. W., 185, 193 Malone, P. S., 114, 115, 123 Maner, J. K., 6, 17, 51, 53, 55, 58, 66, 68 Manusov, V., 103 Mappes, T., 47 Marcus, D. K., 204, 208 Mark, M. M., 111, 124 Marker, L., 183, 193 Marlowe, F., 117, 220, 228 Martin, L. L., 122, 124 Martin, M., 136, 141 Martindale, C., 249, 260 Martinez, R., 301, 306 Massar, K., x, xv, 3, 102, 213, 218, 219, 221, 224, 227, 231 Mathes, E. W., 213, 227 Mathews, A., 133, 141 Matochik, J. A., 185, 193 Matsumoto, D., 74, 87 Matter, R. M., 303, 306 Matthews, D., 163, 175, 280, 289 Maul-Pavicic, A., 47 May, R. M., 183, 193 Mayer, J., 108, 122 Maynard Smith, J., 163, 177, 231, 232, 233, 242 McArthur, L. A., 259 McCauley, C., 304, 307 McCollough, J. K., 48, 177 McDonald, K. B., 83, 86 McDonel, E. C., 266, 276 McGuire, M. T., ii, 125, 126, 141 McNamara, J. M., 36, 41 Medvec, V. H., 269, 276 Melara, R. J., 249, 261 Mendes, W. B., 109, 122 Mentzer, S. J., 300, 307 Metalsky, G. I., 131, 141 Mignault, A., 255, 262, 303, 307 Mikulincer, M., ii, 99, 105, 205 Milgram, S., 4, 17 Miller, E. M., 41, 47. Miller, G. F., 150, 152, 159, 167, 171, 176, 225, 227, 265, 271, 277 Miller, R. G. Jr., 157, 159 Miller, R. R., 233, 243


Miller, R. S., 48, 177, 204, 208 Millman, Z., 296, 307 Millon, T. H., 269, 277 Miranda, J., 129, 141 Mitchell, P., 23, 31 Moffitt, T. E., 209 Møller, A. P., 248, 261 Monroe, S. M., 130, 131, 141 Montepare, J., 298, 307 Moore, K., 249, 260 Moriarty, D., 294, 306 Morrison, B., 241 Morton, N. E., 183, 193 Moylan, S. J., ii, 114, 115, 123 Muller, H. J., 183, 193 Müller-Ruchholtz, W., 47 Murphy, T., 301, 306 Murray, E. N., 255, 259 Murray, L. K., 260 Musselman, L., 246, 260 Mussen, P., 105 Mussweiler, T., ii, 218, 227 Nadler, A., 216, 228 Nakamura, M., 100, 103 Navarrete, C. D., 186, 190, 193, 304, 307 Nealson, K. H., 96, 104 Near, D., 233, 243 Nee, S., 165, 177 Neel, J. V., 182, 183, 192 Nemeroff, C., 296, 307 Nesse, R., 71, 76, 77, 78, 79, 84, 85, 86, 87, 126, 127, 141, 271, 277, 297, 307 Nettle, D., 8, 16, 28, 30, 31, 137, 141, 282, 287, 290, 297, 306 Neuberg, S. L., iii, viii, xvi, 3, 6, 17, 49, 51, 66, 67, 68, 218, 230, 263 Newton, N., 282, 291 Nicastle, L. D., 61, 67 Niedenthal, P., ii, 109, 124 Nisbett, R., 146, 159 Nissinen, K., 47 Nitzberg, R. A., 99, 105 Noller, P., 176 Noonan, K. M., 40, 47, 166, 175 Nosofsky, R. M., 255, 260 Nowicki, S., 34, 35, 40, 44, 47 Nunn, C. L., 41, 48

Nuss, C. K., 288, 289 O’Brien, S. J., 183, 193 O’Neal, E., 255, 261 O’Sullivan, M., 74, 87 Oates, K., 185, 193 Oatley, K., 126, 141 Öhman, A., 126, 141 Oliver, M. B., 165, 177 Olson, M. A., 267, 276 Oppenheimer, N. J., 104 Ortony, A., 74, 87, 109, 124 Öst, L. G., 126, 141 Otto, J., 213, 227 Oubaid, V., 222, 226 Oum, R. E., 180, 190, 193 Overall, N. C., 195, 200, 201, 202, 205, 206, 208 Packer, C., 236, 241 Paepke, A., 40, 48, 191, 194 Page, R., 266, 277 Pagel, M., 41, 46, 47 Pai, R. A., 183, 193 Palermo, R., 253, 261 Palmer, G. J., 239, 242 Panksepp, J., 73, 87, 91, 105 Park, B., 267, 277 Park, J. H., 7, 16, 17, 65, 67, 68, 191, 193, 294, 299, 300, 301, 302, 304, 306, 307 Parker, G. A., 177 Parrot, W. G., 111, 124, 213, 214, 223, 228 Patterson, M., 103 Pause, B. M., 40, 47 Pawlowski, B. P., 31 Penke, L., 157, 159 Penn, D. J., 40, 47 Penton-Voak, I. S., 38, 47, 167, 171, 173, 176, 177, 247, 260, 302, 306 Perrett, D. I., 38, 47, 167, 176, 177, 247, 260 Pervin, L. A., 109, 124 Peters, M., 38, 47, 261 Peterson, R. S., 242 Petrullo, L., 277 Petty, R. E., ii, 110, 111, 118, 124


Pfiefer, J. B., 42, 46 Phillips, K., 219, 227 Piaget, J., 102, 105 Pickett, C. L., ii, iii, 288, 291 Pietromonaco, P. R., 133, 138, 141 Pietrzak, R. H., 223, 228 Pike, A. W., 295, 306 Pike, C. L., 247, 259 Pillsworth, E. G., 43, 47 Pinker, S., 63, 68, 76, 79, 86, 87 Platt, M. L., 129, 140 Plutchik, R., 73, 74, 75, 76, 86, 87 Polonsky, M., 100, 103 Pomiankowski, A., 165, 177 Poore, J. C., 39, 46 Porter, L. W., 140 Porter, R. H., 185, 193 Posner, M. I., 252, 260, 261 Potts, W. K., 40, 47 Power, M. J., 127, 140 Powers, S. R., 91, 103 Preisser, E., 294, 306 Price, D. D., 282, 291 Price, J. S., 126, 128, 138, 141, 142 Puts, D. A., 39, 47, 167, 177 Pyszczynski, T., iii, 283, 290 Queller, D. C., 97, 105 Rabie, T., 297 Raleigh, M. M., 126 Randall, D. W., 256 Ranganath, C., 50 Rapoport, A., 233 Rather, P., 96 Rawins, 301 Reber, 252, 253 Reddington, K., 237 Redfield, R. J., 96 Regan, D. T., 266 Reiber, C., 219 Reidl, L., 213 Reierson, G. W., 38 Repp, B. H., 249, 261 Rescorla, R. A., 255 Rhodes, G., 200, 245, 247, 249, 250, 250, 253, 254, 303 Rice, W. R., 42

Richardson, R., 286 Richerson, P. J., 3 Ridgeway, V., 133 Rikowski, A., 167 Robertson, T. E., 218 Robins, R. W., 198 Robinson, W., 237 Rockenbach, B., 21 Roelke, M. E., 183 Roese, 266 Roggman, L. A., 245, 250, 302 Rohde, P., 128, 130 Rohrmann, B., 135, 136 Rook, K. S., 133, 134, 138 Roseman, I., 77 Ross, M., 146, 269 Roth, D., 22 Rowe, L., 35 Rozin, P., 72, 189, 296, 304 Rubenstein, A. J., 245, 248 Rucker, D., 111 Ruter, K., 218 Ruxton, G. D., 235 Ryan, M. J., 37 Sadalla, E. K., 155 Sadowski, M., 118 Saenz, D. S., 268 Sagarin, B. J., 223 Salmon, C., 7, 169, 180 Salmond, G. P., 96 Salovey, P., 223 Sanchez-Burks, J., 272 Sandison, A. T., 295 Savitsky, K., 269 Sawaguchi, T., 264 Schachter, S., 62, 92 Schaller, M., 5, 6, 7, 8, 11, 15, 65, 189, 191, 240, 254, 255, 281, 287, 294, 295, 299, 300, 301, 304, 305 Schelling, T., 231, 236, 237 Scherer, K. R., 72, 77, 80, 85 Scheyd, G., 200 Schiefenhovel, W., 304 Schlosberg, H., 75 Schmitt, D. P., 174 Schooler, J., 267 Schotter, A., 149


Schultz, S., 25 Schwarz, N., 110, 252 Searcy, W. A., 34, 35, 40, 44 Seebeck, T., 40, 191 Seeley, J. R., 130 Seely, E., 223 Segal, Z. V., 129 Sellen, D. W., 40 Semmelroth, J., 222 Shackelford, T., 5, 215 Shakespeare, W., 28 Shapira, Y., 95 Shaver, P., 99 Shaw, S., 282 Shepard, R. N., 255 Shepher, J., 187 Sher, T., 287 Sherman, P. W., 266 Sherry, D. F., 62 Shettel-Neuber, J., 216 Shiffrin, R.M., 50 Shimkets, L. J., 97 Siegel, S., 255 Sigall, H., 266 Silverman, I., 66, 96, 186, 187 Simao, J., 154 Simmons, L., 38 Simon, H. A., 147, 149 Simpson, J. A., 43, 50, 53, 161, 162, 165, 167, 170, 173, 174, 196, 199, 200, 203, 214, 219, 232, 248, 280, 299 Sinclair, R. C., 111 Singer, J., 92, 109 Singh, D., 38, 42, 220 Slatter, P. E., 249 Sloman, L., 128 Smart, L., 129 Smith, A., 284 Smith, C. A., 77, 78, 109 Smith, D. J., 249 Smith, E. R., 253 Smith, J., 163 Smith, R. H., 220, 223 Smith, R. L., 41 Smurda, J., 39 Snyder, M., 300 Sober, E., 233, 234, 240 Solomon, R. C., 73, 283

Sommer, K., 284, 285, 288 Sorentino, R. M., 237 Sornette, D., 26 Sperber, D., 83 Spies, K., 137 Spiro, M. E., 187 Spoor, J. R., 15, 27, 91, 109, 132, 281 Sprecher, S., 197 Sprengelmeyer, R., 191 Stahl, G., 137 Stapel, D. A., 218 Sternberg, E., 305 Stevens, D. A., 223 Stewart, G. S., 96 Stiller, J., 23, 27, 28 Stogdill, R. M., 230 Stone, G. O., 61 Streeter, S. A., 38 Strenta, A., 300 Sturgill, G., 96 Sumer, N., 205 Sundie, J. M., 61 Swaddle, J. P., 38 Swaminathan, M. S., 183 Swarbrick, R., 27 Swift, S., 96 Symons, D., 216 Tagiuri, R., 264 Talmon, G. Y., 187 Tambor, E. S., 139 Tarabrina, N. V., 213 Terdal, S. K., 139 Thomas, G., 200, 204 Thompson, N. S., 223 Thornhil, R., 8, 9, 14, 35, 38, 41, 42, 43, 61, 95, 148, 162, 163, 165, 167, 181, 185, 214, 219, 231, 232, 247, 248, 297, 302 Throup, J. P., 96 Tither, J. M., 200 Todd, P. M, 4, 14, 147, 148, 150, 152, 154, 155, 158, 299 Tomkins, S. S., 72, 73 Tong, E. M. W., 76, 80 Tooby, J., 2, 4, 8, 62, 63, 66, 71, 72, 73, 75, 76, 77, 78, 79, 82, 83, 85, 110, 126, 134, 147, 163, 181, 182, 183, 184,


187, 188, 191, 230, 236, 245, 248, 263, 280, 287 Trafimow, D., 266 Trapnell, P. D., 130, 138 Treasure, J. L., 301 Tremewan, T., 245, 250 Trent, D., 128 Trivers, R. L., 5, 8, 163, 164, 166, 167, 185, 195 Troisi, A., 125, 126 Troop, N. A., 301 Trost, M. R., 155 Tulving, E., 50 Turner, T. J., 74 Twenge, J. M., 288 Tyler, T. R., 237 van Schaik, C. P., 41 Vanable, P. A., 268, 269 VanLear, C. A., 92 van Vugt, M., 6, 15, 65, 216, 230, 231, 233, 237, 238, 239, 240, 281 Vargas, P., 116 Vaughn Becker, D., 218 Verdolini, K., 39 Veroff, J., 198 Vieira, E. T., Jr., 100 von Hippel, W., 65 Vorauer, J. D., 269 Vujic, T., 287 Wakefield, J., 5 Warburton, W., 285 Wason, P., 133 Waters, C. M., 96, 97 Watson, D., 134, 135, 136, 197 Weary, G., 133 Webster, G. D., 3 Wedekind, C., 40, 191 Weeden, J., 155 Wegener, D. T., 110 Wegner, D. M., 217 Wei, J., 268 Weiner, D., 198 Wells, S. A., 204

Welton, K. E., 252 West-Eberhard, M. J., 232 Westen, D., 222 Westermann, R., 137 Westermarck, E. A., 186 Wheye, D., 62 Whitfield, T. W., 249 Whittlesea, B. W., 253 Wiggins, J. S., 130, 138 Wilcoxon, H. C., 62 Williams, K. D., 4, 15, 34, 65, 79, 91, 96, 132, 163, 186, 187, 279, 281, 282, 283, 284, 285, 286, 287, 288 Williamson, D., 133 Willis, C. A., 221 Wilson, D. S., 233, 234, 235, 237, 240 Wilson, E. O., 92, 180 Wilson, M., 7, 97, 165, 185, 222, 233, 236 Wilson, T. D., 267, 288 Winans, S. C., 96 Winkielman, P., 252 Wittenbrink, B., 267 Wolf, A. P., 186, 187 Wu, C., 247 Yamazaki, K., 40 Ybarra, O., 8, 9, 23, 235, 239, 266, 272 Yukl, G. A., 230 Zadro, L., 281, 283, 286 Zahavi, A., 35, 36, 165 Zajonc, R. B., 78, 109, 253 Zander, A., 236, 237 Zebrowitz, L. A., 247, 248, 255, 256, 298, 303 Zentner, M. R., 197 Zhou, W.-X., 26 Zhu, W.-X., 97 Zimbardo, P. G., 268 Zinner, D. P., 41 Ziomkiewicz, A., 38 Zuckerman, M., 137 Zuk, M., 165

Subject Index Accuracy and partner serving bias, 197 Accuracy of judgment of mate preferences, 204 Adaptations and special design, 1, 6–7, 63, 163, 247–248, 250–251, 258, 280, 287, 289 Adaptive significance of moods, 107–122 Affect as a domain-specific adaptation, 110 Affect as a feedback signal, 109 Affect infusion model, 120 Affect, 109, 120, 251–254, 259, 266 Agency, 130, 131, 135, 138 Aggression, 6, 129, 240, 285, 288, 305 Altruism, 89, 95–97 Anger, 58, 76, 79–84, 86, 113 Anxiety, 99, 126, 129 Appraisal theories, 14, 66–88 Assimilative versus accommodative processing, 112–116 Attachment theory of depression, 129–130 Attachment, 91, 91, 99–100, 129–130, 202 Attention, 5, 11, 49, 50, 51, 58 Attitude–behavior relationship, 265–267 Attitudes, xv, 1, 58, 99, 115, 120, 165 Attraction, 6, 13, 40, 51, 78, 167, 247; see also Mate selection Attributions, 7, 80, 99, 114–116, 301 Ball-toss paradigm, 284–285 Basic emotions, 73–74, 82–83, 107, 113 Behavioral immune system, 293–305 Belonging, need for, 283 320

Betrayal, 223–224 Big Five personality traits, 202, 204–205 Biological fitness and ostracism, 281, 282, 287, 288 Biology, 8–9, 12, 179 Bipolar disorder, 27 Body build, 213, 219–221 Bogus pipeline, 266–267 Bounded rationality, 147 Bower’s associate network model, 110 Brain imaging, 267, 282 Categorization, 180, 185, 186, 251–252 Changes in mate preferences across the ovulatory cycle, 167–175 Cheating, 280; see also Deception Coercion, 238 Cognitive adaptations and leadership, 230, 234, 239 Cognitive tuning, 110 Communication skills and leadership, 237, 239 Communication, 3, 34, 43, 89, 92–98, 113, 120 Communion, 130, 138 Competition, 270–273 Computational theory of mind, 180–181 Computer simulations of mate search, 147, 152–153, 158 Conditional strategies models, 232 Conservation of resources and depression, 126, 130, 138 Conspecifics, 127, 129, 235, 294 Control, need for, 283 Cooperation and competition, 3, 4, 8, 21,


38, 128, 133, 138, 151, 153, 156, 166, 214, 270–273, 280, 282, 283 Coordinating leadership, 231, 234, 236, 237, 238 Coordination problems, 230–240 Cross-cultural studies, 2, 174, 187, 303 Culture, 3, 12, 30, 149 Cyberball, 283, 284, 285, 286, 287 Danger, 254–255 Deception detection and mood, 113–114 Deception, 25, 34, 43–45, 113, 264, 272 Decision mechanisms, 8, 146–148, 154–155, 158 De-escalating strategies and theories of depression, 128 Deleterious recessive mutations, 182–183 Depression and neuroticism, 137 Depression and positive affect, 134–135 Depression, 111, 114, 125–139 Developmental psychology, 22, 23 Disability, 7 Disgust and inbreeding avoidance, 181, 187–191 Disgust, 296, 297, 299, 300, 301, 304, 305 Domain-specific mechanisms, 14, 66, 72, 74, 85, 86, 248, 258, 280 Dominance and leadership, 231, 233, 234, 236, 238 Dominance, 15, 89, 94, 98, 99, 101–102, 214–218, 220–225 Dowry problem and mate search, 149 Dysphoria, 108, 111, 116 Ecological rationality, 147 Ecologically contingent mate search, 164 Ecologically valid cues to kinship, 185 Emotional appraisals, 81, 109 Emotional communication, 2, 22, 33, 45, 94 Emotional fidelity, 222, 223, 224 Emotions, xv, 2, 3, 14, 71–88, 90–94 Empathy and leadership, 237, 239 Environmental conditions and mate preferences, 174 Equilibrium outcomes, 232–234

Error management theory, 51, 282, 283, 297 Ethnocentrism, 303–304 Ethology, 94 Evolution and social cognition, 1–20 Evolutionary benefits of group inclusion, 280–281 Evolutionary functional analysis, 163–164 Evolutionary game theory and leadership, 229–240 Evolutionary theories of depression, 126–139 Eyewitness memory and mood, 116 Facial attractiveness, 303 Facial masculinity, 38, 39 Familiarity, 13, 27, 112, 253, 254, 257–258 Fear, 58, 72, 76, 83 Feminine features, 247 Fertility, 34, 37, 41, 42, 43, 167–169, 172 Flashbulb memory, 58, 60 Followership, 231–236, 238–240; see also Leadership Free-riders, 22, 270 Frequency dependent selection, 233 Frontal lobe, 24, 26 Fundamental attribution error, 114, 116 Game theory and leadership, 229–240 Gender differences, 2, 12, 27, 173, 199, 203, 223–225 Genetic Fitness, 165, 295, 296, 297, 299, 303 Goals, 11, 51, 57, 64, 78, 79, 84, 127, 231, 234, 236, 237, 238, 240 Good genes sexual selection, 162–175 Good provider model of sexual selection, 162–163, 166, 173 Group inclusion, 280–289 Group living, costs and benefits of, 270–271 Handicapping and honest signaling, 165–166 Heuristics, 6, 57, 66, 85, 110–112, 146–149, 155, 158, 187, 198, 254–255, 300–303



Hierarchies, 238 Honest signaling, 35–36, 39–40, 44, 45, 165–166, 173 Human nature, 12, 107, 108 Ideal Standards Model of relationships, 199, 203 Immune system, 295–305 Implicit attitudes, 266–267 Inbreeding avoidance, 4, 179–192 Incest avoidance and pathogens, 183–184 Inferences, 14, 33–48 Information processing, 6, 62, 181–184 Initiative and leadership, 233, 235, 237, 239 Intelligence, 237, 239 Intentionality, 23, 24, 28, 29, 30 Intergroup behavior, 15, 58, 240, 297 Interpersonal benefits of moods, 118–121 Interpersonal communication, 3, 102, 111, 113, 305 Intersexual attraction cues, 162–163 Intimacy, 27, 93, 224 Intrasexual competition, 162, 163, 214, 216, 217, 219, 224 Introspective awareness of mood effects, 115, 117 Jealousy, 6, 12, 15, 213–215, 219, 220, 221, 222, 225 Kinship, 4, 6, 34, 65, 95, 180–191 Language, 90, 91, 93, 100, 180, 185, 239 Leadership and social cognition, 239–241 Leadership in non-human species, 235–236 Leadership, 3, 6, 229–241 Leadership, definition of, 230 Learned helplessness and depression, 127 Life history and sexual rivalry, 220–221 Life history theory, 220 –, 221 Long-term versus short-term mate preferences, 161–175 Machiavellianism, 237 Manipulation, 265–266 Masculine features, 247

Mate preferences, 33, 53, 145–158, 161–175, 216–217, 225 Mate preferences and ovulatory cycle, 161–162, 168–175 Mate selection, xv, 2, 3, 4, 14, 15, 33, 37, 162, 170, 185–207; see also Jealousy Maternal perinatal association and inbreeding avoidance, 187–188 Mating strategies, 25, 63, 161–166, 169, 173–175, 188, 199 Meaningful existence, need for, 283 Memory, 12, 49, 50, 51, 54, 58, 66, 110, 116, 117 Menstrual cycle, 5, 219 Mimicry, 285, 288 Misinformation effect, 116 Mixed mate search, 164 Mood, 83, 107–122 Mood congruence, 110 Mood effects as Darwinian adaptations, 121–122, 126–139 Mood effects on persuasion, 118–121 Mood effects on skepticism, 112–113 Mood-maintenance hypothesis, 110 Morality, 3, 14, 89–105 Motivation, 7, 51, 89, 90, 120, 188, 190, 192 Multi-level selection theory, 233, 234, 240 Mutations and pathogens and sexual selection, 165 Mutual mate search, 152–156 Negative affective states and depressed mood, 107–139 Neocortex size, 264 Neuropsychology, 3, 4, 65, 122 Neuroticism, 137 Nonverbal communication, 3, 113 Obesity, 301 Ostracism, 15, 72, 91, 279–289 Ostracism detection, 281–283, 286–287 Outgroup homogeneity, 51, 58 Overgeneralization, 255–256, 298 Ovulation and mate preference, 161–175 Ovulatory cycle, 12, 15, 41, 43, 61, 161–175, 219


Pain, 281, 282, 285, 288 Pairing game, 152 Paranoid cognition, 268–269 Parasites, 295, 296, 297, 298, 302, 303, 305 Parental investment, 8, 25, 156, 164, 180, 191, 195, 225 Parking problem, 152, 157–158 Partner selection see Mate selection Partner-serving bias, 196 Pathogen detection, 296, 298, 303 Pathogens, 6–7, 40, 42, 165, 174–175, 182–184, 188–191 Perceived vulnerability to disease, 300–301 Person perception, 111, 180, 251, 270 Physical attractiveness, 12, 15, 52, 53, 168–173, 204, 214–217, 222–225, 246–247, 249–250, 252–254, 256–258 Positive affect, 108–111, 115–116, 131, 134–139, 252–253 Positive affect and depression, 134 Predictability, 264–265, 271, 273–274 Primates, 21, 25, 26, 41, 99, 128, 130, 238 Priming, 12, 58, 218, 221, 222, 224 Processing disjunctions, 49–66 Processing style, 110–116 Prototypicality, xvii, 13, 14, 15, 245–259 Prototypicality bias as an adaptation, 247–259 Proximal versus distal mechanisms, 5, 11, 12, 131 Psychodynamic theories of affect, 108 Psychological immune system, 6, 7, 11, 15, 293–305 Reasons for partner similarity in relationships, 197–198 Reciprocal altruism, 8 Recognition heuristic, 148 Regulation in relationships, 199–200, 206–207 Rejection, 4, 112, 152–154, 198; see also Ostracism Relational leaders, 237 Religion, 28, 29, 30, 102, 237

Reproductive fitness, 247–248, 251, 254–255, 259 Resource conservation theory of depression, 126–127, 138 Romantic love, 39, 40; see also Attraction; Sexuality Romantic relationships, 3, 200, 213–226 Romantic rivalry, 214–225 Satisficing, 149, 151, 157–158 Schizophrenia, 27 Self-esteem, need for, 283 Self-esteem, 136–139, 153–155, 202, 237–239, 282–283, 289 Self-perceptions and relationship satisfaction, 205 Self-perceptions of mate value, 196–204 Self-regulation, 288 Sequential decision-making, 148–151 Sexual fidelity, 185, 219, 222–225 Sexual orientation, 216–217, 225 Sexual selection, 162–175 Sexual Strategies Theory, 173–175 Sexual versus emotional fidelity, 222–224 Sexuality, 3, 4, 5, 6, 33, 41, 42, 61, 162–163, 165–166, 172–173, 186–191, 190, 191, 216–217, 224–225 Signal detection, 297 Signaling systems, 8, 9, 34–39, 44 Similarity of partners in relationships, 197, 204–205 Skepticism, factual versus interpersonal, 111, 112 Social attention holding power, 128–129 Social brain hypothesis, 14, 21–32, 93, 99, 127 Social cognition, xvii, 179–192, 217, 251, 258, 284, 299 Social cognition and evolution, 1–20 Social cognition and leadership, 239–241 Social competition and depression, 127, 128–129 Social exclusion and depression, 131–132 Social groups, xvii, 14, 25, 26, 30, 51, 230, 235 Social influence, 4, 120, 125, 170, 231 Social networks, 26–28



Social perception, 3, 15, 34, 132, 181, 182, 264, 298; see also Social cognition Social inference, 33–45, 115, 136, 180, 187, 303 Social rank and depression, 128 Social rejection, 294; see also Ostracism Social relationships, 27, 43, 132, 138 Social risk-taking and depression, 130–139 Social uncertainty, 268, 271 Social value and depression, 132 Social versus non-social influences on persuasion, 125, 127, 136 Speed dating and mate search, 155–156 Spotlight effect, 269 Status, 12, 64, 65, 127, 128, 175, 214, 215, 217, 223, 225, 231 Stereotypes, 7, 12, 111, 267, 301 Stigma, 7, 297, 301, Strategic leadership, 237–240 Strategic pluralism, 162, 166, 199–200 Stress, 5, 102, 131, 134 Subjective familiarity, 253, 258

Subliminal perception of sexual rivals, 217–219 Subliminal priming, 218, 221–222, 224 Sunk costs theory of depression, 127 Suppression and amplification, 58–60 Theory of mind, 22, 23, 24, 29, 179, 181–182 Threat, 214, 218, 219, 222, 223, 224, 225 Top-down versus bottom-up processes, 51, 110 Trade-offs in mate preferences, 203 Uncertainty, 286, 271 Waist-to-hip ratio, 38, 71, 79, 220, 221, 225 Wason card selection task, 133–135, 280 Westermarck Hypothesis and inbreeding avoidance, 186–187 Xenophobia, 303–304