Neurobiology of Exceptionality (The Springer Series on Human Exceptionality)

Citation preview

Neurobiology of Exceptionality

Neurobiology of Exceptionality Edited by

Con Stough Swinburne University of Technology, Hawthorn, Victoria, Australia

Kluwer Academic / Plenum Publishers New York, Boston, Dordrecht, London, Moscow

Library of Congress Cataloging-in-Publication Data Neurobiology of exceptionality : the biology of normal and abnormal traits / [edited] by Con Stough. p. cm. — (Plenum series on human exceptionality) Includes bibliographical references and index. ISBN 0-306-48476-5 1. Neuropsychology. 2. Neuropsychiatry. I. Stough, Con. II. Series. QP360.N4924 2005 612.8—dc22 2004054837

ISBN 0-306-48476-5
C 2005 Kluwer Academic/Plenum Publishers, New York 233 Spring Street, New York, New York 10013

http://www.kluweronline.com 10 9 8 7 6 5 4 3 2 1 A C.I.P record for this book is available from the Library of Congress. All rights reserved No part of this book may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, microfilming, recording, or otherwise, without written permission from the Publisher, with the exception of any material supplied specifically for the purpose of being entered and executed on a computer system, for exclusive use by the purchaser of the work. Permissions for books published in Europe: [email protected] Permissions for books published in the United States of America: [email protected] Printed in the United States of America

Preface Over the last decade there has been considerable progress in our understanding of the neurobiological basis of many psychologically related phenomena. Significant research endeavors have been mounting in both basic cellular and animal neuroscience. In terms of human behavioural research in neuroscience new and exciting research is now emerging in understanding the causes of the more common psychiatric traits. Certainly drug research in psychiatric disorders has grown exponentially over the last few years. In terms of psychiatry, more often than not, much of this research has focused on the most prevalent psychiatric disorders such as schizophrenia and depression. As a researcher involved in understanding the neurobiological basis of both psychological and psychiatric traits I am often asked to provide sources of information and references for integrated reviews and expert opinions that focus on the neurobiology of what I might call less frequently studied but important psychological traits and psychiatric disorders. Such traits are often but not exclusively related to childhood behaviours and disorders and invariably involve an understanding of important psychological processes. Unfortunately there is much less research on the neurobiology of constructs such intelligence, personality and creativity and disorders such as ADHD, autism, mental retardation and antisociality. Moreover the research in this field is not easily accessed. Although there are active research groups studying these phenomena, there is not the same sort of resources allocated for research into the large adult disorders such as depression and schizophrenia as understanding human intelligence. This is a shame in many ways, because clearly there is a need for research on the biological basis of important traits such as intelligence and creativity and childhood disorders such as autism. One of the main aims of this book is to provide some coverage of the neurobiology of lesser researched and profiled psychological and psychiatric traits. Although there are select individual sources of information on some of the topics covered in this book available elsewhere, there is no one single source v

vi

Preface

that provides up to date accounts, that are easily accessible to researchers, psychologists, teachers, students and parents. Indeed the chapters in this book cover a wide range of research on the neurobiology of fascinating psychological and psychiatric traits and are intended to help readers quickly understand our current knowledge of the biological processes for each of these different areas. In this regard I believe the book will be useful to both researchers, educators and parents. In this book I have invited leading researchers in different areas to write comprehensive reviews on topics that I believe will be of great interest to researchers, students, educators, parents and psychologists. Indeed I believe that such a book is important for several reasons. First we must continue to attract a new generation of researchers into studying the neurobiological basis of these traits which have traditionally been under-studied. Second, the information contained in this book is long over due for parents who are interested in not just the behavioural information relating to childhood and other disorders but the underlying biological basis of these behaviours in their children. Often, parents make important decisions for their children without the requisite knowledge to make these decisions. This is not a criticism of parents. Up until recently such information was not easily accessible. Perhaps the information contained in the chapters in this book may assist parents in better understanding these disorders. Third, and perhaps most importantly, both psychologists and teachers often have a profound misunderstanding of the biological basis of both key psychological traits such as intelligence, personality and creativity and abnormal psychological traits that are inherent in childhood disorders such as such ADHD. This often stems from a misunderstanding of the difference between nature and nurture. Many teachers and psychologists still confuse genetic influences on our behaviours with the neurobiological processes that underpin our behaviours. Indeed our biology represents both genetic and environmental influences and underpins all of our behaviours, thoughts and actions. Clearly an understanding of only our children’s behaviours without an understanding of the underlying biological basis for these behaviours is rather limiting. Probably the other reason that both psychologists and educators commonly do not understand the neurobiology of important psychological and psychiatric constructs is that often neurbiological techniques are highly complicated, confusing and technical. To remediate this latter problem, Aina Puce in Chapter One provides an excellent overview and description for psychologists and educators not involved in neuroscience, explaining the basis of current neurobiology methodologies and techniques. The knowledge expertly outlined in this chapter will greatly facilitate the information contained in the chapters to follow. In Section II, the chapters present current reviews of the neurobiological basis of psychological traits spanning constructs such as intelligence, creativity and personality. In Section III, several chapters are presented that deal with our current understanding of the neurobiology of psychiatric traits,

Preface

vii

particularly related to childhood disorders. Although the book is not intended as a comprehensive coverage of all areas in psychiatry and psychology, the book emphasizes areas that are not often covered in both of these areas. Overall the book is concerned with the neurobiology of exceptional psychological traits and psychiatric disorders.

Contents Part I: An Overview of Neurobiological Methods 1. Neurobiological Techniques: Overview of Terms, Procedures, and Technologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Aina Puce

3

Part II: Neurobiology of Psychological Traits 2. The Neurobiology of Impulsive Sensation Seeking: Genetics, Brain Physiology, Biochemistry, and Neurology . . . . . . . . . . . . . . . . . . . . . . . . Marvin Zuckerman

31

3. Neurobiology of Creativity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . David Camfield

53

4. Neurobiology of Intelligence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cindy Van Rooy, John Song, and Con Stough

73

Part III: Neurobiology of Abnormal Traits 5. Neurobiology of Antisociality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107 Lisa J. Cohen 6. Neurobiology of Autism, Mental Retardation, and Down Syndrome: What Can We Learn about Intelligence? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125 Christopher J. Lawrence, Ira Lott, and Richard J. Haier ix

x

Contents

7. Neurobiology of ADHD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143 Maree Farrow, Florence Levy, and Richard Silberstein 8. Neurobiology of Savant Syndrome . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199 Robyn Young Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217

1 Neurobiological Techniques Overview of Terms, Procedures, and Technologies Aina Puce “Whilst part of what we perceive comes through our senses from the object before us, another part always comes out of our own head.” —William James’s ‘The General Law of Perception’ (1890) in The Principles of Psychology.

INTRODUCTION Intelligence, personality, emotion, creativity are all qualities that all in a sense ‘come out of our heads’. The challenge is to study them objectively and scientifically so that we may understand the neurobiology underlying human exceptionality. Humans have always obsessed about their ability or inability to do something relative to others in their peer group or population. The idea that individuals who possess some superior skill or capability, or are different has always encouraged closer scrutiny of that individual’s make-up. During the 20th century this took the form of studying the brains of exceptional individuals post-mortem (e.g. Donaldson & Canavan 1928). Indeed, the brain of Albert Einstein was removed and preserved in formalin within 7 hours of his death! It has subsequently been reexamined in terms of its structural idiosyncracies, and found to have differences in the parietal lobes that supposedly differ from the normal population (Witelson, Kigar & Harvey 1999). These findings have been controversial. Firstly, it has been claimed that these structural idiosyncracies are not different from anatomical variants seen in the normal population (Galaburda 1999). Secondly, and more 3

4

aina puce

importantly, these studies cannot address the issue of whether focal brain regions become enlarged through use rather than constitution (Seitz 1999). Hence, the current focus on understanding the differences underlying exceptional individuals and us ‘lesser mortals’ will probably only be clarified by studying the living, active brain. During the latter half of the 20th century major advances in digital technology enabled the development of a number of neuroimaging methods effectively providing a ‘window onto the brain’. These kinds of techniques have the potential to detect what may be different about the brains of exceptional individuals using in vivo studies. In this first chapter of The Neurobiology of Exceptionality I present a brief background of procedures and technologies most commonly used to study what ‘comes out of our heads’. The purpose of this chapter is not to provide a comprehensive review of all procedures and technologies, but to give the reader: r a brief explanation of the theoretical basis for some of the more commonly encountered procedures and technologies in studying human cognition; r an outline of how the procedure and technology are practically applied to study human participants in vivo, both from data collection and data analysis points of view; r the advantages and disadvantages of using the procedure and technology, both from the point of view of sampling and acquiring data, and of risks and discomforts for the participant. I will attempt, wherever possible, to provide the reader with suggestions for further reading both at a basic and expert level. Technical terms are used sparingly and are accompanied by a definition. Traditionally, the study of the human brain has been divided along two lines, based on structure versus function. Today, this line is blurred somewhat, as many techniques that assess brain function in vivo also use these structural methods as an overlay to display their output.

TECHNIQUES THAT ASSESS BRAIN STRUCTURE Traditionally, brain structure could only be studied post-mortem e.g. Donaldson & Canavan (1928). An exciting development occurred in the latter part of the 20th century, when brain structure could finally be studied in living individuals. Not only was this important for the study of the normal brain, but it also revolutionised assessment techniques in neurology and neurosurgery. Tumours could be imaged and identified and the regions of permanently damaged brain tissue following stroke could be easily seen with these new techniques, enabling the development of more specific and efficient therapeutic interventions. For readers interested in the history of these developments and their clinical applications a comprehensive

Neurobiological Techniques

5

3-part article in Investigative Radiology (Hemmy, Zonneveld, Lobregt, & Fukuta 1994, Zonneveld 1994, Zonneveld & Fukuta 1994) gives a concise and clear review of this area. Here, I present two commonly used methods for assessing brain structure in vivo that have also been applied to the study of the diseased and healthy brain.

Computerised Tomography Computerised Tomography (CT), sometimes known as Computer Assisted Tomography (CAT), uses X-radiation to scan the brain or organ of interest. X-rays are produced by chemically unstable substances as a result of a chemical reaction, whereby the substance reaches its chemically stable state. CT scanning allows images of the brain or body to be generated as different tissues will transmit or absorb X-radiation depending on different attenuation coefficients (related to the density) of the various bodily tissues. The patient is placed between an X-ray source and an X-ray detector array. The X-ray tube and detector array trace a circular path around the patient. Multiple samples are taken across a series of orientations around the patient. At each sampled location a profile of different X-ray intensities is obtained. These sampled X-ray profiles are then filter back-projected, or processed using a specific mathematical algorithm, so that a reconstructed image can be produced i.e. a picture which represents the sampled tissue. For detail on the principles behind filtered back projection see Anderson and Gore (1997). CT scanning was introduced by Sir Godfrey Hounsfield in 1972 (Hounsfield 1973), however, it was only about 7 years later that three-dimensional rendering techniques were first used (Herman & Liu 1979) and then pioneered clinically in cranio-facial surgery in Australia (Hemmy 1987; Hemmy et al 1994). CT scanning has been successfully used for demonstrating abnormalities in bone (particularly around the skull base), detecting acute haemorrhages (following a stroke) and highlighting brain tumours (Fig. 1). On the other hand, the main disadvantage with this technique is that the fine structural details of the soft tissues of the brain are not well seen. Another disadvantage is that X-radiation is used, so radiation safety issues become important. A contrast agent may need to be injected into a peripheral blood vessel for some studies where the integrity of the blood-brain barrier is being investigated (e.g. tumours, acute stroke), hence adding an invasive element to the investigation. This method of investigation may sometimes remain the only safe viable option, as the patient may have previously been implanted with a pacemaker, or other metallic device that prevents them from being safely scanned using magnetic resonance imaging (see below).

Magnetic Resonance Imaging (MRI) Unlike CT, Magnetic Resonance Imaging (MRI) does not use radioactivity. The MRI scanner itself consists of a very large permanent magnet and set of

6

A

aina puce

B

C

Figure 1. CT scans of the head (in axial orientation) showing regions of abnormality (enclosed by broken circles). A. Residual brain injury following a stroke where the blood supply was deprived to the injured region. B. An abnormality of bone: erosion of bone has taken place as a result of a tumor. C. A tumour has invaded both brain tissue as well as bone. Fluid build-up (oedema) in the surrounding tissue has created some compression of the invaded cerebral hemisphere, as seen by the compressed cerebral ventricle on the affected side. The other cerebral hemisphere has also been compressed across the midline.

cylindrical metallic coils that are capable of modifying the magnetic field experienced in the magnet with a high degree of precision in three dimensions. The broad principles outlining MRI are outlined below. A detailed explanation of the principles underlying MRI can be found in Brown and Semelka (1999). Every nucleus possesses a spin, or wobble, around an axis. If a body (or tissue) is placed in the permanent magnet of the MRI, overall the spins of the nuclei will align themselves with the magnetic field. The most studied nucleus in MRI is hydrogen, an element that is present in about 90% of our bodies. Each nucleus has a characteristic, so called, magnetic resonance frequency of spin (or wobble). For the hydrogen nucleus, or proton, this frequency is 63.86 MHz in a static magnetic field of 1.5T (or Tesla)1 . A static field strength of 1.5T is currently the most commonly used field strength in clinical MRI, although recently scanners of 3T and 4T are becoming increasingly used in both research activation as well as clinical studies. MRI began to be routinely used in clinical practice in the 1980s, quite a considerable time after the phenomenon of nuclear magnetic resonance was first documented (Gabillard 1952). In 1973, Sir Peter Mansfield at Nottingham University in the UK and Paul Lauterbur at SUNY Stonybrook independently created a two dimensional map of nuclear spin densities within a material sample (Lauterbur 1973, Mansfield & Grannell 1973). It was only in the late 1970s that scientists began to apply magnetic resonance techniques to living tissue (Mansfield & Pykett 1978).

Neurobiological Techniques

7

During an MRI scan the patient lies in the circular bore, or tunnel, of the magnet while the magnetic field characteristics are changed by rapidly changing magnetic fields in the Radio Frequency (RF) range2 . The brief RF pulses of energy are briefly absorbed as nuclei change their alignment relative to the static field of the MRI scanner. After a brief period the RF energy is released from the brain or body tissue, as the sampled tissue returns to equilibrium. The special microphone, or receiver, records this RF emission, or MR signal. The receiver is usually just another coil, usually worn around the head or body. MRI images are generated after the released RF energy is sampled by the receiver, which itself is connected to a powerful analog-to-digital converter and signal processing computer. Essentially, the important measurements that need be made on any MR signal are: (I) its size or magnitude; (II) its frequency; (III) its phase, or time difference, relative to the original RF pulse. These measurements are made rapidly and sampled using specialised analogdigital converters whose output is rapidly sampled and stored on computer. How are MR signals sampled and processed to produce the kinds of pictures that we are used to seeing the brain’s anatomy in fine detail? First, different tissues have different relaxation times i.e. times of emitting the absorbed RF energy pulse, which form the basis of contrast in the MR image. For example, fat has a shorter relaxation time than does tissue or water. Pulse sequences are a series of RF pulses and gradients applied in a precise reproducible manner which are varied to emphasize different tissue types relative to their respective relaxation times (Fig. 2). So-called T1-weighted images emphasize grey-white brain tissue

A

B

C

D

Figure 2. MRI scans (sagittal view) of the brain at around the level of the eye produced with different sequences to emphasize different tissue type and detail. A. High-resolution T1-weighted image shows excellent differentiation between grey and white matter. B. Low resolution T1-weighted image shows less of the detail. C. Magnetic Resonance Angiography (MRA) image shows major blood vessels and spaces with cerebrospinal fluid. D. Gradient echo echoplanar image acquired within a single shot and high speed in a functional MRI study. Note the difference in resolution between the structural MRI scans (A–C) and the functional MRI image (D).

8

aina puce

Figure 3. A structural MRI scan (axial MRI T2-weighted) of the brain (top to bottom presented from left to right) showing a malignant high-grade tumour with associated oedema which has invaded most of the left cerebral hemisphere (shown as a large bright region on the right side of the image) in scans taken from different levels through the brain.

differences (Fig. 2A and 2B). Fat appears white and water appears black. So-called T2-weighted images emphasize fluids such as water, and blood which appear white (Fig. 3 and Fig. 2D) and deemphasize fat and bone which appear dark in the image. Magnetic resonance angiography (Fig. 2C) also has this emphasis and is designed to specifically image blood vessels. The RF pulses and other static magnetic field gradients are used so that each point in 3D space can be coded uniquely. Sometimes a contrast agent (gadolinium-DTPA) is injected into the bloodstream when information about the integrity of the blood-brain barrier is sought. The MRI scanner can therefore detect a large variety of suspected abnormalities, some examples of which are shown in Figure 3. The advantages of MRI over CT scanning are clear. MRI has better tissue resolution and uses non-ionizing radiation—attractive from a radiation safety point of view. However, MRI scanning comes with its own potential safety hazards. Any person entering the MRI scanner room must be screened for any metallic objects either on their person, or within the body itself. Some metals have magnetic properties (e.g. recall a school science experiment with iron filings and a bar magnet) and may be strongly attracted to the very strong magnetic field in the MRI scanner. They will be propelled towards the center of the MRI scanner at breakneck speeds and could injure anyone in the vicinity of the magnet’s bore. Similarly, any devices that have been surgically implanted in the body, or accidentally embedded e.g. schrapnel, that are themselves magnetic may begin to move within the body and cause internal injury. Additionally, there is the problem of the strong magnetic field erasing credit cards, stopping watches, and reprogramming implanted pacemakers that has to be watched for. Hence, all facilities housing MRI scanners will have a thorough screening procedure, including a safety questionnaire, which is completed before a patient enters the magnet. Despite these hazards, being in the presence of the strong magnetic field itself is not known to be associated with any health risks.

Neurobiological Techniques

9

TECHNIQUES THAT ASSESS BRAIN FUNCTION Direct Measures of Neuronal Output The following methods give us the most direct assessment of brain function by providing direct measures of neuronal output, usually by sampling the electrical activity of the brain produced in the course of neurons at work. Electroencephalography (EEG) Our brains possess, on average, about 1010 nerve cells, or neurons (Shepherd 1998). The neurons (pyramidal cells) are arranged in the cerebral cortex in very organised manner. They usually align their bodies and longitudinal axes perpendicular to the brain’s (cortical) surface. The summed electrical activity from millions of our neurons is continually spontaneously generated regardless of whether we are at work, play or asleep. This electrical activity can be non-invasively recorded from recording electrodes placed on the scalp, as well as from invasive recordings made from inside the brain itself, or on its surface. This technique is called Electroencephalography, or (EEG), and was first performed on human subjects by Hans Berger in Germany in 1929. Many years earlier, Richard Caton had already demonstrated that this was feasible in recordings made directly from the brain’s surface in rabbits as early as 1875 in England. A fascinating history of the developments in this field and in neuroscience in general is given by Stanley Finger in his Origins of Neuroscience (1994). The EEG is biased to record mainly the activity of neurons that are located in the smooth surface, or gyri, of the cerebral cortex. The EEG, as recorded from the scalp, is measured using specialised amplifiers as it is of the order of around one tenthousandth of a volt (100µ Volts). Unfortunately, the neuronal activity that is picked up in the form of electrical signals on the scalp is attenuated, distorted and ‘smeared’ by the fluid bathing the brain (the cerebrospinal fluid, or CSF), the skull and the scalp, so that the exact source of activity can be difficult to determine (Allison, Wood & McCarthy 1986). Recordings made direct from the surface of the brain are usually of the order of around one thousandth of a volt and are used only in specialised neurosurgical applications, most typically epilepsy surgery. We can study the changes in EEG with a great degree of accuracy in time – in the order of thousandths of seconds (milliseconds). The EEG itself is composed of a range of activity spanning frequencies of around 1–40 Hz, and can be sorted and classified into activity in various frequency bands known as delta (1–3.5 Hz), theta (3.5 Hz–7.5 Hz), alpha (7.5–12.5 Hz) and beta (12.5–40 Hz). Activity in the alpha band characteristically occurs at the back of the head, over the visual areas of brain, and can be seen most clearly on eye closure (Fig. 4). In clinical laboratories the EEG is recorded from a standardised set of recording electrodes, based on lines of electrodes proportionally spaced in 10% and 20%

10

aina puce

Figure 4. EEG rhythms recorded during EYES OPEN (left) and EYES CLOSED (right). Recordings have been made using an array of electrodes beginning at the front of the head (top) and ending at the back of the head (bottom). In the EYES CLOSED condition alpha activity is seen usually at the back of the head (enclosed by broken circle), and is absent in the EYES OPEN condition. The large deflection in the EYES OPEN condition (enclosed by the broken circle) at the front of the head is an eyeblink, illustrating that other electrical signals from the body can occur as unwanted signals, or artifacts, in the EEG.

increments of distances from the front-to-back, and side-to-side of the head (using the so-called International 10–20 system). All together there are around 30 or so electrodes. In research laboratories, it is common to use more dense arrays of electrodes, with 64 and 128 electrodes, for example. The electrode positions in these instances can be either related to 10–20 system sites, or can be placed in an array in which all electrodes have the same distance to their nearest neighbour (geodesic placement). Power Spectral Analysis of the EEG. The frequency content of the EEG can be charted using Power Spectral Analysis. Here the EEG is essentially displayed in an alternative format: instead of looking at the EEG waveforms in time (Fig. 4), the same data can be displayed in terms of frequency (Fig. 5). Sometimes, the display of EEG data in the frequency domain can highlight rhythmic features in the EEG that are not as clearly seen when it is displayed in the time domain (Fig. 5). Spectral EEG analysis relies on the mathematically based technique of Fourier transformation, based on a branch of mathematics made famous by Jean Fourier in

Neurobiological Techniques

EYES OPEN

11

EYES CLOSED

Figure 5. Power spectra of the EEG activity seen from the bottom trace of Figure 4 for the eyes open (left) and eyes closed (right) conditions. Alpha activity appears as the most prominent peak in the trace (right, arrow). The y-axis displays power and the x-axis displays frequency ranging from 0 to 60 Hz.

the 19th century. It allows any signal that changes in time to be expressed as a function of frequency by expressing the signal as a series of Fourier coefficients, which effectively describe the amount of signal that is present at particular frequencies. The Fourier Transform of a signal, therefore consists of a theoretically infinite series of summed Fourier coefficients. The computer algorithm which calculates the Fourier transform uses some mathematical short-cuts and is often called the Fast Fourier Transform, or FFT. The behaviour of the EEG can be concisely and accurately plotted using spectral analysis and changes may be monitored over time and specified as either changes in the overall power (or energy) of the EEG signal, or in terms of the relative power in the various EEG frequency bands. EEG Coherence. Another common way of analysing the EEG signal is known as Coherence Analysis (Nunez, Srinivasan, Westdorp, Wijesinghe et al., 1997, Nunez, Silberstein, Shi, Carpenter et al., 1999). Here the degree to which various brain regions generate synchronous EEG signals i.e. are coherent can be calculated. The coherence between individual electrodes sites on the scalp, or between regions of brain can be calculated. For accurate measures of coherence to be calculated, usually a large number of recording electrodes in the scalp are used—typically 128 or greater. Evoked (and Event-Related) Potentials (ERPs) It has been known for some time that the EEG could change predictably and reproducibly in response to sensory stimulation (e.g. Berger’s work in the 1930s). However, considerable time passed before an approach was developed in which these changes in the EEG could be reliably seen. In 1949 Dawson working at Cambridge University was able to use a cathode ray oscilloscope (the first type

12

aina puce

of TV tube) to store multiple traces of nerve action potentials in the periphery, Erb’s point and the scalp EEG in response to peripheral electrical stimulation of the median nerve of the hand (Dawson & Scott 1949). The final display indicated that there were reliable changes in the EEG that were time-locked, or occurred at fixed times, to the brief electrical stimulus. These time-locked changes in the EEG are known as Evoked Potentials, or EPs. Today, EPs are recorded routinely by digitally sampling and storing the EEG. The EEG is basically cut into brief segments known as epochs that begin at the time that the stimulus is delivered. Then the epochs are all summed together, or averaged. The resulting signal average is a display consisting a serious of voltage ‘bumps’ that change over time (Fig. 6). Averaging can be performed by a special device

occipital cortex

non-specific visual response

temporal cortex

face-specific response

limbic system

target detection response

+ 100 ms

100 µV

Figure 6. Averaged ERP waveforms recorded from within the brain itself using special brain surface and depth electrodes. Note the large size of the ERPs, which are about an order of magnitude greater than those usually recorded from the scalp. Here data from a target detection task (subjects made a button press to grey squares) are shown. Early ERP activity is seen to all stimulus categories (top) in occipital cortex. Later category-specific ERP activity occurs to only one stimulus category i.e. task irrelevant faces (middle) in visually sensitive temporal cortex. Late ERP activity in response to the target is seen in the limbic system (bottom).

Neurobiological Techniques

13

such as an EP Averaging System which is routinely used in clinical applications. In the research environment it is more common to digitally store the EEG for subsequent analysis. The EEG data can then be averaged and analysed using additional methods that are not available on a clinical EP system. Typically, in research studies we are interested in examining EPs to various stimulus types in a single experiment, and it is not usually possible to separate responses to different stimulus types on most clinical EP systems. In an experiment using multiple stimulus types we are often interested in studying not only the perception of the stimulus, but also cognition. The perceptual EP components are often called exogenous potentials, as their structure or morphology, and timing is predictable, pretty invariant and can be attributable to various structures in primary cortical or subcortical sensory structures in the central nervous system. EP components, which typically occur after the activity in sensory structures has occurred, and are related more to cognition are known as endogenous potentials. They are called endogenous as they are internally generated mental events that may occur even in cases of stimulus omission (Ruchkin, Sutton, Stega 1980)! Hence, it is common to speak of Event-Related Potentials, or ERPs, in this type of context. An ERP experiment will generate potentials that are both exogenous and endogenous, and typically uses an experimental design where multiple stimulus types are presented in random order. ERP experiments can often clearly differentiate responses to different stimulus categories as shown in Figure 6. Topographical Mapping Techniques Any aspect of the EEG (power, coherence) or ERP (peak voltage at a certain time point) can be displayed across the recording electrode space, in a topographic map which shows how the displayed parameter behaves across the head. Similarly, current source density maps, calculated from the first mathematical derivative of the EEG voltage are also a common form of displaying the EEG or ERP activity. Additionally, the results of statistical tests may also be displayed in the way, where the results of contrasting 2 test conditions may be displayed to show the variation in activity across the head. These types of mapping technique can also be applied not only to scalp EEG and ERP data, but to data recorded directly from the brain’s surface itself (Fig. 7). Additionally, the scalp EEG and ERP data may be plotted out in terms of what the voltages would look like on the surface of the brain. Here, a mathematical algorithm takes into account the attenuation and smearing effects of the scalp, skull and CSF and generates a plot of the theoretically calculated profile of voltages on the brain’s surface. These techniques are known as Laplacians (Nunez, Silberstein, Cadusch, Wijesinghe et al., 1994), named after a branch of mathematical transformation credited to the French mathematician Pierre-Simon Laplace in the late 19th century.

14

aina puce

Figure 7. Example of a voltage topography map created from a grid of intracranial electrodes sited on the surface of the brain, seen here from its underside in an MRI based reconstruction (left). The voltage distributions elicited in response to faces and face parts over the occipitotemporal cortex show both positive and negative voltages at key time points following stimulus onset. The gray scale displays voltages from ±200 µV.

Steady-State Probe Topography (SSPT) Like conventional ERP techniques, SSPT also relies on demonstrating changes in the EEG that occur as a function of stimulus presentation, however, the approach is conceptually quite different. The SSPT technique uses a ‘probe’ stimulus which consists of a continuously presented sinusoidally flickering light stimulus, which is essentially presented as a ‘background’ stimulus while the subject watches a stimulus display and performs a cognitive task or participates in an experiment with a pharmacological manipulation. The sinusoidal flicker stimulus is delivered through a set of goggles with a semi-transparent mirror, and the subject watches the stimulus display that is associated with the task on a computer screen. By examining changes in the amplitude (size) and phase difference (expressed as a change in latency) of the so-called steady-state visual evoked potential (SSVEP), it is possible to infer dynamic changes in brain activity and processing mode associated with a cognitive task. The SSVEP was originally recorded from depth electrodes implanted in visual cortex (Kamp, Sem-Jacobsen & Van Leeuwen 1960), and subsequently from the scalp (Van der Tweel & Verduyn Lunel 1965) in recordings made from small sets of electrodes. In these studies the sinusoidal flicker was the sole stimulus and the response properties of the SSVEP were examined. Today, SSVEP recordings are

Neurobiological Techniques

15

typically made using large sets of electrodes using the flicker stimulus as a probe (Silberstein, Schier, Pipingas, Ciorciari et al., 1990), and the data are displayed as topographic maps of SSVEP amplitude and latency changes across conditions (Fig. 8A and 8B). The approach used was based on probe evoked potential studies where a well-defined, repetitive stimulus was used to indirectly study perception and cognition (Papanicolaou & Johnstone 1984). The amplitude of the SSVEP to a 13 Hz flicker stimulus changes similarly to alpha EEG activity (10–13 Hz) in that decreases in conditions of increased visual vigilance (Silberstein et al., 1990, Nield, Silberstein, Pipingas, Simpson, Burkitt 1998). SSVEP latency changes are interpreted as highlighting changes in neural information processing speed in the neural generators of the SSVEP, which in turn index regional variations in excitatory and inhibitory tone (Silberstein, Farrow, Levy, et al., 1998). Additionally, the statistical reliability of these effects may be evaluated using statistical measures that takes

amplitude latency Hotelling’s T

Figure 8. Topographic maps of SSVEP changes during nicotine administration relative to a placebo condition. A. Normalised amplitude difference (in µV) with cool colors representing increased SSVEP amplitudes in the nicotine condition relative to placebo. B. SSVEP latency differences (in msec) with warm colors representing latency decreases for the nicotine condition, relative to placebo. C. Hotelling’s T statistic showing significant differences between nicotine and placebo conditions for both SSVEP amplitude and latency. (Modified from Thompson, Tzambazis, Stough, Nagata, Silberstein, 2000.)

16

aina puce

into account both the SSPT magnitude and phase changes, the Hotellings-T test (Fig. 8C and see Silberstein, Ciorciari & Pipingas 1995), to display significantly activated brain regions. Alternatively, the SSPT magnitude and latency differences between conditions can be tested separately using regular t-tests or other statistics. Magnetoencephalography (MEG) Like EEG and ERP studies, magnetoencephalography detects rapid changes in brain activity over time. Rather than measuring voltages from recording electrodes placed on the scalp, MEG uses electromagnetic sensors to sample the changes in the magnetic field that are emitted from the brain as a function of time (George, Aine, Mosher, Schmidt et al., 1995, Lounasmaa, Hamalainen, Hari & Salmelin et al., 1996). It was first described by Cohen in the late 1960s-early 1970s (Cohen 1968, 1972). This method uses principles similar to those described in the MRI section: the sensors themselves consist of wires that detect small currents that are induced in them as a function of the changing magnetic fields that the brain produces. These tiny currents are difficult to detect, and the MEG sensor array is also cooled by liquid helium and nitrogen, just as an MRI scanner’s magnet. This ensures that extremely small currents can be detected, however, it makes the system susceptible to noise (unwanted magnetic fields that are not generated by the subject’s brain). For this reason MEG recordings are usually performed in a shielded room, i.e. a purpose built room that has been specially designed to screen out stray magnetic fields from outside. In addition, no electrical equipment or other sources unwanted magnetic fields are in the room itself. The internal structures of the room are usually made of wood and other non-metallic materials, and the participants themselves remove all items of metal on their person3 . MEG has the advantage over EEG techniques, in that the magnetic fields emitted from the brain are not distorted or smeared by the cerebrospinal fluid, skull and scalp (Cuffin & Cohen 1972). On the other hand, as the MEG sensors themselves are coils of wire that sit parallel to the scalp’s surface, they will be biased to sample activity that comes mainly from neurons that are oriented perpendicular to them4 . Hence, EEG and MEG techniques sampled potentially complimentary brain activity, with the EEG biased to recording activity from the flat surfaces or gyri of the brain, and MEG recording activity of neurons in the folds or sulci. The detected MEG activity comes mainly from so-called radial sources5 , and the EEG activity comes mainly from tangential sources. The collected MEG data can then be mapped as a series of sources and sinks of current emanating from the head (Fig. 9B). Given that the positions of the sensors are known exactly (Fig. 9A), the neuronal populations within the brain that produce this activity can be identified: sources and sinks located close together usually indicate that the origin of the neural activity lies close to the cortical surface near the detector. On contrast, large separation distances between sources and sinks usually indicate a deep-seated neuronal generator.

Neurobiological Techniques

17

Figure 9. MEG activity is recorded from an array of sensors placed near the scalp shown in an MRI based reconstruction of the head (A), which then detect the evoked magnetic fields from the scalp in response to stimulation. B. Averaged MEG waveforms at all 37 sensor positions in response to eye gaze deviations in a face (top), and to inward motion of the radial pattern encircling the displayed face (bottom). C. The locations of the active neuronal sources have been modelled and localised to the left occipitotemporal cortex, displayed on the right of the MRI scan in 3 different planes. D. Location of neuronal sources in a volume rendered, cut-away view of brain generated from a high-resolution structural MRI scan. (Modified from Watanabe, Kakigi & Puce 2001.)

Additionally, sources for this neuronal activity can be calculated using specialised (Equivalent Current Dipole) software. These software routines take into account the individual subject’s head and brain structure, and require each subject to also have a high-quality MRI scan of their brain. The position, strength and orientation of the sources can then be modelled taking into account the profile of MEG activity measured across the sensor array and the subject’s brain anatomy (Fig. 9C), and the source can be displayed on a volume rendered cut-away brain (Fig. 9D).

Indirect Measures of Neuronal Output These techniques mostly depend on measuring metabolic activity and/or blood flow. The measured parameter is then related to neuronal output. Neuronal output is said to be coupled to blood flow and metabolic activity i.e. there is a proportional relationship between the two. With positron emission tomography (described below), it is possible to examine brain blood flow, brain blood volume, and glucose and oxygen metabolism. Positron Emission Tomography (PET) This tomographic imaging technique samples the by-products of a chemical reaction generated by short-lived isotopes, or unstable radioactive substances. This method was first used in 1951 in order to image brain tumours in the human brain (Wrenn, Good & Handler, 1951). Ingvar and Lassen (1961) pioneered the inhalation of radioactive xenon gas (Xe133 ) to quantify of regional cerebral blood flow

18

aina puce

(rCBF) for the very first time. This approach was modified to use intravenous injections of radioactive (short-lived) water and other physiological substances such as compounds containing radioactive carbon, fluorine and oxygen for example. Today it is used not only to study disease processes, but also to study perception and cognition in healthy participants. There have been three main applications of the PET technique, those of imaging: (I) cerebral blood flow; (II) glucose metabolism; (III) the distribution of neuroreceptors (for clinical applications see Mazziotta, Toga, Frackowiak 2000). The isotopes that are used in PET studies are radioactive, and therefore chemically unstable, and degrade to their non-radioactive stable state by emitting a positively charged atomic particle, or positron. Once the positron is released it soon encounters a negative particle, an electron, and the two oppositely charged particles annihilate one another. In doing so, they produce a burst of γ-ray energy, consisting of two photons i.e. packets of radiation, which leave the annihilation site in opposite directions. These photons have a discrete measurable energy or signature in the so-called γ-ray range that can be sampled by an array of detectors that detect these two coincident (i.e. simultaneously emitted at 180 degrees to one another) photons. The location of the annihilation site is inferred from the final destination of the two photons, and can be plotted over time for all detected photons. These data provide a picture of the blood flow pattern in the brain at the time of sampling, as regions with higher radioactive emissions will correspond to sites that

A

B

Figure 10. Functional neuroimaging studies in a patient with temporal lobe epilepsy. A. Imaging performed between seizures and at quiet rest. A PET fluorodeoxyglucose study shows large area of glucose hypometabolism (low metabolism) in an axial image through the left temporal lobe (enclosed by broken circle) B. SPECT 99 Tc-HMPAO studies show the blood flow profile during a seizure. Now the same region in the temporal lobe shows a large increase in blood flow, or hyperperfusion, during the seizure, which at the point in time of the injection of tracer is largely confined to the left temporal lobe (enclosed by broken circle).

Neurobiological Techniques

19

have increased blood flow. Sites with increased blood flow are inferred to be more active relative to other brain regions while the subject performs a certain cognitive task. For cerebral tissue an active state, the supply of delivered oxygenated blood greatly exceeds the metabolic demand (Fox & Raichle 1986). The different isotopes that are used in PET scanning usually have a very short half-life, or time period in which it takes the measured emitted radioactivity to decrease by half of its original value. Examples of some commonly used isotopes and their half-lives are: Oxygen15 2.1 minutes, Nitrogen13 10 minutes, Carbon11 20.1 minutes and Fluorine18 110 minutes. These isotopes exist as tracers when they exist as a compound e.g. Oxygen15 is delivered to the participant as radioactive 18 water, H15 2 O and is used for determining cerebral blood flow. Fluorine is delivered as fluorodeoxyglucose and is used for measuring cerebral metabolic rates for glucose. Carbon11 is used to label quantitate biochemical changes in the brain, such as the determination of the density of various neuroreceptor types in the brain (e.g. benzodiazepine and opiate). The very short time in which these compounds degrade makes it necessary to produce them on-site in a device known as a cyclotron. In a cyclotron a reagent substance is bombarded with a high-energy beam of electrons, or negatively charged particles, which produces the desired chemical reaction and desired isotope. Consequently, this is an expensive technology—the continued production of radioactive reagents requires a local on-site isotope production facility known as an accelerator, and the PET scanner itself is a costly piece of equipment consisting of arrays of γ-ray detectors, amplifiers, and signal processing and computing equipment. The γ-ray detectors are known as scintillation crystals. They get their name from the process that occurs when an incoming photon hits the structure of the crystal. A scintillation, or small flash of light, occurs. These flashes of light are too small to detect reliably, hence they are next detected by an array of photomultipliers, which boost these signals so that they can be recorded and processed to create images of the brain (or body part of interest). The pairs of scintillation detectors are housed in a set of rings in the scanner surrounding the patient. They detect the pairs of oppositely travelling photons and allow a distribution of the tracer concentration in the brain to be measured. The reconstruction of the imaged object takes place using a similar back-projection approach outlined in the section on CT above. The focal changes in blood flow (measured with H15 2 O) that occur when participants perform a certain cognitive task can only be identified when they are contrasted directly i.e. subtracted, from the blood flow pattern obtained to another, so-called ‘control’ task. The ‘control’ task is usually chosen to feature similar attributes to the activation task of interest, yet differs on the essential putative function under study. The reliability of the changes in activation can be assessed using a number of statistical techniques, the most popular one being Statistical Parametric Mapping (SPM), which was developed by Karl Friston and colleagues

20

aina puce

working in the Functional Imaging Laboratory at Queens Square in London in the late 1980s (Friston, Frith, Passingham, Liddle & Frackowiak 1991). There have been a number of refinements of the technique in regular software updates. The activation maps, and SPM maps must be overlayed onto a structural image of the brain and usually require the participants to also have a separate highresolution MRI scan. The PET activation data must first be aligned exactly i.e. coregistered with the structural MRI images. The coregistration process involves aligning common anatomical landmarks into a common three-dimensional space. One disadvantage of PET scanning is its relatively poor spatial and temporal resolution (cf. functional MRI described below). PET images typically can be resolved to only 5mm detail. Given that the radioactivity must grow and then die away following the performance of the activation and control tasks, this adds a time constraint to the studies. There must be a ‘wash-out’ period between experiments, or sufficient time to allow the absorbed radioactivity in the tissue of interest to fall to a undetectable value. Additionally, there are radiation safety considerations for participants. Participants must receive ‘safe’ doses of radiation i.e. be within well standardised and enforced safety limits. This also precludes the performance of multiple serial studies in the same participant. Fortunately, the very short half-life of the tracer isotopes means that the radioactivity dies away rapidly. Experimenters are also exposed to the effects of radiation if they are in the participant’s immediate vicinity during the course of the experiment. Single Photon Emission Computerised Tomography (SPECT) This method relies on the detection of single photons emitted from a radioactive decay reaction in which a γ-ray photon is produced. Unlike PET, where radioactive uptake of tracer is measured over a period of time, SPECT imaging is based on using a different type of brain perfusion contrast agent. These substances were initially made from iodine-based compounds (e.g. Winchell, Baldwin & Lin 1980; Kung, Tramposh & Blau 1983). Today Technetium (99m Tc) based compounds remain the most widely used (Tikofsky, Ichise, Seibyl, & Verhoeff 1999), such as 99m Tc-hexamethyl-propylamine-oxime (99m Tc-HMPAO), more commonly known by the name of CERETEC. CERETEC effectively produces a ‘snapshot’ of the pattern on blood flow in the brain at the time of injection (within around 1–2 minutes of injection), which remains stable and unaltered. The patient may then be scanned at a convenient time, for a period of several hours following the injection. The γ-ray emissions are detected with a γ-camera, which uses an array of scintillation detectors in a similar manner to the described in PET scanning. SPECT, by nature of its single photon emission, has poorer spatial resolution relative to the dual photon PET, however, it is less costly, in that contrast agents do not have to be produced on site and the γ-camera itself is cheaper relative to a PET scanner.

Neurobiological Techniques

21

An excellent detailed review of this methodology covering tracers, data sampling and analysis, signal processing issues, as well as applications has recently been published (Tikofsky et al. 1999) and should interest the more technically minded reader. Functional Magnetic Resonance Imaging ( fMRI) Functional Magnetic Resonance Imaging (fMRI) uses similar principles to those described in the section on MRI. Instead of generating structural images of the brain, so-called functional images are generated. Here the sampled MR signal effectively reflects blood flow in the brain or organ of interest. For an excellent review of the historical developments in this area see Chapter 1 of Schmitt, Stehling and Turner (1998). The most commonly used fMRI method relies on the so-called Blood Oxygen Level Dependent (BOLD) effect. This method uses the naturally occurring concentration of deoxyhemoglobin (in deoxygenated or venous blood)—a substance that is paramagnetic i.e. it introduces local magnetic field inhomogeneities (MR distortions or signal decreases) in its presence. When a local brain region becomes active, there is a great influx of oxyhemoglobin (in oxygenated or arterial blood). Oxyhemoglobin is nonparamagnetic i.e. does not produce local disturbances in the local magnetic field. Instead, the MR signal increases as the deoxyhemoglobin is washed away. This is the effect that is measured. This is somewhat paradoxical— as deoxyhemoglobin is the actual by-product of the activation process. However, cerebral activation produces a reaction in the vascular system that ensures that more than enough oxygenated blood flows in to the active region in industrial quantities. The first fMRI studies using blood as an endogenous contrast agent were performed in rodents (Ogawa, Lee, Kay, & Tank 1990; Ogawa, Lee, Najak, & Glynn 1990). The signal changes that are typically seen in fMRI studies of 1.5T (currently most commonly used and available clinical MRI scanners) are usually of the order of 4% in primary cortex (sensory, motor, visual), and may be less in higher order cortical regions (McCarthy, Puce, Luby, Belger & Allison 1996). Recently, fMRI studies at higher field strengths (3T and 4T) are being used, primarily in a research environment. At these higher field strengths better signal-to-noise ratios (higher signal strengths) have been reported. The high signal also allows smaller volume elements (voxels) to be sampled, allowing much high spatial resolution for activation studies. It has been claimed that at 4T cortical columns can be visualized in occipital cortex (Menon, Ogawa, Hu, Strupp & Ugurbil 1997). Most typical fMRI activation studies in some way compare functional images that have been acquired in two (or more) different states. The data across the states are then compared statistically to establish whether there are reliable and definite differences between the states. Some examples of ‘different states’ can be rest vs

22

aina puce

ANIMALS

TOOLS

Figure 11. fMRI activation study of visually sensitive cortex performed at high-field strength (3T) in a single participant seen in successive axial views (bottom to top from left to right). Activated voxels to viewing centrally presented pictures of animals (top) and tools (bottom) relative to a control condition with a central fixation cross are shown. Activated voxels were identified by using a KolmogorovSmirnov (non-parametric statistical) test between animals vs control (top) and tools vs control (bottom), thresholded at a Bonferoni corrected probability of 2.42 × 10−7 , and then overlayed on a structural MRI scan.

a perceptual/cognitive task, one task vs another task, or perhaps a drug-free vs a drug test condition. Figure 11 displays activation data from a visual task in which participants viewed centrally presented pictures of animals (Fig. 11, top) and tools (Fig. 11, bottom) relative to a centrally displayed fixation cross. Visually sensitive regions in both the occipital and temporal lobes have been highlighted. Notice the consistency of activation across the two viewing conditions—the same cortical regions have been activated to both visual stimulus categories. In order to perform an activation study the stimuli must be delivered to the subject. This poses a greater technical challenge than in PET studies (see below). The introduction of other electrical signals in the magnet room can produce RF interference and can substantially degrade the quality of the MR images. Therefore, care must be taken with introducing stimulus delivery devices, as well as other physiological monitoring gear, into the magnet room. There are so-called MR-friendly pieces of equipment that do not make this a problem. Alternatively, the stimulus delivery devices can sit outside in the control room. Any necessary

Neurobiological Techniques

23

signal that needs to be introduced, or taken, from the magnet room is channelled by fiber-optic cable through special holes in the wall of the MRI scanner room known as wave guides. The wave guide, or metallic tube, acts as an RF filter and attenuates RF with a wavelength equivalent to the length of the wave guide tube itself. Other challenges to stimulus delivery are physical constraints. First, visibility in the MRI scanner is limited for the participant. Visual stimuli can be presented to the participant via special MR-friendly goggles, that allow an almost virtual reality display to be viewed. These are expensive. Alternatively, a mirror can be mounted on the head coil or the scanner and angled so that the participant can see out. The stimulus display is then projected on a display screen, usually mounted near the feet of the participant. Second, fMRI scanning is a noisy procedure. Participants usually wear headphones to shield them from scanning noise. Hence, delivery of auditory stimuli can be a challenge. Usually, special MR-friendly headphones are used to deliver the stimuli and shield the participant from scanner noise. On some scanners the scanning protocol is such that all functional images are acquired at the beginning of each TR, hence there is sufficient silent time for the participant to listen and hear the stimuli without difficulty. On our MRI scanners this can cause a problem. Sensorimotor and other studies requiring participants to perform motor responses usually require purpose-built devices that are not only MR-friendly, but also fit into the relatively narrow tubular space that the participant lies in. The most popular design to date for fMRI studies has been the so-called block design, in which each activation task or condition is repeatedly presented in blocks, enabling the images acquired in each condition to be grouped and analysed statistically. More recently, activation tasks using event-related designs are being used, whereby the blood flow responses to single events can be reliably imaged (Rosen, Buckner & Dale, 1998).

EPILOGUE There are a variety of neurobiological techniques that are available to the cognitive neuroscientist for the study of exceptionality. The technique of choice depends largely on the required spatial and temporal resolution, which depend on the scientific question being asked. Figure 12 shows the relative resolutions of methods used in cognitive neurosciences relative to one another, with respect to temporal and spatial scale. Traditionally, one technique has usually been used in conjunction with behavioural measures. More recently, however, cognitive neuroscientists are beginning to combine techniques, so as to attempt to build up a more exact picture of the temporal behaviour of the active structures making up a cognitive network. For instance, ERP and fMRI data are being combined, with

Space (log millimetres)

24

aina puce

SPECT

3 brain

MEG & scalp EEG

2 PET

map

1

column

0

lesions

fMRI microlesions 2DG layer neuron dendrite

−1 −2 −3

intracranial EEG

optical dyes

single-unit electrophysiology patch clamp

light microscopy

synapse

−4 −4 −3 −2 millisecond

−1

3 4 1 2 0 minute hour second

5 day

6

7

Time (log seconds) Figure 12. Relative spatial and temporal resolution of various techniques used in cognitive neuroscience research in terms of both actual units of measurement, and across different levels within the central nervous system. This illustration has been adapted from Churchland and Sejnowski (1988) to now include some additional neuroscientific techniques (broken rectangles).

the ultimate endpoint being more accurate characterisation and modelling of the active neuronal sources (e.g. Liu, Belliveau & Dale 1998; Dale, Liu, Fischl, Buckner et al. 2000). Here, the individual participant’s anatomy of the head (e.g. skull thickness, and unique pattern of folding in the brain) provides a means to reduce the number of possible (and potential infinite) number of solutions of the pattern of neural activity seen across the head at any instant in time. Gifted individuals form a very small proportion of the general population. The recent developments in cognitive neuroscience will make it possible to gain the required insight into individuals with exceptional ability in the beginning of the 21st century.

ACKNOWLEDGEMENTS Mr. Ari Syngeniotis of the Brain Research Institute, Austin & Repatriation Medical Centre, Melbourne, Australia generated the CT and structural and functional MRI images (from a 1.5T and 3.0T MRI scanner).

Neurobiological Techniques

25

Dr. Sam Belangieri, of the Department of Nuclear Medicine, and Centre for Positron Emission Tomography, Austin & Repatriation Medical Centre furnished the PET and SPECT activation images. Mr. James Thompson of the Brain Sciences Institute, Swinburne University of Technology graciously generated the images of SSVEP topography. The MEG data were acquired by Dr. Shoko Watanabe in the Laboratory of Professor Ryusuke, Department of Integrative Physiology, National Institute for Physiological Sciences, Okazaki, Japan. The intracranial ERP data were acquired in the Neuropsychology Laboratory headed by Professor Truett Allison, Department of Neurology, Yale University School of Medicine, New Haven, CT, USA. The fMRI activation data were acquired using the visual stimuli of Professor Glyn Humphreys, Department of Psychology, University of Birmingham, UK.

NOTES 1. A Tesla is the standard unit of measurement for magnetic field, named after a physicist named Nicola Tesla who was born in Croatia in 1856. Tesla first observed the relationship between alternating current in a set of wires and the changes in the magnetic field that were produced. 2. The RF range is usually expressed in 10s and 100s of MHz, where 1 Hertz (Hz) = 1 cycle/second, and 1 MHz = 1,000,000Hz. 3. The presence of metal objects in the room can distort or alter the measured magnetic fields. 4. This is because a current induced in a wire only occurs when there is a magnetic field that is perpendicular to it. There is a physical law allows the size and the orientation of the induced current and the magnetic, as well as electric, field to be calculated. 5. By contrast, EEG electrodes that sit of the scalp’s surface sample activity mainly from neurons that are tangential to the cortical surface (tangential sources). Hence, EEG and MEG actually give complementary information about the neural activity that is seen in preferentially in the surface (gyri) and folds (sulci) of the cerebral cortex.

REFERENCES Allison, T., Wood, C.C., McCarthy, G. (1986). The Central Nervous System. In G.H. Coles, E. Donchin, & S.W. Porges. (Eds.), Psychophysiology. (pp 1–25). New York: The Guilford Press. Anderson, A.W., Gore, J.C. (1997). The physical basis of neuroimaging techniques. Child and Adolescent Psychiatric Clinics of North America, 6, 213–264. ¨ Berger, H. (1929). Uber das Elektroenkephalogramm des Menschen. Archiv f¨ur Psychiatrie und Nervenkrankheiten, 87, 527–570. Brown, M.A., Semelka, R.C. (1999). MRI Basic Principles and Applications. Second Edition. New York: Wiley-Liss. Caton, R. (1875). The electric currents of the brain. British Medical Journal, 2, 278. Churchland, P.S., Sejnowski, T.J. (1988). Perspectives on cognitive neuroscience. Science 242, 741– 745. Cohen, D. (1968). Magnetoencephalography: Evidence of magnetic fields produced by alpha-rhythm currents. Science, 161, 784–786.

26

aina puce

Cohen, D. (1972). Magnetoencephalography: Detection of the brain’s electrical activity with a superconducting magnetometer. Science, 175, 664–666. Cuffin, B.N., Cohen. D. (1979). Comparison of the magnetoencephalogram and electroencephalo-gram. Electroencephalography and Clinical Neurophysiology, 47, 132–46. Dale, A.M., Liu, A.K., Fischl, B.R., Buckner, R.L., Belliveau, J.W., Lewine, J.D., Halgren, E. (2000). Dynamic statistical parametric mapping: combining fMRI and MEG for high-resolution imaging of cortical activity. Neuron, 26, 55–67. Dawson, G.D., Scott, J.W. (1949). The recording of nerve action potentials through skin in man. Journal of Neurology, Neurosurgery and Psychiatry, 12, 259–267. Donaldson, H.H., Canavan, M.M. (1928). A study of the brains of three scholars. Granville Stanley Hall, Sir William Osler, Edward Sylvester Morse. Journal of Comparative Neurology, 46, 1–95. Finger, S. (1994). Origins of Neuroscience. A History of Explorations into Brain Function. New York: Oxford University Press. Fox, P.T., Raichle, M.E. (1986). Focal physiological uncoupling of cerebral blood flow and oxidative metabolism during somatosensory stimulation in human subjects. Proceedings of the National Academy of Sciences (USA), 83, 1140–1144. Friston, K.J., Frith, C.D., Liddle, P.F., Frackowiak, R.S.J. (1991). Comparings functional (PET) images: The assessment of significant change. Journal of Cerebral Blood Flow and Metabolism, 11, 690– 699. Gabillard, R. (1952). A steady state transient technique in nuclear magnetic resonance. Physiology Reviews, 85, 694–705 Galaburda, A. (1999). Albert Einstein’s brain [letter]. Lancet, 354, 1821; discussion 1822. George, J.S., Aine, C.J., Mosher, J.C., Schmidt, D.M., Ranken, D.M., Schlitt, H.A., Wood, C.C., Lewine, J.D., Sanders, J.A., Belliveau, J.W. (1995). Mapping function in the human brain with magnetoencephalography, anatomical magnetic resonance imaging, and functional magnetic resonance imaging. Journal of Clinical Neurophysiology, 12, 406–31. Hemmy, D.C. (1987). Three-dimensional imaging in craniofacial disorders: A brief review. Australian & New Zealand Journal of Surgery, 57, 101–104. Hemmy, D.C., Zonneveld, F.W., Lobregt, S., Fukuta, K. (1994). A decade of clinical threedimensional imaging: A review. Part I. Historical development. Investigative Radiology 29, 489– 496. Herman, G.T., Liu, H.K. (1979). Three-dimensional display of human organs from computed tomograms. Computer Graphics and Image Processing, 9, 1–21. Hounsfield, G.N. (1973). Computerized transverse axial scanning (tomography): Part I. Description of a system. British Journal of Radiology, 46, 1016–1922. Ingvar, D.H., Lassen, N.A. (1961). Quantitative determination of regional cerebral blood flow in man. Lancet, 2, 806–807. James, W. (1890). The Principles of Psychology. Reprinted in 1983 by the Harvard University Press, Cambridge MA, USA. Kamp, A., Sem-Jacobsen, C.W., Van Leeuwen, W.S. (1960). Cortical responses to modulated light in the human subject. Acta Physiologica Scandinavica, 48, 1–12. Kung, H.F., Tramposh, K., Blau, M. (1983). A new brain imaging agent: (123 I) HIPDM: N,N,N=-trimethyl-N=-(2-hydroxy–3-methyl–5-iodobenzyl)–1,3-propanediamine. Journal of Nuclear Medicine, 24, 66–72. Lauterbur, P.C. (1973). Image formation by induced local interaction: examples employing nuclear magnetic resonance. Nature, 243: 190–191. Liu, A.K., Belliveau, J.W., Dale, A.M. (1998). Spatiotemporal imaging of human brain activity using functional MRI constrained magnetoencephalography data: Monte Carlo simulations. Proceedings of the National Academy of Sciences (USA), 95, 8945–8950.

Neurobiological Techniques

27

Lounasmaa, O.V., Hamalainen, M., Hari, R., Salmelin R. (1996). Information processing in the human brain: magnetoencephalographic approach. Proceedings of the National Academy of Sciences (USA), 93, 8809–8815. Mansfield, P., Grannell, P.K. (1973). NMR ‘diffraction’ in solids? Journal of Physics C, 6: 422–426. Mansfield, P., Pykett, I.L. (1978). Biological and medical imaging by NMR. Journal of Magnetic Resonance, 29, 355–372. Mazziotta, J.C., Toga, A.W., Frackowiak, R.S.J. (2000). Brain Mapping. The Disorders. San Diego: Academic Press. McCarthy, G., Puce, A., Luby, M., Belger, A., Allison, T. (1996). Magnetic resonance imaging studies of functional brain activation: Analysis and interpretation. In: Visualization of Information Processing in the Human Brain: Recent advances in MEG and functional MRI. Electroencephalography and Clinical Neurophysiology (Suppl) 47, 15–31. Menon, R.S., Ogawa, S., Hu, X., Strupp, J.P, Ugurbil, K. (1997). Ocular dominance columns in human V1 demonstrated by functional magnetic resonance imaging. Journal of Neurophysiology, 77, 2780–2787. Nield, G., Silberstein, R.B., Pipingas, A., Simpson, D.G., Burkitt, G. (1998). Effects of visual vigilance tasks on gamma and alpha frequency range staeady-state potential (SSVEP) topography. Brain Topography Today. Eds. Y. Koga, K. Nagata, & H. Hirata. Elsevier. pp. 189–194. Nunez, P.L., Silberstein, R.B., Cadusch, P.J., Wijesinghe, R., Westdorp, A.F., Srinivasan, R. (1994). A theoretical and experimental study of high-resolution EEG based on surface Laplacians and cortical imaging. Electroencephalography and Clinical Neurophysiology, 90, 4–57. Nunez, P.L., Srinivasan, R., Westdorp, A.F., Wijesinghe, R.S., Tucker, D.M., Silberstein, R.B., Cadusch, P.J. (1997) EEG coherencey I: Statistics, reference electrode, volume conduction, Laplacians and cortical imaging. Electroencephalography and Clinical Neurophysiology, 103, 516–527. Nunez, P.L., Silberstein, R.B., Shi, Z., Carpenter, M.R., Srinivasan, R., Tucker, D.M., Doran, S.M., Cadusch, P.J., Wijesinghe, R. (1999). EEG coherency II: Experimental measures of multiple EEG coherence measures. Electroencephalography and Clinical Neurophysiology, 110, 469–86. Ogawa, S., Lee, T., Kay, A., Tank, D. (1990). Brain magnetic resonance imaging with contrast dependent on blood oxygenation. Proceedings of the National Academy of Sciences (USA), 87, 9868– 9872. Ogawa, S., Lee, T.-M., Nayak, A.S., Glynn, P. (1990). Oxygenation-sensitive contrast in magnetic resonance image of rodent brain at high magnetic fields. Magnetic Resonance in Medicine 14, 68–78. Papanicolaou, A.C., Johnsone, J. (1984). Probe evoked potentials: Theory, method and applications. International Journal of Neuroscience, 24, 107–131. Rosen, B.R., Buckner, R.L., Dale, A.M. (1998). Event-related functional MRI: Past, present and future. Proceedings of the National Academy of Sciences (USA), 95, 773–780. Ruchkin, D.S, Sutton, S. Stega, M. (1980). Emitted P300 and slow wave event-related potentials in guessing and detection tasks. Electroencephalography and Clinical Neurophysiology, 49, 1–14. Seitz, J. (1999). Albert Einstein’s brain [letter]. Lancet, 354, 1822–3. Schmitt, F., Stehling, M.K., Turner, R. (1998). Echo-Planar Imaging, Theory, Techniques and Application. Berlin: Springer. Shepherd, G.M. (1998). The synaptic organisation of the brain. New York: Oxford University Press. Silberstein, R.B., Ciorciari, J., Pipingas, A. (1995). Steady-state visually evoked potential topography during the Wisconsin card sorting test. Electroencephalography and Clinical Neurophysiology, 96, 24–35. Silberstein, R.B., Farrow, M.A., Levy, F., Pipingas, A., Hay, D.A., Jarman, F.C. (1998). Functional brain electrical activity mapping in boys with attention deficit hyperactivity disorder. Archives of General Psychiatry, 55, 1105–1122.

28

aina puce

Silberstein, R.B., Schier, M.A., Pipingas, A., Ciorciari, J., Wood, S.J., Simpson, D.G. (1990). Steadystate visually evoked potential topography associated with a visual vigilance task. Brain Topography, 3, 337–347. Thompson, J.C., Tzambazis, K., Stough, C., Nagata, K., Silberstein RB. (2000). The effects of nicotine on the 13Hz steady-state visually evoked potential. Clinical Neurophysiology, 111, 1589–1595. Tikofsky, R.S., Ichise, M., Seibyl, J.P., Verhoeff, N.P.L.G. (1999). Functional brain SPECT imaging: 1999 and beyond. In L.M. Freeman (Ed.), Nuclear Medicine Annual 1999 (pp 193–263). Philadelphia: Lippincott Williams & Wilkins. Van der Tweel, L.H., Verduyn Lunel, H.F.E. (1965). Human visual responses to sinusoidally modulated light. Electroencephalography and Clinical Neurophysiology, 18, 587–598. Watanabe, S., Kakigi, R., Puce, A. (2001). Occipitotemporal activity elicited by viewing eye movements: a magnetoencephalographic study. Neuroimage, 13, 351–363. Winchell, H.S., Baldwin, R., Lin, T.H. (1980). The development of 123 I-labeled amines for brain studies: localization of 123 I iodophenylakyamines in rat brain. Journal of Nuclear Medicine, 21, 940–946. Witelson, S.F., Kigar, D.L., Harvey, T. (1999). The exceptional brain of Albert Einstein. Lancet, 353, 2149–2153. Wrenn, E.R., Good, M.L., Handler, P. (1951). The use of positron emission radioisotopes for the localization of brain tumors. Science, 113, 525–527. Zonneveld, F.W. (1994). A decade of clinical three-dimensional imaging: A review. Part III. Image analysis and interaction, display options, and physical models. Investigative Radiology, 29, 716– 725. Zonneveld, F.W. & Fukuta, K. (1994). A decade of clinical three-dimensional imaging: A review. Part II. Clinical applications. Investigative Radiology, 29, 574–589.

2 The Neurobiology of Impulsive Sensation Seeking Genetics, Brain Physiology, Biochemistry, and Neurology Marvin Zuckerman

TRAIT DEFINITION I was originally asked to write this chapter on the “Neurobiology of Personality,” but on reflection realized that this was too broad a topic to be covered in a short chapter. I therefore decided to focus on a specific personality trait which I have studied for about 40 years. Those interested in the broader field of the “Psychobiology of Personality” may consult my book on this subject (Zuckerman, 1991) or my chapter in the recently published Handbook of Psychology (Zuckerman, 2003). Another volume devoted to sensation seeking and its “behavioral correlates and biosocial bases” can be consulted for a more extensive exposition on the trait (Zuckerman, 1994). The definition of sensation seeking in this book is: “ . . . a trait defined by the seeking of varied, novel, complex, and intense sensations and experiences, and the willingness to take physical, social, legal, and financial risks for the sake of such experience” (p. 27). It must be emphasized that risk per se is not the goal of sensation seeking, although it may enhance the arousal produced by the sensations themselves. Impulsive sensation seekers hardly consider the risks, but more planful ones evaluate them and do what they can to reduce them. Low sensation seekers simply do not

Marvin Zuckerman

•

Professor Emeritus, University of Delaware

31

32

marvin zuckerman

see the point in taking any risks that are unneccesary and they tend to overestimate risk, as highs underestimate it (Horvath & Zuckerman, 1993; Zuckerman, 1979). Other investigtors have used terms like “novelty seeking,” “thrill-seeking,” “monotony avoidance,” and “venturesomeness” to describe this trait, but these are unsatisfactory in that they only describe part of the broader concept, including what qualities of sensation are sought (intensity and complexity as well as novelty) and what forms of expression are found. Factor analyses of the extended scale items revealed four forms of expression which were translated into subscales in forms IV and V of the Sensation Seeking Scale (SSS) (Zuckerman, 1971; Zuckerman, Eysenck, & Eysenck, 1978): 1. Thrill and Adventure Seeking (TAS): expressed in a desire to try risky but exciting physical activities like sky diving, scuba diving, hang-gliding etc. 2. Experience Seeking (ES) or the seeking of excitement through travel, art, music, and leading an unconventional life-style with like-minded friends. 3. Disinhibiton (Dis) a social expression, seeking excitement through social drinking, drugs, parties, and sexual activities with a variety of partners. 4. Boredom Susceptibility (BS) is an aversion to lack of variety in stimulation or friends and restlessness when there is a lack of variation and change. A total score based on all four factors is used in form V as contrasted with a General scale in form IV of the SSS. More recently, based on the factor analyses of scales and items, we developed a scale combining impulsivity and sensation seeking items called “Impulsive Sensation Seeking” (Zuckerman, 2002; Zuckerman, Kuhlman, Joireman, Teta, & Kraft, 1993).

PSYCHOBIOLOGY The reliability and construct validity of scales used to measure sensation seeking and the many kinds of phenomena associated with the trait are described in previous books (Zuckerman, 1979; 1994). This chapter will focus on the biological correlates. My approach is a top-down one, from traits to genes with all levels between (Zuckerman, 1993), However, I have drawn rather extensively from studies of exploration and novelty seeking in other species in a comparative approach (Zuckerman, 1984). In this chapter I will describe studies which link behavioral expressions of sensation seeking in humans and other species using a common psychophysiological index found in both. I will begin, however, with the area that links the top (traits) with the bottom (genetics) in the science of human behavior genetics.

The Neurobiology of Impulsive Sensation Seeking

33

GENETICS OF SENSATION SEEKING Biometric studies of the genetics of personality and intelligence have used several methods including studies comparing identical and fraternal twins, adoption studies comparing children with biological and adoptive parents and siblings, and family studies of first degree relatives from intact families. The methodology for these studies is aptly described in the most recent edition of “Behavioral Genetics” (Plomin, DeFries, McClearn, & Rutter, 1997). The first genetic study of sensation seeking analyzed the genetic and environmental contributions to the trait in a large sample of English twins (Fulker, Eysenck, & Zuckerman, 1980). The correlations for identical (MZ) and fraternal (DZ) are shown in Table 1. Heritabilities calculated for these data showed that 58% of the variation in the trait, as measured by the Total score on the SSS V, could be accounted for by genetic factors. The heritability would be 69% if corrected for the unreliability of the SSS measure. This is a high heritability for personality traits which usually average at about 40% for the uncorrected heritability (Bouchard, 1994; Loehlin, 1992). The remaining variance was due to the specific or nonshared environment and error of trait measurement. There was no detectable effect for shared environment, or that part of the environment shared by virtue of having the same parents and growing up in the same general social environment. There is always some doubt as to whether the comparison between identical and fraternal twins really controls for shared environment since parents do tend to treat identical twins more similarly than they do fraternal twins. The adopted twin method is a more stringent control because all twins are separated soon after birth and adopted into different families. The SSS V was given to the separated twins in the Minnesota study. The correlation for the identical twins separated during their formative years (Table 1) was .54 and that for separated fraternal twins was .32. The correlation of the identicals was not very different for those found in the Fulker et al. study of twins from intact families. Since the correlation between separated identical twins is the heritability it is 54% for the purely genetic factor. Because fraternal twins Table 1. Twin Correlations and Heritabilities of Twins on SSS Total Score

Fulker et al. (1980) Twins raised together Hur & Bouchard (1997) Twins separated

MZ

DZ

Males

.63

.21

Females

.56

.21

Males Female

.54

.32

.54

.64

h2

.58

.59

34

marvin zuckerman

share only half of their genes their correlation must be doubled and thus yields a heritability of 64%. If we average these two estimates we get 59%, almost exactly the same 58% heritability found in the Fulker et al. study of twins raised together. This would indicate that the shared or family environment of twins does not account for the variance in the trait. The heritability data for the subscales in the Fulker et al study were analyzed by Eysenck (1983). A replication study in a Dutch population (Koopmans et al., 1995), and the separated twin study (Hur & Bouchard, 1997) also analyzed the data by the SSS subscales. The results from these studies are shown in Table 2. The heritabilities for the three subscales Dis, TAS, and ES were fairly similar across the five groups in the three studies, averaging .50 to .57. That for BS was lower, probably because of the lower reliability of this subscale. None of the studies showed effects of shared environment and the remaining variance was due to the non-shared environment and error. Eysenck found specific genetic factors accounted for the predominance of the genetic variance for the subscales, except for ES which was almost entirely accounted for by the general genetic factor found in all of the scales. Koopmans et al. could not find one genetic factor underlying all 4 subscales, but did find genetic coveriance between Dis and BS, and another genetic factor underlying TAS and ES. Dis and BS are the most elevated scales of sensation seeking found in antisocial personalities of the primary type (Emmons & Webb, 1974). The Hur and Bouchard study also included the Control scale (an inverse measure of impulsivity) of the Multiple Personality Questionnaire to their separated twins. The Control scale correlated negatively with all of the SSS subscales, and for all but the TAS scale the major part of the covariance was due to shared genetic variance by Control and SSS. This genetic relationship between impulsivity and sensation seeking is supportive of the new scale combining the two (ImpSS). In all of these studies there was little evidence for an influence of shared environment. However, another analysis of the Koopmans et al. study of Dutch twins shows the influence of family environment (Boomsa, de Geus, van Baal, &

Table 2. Heritabilities for the SSS Subscales fromThree Studies Koopmans et al. (1995)

Hur & Bouchard Mean

SSS

Eysenck (1983) Males

Females

Males

Females

M&F

Dis TAS ES BS

.51 .45 .58 .41

.41 .44 .57 .34

.62 .62 .56 .48

.60 .63 .58 .54

.46 .54 .55 .40

All 5 .50 .54 .57 .43

Note. SSS = Sensation Seeking Scale; Dis = Disinhibition; TAS = Thrill and Adventure Seeking; ES = Experience Seeking; BS = Boredom Susceptibility

The Neurobiology of Impulsive Sensation Seeking

35

Table 3. Twin Correlations and Proportions of Variance due to Genes, Shared Environment, and Non-Shared Environment∗ Males

Females

Upbringing

MZ

DZ

MZ

DZ

Religious Non-Religious

.62 .62

.62 .35

.61 .58

.50 .35

Mean Dis

Genetic

Shared E

Non-Shared E

29.45 31.86 34.21 36.53

37 61 00 49

25 00 62 11

38 39 38 40

Percents of variance Gender/Upbringing Female-Religious Female-Non-Religious Male-Religious Male-Non-Religious ∗

From D. I. Boomsa, et al. (1999) Twin Research, V.2, 115–125. Copyright Stockton Press 1999. Reprinted with permission.

Koopmans, 1999). The overall heritability for the Dis scale was high (.60–.62). Boomsa et al. divided the sample of twins into those with a religious upbringing and those with no religious upbringing. Most of the twins continued to practice or not practice their religion as a function of their upbringing. Both upbringing and current religion showed high concordance in twins regardless of zygosity indicating that only shared environment influenced past or current religion. As expected, those with a religious upbringing had lower scores on Dis than those with no religious upbringing. Table 3 shows the twin correlations on Dis for twins with and without a religious upbringing. For males with a non-religious upbringing the analysis showed a moderate effect of an additive genetic factor (49%), a small effect of shared environment (11%), and a moderate effect of non-shared environment (40%) on Dis. But for males raised in a religious environment the correlations for identical and fraternal twins were the same and both were high. Genetic factors accounted for none of the variance and all of the variance on Dis was due to shared (62%) and non-shared environments (38%). Females with a non-religious upbringing showed a strong genetic effect and no effect of shared environment, whereas those with a religious background showed a lower effect for genetic factors and a significant influence of shared environment as well as non-shared environment. In sum, a religious upbringing attenuates the genetic influence and increases the shared environmental influence as well as lowering the overall scores on Dis. In the absence of a religious upbringing the genetic factor becomes paramount. The overall result for the total population conceals what appears to be a real genotype X environment interaction.

36

marvin zuckerman

Molecular Genetics Recent developments in molecular genetics offer the possibility of finding specific genes of major effect for personality traits as well as forms of psychopathology. Ebstein et al. (1996) first reported an association between alleles of the Dopamine Receptor D4 gene and the trait of novelty seeking as measured by a scale developed by Cloninger (1987). There are two common forms of the gene in Western populations: a short form with 4 repeats of the base sequence and a long form with 7 repeats. The longer form is associated with high scores on novelty seeking and the short form with low to moderate scores on this scale. Novelty Seeking (NS) correlates very highly (r about .7) with the ZKPQ scale Impulsive Sensation Seeking (Zuckerman & Cloninger, 1996). Soon there were some successful replications and some unsuccessful ones leading Baron (1998) to conclude that the association between the trait (sometimes measured by other tests) and gene was non-existent. However, a more recent review by Prolo and Licino (2002) reports 11 successful and 10 unsuccessful attempts to replicate. When positive results are obtained the gene only accounts for about 10% at maximum of the genetic variance so that other genes must be involved. Comings, Saucier, & MacMurray (2002) found that 4 dopamine receptor genes contributed to 5.25% of the variance of novelty seeking in an additive manner with each gene contributing a small part of the total genetic variance. Noble et al. (1998) found similar additive effects of the DRD2 and DRD4 genes. Apart from additive genetical effects, there may be interactive effects which could decrease the chances of finding the main effect of D4 in some studies. Strobel, Lesch, and Brocke (2003) found that while the DRD4 long allele was not associated with high NS scores alone; but when the short allele of the serotonin transporter gene and an allele of the catechol-O-methyltransferase (COMT) were present the association of NS with the DRD4 alleles was significant. A similar kind of interaction was found for genetic effects on infant exploratory behavior, involving the DRD4 receptor and a serotonin transporter gene (Ebstein & Auerbach, 2002; Ebstein et al., 1998). The DRD4 is also associated with disorders characterized by impulsivity and sensation seeking such as drug and alcohol abuse and attentiondeficit/hyperactivity disorder (Ebstein & Kotler, 2002). The DRD2 dopamine gene has also been associated with alcoholism, ADHD, pathological gambling, and smoking in some, but not all, studies (Comings et al., 2002). Several dopamine receptors seem to play a significant role in sensation seeking and impulsivity and the serotonin transporter gene may play an interactive role. Impulsive sensation seeking may be involved in approach behavior, mediated by the dopamine neurotransmitters, whereas serotonin may inhibit or delay approach behavior. This hypothesis will be further elaborated in subsequent parts of this chapter.

The Neurobiology of Impulsive Sensation Seeking

37

PSYCHOPHYSIOLOGY The original concept of sensation seeking was based on the construct of individual differences in “optimal levels of arousal”, suggesting that high sensation seekers might be tonically underaroused and seek novel or intense stimulation in order to increase arousal. Measures of tonic arousal, using EEG or skin conductance (SC), did not reveal arousal differences between high and low sensation seekers (Zuckerman, 1990). However, 5 out of 8 studies found that high sensation seekers gave a stronger SCR to the first presentation of a visual or auditory stimulus (Orienting Reflex, OR). These differences usually disappeared on subsequent trials when the stimulus was no longer novel. The differences were enhanced when the stimuli had content of interest to high sensation seekers, like aggressive and sexual words (Smith, Perlstein, Davidson, & Michael, 1986, Smith, Davidson, Smith, Goldstein, & Perlstein, 1989). Similar findings were obtained for heart rate (HR). in response to tones of moderate intensity in which the OR is a deceleration of HR. The high sensation seekers showed the greater deceleration in response to the stimulus on first presentation. These differences could be a reflection of the greater sensitivity to novel stimuli and the preference such stimuli in high sensation seekers. Buchsbaum’s (1971) method of defining a continuum of cortical augmenting to reducing involves measurement of cortical evoked potentials (EP) in response to stimuli across a range of intensities. Augmenting is defined by the extreme in which the EP amplitude strongly increases in direct relationship to the stimulus intensity. Reducing is shown when the EP amplitude shows little increase across intensities and sometimes a significant decrease at the highest intensities. Figure 1 shows the relationship between the EP measure and the Disinhibiton (Dis) subscale of the SSS in the first study by Zuckerman, Murtaugh, & Siegel (1974). Figure 2 shows the same relationship in a study using the auditory EP (Zuckerman, Simons, & Como, 1988). Many attempts have been made to replicate the results using the visual EP or the auditory EP (Zuckerman, 1990, 1994). Only 4 out of 8 attempts to replicate the results for the visual EP with the Dis scale were successful, but 8 out of 10 studies found a significant relationship between auditory EP augmenting and at least one of the SSS subscales or a general total score. Impulsivity has also been related to augmenting of the visual EP (Barratt, Pritchard, Faulk, & Brandt, 1987; Carrillo-de-la-Pena & Barratt, 1993). The EP augmenting-reducing relationship to sensation seeking has been interpreted as representing the capacity of the cortex to function at high intensities of stimulation in contrast to the reaction of low sensation seekers who show evidence of cortical inhibition mechanisms at such intensities. Some of the biochemical correlates of human augmenting include enzymes and neurotransmitters that are related to sensation seeking: low levels of the enzymes monoamine oxidase (MAO) and dopamine beta hydroxylase (DBH), the serotonin metabolite

38

marvin zuckerman

EP AMP HIGH DISINHIBITERS

24 22 20 18 16 14 12

LOW DISINHIBITERS

10 1

2

4

8

16

STIMULUS INTENSITY Figure 1. Visually evoked potentials (VEPs) of high and low scorers on the Disinhibiton subscale of the Sensation Seeking Scale as a function of stimulus intensity. From “Sensation seeking and cortical augmenting-reducing,” by M. Zuckerman et al., 1974, Psychophysiology, 11, 539. Copyright 1974 by the Society for Psychophysiological Research. Reprinted by permission.

EP AMP

HIGH DISINHIBITERS

9.0 8.5 8.0 7.5 7.0

LOW DISINHIBITERS

6.5 6.0 50

65

80

95

STIMULUS INTENSITY (DB) Figure 2. Auditory evoked potentials (EPs) of high and low scorers on the Disinhibition subscale of the Sensation Seeking Scale as a function of stimulus intensity for the short interstimulus interval condition (2 seconds). From “Sensation seeking and stimulus intensity as modulators of cortical, cardiovascular, and electrodermal response: A cross-modality study,” by M. Zuckerman et al., 1988, Personality and Individual Differences, 9, 368. Copyright 1988 by Pergamon Press. Reprinted by permission.

The Neurobiology of Impulsive Sensation Seeking

39

5-hydroxindoleacetic acid (5-HIAA), the dopamine metabolite homovanillic acid (HVA), and endorphins (Haier, Buchsbaum, Murphy, Gottesman, & Coursey, 1980; von Knorring & Perris, 1981). The relation of these to sensation seeking will be discussed in the subsequent section of this chapter. The augmenting-reducing marker has been used to study individual differences in other species. Cats showing the augmenting pattern are more exploratory, active, aggressive, and more likely to approach novel stimuli (Hall, Rappaport, Hopkins, Griffin, & Silverman, 1970; Lukas & Siegel, 1977; Saxton, Siegel, & Lukas, 1987). Saxton et al. (1987), using operant experiment paradigms, found that augmenter cats adapted more easily to the experimental chamber, learned more quickly, and responded to a fixed-interval food reinforcement task more vigorously than reducer cats. However, on a task requiring inhibition of response (differential reinforcement for low rate of response) the reducers were better because the augmenters could not restrain their responses at the low rate required for reinforcement. Augmenter cats may be a model for impulsive sensation seeking humans. Siegel, Sisson, and Driscoll (1993) extended the paradigm to two selectively bred strains of rats: Roman High Avoidance (RHA) and Roman Low Avoidance (RLA). RHA rats were bred for their capacity to actively avoid shock and RLA rats were bred for their general inhibition in reaction to shock; they tend to freeze. In a comparison of the two strains on the visual EP nearly all of the RHA rats were EP augmenters and almost all of the RLA rats were EP reducers. All members of a highly inbred strain are genetically homogeneous, almost like a colony of clones. Comparisons of the two strains on behavioral and physiological variables from other studies (Table 4) indicate likely correlates of augmenting-reducing. RHA rats (augmenters) are more exploratory and active in the open-field test whereas RLA’s (reducers) are inactive and more fearfully aroused. Like the augmenting cats, the RHA rats found it difficult to inhibit responses in a DRL reinforcement paradigm. RHA’s are more ready to drink alcohol in solution whereas the RLA’s are abstainers. RHA’s have a higher tolerance for barbiturates. RHA females are less maternal with their pups, spending less time on the nest than RLA’s The lateral hypothalamus is part of a mesolimbic reward system; animals will tend to self-stimulate themselves in these areas using electrodes planted in them connected to response keys. Dopamine neurons are involved in the brain selfstimulation response; blocking these neurons eliminates the incentive motivation for self-stimulation. The rate of self-stimulation, however, depends on the intensity of the stimulation itself. The RHA’s are less responsive to low and more responsive to high intensities of stimulation reward. The RLA’s actually show escape responses at the higher intensities, shutting off the reward, a voluntary parallel of the EP reducing phenomenon. The hormonal and neurotransmitter responses to stress are also different in the RHA and RLA strains. In response to stress the RHAs show increases in dopamine

40

marvin zuckerman

Table 4. Differences between RHA and RLA Strains VARIABLE

STRAIN DIFFERENCE

VEP Augmenting-Reducing Exploration (open-field test) Shock Induced Aggression Alcohol Drinking Barbiturates Maternal Behavior Hypothalamic

RHAs augmenters, RLAs reducers RHAs more active, less emotional RHAs more aggressive RHAs drink alcohol (in solution), RLAs abstain RHAs high tolerance, RLAs low tolerance RHA females spend less time in nest with pups. RHAs less sensitive to high intensity stimulation but more responsive to high intensity. RLAs more sensitive to low intensity, avoid high intensity

STRESS EFFECTS: MONOAMINES & HORMONES Prefrontal Cortex Hypothalamus

Pituitary

STRAIN DIFFERENCES RHAs increased dopamine, RLAs no change RLAs increased serotonin (5HT), RHAs less change. RLAs increased corticotropin releasing factor (CRF), RLAs less change in CRF RLAs increased adrenocorticotropic hormone (ACTH), RHAs less change in ACTH

Note. From M. Zuckerman (2002) Personality and psychopathy: Shared behavioral and biological traits. In Neurobiology of Criminal Behavior. J. Glicksohn (Ed.). Boston, MA: Kluwer Academic Publishers. Copyright 2002. Reprinted by permission.

levels in the prefrontal cortex, whereas the RLAs show no change. The mesolimbic dopamine system terminates in the prefrontal cortex. The RLA’s show more activation of the Hypothalamic-Pituitary-Adrenocortical (HYPAC) system in response to stress. The RLAs have increased serotonergic increase in the hypothalamus, increased response of the corticotropin releasing factor (CRF), leading to increased activation of the pituitary releasing adrenocorticotropic hormones (ACTH). The RHA’s show less response in this hormonal stress pathway. A group of French investigators have developed an animal model for sensation seeking utilizing the reactivity to novelty in rats as an analogue of the human trait (Dellu, Piazza, Mayo, LeMoal, & Simon, 1996). The high reactives (HRs) show novelty seeking behaviors in several kinds of experimental situations and the HRs showed more amphetamine self-admistration and response for food reinforcement than the LRs. In response to stress the HRs showed more increase in dopamine concentration in the nucleus accumbens (NA), a locus for self-stimulation of stimulant drugs. Locomotor response to novelty also correlated positively with DOPAC (a metabolite of dopamine) in the NA in a subsequent basal condition (post-mortem). Exposure to novelty led to equivalent increases in corticosterone 30 minutes after exposure, but the response of the HR’s remained elevated 2 hours after exposure whereas that of the LRs returned to baseline. Their data support the role of

The Neurobiology of Impulsive Sensation Seeking

41

dopamine reactivity in the RHA/RLA model, but their results are contradictory in regard to HYPAC activity.

BIOCHEMISTRY Hormones The end result of HYPAC activity is cortisol release in humans, often seen in situations of prolonged stress. Ballenger et al. (1983) found that CSF cortisol correlated negatively with scales in the P-ImpUSS personality factor including the EPQ P scale, the SSS Disinhibition scale, the MMPI Hypomania scale and the number of reported previous and present sexual partners. The low levels of cortisol found in disinhibited antisocial traits in the normal population is consistent with the low levels of urinary cortisol found in habitually psychopathic and violent prisoners (Virkunen, 1985). Low levels of CSF corticotrophin are also typical in alcoholic offenders with psychopathic personality traits (Virkkunen et al., 1994). Cortisol is high in persons with anxiety states and severe depression, but lower in manic states. The low levels of cortisol related to psychopathy may indicate the lack of anxiety and stress reactivity in these personality types. Adrenalin or norepinephrine produced in the adrenal medulla is another hormone indicative of states of stress or anxiety arousal. Low urinary levels of epinephrine in boys of 13 predicted criminal activity when they reached 18–26 years of age (Magnusson, 1987, 1996; Olweus, 1987). The low states of NE are even found in prisoners in stressful situations like awaiting a criminal trial (Lidberg, Levander, Schalling, & Lidberg, 1978). Testosterone from blood is correlated with sensation seeking, particularly that of the experience seeking and disinhibitory types (Aluja & Torrubia, in press; Daitzman & Zuckerman, 1980), although Bogaert and Fisher (1995) and Dabbs (2000) found only nonsignificant tendencies toward association using salivary testosterone. Hypogonadal men with very low testosterone referred for complaints of erectile dysfunction were lower on sensation seeking than men with normal levels of testosterone (O’Carroll, 1984). Testosterone in young males correlates with their sexual experience, as defined by the number of sexual partners they have had (Bogaert & Fisher, 1995; Dabbs, 2000; Daitzman & Zuckerman, 1980). Other corelates of testosterone in males include assertiveness, impulsivity, and low self-control. A history of antisocial behavior, beginning in childhood, is found in men with high testosterone levels (Dabbs, 2000).

Neurotransmitters and their Enzymes Monoamine Oxidase (MAO) is an enzyme found in two forms, A and B. MAO-B is obtained from blood platelets in living organisms and is primarily involved in the metabolism of dopamine, regulating its level in the dopamine neuron

42

marvin zuckerman

Table 5. Psychopathology and Monoamine Oxidase (MAO) Low MAO found in: High normal sensation seekers Attention-Deficit Hyperactivity Disorder Antisocial Personality Disorder Chronic Criminality Borderline Personality Disorder Alcoholism Relatives of alcoholics Drug Abusers Bipolar Mood Disorder Relatives of Bipolar Disorders Paranoid Schizophrenics

Zuckerman (1994) Shekim et al. (1986) Lidberg et al. (1985) Klinteberg (1996) Reist et al. (1990) Major & Murphy (1988) Shukitt (1994), Sher (1993) von Knorring et al. (1987) Murphy & Weiss (1972) Leckman et al. (1977) Zureik & Meltzer (1990)

by catabolizing this neurotransmitter after uptake (Murphy, Aulakh, Garrick, & Sunderland, 1987). MAO-A is involved in the intracellular metabolism of serotonin while MAO-B metabolizes extracellular serotonin (Azmitia & Whitaker-Azmtia, 1995). MAO-B is positively correlated with cortisol activity (Schatzberg & Schildkraut, 1995) and with serotonin activity in particular brain areas (Adolfsson, Gottfries, Oreland, Roos, & Winblad, 1978). Twin studies suggest that MAO-B is under nearly total genetic control involving one or two major genes. Behavioral differences between low- and high-MAO babies have been found in the first three days of life consisting of higher activity, motor reactivity and motor maturity in the low MAO babies (Sostek, Sostek, Murphy, Martin, & Born, 1981). MAO-B has been correlated negatively with sensation seeking in many studies; high sensation seekers tend to have low levels of platelet MAO (Zuckerman, 1994). Although the mean relationship is not high (about .24 for Total SS and the Dis subscale) it was significant in 9 of 13 studies reviewed in 1994). Low MAO found in various disorders also characterized by high levels of sensation seeking, particularly disinhibition, and impulsivity (Table 5). Among alcoholics, low MAO is especially associated with the type II alcoholic characterized by younger onset, more antisocial behavior when intoxicated, and a family history of alcoholism (Pandey et al., 1988). The suggestion that the MAO link with psychopathology is genetically transmitted is supported by the finding of low MAO in the relatives of alcoholics and bipolar patients who may not manifest the disorder itself. Many studies have found associations between low MAO and tobacco, alcohol, and drug use in non-clinical populations. Convictions for felony offenses are not common in a college population, but 37% of students selected from a very low MAO group had such convictions, contrasted with only 6% of the high MAO group (Coursey, Buchsbaum, & Murphy, 1979).

The Neurobiology of Impulsive Sensation Seeking

43

Low MAO monkeys living in a natural environment were more playful, active, social, aggressive, dominant, and sexually active than monkeys with high MAO levels (Redmond, Murphy, & Baulu, 1979). College students with low MAO reported more hours spent socializing than high MAO students (Coursey et al., 1979) but the results with self-report measures of extraversion have been inconsistent. All of the above studies have used MAO-B from blood platelets. Recent data have implicated MAO type A in antisocial behavior and aggression, and studies on mice have shown a relationship between aggression and the absence of the gene. Caspi et al. (2002) have shown an interaction between childhood maltreatment and alleles of the gene encoding MAO-A. This is a longtudinal study so that the records of childhood maltreatment do not depend on retrospective accounts. Among those with the form of the gene that results in low MAO-A activity those children who also experienced severe maltreatment had a high incidence of conduct disorder during childhood, and antisocial personality disorder with violence at age 26. But among those with the form of the gene producing high MAO-A activity there was no difference in incidence of either of these disorders or in violence between those maltreated and those not maltreated during childhood. Neither the form of the gene nor childhood maltreatment alone were sufficient to produce antisocial personality, but the genetic-experience interaction did so.

Monoamines The influence of both types of MAO genes points to the importance on the monoamine neurotransitters regulated by these enzymes: the catecholamines dopamine and norepinephine and the indoleamine, serotonin. A large body of comparative research suggests some broad functions of the monoamines in behavioral reactions (Zuckerman, 1991). In general, dopamine circuits in the brain are involved in appetitive approach and exploration, serotonin in behavioral control, and inhibition of approach in response to signals of danger, and norepinephrine in cortical and autonomic arousal, activity, and fearful reactions to novel or threatening stimuli. This is an oversimplistic picture. In fact, all three monoamines may be activated by threat. Serotonin is associated with the inhibition of all ongoing behavior in fear (freezing), and dopamine with the active avoidance (running). Norepinephrine may not only be associated with fear arousal but with pleasurable arousal as produced by sex, stimulant drugs, or a non-threatening but novel stimulus. My model proposes that all three behavioral mechanisms, therefore all three monoamines, are involved in impulsive sensation seeking as shown in Figure 3. Note that there are interactions suggested between behavior mechanisms and between the monoamines at each level. In contrast, the model of Cloninger (1983) proposes one neurotransmitter underlying each specific trait: dopamine

44

marvin zuckerman E-Sociability

+

+ Approach

+

− −

Dopamine

− Gonadal Hormones

−

− Inhibition

N-Anxiety

− +

Arousal

+

− + Serotonin − Norepinephrine − GABA DBH + − +

+

+

P-ImpUSS

MAO Type B

Endorphins

Figure 3. A psychopharmacological model for Extraversion-Sociability (E-Sociability), Impulsive Unsocialized Sensation Seeking (P-ImpUSS), and Neuroticism-Anxiety (N-Anxiety) showing underlying behavioral mechanisms (approach, inhibition, and arousal) and neurotransmitters, enzymes and hormones involved. Agonistic interactions between factors are indicated by a plus sign and antagonistic interactions by a minus sign. MAO = monoamine oxidase; DBH = Dopamine-beta-hydroxylase; GABA = gamma-aminobutyric acid. From “Good and bad humors: biochemical bases of personality and its disorders.” by M. Zuckerman in Psychological Science, 1995, 6, 331. Copyright 1995 by American Psychological Society. Reprinted by permission.

for novelty seeking, serotonin for harm avoidance, and norepinephrine for reward dependence. The study of the role of the monoamines in human personality is much more difficult than studying their relationship to behavior in rats because one does not have direct access to the brain or the ability to produce specific experimental alterations through chemical or neurological lesioning or the ability to verify these through autopsy. Most of the work is correlational rather than experimental and based on indirect measures of monoamine activity. The first attempts were simple correlational studies between the metabolites of the monoamines in cerebrospinal fluid (CSF), blood, and urine. Those from CSF were considered better indicators of brain levels of activity although the necessity of doing a lumbar puncture to obtain them raised questions about the role of stress. The invasiveness of the procedure precludes doing repeat measurements after some experimental treatment. Ballenger et al. (1983) found significant negative correlations between CSF norepinephrine (NE) and plasma dopamine-beta-hydroxlase (DBH) and the SSS General scale in normal controls. DBH is the enzyme that converts dopamine to NE within the NE neuron and CSF NE was correlated positively with plasma DBH. Two subsequent studies replicated the DBH-SSS correlation, but three others actually found positive rather than negative correlations and one found no relationship between DBH and the SSS. The metabolites of dopamine (HVA) and serotonin (5-HIAA) were not correlated with the SSS. A study by Limson et al. (1991) found no significant correlations between NE or any other monoamine metabolites in

The Neurobiology of Impulsive Sensation Seeking

45

CSF and any of the novelty seeking, harm-avoidance, or reward dependence scales of the TPQ. Another approach is to test the effects of a neurotransmitter agonist as a function of personality traits. Hutchison, Wood, and Swift (1999) gave d-amphetamine on one occasion and a placebo on another to normal subjects. The TPQ and SSS were used to measure sensation or novelty seeking. Subjective effects of the treatments were measured using adjective rating scales. The drug had a main effect compared to the placebo on self-reported stimulation, elation, vigor, and positive affect. The stimulation effect showed an interaction with both Novelty Seeking and Disinhibition (Dis) subscales with those scoring high on these scales showing more subjective arousal. The Dis subscale also interacted with the drug effect on elation and positive affect. Other SSS subscales also showed similar interactions with the drug. The increased response of high sensation seekers to amphetamine could indicate a greater dopaminergic reaction since dopamine, not NE, mediates the reward effects in the mesolimbic dopaminergic system in rats. A similar finding was obtained with nicotine administered by a nasal spray in non-smokers (Perkins, Gerlack, Broge, Grobe, & Wilson, 2000). Subjects who scored high on the Dis and ES subscales of the SSS reported a stronger head rush, vigor, and arousal, even at lower doses of nicotine. Like amphetamine, nicotine stimulates dopamine release. Taken together the two studies indicate a greater dopaminergic reactivity in high sensation seekers which may explain their vulnerability to drug abuse. Even if low sensation seekers try drugs they are less likely to experience the kind of arousal that “hooks” the highs. Netter, Hennig, and Roed (1996) found that a dopamine agonist increased the craving for nicotine more in high sensation seekers than in lows. A third line of research is to use drugs that stimulate (agonists) or inhibit (antagonists) activity in a particular neurotransmitter system and measure their effects on the neurotransmitter itself, using hormonal indices of activation. Netter et al. (1996) and Depue (1995) both found that a prolactin index of serontergic response to a serotonin agonist was negatively related to sensation seeking and impulsivity scales, and to the EPQ Psychoticism scale indicating a lack of responsivity of this system or subsensitivity of serotonin receptors. Another study found that disinhibited sensation seeking and impulsivity were related to a response indicating low serotonergic responsivity to a serotonergic antagonist (Hennig et al., 1998). The low reactivity or sensitivity of the serotonergic system in impulsive sensation seeking and psychoticism (psychopathy) traits are consistent with the finding of low serotonin activity in personality disorders characterized by impulsivity and aggression directed outward (assault, homicide) or inward (suicide) (Coccaro & Siever, 1995; Mann, 1995). Clinical studies have naturally concentrated on impulsive disorders, but the findings on personality suggest that lowered serotonin is associated with disinihibition and impulsivity in the normal range of personality.

46

marvin zuckerman

These studies did not find an association of a dopamine-induced decrease of prolactin with sensation seeking or impulsivity scores, although Depue did find the association with the EPQ P scale suggesting high dopaminergic reactivity in high P scorers. Contrariwise, the Eysenck Venturesomeness and Risk-Taking scales correlated positively with prolactin response indicating low reactivity (dopamine inhibits prolactin response).

NEUROLOGY Hardly any work has been done using brain imaging to study normal personality traits and what has been done used small samples with comparisons of many brain areas increasing the likelihood of Type I error (Haier, Sokolski, Katz, & Buchsbaum, 1987). Most of the neurological models of sensation seeking point to the mesolimbic dopaminergic system beginning in the ventral tegmental area and projecting to the nucleus accumbens, other limbic areas through the lateral hypothalamus and terminating in the lateral and medial prefrontal cortex. This has been called a “reward system” because when electrodes are planted in many of these areas and connected to keys that allow the animal to self-stimulate the brain the subjects show high and persistent rates of self-stimulation. These are also areas, particularly the nucleus accumbens and ventral tegmentum, where drugs have their powerful rewarding effects. The mesolimbic dopamine system has been suggested as a prime brain substrate for personality traits related to approach behavior (Depue & Iacono, 1989; Depue et al., 1994; Gray, 1987; LeMoal, 1995; Zuckerman, 1983, 1995). Studies of rats have shown that those animals who show novelty seeking as a trait are also willing to ingest alcohol and self-administer drugs like amphetamine (Bardo, Donohew, & Harrington, 1996; Dellu et al., 1996; Fern´andez-Teruel, et al., 2002). This connection models the one between sensation or novelty seeking and drug use in humans. Exposure to novelty in rats stimulates release of dopamine in the nucleus accumbens and injection of a dopamine antagonist into that area blocks novelty seeking activity, independent of its effect on motor activity (Bardo et al., 1996). Higher basal levels of DOPAC (a dopamine metabolite) were found in the brains of high novelty responding rats compared to low responders. High responders also showed lower basal levels of the serotonin metabolite 5-HIAA in the same brain areas indicating an interaction of dopamine and serotonin in sensation seeking as shown in Figure 3. These results from comparative studies argue against Cloninger’s (1993) hypothesis that novelty seekers have low basal levels of dopamine and therefore seek novel activities and stimuli which increase dopaminergic reactivity. At the human level those who suffer from Parkinson’s disease (PD) are lower on Cloninger’s Novelty Seeking scale than controls suffering from rheumatoid and

The Neurobiology of Impulsive Sensation Seeking

47

other orthopedically impairing disorders (Menza, Golbe, Cody, & Forman, 1993). The groups did differ on a special geriatric depression scale but this scale was not correlated with the novelty seeking one. In PD there are decreases of up to 75% of the dopamine in ventral-tegmental and nigrostriatal neuronal systems. The latter system is what causes the motor deficits in PD patients. It could be argued that it is the motor impairment that leads to a secondary loss of interest in the environment in general. This is why the authors used orthopedically impaired controls. But it is also possible that apart from motor impairment the nigrostriatal system is directly involved in the desire to initiate novelty seeking activities. It is interesting that in Haier et al.’s PET study of personality the ES and Dis subscales of the SSS both correlated positively with glucose uptake in the caudate and putamen parts of the striatal dopamine system. The problem is that this finding was in a group of patients with generalized anxiety disorder. The control group was too small to expect much in the way of significant correlational results.

CONCLUSIONS Over the years since the development of the first sensation seeking scale (Zuckerman, Kolin, Price, & Zoob, 1964) investigators have explored this trait at all levels of psychobiology from the trait and its behavioral expressions to the genes that shape the biological systems involved in the trait. Comparative studies in animals and correlational studies in psychopathology and normal personality variations in humans have shown connections between levels such as those between psychophysiology, behavior, and psychopharmacology. Many of the pieces of the total picture have been fitted into the frame of the total psychobiological picture for the trait. Many gaps remain and some pieces do not seem to fit anywhere. Some might object to studying personality at any level but the social and using any species besides the human, misconstruing the purpose as reductionism. However as Crick (1988) comments: “In nature hybrid species are usually sterile, but in science the reverse is often true. Hybrid subjects are often astonishingly fertile, whereas if a scientific discipline remains too pure it usually wilts.” (p. 150).

As long as we regarded the trait as a mere predictor of social behavior its scientific significance was limited by the modest criterion validity coefficients (Mischel, 1968). However as the construct validity of personality traits is found to be rooted in genes and biological systems their function as scientific constructs is enhanced. The goal is not to exclude life experiences as explanations but to see how they intertwine with the biological stakes.

48

marvin zuckerman

REFERENCES Adolfsson, R., Gottfries, C. G., Oreland, L., Roos, B. E., & Winblad, B. (1978). Monoamine oxidase activity and serotogenic turnover in human brain. Progress in Neuropsychopharmacology, 2, 225–230. Alm, P. O., Alm, M., Humble, K., Leppert, J., S¨orensen, S., Lidberg, L., & Oreland, L. (1994) Criminality and platelet monoamine oxidase activity in former juvenile delinquents as adults. Acta Psychiatrica Scandinavica, 89, 41–45. Aluja, A., & Torrubia, R. (in press). Hostility-aggressiveness, sensation seeking, and sexual hormones in men: reexploring their relationship. Neuropsychobiology. Azmitia, E. C., & Whitaker-Azmita (1995). Anatomy cell biology and plasticity of the serotonergic system: Neuropsychopharmacological implications for the actions of psychotropic drugs. In F. E. Bloom & D. J. Kupfer (Eds.) Psychopharmacology: The fourth generation of progress (pp. 443–449). New York: Raven Press. Ballenger, J. C., Post, R. M., Jimerson, D. C., Lake, C. R., Murphy, D. L., Zuckerman, M., & Cronin, C. (1983). Biochemical correlates of personality traits in normals: An exploratory study. Personality and Individual Differences, 4, 615–625. Bardo, M. T., Donohew, R. L., & Harrington, N. G. (1996). Psychobiology of novelty seeking and drug seeking behavior. Behavioral Brain Research, 77, 23–43. Baron, M. (1998). Mapping genes for personality: Is the saga sagging? Molecular Psychiatry, 3, 106– 108. Barratt, E. S., Pritchard, W. S., Faulk, D. M., & Brandt, M. E. (1987). The relationship between impulsiveness subtraits, trait anxiety and visual N100-augmenting-reducing: A topographic analysis. Personality and Individual Differences, 8, 43–51. Bogaert, A. F., & Fisher, W. A. (1995). Predictors of university men’s number of sexual partners. Journal of Sex Research, 32, 119–130. Boomsma, D. I., de Geus, E. J. C., van Baal, GCM, & Koopmans, J. R. (1999). A religious upbringing reduces the influence of genetic factors on disinhibition: evidence for interaction between genotype and environment on personality. Twin Research, 2, 115–125. Bouchard, T. J. Jr. (1994). Genes, environment and personality. Science, 264, 1700–1701. Buchsbaum, M. S. (1971). Neural events and the psychophysical law. Science, 172, 502. Carrilo-de-la-Pena, M. T., & Barratt, E. S. (1993). Impulsivity and the ERP augmenting/reducing. Personality and Individual Differences, 15, 25–32. Caspi, A., McClay, J., Moffitt, T. E., Mill, J., Martin, J., Craig, T. W., Taylor, A., & Poulton, R. (2002). Role of genotype in the cycle of violence in maltreated children. Science, 297, 851–854. Cloninger, C. R. (1987). A systematic method for clinical description and classification of personality variants. Archives of General Psychiatry, 44, 573–588. Cloninger, C. R., Svrakinc, D. M., Przybeck, T. R. (1993). A psychobiological model of temperament and character. Archives of General Psychiatry, 50, 975–990. Coccaro, E. F. & Siever, L. J. (1995). The neuropsychopharmacology of personality disorders. In F. E. Bloom & D. J. Kupfer (Eds.) Psychopharmacology: The fourth generation of progress (pp. 1567–1579). New York: Raven Press. Comings, D. E., Saucier, G., & MacMurray, J. P. (2002). Role of DRD2 and other dopamine genes. In J. Benjamin, R. P. Ebstein, & R. H. Belmaker (Eds.) Molecular genetics and the human personality (pp. 165–191). Washington, DC: American Psychiatric Publishing. Coursey, R. D., Buchsbaum, M. S., & Murphy, D. L. (1979). Platelet MAO activity and evoked potentials in the identification of subjects biologically at risk for psychiatric disorders. British Journal of Psychiatry, 134, 372–381. Crick, F. (1988) What mad pursuit: A personal view of scientific discovery. New York: Basic Books. Dabbs, J. M. (2000) Heroes, rogues and lovers. New York: McGraw-Hill.

The Neurobiology of Impulsive Sensation Seeking

49

Daitzman, R. J., & Zuckerman, M. (1980) Disinhibitory sensation seeking, personality and gonadal hormones. Personality and Individual Differences, 1, 103–110. Dellu, F., Piazza, P. V., Mayo, W., Le Moal, M., & Simon, H. (1996) Novelty-seeking in rats- Biobehavioral characteristics and possible relationship with the sensation-seeking trait in man. Neuropsychobiology, 34, 136–145. Depue. R. A. (1995). Neurobiological factors in personality and depression. European Journal of Personality, 9, 413–439. Depue, R. A., & Iacono, W. G. (1989). Neurobehavioral aspects of affective disorders. Annual Review of Psychology, 40, 457–492. Depue, R. A., Luciana, M., Arbisi, P., Collins, P., & Leon, A. (1994). Dopamine and the structure of personality: Relationship of agonist induced dopamine activity to positive emotionality. Journal of Personality and Social Psychology, 67, 485–498. Ebstein, R. P., & Auerbach, J. G. (2002). Dopamine D4 receptor and serotonin transporter promoter polymorphisms and temperament in early childhoood. In J. Benjamine, R. P. Ebstein, & R. H. Belmaker (Eds.) Molecular genetics and the human personality (pp. 137–149). Washington, DC: American Psychiatric Publishing. Ebstein, R. P. & Kotler, M. (2002). Personality, substance abuse, and genes. In J. Benjamin, R. P. Ebstein, & R. H. Belmaker (Eds.) Molecular genetics and the human personality. (pp. 151–163). Washington, DC: American Psychiatric Publishing. Ebstein, R. P., Levine, J., Geller, V., Auerback, J., Gritsenko, I., & Belmaker, R. H. (1998). Dopamine D4 receptor and serotonin transporter promoter in the determination of neonatal temperament. Molecular Psychiatry, 3, 238–246. Ebstein, R. P., Novick, O., Umansky, R., Priel, B., Osher, Y., Blaine, D., Bennett, E. R., Nemanov, L., Katz, M., & Belmaker, R. H. (1996). Dopamine D4 receptor (D4DR) exon III polymorphism associated with the human personality trait of novelty seeking. Nature Genetics, 12, 78–80. Emmons, T. D., & Webb, W. W. (1974). Subjective correlates of emotional responsivity and stimulation seeking in psychopaths, normals, and acting-out neurotics. Journal of Consulting and Clinical Psychology, 42, 620–625. Eysenck, H. J. (1983). A biometrical-genetical analysis of impulsive and sensation seeking behavior. In M. Zuckerman (Ed.) Biological bases of sensation seeking, impulsivity and anxiety. Hillsdale, NJ: Erlbaum. Fernandez-Teruel, A., Driscoll, P., Gil, L., Arguilar, R., Tobena, A., & Escorihuela, R. M. (2002). Enduring effects of environmental enrichment on novelty seeking, saccharin and ethanol intake in two rat lines (RHA/Verh and RLA/Verh) differing in incentive seeking behavior. Pharmacology, Biochemistry and Behavior, 73, 225–231. Fulker, D. W., Eysenck, S. B. G., & Zuckerman, M. (1980). The genetics of sensation seeking. Journal of Personality Research, 14, 261–281. Gray, J. A. (1987). The neuropsychology of emotion and personality. In S. M. Stahl, S. D. Iverson, & E. C. Goodman (Eds.) Cognitive Neurochemistry (pp. 171–190). Oxford, UK: Oxford University Press. Haier, R. J., Buchsbaum, M. S., Murphy, D. L., Gottesman, I. I., & Coursey, R. D. Psychiatric vulnerability, monoamine oxidase and the average evoked potential. Archives of General Psychiatry, 37, 340–345. Haier, R. J., Sokolski, K., Katz, M., & Buchsbaum, M. S. (1987). The study of personality with positron emission tomography. In J. Strelau & H. J. Eysenck (Eds.) Personality dimensions and arousal (pp. 251–267). New York: Plenum. Hall, R. A., Rappaport, M., Hopkins, H. K., de Griffin, R. B., & Silverman, J. (1970). Evoked response and behavior in cats. Science, 170, 998–1000. Hennig, J., Kroeger, A., Meyer, B., Prochaska, H., Krien, P., Huwe, S., & Netter, P. (1998). Personality correlates of +/− Pinodol induced decreases in prolactin. Psychopharmacopsychiatry, 31, 19–24.

50

marvin zuckerman

Horvath, P., & Zuckerman, M. (1993). Sensation seeking, risk appraisal, and risky behavior. Personality and Individual Differences, 14, 41–52. Hur, Y., & Bouchard, T. J. Jr. (1997). The genetic correlations between impulsivity and sensation seeking traits. Behavior Genetics, 27, 455–463. Hutchison, K. E., Wood, M. D., & Swift, R. (1999). Personality factors and psychophysiological responses to d-amphetamine in humans. Experimental and Clinical Psychopharmacology, 7, 493–501. Koopmans, J. R., Boomsa, D. I., Heath, A. C., & Lorenz, J. P. D. (1995). A multivariate genetic analysis of sensation seeking. Behavior Genetics, 25, 349–356. Leckman, J. F., Gershon, E. S., Nichols, A. S., & Murphy, D. L. (1977). Reduced MAO activity in first degree relatives of individuals with bipolar affective disorders. Archives of General Psychiatry, 34, 601–606. Le Moal, M. (1995). Mesocorticolimbic dopaminergic neurons: Functional and regulatory roles. In F. E. Bloom & D. J. Kupfer (Eds.) Psychopharmacology: The fourth generation of progress (pp. 283–294). New York: Raven Press. Lidberg, L., Levander, S. E., Schalling, D., & Lidberg, Y. (1978). Urinary catecholamines, stress, and psychopathy: A study of arrested men awaiting trial. Psychosomatic Medicine, 40, 116–125. Lidberg, L., Modlin, I., Oreland, L., Tuck, J. R., & Gillner, A. (1985). Platelet monoamine oxidase and psychopathy. Psychiatry Research, 16, 339–343. Limson, R., Goldman, D., Roy, A., Lamparski, D., Ravitz, B., Adinoff, B., & Linnoila, M. (1991). Personality and cerebrospinal monoamine metabolites in alchoholics and normals. Archives of General Psychiatry, 48, 437–441. Loehlin, J. C. (1992). Genes and environment in personality development. Newbury Park, CA: Sage Publications. Lukas, J. H., & Siegel, J. (1977). Cortical mechanisms that augment or reduce evoked potentials in cats. Science, 196, 73–75. Magnusson, D. (1987). Individual development in an interactional perspective. In D. Magnuss (Ed.) Paths through life. Hillsdale, NJ: Erlbaum. Magnusson, D. (1996). The patterning of antisocial behavior and autonomic reactivity. In D. M. Stoff, & R. B. Cairns (Eds.) Aggression and violence. Genetic neurobiological, and biosocial perspectives (pp. 291–308). Mahwah, NJ: Erlbaum Major, L. F. & Murphy, D. L. (1978). Platelet and plasma amine oxidase activity in alcoholic individuals. British Journal of Psychiatry, 132, 548–554. Mann, J. J. (1995). Violence and aggression. In F. E. Bloom & D. J. Kupfer (Eds.) Psychopharmacology: The fourth generation of progress (pp. 1919–1928). New York: Raven Press. Menza, M. A., Golbe, L. I., Cody, R. A., & Forman, N. E. (1993). Dopamine-related personality traits in Parkinson’s disease. Neurology, 43, 505–508. Mischel, W. (1968). Personality and assessment. London: Wiley. Murphy, D. L., Aulakh, C. S., Garrick, N. A., & Sunderland, T. (1987). Monoamine oxidase inhibitors as antidepressants: Implications for the mechanism of action of antidepressants and the psychobiology of the affective disorders and some related disorders. In H. Y. Meltzer (Ed.) Psychopharmacology: The third generation of progress (pp. 545–552). New York: Raven Press. Murphy, D. L., & Weiss, R. (1972). Reduced monoamine oxidase activity in blood platelets from bipolar depressed patients. American Journal of Psychiatry, 128, 1351–1357. Netter, P., Hennig, J., & Roed, I. S. (1996). Serotonin and dopamine as mediators of sensation seeking behavior. Neuropsychobiology, 34, 155–165. O’Carroll, R. E. (1984). Androgen administration to hypogonadal and eugonadal men: Effects on measures of sensation seeking, personality and spatial ability. Personality and Individual Differences, 5, 595–598.

The Neurobiology of Impulsive Sensation Seeking

51

Olweus, D. (1987). Testosterone and adrenaline: Aggressive antisocial behavior in normal adolescent males. In S. A. Mednick, T. E. Moffitt, & S. A. Stack (Eds.) The causes of crime: New biological approaches (pp. 263–282). Cambridge, UK: Cambridge University Press. Pandy, G. N., Fawcett, J., Gibbons, R., Clark, D. C., & Davis, J. M. (1988). Platelet monoamine oxidase in alcoholism. Biological Psychiatry, 24, 15–24. Perkins, K. A., Gerlach, D., Broge, M., Grobe, J. E., & Wilson, A. (2000). Greater sensitivity to subjective effects of nicotine in nonsmokers high in sensation seeking. Experimental and Clinical Psychopharmacology, 8, 462–471. Plomin, R., DeFries, J. C., McClearn, G. E., & Rutter, M. (1997). Behavioral Genetics (3rd ed.) New York: W. H. Freeman & Co. Prolo, P., & Licinio, J. (2002). DRD4 and novelty seeking. In J. Benjamin, R. P. Ebstein, & R. H. Belmaker (Eds.) Molecular genetics and the human personality (pp. 91–107). Washington, DC: American Psychiatric Publishing. Redmond, D. E. Jr., Murphy, D. L., & Baulu, J. (1979). Platelet monoamine oxidase activity correlates with social affiliative and agonistic behaviors in normal rhesus monkeys. Psychosomatic Medicine, 41, 87–100. Reist, C., Haier, R. J., De Met, E., & Cicz-De Met, A. (1990). Platelet MAO activity in personality disorders and normal controls. Psychiatry Research, 30, 221–227. Saxton, P. M., Siegel, J., & Lukas, J. H. (1987). Visual evoked potential augmenting/reducing slopes in cats-2. Correlations with behavior. Personality and Individual Differences, 8, 511–519. Schatzberg, A. F., & Schildkraut, J. J. (1995). Recent studies on norepinephrine systems in mood disorders. In P. E. Bloom & R. J. Kupfer (Eds.) Psychopharmacology: The fourth generation of progress (pp. 911–920). New York: Raven Press. Schukit, M. A. (1994). Familial alcoholism. In T. A. Widiger, A. J. Francis, H. A. Pincus, M. B. First, R. Ross, & W. Davis (Eds.) DSM IV Sourcebook, Vol. I (pp. 159–167) Washington, DC: American Psychiatric Association. Shekim, W. O., Bylund, D. B., Alexson, J., Glaser, R. D., Jones, S. B., Hodges, K., & Perdue, S. (1986). Platelet MAO and measures of attention and impulsivity in boys with attention deficit disorder and hyperactivity. Psychiatry Research, 18, 179–188. Sher, K. J. (1993). Children of alcoholics and the intergenerational transmission of alcoholism: A biopsychosocial perspective. In J. S. Baer, A. Marlatt, & R. J. McMahon (Eds.) Addictive behaviors across the life span (pp. 3–33). Newbury Park, NJ: Sage Publications. Siegel, J., Sisson, D. F., & Driscoll, P. (1993). Augmenting and reducing of visual evoked potentials in Roman high- and low-avoidance rats. Physiology and Behavior, 54, 707–711. Smith, B. D., Davidson, R. A., Smith, D. C., Goldstein, H., & Perlstein, W. (1989). Sensation seeking and arousal: Effects of strong stimulation on electrodermal activation and memory task performance. Personality and Individual Differences, 10, 671–679. Smith, B. D., Perlstein, W. M., Davidson, R. A., & Michael, K. (1986). Sensation seeking: Differential effects of relevant novel stimulation on electrodermal activity. Personality and Individual Differences, 4, 445–452. Sostek, A. J., Sostek, A. M., Murphy, D. L., Martin, E. B., & Born, W. S. (1981). Cord blood amine oxidase activities relate to arousal and motor functioning in human newborns. Life Sciences, 28, 2561–2568. Strobel, A., Lesch, K. P., & Brocke, B. (2003). Dopamine D4 receptor gene polymorphisms and novelty seeking: Evidence for a modulatory role of additonal polymorphisms. Paper presented at the 11th biennial meeting of the International Society for the Study of Individual Differences, Graz, Austria. Virkkunen, M. (1985). Urinary free cortisol secretion in habitually violent offenders. Acta Psychiatrica Scandinavica, 72, 40–44.

52

marvin zuckerman

Virkkunen, M., Rawlings, R., Tokola, R., Poland, R. E., Guidotti, A., Nemeroff, C., & Bissette, G. (1994). CSF biochemistry, glucose metabolism, and diurnal activity in alcoholic, violent offenders, fire setters, and healthy volunteers. Archives of General Psychiatry, 51, 20–27. Von Knorring, L., Oreland, L., & von Knorring, A. L. (1987). Personality traits and platelet MAO activity in alcohol and drug abusing teenage boys. Acta Psychiatrica Scandinavica, 75, 307– 314. Von Knorring, L., & Perris, C. (1981). Biochemistry of the augmenting reducing response in visual evoked potentials. Neuropsychobiology, 7, 1–8. Zuckerman, M. (1971). Dimensions of sensation seeking. Journal of Consulting and Clinical Psychology, 36, 45–52. Zuckerman, M. (1979). Sensation seeking and risk-taking. In C. E. Izard (Ed.) Emotions in personality and psychopathology (pp. 163–197). New York: Plenum Zuckerman, M. (1983). A biological theory of sensation seeking. In M. Zuckerman (Ed.) Biological bases of sensation seeking, impulsivity, and anxiety (pp. 37–76). Hillsdale, NJ: Erlbaum. Zuckerman, M. (1984). Sensation seeking: A comparative approach to a human trait. Behavioral and Brain Sciences, 7, 413–471. Zuckerman, M. (1990). The psychophysiology of sensation seeking. Journal of Personality, 58, 313– 345. Zuckerman, M. (1991). Psychobiology of personality. Cambridge, UK: Cambridge University Press. Zuckerman, M. (1993). Personality from top (traits) to bottom (genetics) with stops at each level between. In J. Hettema & I. J. Deary (Eds.) Foundations of Personality (pp. 73–100). Dordrecht, Netherlands: Kluwer. Zuckerman, M. (1994). Behavioral expressions and biosocial bases of sensation seeking. New York: Cambridge University Press. Zuckerman, M. (1995). Good and bad humors: Biochemical bases of personality and its disorders. Psychological Science, 6, 325–332. Zuckerman, M. (2002). Zuckerman-Kuhlman Personality Questionnaire (ZKPQ): An alternative fivefactor model. In B. DeRaad & M. Perugini (Eds.) Big five assessment (pp. 377–396). Seattle, WA: Hogrefe & Huber Publishers. Zuckerman, M. (2003). Biological bases of personality. In T. Millon & M. J. Lerner (Eds.) Handbook of psychology, Vol. 5: personality and social psychology (pp. 85–116). Hoboken, NJ: Wiley. Zuckerman, M., & Cloninger, C. R. (1996). Relationships between Cloninger’s, Zuckerman’s, and Eysenck’s dimensions of personality. Personality and Individual Differences, 21, 283–285. Zuckerman, M., Kolin, E. A., Price, L., & Zoob, I. (1964). Development of a sensation seeking scale. Journal of Consulting Psychology, 28, 477–482. Zuckerman, M., Kuhlman, D. M., Joireman, J., Teta, P., & Kraft, M. (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. Zuckerman, M., Murtaugh, T.T., & Siegel, J. (1974). Sensation seeking and cortical augmentingreducing. Psychophysiology, 11, 535–542. Zuckerman, M., Simons, R. F., & Como, P. G. (1988). Sensation seeking and stimulus intensity as modulators of cortical, cardiovascular, and electordermal response: A cross modality study. Personality and Individual Differences, 9, 361–372. Zureik, J. L., & Meltzer, H. Y. (1988). Platelet MAO activity in hallucinating and paranoid schizophrenics. A review and meta-analysis. Biological Psychiatry, 24, 63–78.

3 Neurobiology of Creativity David Camfield

The following chapter will review the current state of research into the relationship between creativity and neural processes. The review will begin with an overview of current conceptual formulations of creativity as a cognitive ability and the proposed role of inhibition in creative thought. The focus will then shift to more recent research on patterns of electrophysiological activity in the brain that have been found to be associated with creativity. Studies concerning EEG frequency, EEG coherence, EEG complexity and P300 evoked potentials will be discussed. The use of positron emmision tomography (PET) and magnetic resonance imaging (MRI) in the study of creativity will then be reviewed, with special emphasis on studies of verbal fluency and enhanced creativity in frontotemporal dementia. The review will finish with a discussion of hemispheric asymmetry in creativity, the comparison of data from a range of imaging modalities, and a discussion of the role of the personality trait Openness to Experience in the understanding of creativity.

THE MEASUREMENT OF CREATIVE COGNITION There has been general consensus over the last 30 years in defining creative thought or behaviour as that which is both novel-original and useful-adaptive (Feist, 1998). In order for a thought or behaviour to be considered creative, it is not enough for it to be only original, it must also fulfil the additional criteria of being socially useful or adaptive. An important distinction must also be made between Trait David Camfield Australia.

•

Centre for Neuropsychology, Swinburne University of Technology, Melbourne,

53

54

david camfield

creativity; the latent trait underlying creative behaviour, and Achievement creativity; socially useful/acceptable products resulting from trait creativity, intelligence and other components (Eysenck, 1995). While trait creativity describes an ability inherent to the individual, the manifestation of creative achievement is dependent on an interaction between creative ability and many other factors, both internal and external (e.g. cognitive, environmental and motivational). Trait creativity is a necessary, but not sufficient, condition for creative achievement (Eysenck, 1993). For this reason it is important to be able to measure trait creativity independently of creative achievement. One approach to measuring trait creativity has been through the use of cognitive ability tests. The use of cognitive ability tests in the measurement of creativity can be traced back to the pioneering work of Hargreaves (1927). Through analysis of a variety of cognitive test responses, Hargreaves (1927) found that the responses from tests requiring imaginative answers formed a factor that was independent of intelligence. Numerous tests of creative ability have been formulated since then, with a general consensus that creative ability can best be operationalised as divergent thinking (Guilford, 1967a). Divergent thinking tests provide a measure of the novel-original aspect of creativity. In a typical divergent task participants are asked to provide as many appropriate answers as possible in response to a particular stimulus (e.g. word, sentence, picture, situation). These tests generally measure one of two things: (i) fluency; the quantity of associations that can be produced in response to a stimulus and (ii) originality; original and unusual associations in response to a stimulus. The basis of divergent thinking tests is the theory that creativity entails the formation of associative links between otherwise unrelated concepts. Mednick (1962), a key proponent of the associationist view of creativity, defined the creative process as “the forming of associative elements into new combinations which either meet specified requirements or are in some way useful. The more mutually remote the elements of the new combinations, the more creative the process or solution.” (Mednick, 1962, p. 221). Eysenck (1993) has used the term “associative hierarchy” to describe the probability an individual has of producing unique associations in comparison to common ones. An individual with a steep gradient to their associative hierarchy is much more likely to give common associations compared to uncommon associations. However, an individual with a flat gradient will make uncommon associations with equal frequency to common associations. A related concept is that of associative horizons, which refers to the extent with which a stimulus word, for example ‘shoe’, is associated with other words (Eysenck, 1993). Responses with words directly related such as ‘foot’ would reflect a narrow horizon. Words such as ‘hand’ or ‘toe’ would reflect a slightly wider horizon. However, words such as ‘soldier’ or ‘snow’ would indicate a wide associative horizon, also termed ‘over-inclusiveness’ by some authors (Cameron, 1938; Cameron, 1947).

Neurobiology of Creativity

55

CREATIVITY AND DISINHIBITION Eysenck (1993; 1995) proposed that the cognitive style of over-inclusive thinking lies at the basis of creativity, and was also related to the disordered thought processes observed in Schizophrenia. In normal cognitive functioning, stored memories of past experience are used to guide current perception, thus reducing information load. However, in schizophrenia it is proposed that the influence of stored memories over current perception is weakened. Expressed differently, there is a weakening of inhibitory processes that would otherwise narrow down the response options (Eysenck, 1993). According to Eysenck’s (1993) model of creativity, over-inclusive thinking is the result of a similar failure to filter/inhibit irrelevant stimuli. This theory has been tested using experimental designs that provide a measure of the strength of inhibitory processes within the individual. The two most common designs are negative priming and latent inhibition. In negative priming, a distractor stimulus is presented to subjects, in a context in which it is irrelevant to successful task completion. Subsequently, the same distractor is re-presented as a target object that must be dealt with. The common finding is that reacting to a stimulus that had previously been inhibited takes longer than when there had not been any prior exposure to it. The Stroop colour-naming task is a typical example of the negative priming design. A colour word (e.g. red) may be presented in green ink, with the aim of the task being to disregard the word and name the colour of the ink. If the next task consists of a word printed in red ink, the response of a normal subject will be significantly slower, because the previously ignored word red and associations with it have now acquired negative salience. However, when schizophrenic or schizotypal subjects are tested in this manner, negative priming fails to significantly alter their response times (Beech, Powell, McWilliam, & Claridge, 1989; Beech, Baylis, Smithson, & Claridge, 1989). As mentioned above, latent inhibition is another technique that has also been used to measure the strength of inhibitory processes. Latent inhibition refers to the finding that non-reinforced pre-exposure to a stimulus will reduce the ability of that stimulus to act as a conditioned stimulus in subsequent conditioning procedures (e.g. classical or instrumental). When pre-exposure fails to bring about inhibition in subsequent conditioning, this is indicative of attentional deficits. Numerous studies have documented a lack of latent inhibition in both schizophrenics and individuals scoring high in Psychoticism (Eysenck, 1995). Lubow (1989) has explained the lack of latent inhibition in shizophrenia as being due to over activity of the mesolimbic-dopamine system. He demonstrated this by studies showing amphetamine (a dopamine agonist) would decrease inhibition, while haloperidol (a dopamine antagonist) would increase inhibition. Similarly, Lubow (1989) demonstrated that serotonin depletion also decreases latent inhibition and points to the involvement of the hippocampus in the process. These findings raise the possibility that perhaps creativity may be enhanced by increased levels of dopamine or

56

david camfield

decreased levels of serotonin. It is important to note that neither negative priming or latent inhibition have been studied directly in relation to creativity. However, the theory of disinhibition proposed by Eysenck has gained indirect evidence due to the link between psychoticism and creativity. Further, this theory is appealing as it provides a precise experimental procedure with which a direct link can be tested.

CREATIVITY AND EEG FREQUENCY Whitton, Moldofsky and Lue (1978) investigated EEG power spectral patterns associated with both hallucinatory behaviour in schizophrenic patients and creative responses in normals. A selection of Guilford’s (1967b) divergent classification and cognition classification tests were used in order to elicit a creative response. These tests were chosen because they brought about closure, that is a sudden experience of perceptual resolution. Average percent spectral power in each EEG band (recorded from Cz) was compared between the conditions of resting, 4-seconds preceding a report of a visual or auditory hallucination (schizophrenic patients), and 4-seconds preceding a creative response. It was found that in both the time preceding a creative response and the time preceding a report of hallucination, spectral intensity in the delta and theta bands increased significantly compared to rest, while intensity in the high beta band decreased significantly. Whitton et al. (1978) suggested that the observed increase in low frequency power may be associated with internally directed attention. These results are intriguing in that they are suggestive of a similarity between the intrusive experience of hallucination in schizophrenia and the sudden perceptual resolution experienced with creative achievement. While no changes in alpha power were found to be associated with creative responses by Whitton et al. (1978), research by Martindale and Hasenfus (1978) suggests that creative individuals display differences in alpha power between stages of the creative process. The participants used for the study were students enrolled in a creative writing class, divided into a creative and a non-creative group on the basis of expert ratings of their work. Participants were given the following creative writing task: “A man meets a woman and asks her out on a date. Make up a story about who the people are, how they met, and what will happen. Use your imagination.” They were given 3 minutes to think about what they would write about (inspiration stage) before being allowed to write for 5 minutes (elaboration stage). EEG was recorded from the right posterior temporal area during both resting, inspiration and elaboration stages. No significant differences in alpha power were found between groups at rest. However, for the creative group, a significantly higher percentage of alpha waves were found to be present during the inspiration stage in comparison to the elaboration stage. Further, for the uncreative group, no differences were found.

Neurobiology of Creativity

57

In a follow up experiment, Martindale and Hasenfus (1978) tested a second group of participants, divided into groups according to their scores on the Remote Associatiates Test (Mednick & Mednick, 1967), a measure of originality, and the Alternate Uses Test (Christensen, Guilford, Merrifield, & Wilson, 1970), a measure of fluency. EEG was recorded from above Wernicke’s area on the left hemisphere during rest, 5 minutes of random speech (equivalent to the inspiration stage) and 5 minutes of fantasy speech (equivalent to the elaboration stage, based on the scenario used previously). Half of the participants were instructed to be as original as possible in both tasks, while the other half were not. High scorers on the remote associates test were found to exhibit less alpha activity across all conditions compared to low scorers. Contrary to the findings of the first experiment, percentage alpha levels were found to be generally lower during the inspiration compared to the elaboration stage. However, neither the tasks nor the EEG measurements were strictly equivalent to those used in the earlier experiment. Despite these differences, it is interesting to note that the highest percentage of alpha waves in the inspiration task were found for the group scoring high on the Alternate uses test, with instructions to be original. Further in this condition percentage alpha levels were found to be similar across inspiration and elaboration stages. These findings suggest that the instruction to be original has an important impact on the resulting percentage alpha in creative tasks.

CREATIVITY AND EEG COHERENCE For an investigation into a higher cognitive process such as creativity, electroencephalogram (EEG) coherence analysis is an electrophysiological technique that holds great promise. In comparison to other techniques such as power spectrum analysis, a study of coherence reveals not only the level of activity of various brain regions but also the functional relationships between different areas in the frequency domain (Petsche, 1996). Increases in coherence are generally interpreted as an increase in the cooperation of respective brain regions, while decreases in coherence are interpreted as evidence of functional separation/de-coupling. In this way the pattern of activation across the entire cortex may be analysed for both spatial and temporal patterns that may be associated with creativity. At a conceptual level, coherence analysis is also appealing, in that it can be directly related to the associationist theory of creativity. Here two levels of association can be observed simultaneously: the level of association between concepts and the level of association between neuronal populations. Research by Orme-Johnson and Haynes (1981) measured EEG coherence in a group of participants undergoing training in meditation. Participants were grouped into a high and low meditation proficiency group on the basis of course logs and ratings by their instructors. The participants were also tested for fluency using the

58

david camfield

Unusual Uses subtest of the Torrance Tests of Creative Thinking (Torrance, 1974). EEG was measured using two bilateral frontal (F3, F4) and two bilateral central (C3, C4) sites during rest with eyes closed, with eyes open and during meditation. Mean alpha coherence between all sites was found to be positively correlated with creativity, as was alpha coherence between the right-hemisphere sites (F4 and C4) and alpha coherence between the bilateral frontal alpha sites (F3 and F4). The relationship between creativity and coherence for the pair of channels with the strongest alpha coherence in each individual (dominant alpha) was also examined. The mean of the frequency bands for dominant alpha was also found to be positively related to creativity. Interestingly, with the exception of bilateral frontal alpha coherence, the same pattern of results was found to differentiate the more proficient from the less proficient meditators. Further, the more proficient group scored significantly higher on creativity than the less proficient group. These findings suggest that proficiency at meditation, or familiarity with a meditative state, may be conducive to creativity. Petsche (1996) measured EEG coherence during completion of both verbal, visual, and musical creative tasks. In the verbal task participants were required to construct a short story out of 10 memorised words. In the visual task, they were required to mentally create a picture, and in the musical task they were required to mentally compose a short piece of music of their own choice. For the verbal task, coherence increases were most noticeable in the low frequency, theta and delta bands, but not the higher frequency bands. In the visual task coherence increases were apparent in all frequency bands (theta, alpha1 and beta 2 particularly) while a functional decoupling was observed between the two frontal lobes, particularly in the lower frequency bands. A similar pattern was observed for the musical composition task, with increases in coherence observed in all frequency bands and a functional decoupling between the frontal lobes. The main finding of the study was that across all three creative tasks increases in intra-hemispheric and inter-hemispheric long-distance coherence was observed in comparison to the EEG at rest. This increase in coherence indicates closer cooperation among disparate brain regions during the execution of a creative task. Research by Thatcher, Krause and Hrybyk (1986) suggests that an increase in coherence between distant brain regions indicates the involvement of type-I pyramidal neurons which are needed to transfer information over long distances. An intriguing finding of the research by Petsche (1996) was that coherence changes in the upper alpha band appeared to reflect individual features in the completion of the creative tasks. Petsche, Kaplan, von Stein, and Filz (1997) further investigated coherence changes in the alpha 1 (7.5–9 Hz) and the alpha 2 (9.5– 12.5 Hz) bands during creative tasks. When asked to mentally create a picture, long distance increases in coherence were observed both ipsi-laterally and contralaterally. These changes were more pronounced in alpha 1 than alpha 2. Interhemispheric changes were most noticeable in alpha 1 between the posterior parts of the

Neurobiology of Creativity

59

brain. Interestingly, when college-trained artists were compared to non-artists no significant coherence differences were observed for this task. Petsche et al. (1997) also examined coherence changes in the alpha 1 and alpha 2 band under the task of musical composition, using professional composers. In this task highly individualised patterns of coherence change were observed between each composer in the alpha bands. In contrast, coherence increases in long distance delta, theta and uppermost beta ranges appeared to reflect more general brain processes while composing music. In contrast to the analysis of coherence during the execution of creative tasks, Jausovec and Jausovec (2000a) examined differences in coherence patterns between people rated high and low in creativity while at rest (counting backwards from 400). Pronounced negative correlations between rated creativity and resting coherence were found in both the alpha and beta frequency bands. These results suggested that highly creative individuals display less cooperation between brain areas in these frequency bands during a resting state when compared to less-creative individuals. In a similar study Jausovec and Jausovec (2000b) examined the relationship between EEG coherence and the creativity required to solve a problem. Solutions to a dialectic open-ended essay-writing problem were found to involve far greater intra-hemispheric and inter-hemispheric cooperation in the alpha band, predominantly between far distant brain regions (frontopolar and parieto-occipital sites), in comparison to a divergent problem with a clearly well-defined means of solution. These findings were again suggestive of a greater involvement of the cortico-cortical fibers in the completion of creative, compared to less-creative, tasks.

CREATIVITY AND EEG COMPLEXITY Studies of complexity in the EEG are based on the premise that the observed variations of the electrical field of the brain are the result of a complex yet deterministic system. This is in contrast to the ‘traditional’ methods of EEG analysis (eg. Spectral analysis) which views these variations as random processes. Molle et al. (1996) hypothesised that divergent thinking would increase the dimensional complexity of the EEG in comparison to tasks involving convergent thinking. Their rationale was that the task of producing as many unique ideas as possible would increase the degree of competition among cortical neuronal cell assemblies, leading to an increase in dimensional complexity. The results of their study lent support to this theory, with dimensional complexity observed to increase in divergent thinking, compared to convergent thinking, the increase being most pronounced over central and parietal cortical areas. However, over the fronto-cortical regions, the EEG complexity was comparable during divergent thinking and mental relaxation, with both of these conditions displaying higher complexity than convergent

60

david camfield

thinking. Molle et al. (1996) interpreted these results to suggest that creative thinking was characterized by a ‘destructuring’ of strong mental associative habits, entailing a spread of activation to assemblies rarely activated. The finding of comparable complexity in frontal areas during relaxation and divergent thought, was explained as a ‘loosening of attentional control’, in comparison to a more focussed attention in the convergent task.

CREATIVITY AND THE P300 EVOKED POTENTIAL The P300 (P3) is an endogenous event-related potential (ERP) that occurs between 300 and 400ms post-stimulus and has traditionally been associated with the detection of stimulus novelty. Research by Barcelo, Perianex and Knight (2002), through experimentation using the Wisconsin card sorting test (WCST; Heaton, Chelune, et al., 1993), has provided evidence to suggest that the P3a may provide a general measure of cognitive flexibility in both stimulus and task novelty. The WCST is a test of prefrontal function where a card is presented and the participant is required to match it to one of four key cards according to a previously specified stimulus dimension (either colour, shape, or number of items in the card). The dimension on which the cards are to be matched (the task rule) is systematically varied, and the ability of the participant to successfully adjust to the changes in task rules provides a measure of cognitive flexibility, presumed to be controlled by the prefrontal cortex. Barcelo et al. (2002) found frontally distributed P3a activity to be elicited by feedback cues signalling a change in task rules. In contrast, feedback cues signalling that the task rules were to remain the same elicited a comparatively sharp reduction in P3a amplitude. These results suggested that the P3a reflected the switching of task sets in working memory. The ability to switch task sets in problem solving has been implicated by Russ (1993) as an important aspect of creativity. The P300 is a measure that can also be used to gauge the use of working memory (WM) during the completion of a task (Donchin & Coles, 1988). Lavric, Forstmeier and Rippon (2000), analysed differences in P300 amplitude between analytical and creative tasks, in order to determine whether there were differences in the use of WM between these tasks. The analytical task was a well-defined deductive reasoning task while the creative task involved the novel use of information in order to arrive at a successful solution. In order to gauge the use of WM during the completion of the tasks, a concurrent WM task (CWMT) was administered in addition to the analytical or creative task. The CWMT consisted of counting auditory stimuli presented at pseudorandom intervals. P3a amplitudes were found to be higher in the frontal region in response to counting tones during analytical problem solving as compared to creative problem solving. Further, the later P3b amplitudes were found to be higher across all regions with a left laterality

Neurobiology of Creativity

61

in frontal, central-parietal and temporal regions in analytical compared to creative problem solving. In relation to the higher frontal amplitudes observed in the P3a during the analytical task, Lavric et al. (2000) suggested that this was due to a stronger involvement of WM in this task in comparison to the creative task (i.e. greater competition for attentional resources between the analytical task and the CWMT). These findings suggest that working memory is not as crucial to creative problem solving as to the successful solution of analytical tasks. Future research concerning differences in P300 amplitudes between individuals rated high on creativity compared to those rated low on creativity could be used to identify stable individual differences in brain function (if any) between these individuals. An intriguing aspect to the literature on P300 is that a growing body of evidence from pharmacological studies now implicate neurotransmitter systems in the modulation of P300 amplitude. Two neurotransmitters that have been found to modulate the expression of the P300 are dopamine (Klorman & Brumaghim, 1991; Callaway, 1991; Hansenne et al., 1995) and serotonin (Hansenne, Pitchot, Papart, & Ansseau, 1998). It is possible that differences in the amplitude of the P300 in creative problem solving may in part reflect a different mode of operation in the brain as compared to analytical problem solving, mediated by different patterns of dopaminergic and serotonergic transmission. However, caution is needed in interpreting the meaning of increased P300 amplitudes as the P300 has also been found to be sensitive to motivation and vigilance factors (Johnson, 1993; Hansenne, 1999).

BRAIN IMAGING AND VERBAL FLUENCY As previously mentioned, verbal fluency (the generation of verbal responses in relation to a stimulus word or letter) is commonly used as a cognitive measure of creativity. A number of studies have used modern brain imaging technology to localise the areas of the brain that are most involved during the execution of verbal fluency ( VF ) tasks. Parks et al. (1988) used PET to measure cerebral metabolic rate for glucose (CMRglc) during the execution of a VF task. The VF task required participants to generate as many words as possible in response to a stimulus letter over a 1-minute time period, with an average obtained after random presentation of varying letters over a 30-minute interval. For means of comparison PET was also recorded from a control group of participants during resting conditions. CMRglc activation was found to be 23.3% higher over the entire cortex during administration of the VF task in comparison to resting values. The greatest amount of cortical activation was found to be focussed within the temporal lobes (27%) and the frontal lobes (25%) bilaterally. However, when the metabolic values were normalised to the occipital lobe, only the increase in temporal lobe activation remained significant. In analysis of the relationship between

62

david camfield

CMRglc during activation and the score obtained on the VF task, CMRglc in both the frontal, temporal and parietal lobes was found to be significantly negatively correlated with VF performance. Parks et al. (1988) suggested that participants with greater verbal fluency required less cognitive effort in order to successfully complete the VF task. Boivin et al. (1992) used PET in order to investigate the relationship between performance on a verbal fluency (VF) task and regional glucose metabolism at rest. The VF task consisted of asking participants to quickly name as many words as they could that started with the letter “d” within a 1-minute time limit. The Weschler Adult Intelligence Scale—Revised ( WAIS-R; Wechsler, 1981) was also administered in order to gain a picture of the relative specificity of brain glucose metabolism in verbal fluency as compared to general intelligence. VF was found to be negatively correlated with relative metabolic rate in the bilateral frontal cortical regions and positively correlated with relative metabolic rate in the left temporal lobe. Using a similar rationale to Parks et al. (1988), Boivin et al. (1992) suggested that the negative correlation in the frontal cortical regions was due to a relative economy of effort and enhanced cognitive efficiency in high VF scorers compared to low VF scorers. It was suggested that the positive correlation obtained between VF performance and glucose metabolism in the left temporal areas reflected a verbal memory component. Interestingly none of the WAIS-R sub-scales were found to correlate with regional resting glucose metabolism measures, indicating that these patterns of activation were specific to verbal fluency performance and unrelated to general intelligence. Phelps, Hyder, Blamire and Shulman (1997) investigated areas of cortical activation during verbal fluency tasks with the use of functional magnetic resonance imaging (fMRI). An advantage of fMRI compared to PET is that the former enables a superior degree of spatial resolution. Participants were given three different tasks to complete: Repeat; where they were required to simply repeat a cue word (e.g. man-MAN), Opposite; where they were required to generate the antonym of the cue word (e.g. hot-COLD), and generate; where they were required to generate eight different words beginning with the letter of the cue word (e.g. R, oneRABBIT). fMRI brain scans were obtained under each of the tasks, and these were used to identify differences in patterns of activation between the generate (verbal fluency) condition and the other (simpler) verbal tasks. The areas found to be significantly more active during the verbal fluency task (generate task compared to the repeat and opposite tasks) were the superior/middle frontal gyrus, the inferior frontal gyrus and the anterior cingulate. It was suggested that greater activation in the superior/middle frontal gyrus was due to the effortful search through words from memory. Greater activation in the anterior cingulate was explained as being due to attentional demands of the task, while greater activation in the inferior frontal gyrus appeared to reflect the difficulty of the task and the production of responses based on word structure. While this study does not relate individual differences in

Neurobiology of Creativity

63

VF performance to these brain areas, it suggests that these are perhaps the specific structures were such differences would be expected to be found.

FRONTOTEMPORAL DEMENTIA AND ENHANCED CREATIVITY Frontotemporal dementia ( FTD) is a rare form of dementia specific to the frontal and temporal lobes, that accounts for as much as 25% of the presenile dementias (Miller et al., 1998). A number of studies have now documented the bizarre finding that in relation to the temporal lobe variant of FTD, enhanced artistic ability often emerges amidst an otherwise progressive cognitive decline. In the temporal lobe variant of FTD, the anterior temporal and basal frontal lobes atrophy slowly while dorsolateral frontal areas remain intact (Miller et al., 1998). Miller, Ponton, Benson, Cummings and Mena (1996) describe a 56-year-old businessman with the temporal lobe variant of FTD who began painting for the first time, despite no previous interest in art. The subjective experience described by the patient was one of heightened visual awareness, whereby lights and sounds were experienced as either intensely painful or as producing a euphoria that enhanced creativity. Even more unusual was the finding that over the next decade, as the disease progressed, he drew with increasing precision and detail, to the extent that he started to win awards at local art shows. However, by the age of 67 his work began to deteriorate, and at the age of 68 a number of brain scans were taken. MRI revealed bitemporal atrophy, and a single photon emission computed tomography (SPECT) revealed bilateral temporal hypoperfusion, although frontal, occipital and parietal perfusion were found to be normal. In further case studies of patients with the temporal variant of FTD, Miller et al. (1998) describe another three individuals with similar stories. All of them developing a interest in visual art during the early stages of the disease, displaying increasing proficiency as the disease progressed, and eventually becoming too incapacitated to continue. Miller et al. (1998) comment further on similarities between the patients, highlighting that all of them i) displayed creativity in visual, rather than verbal, forms, ii) they produced art that was largely reconstructed from memory iii) their works of art were realistic copies rather than abstract of symbolic depictions, iv) they displayed an increased interest in fine detail, v) there was an almost obsessional drive for perfection in their work. As with the first patient examined, a common finding from SPECT analysis during the late stages of the disease was that of bitemporal hypoperfusion. Miller et al. (1998) suggest that the reason the visuospatial abilities were spared in these patients was that the dorsolateral prefrontal cortex is spared in the temporal lobe variant of FTD. Further, they suggested that the selective degeneration of the anterior temporal and orbital frontal cortex caused an enhancement of artistic interests and aptitudes

64

david camfield

due to decreases in the number of inhibitory projections to the visual systems involved with perception. This type of behavioural improvement observed in patients with brain injury has been termed “paradoxical functional facilitation” by (Kapur, 1996). The most accurate picture of the brain areas involved in the emergence of artistic ability associated with FTD was provided by Miller, Boone, Cummings, Reade and Mishkin (2000). In this study clinical, neuropsychological and neuroimaging data from 69 patients with a clinical diagnosis of FTD were compared. The data from patients who displayed emerging artistic skills in FTD were compared to those who did not, in order to isolate the important brain areas involved. Interestingly, enhanced musical as well as visual skills were found in this sample, with creativity displayed in a diverse number of ways including invention, bridge or chess playing, composition, piano playing, photography, painting and crafts. However, as with the previous study it is noteworthy that none of these talents were manifest in a verbal form. Twelve FTD patients (17%) were found to display new or preserved musical or visual ability. Nine of these patients displayed left-sided predominant hypoperfusion using SPECT while only 12 of the 45 patients without artistic ability were left-sided. Further, eight of the twelve patients with ability displayed the temporal lobe variant pattern, while only nine of the 45 patients without artistic ability showed the temporal lobe variant. These differences were found to be significant. In summary, most of the FTD patients who displayed artistic ability were found to have a perfusion deficit specific to the temporal lobes, with asymmetrical left hemisphere degeneration. These findings lent further support to the hypothesis that selective degeneration of the temporal cortex leads to paradoxical functional facilitation of visual and musical systems. Specifically, degradation of the left anterior temporal cortex appears to be particularly involved in the disinhibition of these systems. It is also noteworthy that the motivation to engage in artistic ability was also observed to increase dramatically in these patients, which Miller et al. (2000) suggest is due to the reduction of inhibition to the dorsolateral frontal regions involved with working memory.

HEMISPHERIC ASYMMETRY IN CREATIVITY The rationale for expecting to find differences in hemispheric activation in relation to creative processes originates from research regarding the specialisation of the hemispheres for specific tasks. It has been proposed that the left hemisphere is specialised for verbal and analytical processing, while the right hemisphere is specialised for global/parallel processing such as the production of music and mental images (Martindale, 1999). Martindale, Hines, Mitchell and Covello (1984) investigated hemispheric EEG asymmetry in relation to creativity, as assessed by the Alternate uses and Remote Associates tests. No differences in EEG asymmetry

Neurobiology of Creativity

65

were found between participants high and low in creativity when they were at rest. However, when EEG was measured during the performance of a creative task (speaking or writing a fantasy story), the highly creative participants were found to display greater right than left hemispheric activation. This pattern was not found in the less creative participants. In a further experiment, Martindale et al. (1984) compared EEG asymmetry amongst art students compared to artistically untrained participants. Two tasks were given, the drawing of a cow vertebra, and reading of an article on economics. In comparison to the untrained participants, the arts students were found to display greater right than left-hemispheric activation during the drawing task, but not during the reading task. Similarly, no differences in hemispheric activation between groups were observed at rest. The findings from this study suggest that creative individuals use the right hemisphere more than the left hemisphere in the completion of creative tasks. Carlsson, Wendt and Risberg (2000) investigated regional cerebral blood flow (rCBF) during the completion of both an automatic speech, a verbal fluency and a uses of objects task. The automatic speech task involved counting aloud. In the verbal fluency task participants were instructed to say all the words they could think of that began with a specified letter, with a new letter given every minute. The uses of objects task (creative task) required participants to specify as many different uses of a brick that they could think of, both ordinary and uncommon. Participants were divided into low and high creativity groups on the basis of their scores on a test of creative functioning and differences in rCBF were analysed between tasks. In the anterior prefrontal regions the highly creative group were found to have higher levels of blood flow bilaterally in the creative task compared to the verbal fluency task while the less creative group displayed a decrease in the right hemisphere and no change to the left hemisphere. In the frontotemporal regions the highly creative group displayed unchanged levels of blood flow bilaterally in the creative task compared to the verbal fluency task, while the less creative group displayed bilateral decreases. Finally, in the superior frontal regions the highly creative groups displayed a left-sided asymmetry across all tasks, whereas the less creative group displayed a symmetrical pattern. However, for the creative group an increase in blood flow was observed in the right hemisphere and unchanged levels in the left hemisphere for the creative task compared to the verbal fluency task, while the less creative group displayed bilateral decreases. It is also worth noting that in both the left and right superior frontal regions a negative correlation was found between rCBF levels and the level of performance (number of categories) in the creative task, perhaps due to enhanced efficiency (Carlsson et al., 2000). These findings suggest that the superior frontal regions may be particularly important in the solution of creative/divergent tasks, as evidenced by greater left-sided asymmetry in the superior frontal regions of creative people compared to non-creative people. In summary, these findings suggest that the more

66

david camfield

creative individuals utilize bilateral prefrontal regions in the solution of creative tasks, while less creative individuals rely predominantly on the left hemisphere. Further, there may be differences in the functioning of the right superior frontal regions between creative as compared to less-creative individuals. Kurup and Kurup (2003) have proposed a biochemical model to explain asymmetric differences between creative and less-creative individuals. Central to this model is the glycoside Digoxin, produced by the hypothalamus. Digoxin is an endogenous membrane Na+ – K+ ATPase inhibitor that is involved in the modulation of several neurotransmitter systems throughout the human brain (Hisaka, Kasamatsu, Takenaga, & Ohtawa, 1990). Kurup and Kurup (2003) measured differences in blood serum levels of digoxin and neurotransmitters between both left and right hemisphere-dominant and creative (poets) versus less-creative individuals. Digoxin levels were found to be significantly higher in the creative compared to the less-creative individuals and in the right-hemisphere dominant compared to left-hemisphere dominant individuals, as was serotoninegic, strychninergic, and nicotinergic transmission. In contrast, the less creative and the left-hemisphere dominant individuals were found to display higher dopaminergic, morphinergic and noradrenergic transmission in comparison to the creative and right-hemisphere dominant individuals. On the basis of these findings Kurup and Kurup (2003) suggested that elevated levels of digoxin produce a hyperconscious state that is characterised by increased focused attention, perceptual binding, and short-term memory, processes that are inducive to creativity.

A POLYMETHODOLOGICAL STUDY OF CREATIVITY Perhaps the most comprehensive study of the neurophysiology of creative processes has been conducted by researchers at the institute of the human brain at the Russian Academy of Science. Four tasks of varying creative content were specifically developed for EEG and PET analysis. The first task (D: Difficult) consisted of composing a mental story from words spanning several different semantic areas (e.g. to begin, glass, to want, roof, mountain, to keep silence), which was designed to induce subjects to give up stereotyped ways of thinking. The second task (E: Easy) consisted of composing a mental story from words from the same semantic area (e.g. school, to understand, task, to learn, lesson), which was considered to involve more stereotyped thinking than the first task. While the first and second tasks involved making a story out of as many of the listed words as possible, in any order, the third task (R: Reconstruction) consisted of making a mental story out of the listed words without changing their order (e.g. manager, to suggest, to call, boss, department, to demand, to make up, program), requiring less creativity than the previous two tasks. The final task (W: Word memorisation) consisted of simply memorizing a list of words in a mechanical

Neurobiology of Creativity

67

fashion, a task that essentially comprised no creativity (Starchenko, Vorob’ev, Kliucharev, Bekhtereva, & Medevedev, 2000). Bekhtereva, Dan’ko, Starchenko, Pakhomov and Medvedev (2001) analysed PET and EEG data associated with the four creative tasks among two separate samples of young volunteers, the latter consisting of both actor and non-actor subgroups. In regards to the EEG data only the results from the non-actor sub-group will be discussed. Differences in patterns of local cerebral blood flow (LCBF) between the various tasks were measured by means of PET while differences in the level of electrical activity was measured by means of EEG power, and the integration of brain areas between tasks was measured by means of EEG coherence. Results may be grouped according to the tasks compared. i) D-E Comparison. PET revealed that the difficult creative task employed the right frontal lobe more than the easy task. EEG power was found to be significantly lower at F7 and T4 during the difficult task, while no differences in EEG coherence were observed. ii) D-R Comparison. PET revealed that the difficult creative task employed the left frontal lobe and left parieto-occipital area more than the reconstruction task. EEG power was found to be significantly lower at F7, F8, T4, Pz, O1 and O2 during the difficult creative task. EEG coherence was found to increase bilaterally in the frontal areas, particularly in the left hemisphere during the difficult creative task. iii) D-W Comparison. PET revealed that the difficult creative task employed greater activation in bilateral frontal regions, particularly in the left hemisphere, and the left tempero-parietal-occipital regions in comparison to the word memorisation task, the activation was similar to the D-R comparison though slightly more extensive. EEG power was found to be significantly lower at F7, Fz, F4, F8, P3, Pz, O1 and O2 during the difficult creative task, while EEG coherence was found to be higher during the difficult creative task in bilateral frontal areas, and to a lesser extent the posterior areas. iv) E-R Comparison. PET revealed that the easy creative task employed the left frontal lobe and left parieto-occipital areas more than the reconstruction task, similar to the D-R comparison. EEG power was not found to be significantly different between tasks, while EEG coherence was found to be higher during the easy creative task in bilateral frontal areas, particularly in the left hemisphere. v) E-W Comparison. PET revealed that the easy creative task employed the bilateral frontal regions, particularly in the left hemisphere, and the tempro-parietal regions more than the word memorisation task, again similar to the D-W and D-R comparisons. EEG power was not found to be significantly different between tasks, while EEG coherence was found to be higher in frontal and to a lesser extent posterior areas in the easy creative task (a coherence pattern that was more diffuse and less frontally oriented than previous comparisons). In summarising these findings, Bekhtereva et al. (2001) suggest that it is the frontal region of the left hemisphere which is involved to a greater extent in creative tasks in comparison to non-creative tasks, while the frontal region of the

68

david camfield

right hemisphere is required for creative tasks of increased difficulty in comparison to easier creative tasks. In commenting on the patterns found in EEG coherence, they note that an increase in coherence in frontal and anterior-frontal regions of the cortex appears to be particularly important in the solution of creative tasks, in particular an increase in inter-hemispheric coherence. This study is particularly powerful in that it uses converging lines of evidence from both PET and EEG data to arrive at similar conclusions regarding the importance of the frontal lobes in creative tasks.

OPENNESS TO EXPERIENCE Openness to Experience (Openness) is a construct that has it’s roots in the psychodynamic study of art and creative thought as well as research predicting individual differences in hypnotic susceptibility (Schachtel, 1959; Fitzgerald, 1966; Coan, 1972). Empirically openness has been found to be related to a wide range of constructs including Lexical Factor V, Absorption, Boundary Thickness, Sensation Seeking, and the cognitive abilities of General intelligence (‘g’) and divergent thinking (McCrae, 1993–1994). Openness is included by Costa and McCrae (1992) as one of the five dimensions measured by the NEO Personality Inventory (NEO PI-R; Costa & McCrae, 1992). Openness describes a broad and general dimension of personality, “seen in vivid fantasy, artistic sensitivity, depth of feeling, behavioural flexibility, intellectual curiosity, and unconventional attitudes” (McCrae & Costa, 1997). There are six facets (sub-scales) of the Openness domain as it appears in the NEO PI-R: Fantasy, Aesthetics, Feelings, Actions, Ideas, and Values. It is assessed by the NEO PI-R using 48 items, with 8 items for each facet sub-scale. A number of studies in recent years have explored the relationship between creativity and the Five Factor Model ( FFM) of personality. Within this framework, Openness has emerged as the personality trait most strongly related to creativity, using both cognitive and criterion measures of trait creativity (McCrae, 1987; Holland, Dollinger, Holland, & MacDonald, 1995; King, McKee Walker, & Broyles, 1996; Feist, 1998). In the Baltimore Longitudinal Study of Aging (BLSA) McCrae and Costa (1987) found that the Creative Personality Scale (CPS; Gough, 1979) was most strongly and consistently related to Openness. The correlations between the CPS and Openness by self-reports, peer and spouse ratings were 44, .34, and .26 respectively. On the basis of these findings McCrae (1987) suggested that divergent thinking may provide the aptitude for original thinking, whereas Openness provides the inclination to actually be creative. Considering the strong link that Openness has to both creative ability and achievement it is worthwhile studying the relationship between Openness and biological measures. Further, the psychometric properties of Openness are far

Neurobiology of Creativity

69

stronger than the instruments currently used to gauge creative ability, considering almost 20 years worth of validation associated with the NE0 PI-R. Research by Stough, Donaldson, Scarlata and Ciorciari (2001) provides a glimpse into the potential that this dimension may have for understanding creative processes. Using the technique of photic driving, the percentage of EEG ouput in the theta range was found to be positively correlated with NE0 PI-R Openness. This preliminary study suggested that individuals high in Openness produce a greater amount of theta waves in comparison to individuals low in Openness. Drawing on studies indicating a higher degree of theta production during childhood and pleasure states (Maulsby, 1971; Kugler & Laub, 1971). Stough et al. (2001) suggest that theta production may provide a biological explanation for the flexible thinking and liberal values associated with Open individuals.

CONCLUSIONS While the study of the neural bases of creativity is still in it’s infancy, the existing research in this area has already begun to provide a picture of the relationship between creative processes and the human brain. Further, from the existing studies in this area, certain key themes have begun to emerge. The importance of the frontal lobes is central to almost all studies of creative ability. Converging evidence from both PET, MRI and EEG studies suggest that the activation of frontal lobes clearly differentiates creative from non-creative tasks. Further, these studies also suggest that more creative individuals have greater efficiency associated with frontal lobes function. An analysis of EEG coherence also points to both greater intra- and inter-hemispheric cooperation between frontal areas across several frequencies during the execution of creative tasks. Some studies have additionally suggested an increase in long distance coherence such as connections between fronto-polar and parieto-occipital sites. Another recurring theme in the creativity literature has been the presence of hemispheric asymmetry. A number of studies suggest a greater involvement of the right hemisphere frontal regions compared to the left hemisphere, in the solution of creative tasks. Indirect evidence from studies of schizophrenia and psychoticism also suggests that the weakening of inhibitory processes may play a facilatory role in creativity. Similarly, studies of the emergence of artistic talent associated with frontotemporal dementia links the degradation of the left anterior temporal cortex to removal of inhibitory influences on the frontal lobes. Finally, a few studies have demonstrated a link between the presence of low frequency waves and creative processes. Delta and theta waves appear to be associated with the perceptual resolution of a creative task, while theta response to photic driving appears to be related to Openness to Experience, a personality trait with strong ties to creativity. Further research utilizing poly-modal brain imaging techniques

70

david camfield

is sure to considerably expand our understanding of both the neural bases of creative processes and the neural bases of individual differences in these abilities.

REFERENCES Barcelo, F., Perianez, J. A., & Knight, R. T. (2002). Think differently: a brain orienting response to task novelty. Neuroreport, 13(15), 1887–92. Beech, A., Baylis, G. C., Smithson, P., & Claridge, G. (1989). Individual differences in schizotypy as reflected in measures of cognitive inhibition. British Journal of Clinical Psychology, Vol 28(2), 117–129. Beech, A., Powell, T., McWilliam, J., & Claridge, G. (1989). Evidence of reduced “cognitive inhibition” in schizophrenia. British Journal of Clinical Psychology, Vol 28(2), 109–116. Bekhtereva, N. P., Dan’ko, S. G., Starchenko, M. G., Pakhomov, S. V., & Medvedev, S. V. (2001). [Research of brain organization of the creative process. Part III. Brain activation according to data from analysis of local brain blood circulation and EEG]. Fiziol Cheloveka, 27(4), 6–14. Boivin, M. J., Giordani, B., Berent, S., Amato, D. A., & et al. (1992). Verbal fluency and positron emission tomographic mapping of regional cerebral glucose metabolism. Cortex, Vol 28(2), 231– 239. Callaway, E. (1991). Discussion: towards a pharmacology of information processing. Electroencephalogr Clin Neurophysiol Suppl, 42, 173–6. Cameron, N. (1938). Reasoning, regression and communication in schizophrenics. Psychological Monographs, 50, No. 1, 34. Cameron, N. (1947). The psychology of behavior disorders: a bio-social interpretation. Carlsson, I., Wendt, P. E., & Risberg, J. (2000). On the neurobiology of creativity. Differences in frontal activity between high and low creative subjects. Neuropsychologia, 38(6), 873–85. Christensen, P. R., Guilford, J. J., Merrifield, P. R., & Wilson, R. C. (1970). Alternate Uses. Beverly Hills: Sheridan Psychological Service. Coan, R. W. (1972). Measurable components of openness to experience. Journal of Consulting & Clinical Psychology, Vol. 39(2), 346. Costa Jr., P. T., & McCrae, R. R. (1992). Revised NEO Personality Inventory (NEO PI-R) and NEO FiveFactor Inventory (NEO-FFI) Professional Manual. Odessa, Florida: Psychological Assessment Resources. ˙ & Coles, M. G. (1988). Is the p300 component a manifestation of context updating? Donchin, E., Behavioral & Brain Sciences, Vol 11(3), 357–427. Eysenck, H. J. (1993). Creativity and personality: suggestions for a theory. Psychological Inquiry, 32 p. Eysenck, H. J. (1995). Creativity as a product of intelligence and personality. D. H. E. Saklofske, & M. E. Zeidner International handbook of personality and intelligence. (pp. 231–247). New York, NY, US: Plenum Press. Feist, G. J. (1998). A meta-analysis of personality in scientific and artistic creativity. Personality & Social Psychology Review, 2(4), 290–209. Fitzgerald, E. T. (1966). The measurement of openness to experience: a study of regression in the service of the ego. Dissertation Abstracts International, 27(3-B), 955. Gough, H. G. (1979). A creative personality scale for the adjective check list. Journal of Personality & Social Psychology, Vol 37(8), 1398–1405. Guilford, J. P. (1967a). Creativity: yesterday, today, and tomorrow. Journal of Creative Behavior, 1(1), 3–14. Guilford, J. P. (1967b). The nature of human intelligence.

Neurobiology of Creativity

71

Hansenne, M. (1999). P300 and personality: an investigation with the Cloninger’s model. Biol Psychol, 50(2), 143–55. Hansenne, M., Pitchot, W., Gonzalez Moreno, A., Papart, P., Timsit-Berthier, M., & Ansseau, M. (1995). Catecholaminergic function and P300 amplitude in major depressive disorder (P300 and catecholamines). Electroencephalogr Clin Neurophysiol, 96(2), 194–6. Hansenne, M., Pitchot, W., Papart, P., & Ansseau, M. (1998). Serotonergic modulation of the p300 event related brain potential. Human Psychopharmacology Clinical & Experimental, Vol 13(4), 239–243. Hargreaves, H. L. (1927). The “faculty” of imagination. British Journal of Psychology, X, 74. Heaton, R. K., Chelune, G. J., Talley, J. L., Kay, G. G., & Curtiss, G. (1993). Wisconsin Card Sorting Test manual: Revised and expanded. Odessa, FL: Psychological Assessment Resources. Hisaka, A., Kasamatsu, S., Takenaga, N., & Ohtawa, M. (1990). Absorption of a novel prodrug of L-dopa, L-3-(3-hydroxy-4-pivaloyloxyphenyl)alanine (NB-355). In vitro and in situ studies. Drug Metab Dispos, 18(5), 621–5. Holland, D. C., Dollinger, S. J., Holland, C. J., & MacDonald, D. A. (1995). The relationship between psychometric intelligence and the five-factor model of personality in a rehabilitation sample. Journal of Clinical Psychology, Vol 51(1), 79–88. Jausovec, N., & Jausovec, K. (2000a). Differences in resting EEG related to ability. Brain Topogr, 12(3), 229–40. Jausovec, N., & Jausovec, K. (2000b). EEG activity during the performance of complex mental problems. Int J Psychophysiol, 36(1), 73–88. Johnson, R. Jr. (1993). On the neural generators of the P300 component of the event-related potential. Psychophysiology, 30(1), 90–7. Kapur, N. (1996). Paradoxical functional facilitation in brain-behaviour research. A critical review. Brain, 119 (Pt 5), 1775–90. King, L. A., McKee Walker, L., & Broyles, S. J. (1996). Creativity and the five-factor model. Journal of Research in Personality, Vol 30(2), 189–203. Klorman, R., & Brumaghim, J. T. (1991). Stimulant drugs and ERPs. Electroencephalogr Clin Neurophysiol Suppl, 42, 135–41. Kugler, J., & Laub, M. (1971). Lambda waves during sleep. Electroencephalogr Clin Neurophysiol, 30(2), 168. Kurup, R. K., & Kurup, P. A. (2003). Hypothalamic digoxin, hemispheric chemical dominance, and creativity. International Journal of Neuroscience, 13p. Lavric, A., Forstmeier, S., & Rippon, G. (2000). Differences in working memory involvement in analytical and creative tasks: an erp study. Neuroreport: For Rapid Communication of Neuroscience Research, Vol 11(8), 1613–1618. Lubow, R. E. (1989). Latent inhibition and conditioned attention theory. Martindale, C. (1999). Biological bases of creativity. R. J. E. Sternberg Handbook of creativity. (pp. 137–152). New York, NY, US: Cambridge University Press. Martindale, C., & Hasenfus, N. (1978). Eeg differences as a function of creativity, stage of the creative process, and effort to be original. Biological Psychology, Vol 6(3), 157–167. Martindale, C., Hines, D., Mitchell, L., & Covello, E. (1984). Eeg alpha asymmetry and creativity. Personality & Individual Differences, Vol 5(1), 77–86. Maulsby, R. L. (1971). An illustration of emotionally evoked theta rhythm in infancy: hedonic hypersynchrony. Electroencephalogr Clin Neurophysiol, 31(2), 157–65. McCrae, R. R. (1987). Creativity, divergent thinking, and openness to experience. Journal of Personality & Social Psychology, Vol 52(6), 1258–1265. McCrae, R. R. (1993–1994). Openness to experience as a basic dimension of personality. Imagination, Cognition and Personality, 13(1), 39–55.

72

david camfield

McCrae, R. R., & Costa, P. T. Jr. (1997). Conceptions and correlates of openness to experience. R. E. Hogan, J. A. E. Johnson, & et al Handbook of personality psychology. (pp. 825–847). Mednick, S. A., & Mednick, M. T. (1967). Remote Associates Test: Examiners Manual. Boston: Houghton Mifflin. Mednick, S. (1962). The associative basis of the creative process. Psychological Review, 69(3), 220– 232. Miller, B. L., Boone, K., Cummings, J. L., Read, S. L., & Mishkin, F. (2000). Functional correlates of musical and visual ability in frontotemporal dementia. Br J Psychiatry, 176, 458–63. Miller, B. L., Cummings, J., Mishkin, F., Boone, K., Prince, F., Ponton, M., & Cotman, C. (1998). Emergence of artistic talent in frontotemporal dementia. Neurology, 51(4), 978–82. Miller, B. L., Ponton, M., Benson, D. F., Cummings, J. L., & Mena, I. (1996). Enhanced artistic creativity with temporal lobe degeneration. Lancet, 348(9043), 1744–5. Molle, M., Marshall, L., Lutzenberger, W., Pietrowsky, R., Fehm, H. L., & Born, J. (1996). Enhanced dynamic complexity in the human EEG during creative thinking. Neurosci Lett, 208(1), 61–4. Orme-Johnson, D. W., & Haynes, C. T. (1981). EEG Phase coherence, pure consciousness, creativity, and TM-Sidhi experiences. Neuroscience, 13, 211–217. Parks, R. W., Loewenstein, D. A., Dodrill, K. L., Barker, W. W., & et al. (1988). Cerebral metabolic effects of a verbal fluency test: a pet scan study. Journal of Clinical & Experimental Neuropsychology, Vol 10(5), 565–575. Petsche, H. (1996). Approaches to verbal, visual and musical creativity by EEG coherence analysis. Int J Psychophysiol, 24(1–2), 145–59. Petsche, H., Kaplan, S., von Stein, A., & Filz, O. (1997). The possible meaning of the upper and lower alpha frequency ranges for cognitive and creative tasks. Int J Psychophysiol, 26 (1–3), 77–97. Phelps, E. A., Hyder, F., Blamire, A. M., & Shulman, R. G. (1997). Fmri of the prefrontal cortex during overt verbal fluency. Neuroreport: An International Journal for the Rapid Communication of Research in Neuroscience, Vol 8(2), 561–565. Russ, S. W. (1993). Affect and creativity: the role of affect and play in the creative process. Schachtel, E. G. (1959). Metamorphosis: on the development of affect, perception, attention, and memory. Starchenko, M. G., Vorob’ev, V. A., Kliucharev, V. A., Bekhtereva, N. P., & Medevedev, S. V. (2000). [The cerebral organization of creativity. I. The development of a psychological test]. Fiziol Cheloveka. 26(2), 5–9. Stough, C., Donaldson, C., Scarlata, B., & Ciorciari, J. (2001). Psychophysiological correlates of the NEO PI-R openness, agreeableness and conscientiousness: preliminary results. Int J Psychophysiol, 41(1), 87–91. Thatcher, R. W., Krause, P. J., & Hrybyk, M. (1986). Cortico-cortical associations and eeg coherence: a two-compartmental model. Electroencephalography & Clinical Neurophysiology, Vol 64(2), 123–143. Torrance, P. E. (1974). Torrance Tests of Creative Thinking: Norms-Technical Manual. Lexington, Massachusetts: Personnel Press Inc. Wechsler, D. (1981). Wechsler Adult Intelligence Scale-Revised Manual. New York: The Psychological Corporation. Whitton, J. L., Moldofsky, H., & Lue, F. (1978). Eeg frequency patterns associated with hallucinations in schizophrenia and “creativity” in normals. Biological Psychiatry, Vol 13(1), 123–133.

4 Neurobiology of Intelligence Cindy Van Rooy, John Song, and Con Stough

“Biological intelligence is the kind of concept Galton was concerned with; it refers to the structure of the human brain, its physiology, biochemistry, and genetics which are responsible for the possibility of intelligent action on the part of human beings.” (Eysenck, 1988, p3).

Over a century ago Sir Francis Galton (1892) described his seminal study of individual differences in intelligence. Galton hypothesised that microlevel sensory and perceptual processes were inherited and formed the basis for excelling in areas such as the arts, sciences and law. He argued that those persons who excelled in these eminent areas would be more efficient at detecting changes in illumination, reacting to slight tactile pressure, and noticing simple auditory and visual stimuli. Whilst some have argued that Galton’s findings did not support his hypothesis, an early review of the research by R.A. Mc Farland (1928) noted that while the literature was contradictory, there did seem to be some interesting relationships supporting Galton’s position of a link between the efficiency of microlevel processing and measures of general intellectual ability. The contradiction among these studies may have been the result of using unreliable measures, and employing procedures that were not uniformly rigorous (Ceci, 1990). Since Galton’s original research, there has been a wide range of research attempting to identify the biological basis of intelligence.

Cindy Van Rooy and John Song • Brain Sciences Institute, Swinburne University, Melbourne, Victoria, Australia. Con Stough • Swinburne Center for Neuropsychology, Swinburne University, Melbourne, Victoria, Australia.

73

74

cindy van rooy et al.

Figure 1. Relationship between Biological Intelligence, IQ and Social Intelligence. (from Eysenck, 1988, p 4)

Eysenck (1988) proposed that while culture, family upbringing, socioeconomic status, and education all play a role in determining psychometric intelligence (IQ), the most influential role is played by biological intelligence (see Figure 1). In our search for the biological basis of intelligence it would make sense to focus our attention on the brain, as the brain controls all aspects of human function and is widely responsible for the diversity of human thought and behaviour. The human brain is a highly complex organ that consists of several billion cells that are connected by 150,000 km of neural fibres. Ultimately, it is the properties of these cells and neural fibres, which work endlessly in complex neural firing patterns, that are responsible for our behaviours and thoughts (Chen & Buckley, 1988). Researchers have utilized a range of brain imaging techniques to investigate the properties of these cells and neural fibres in their search for the biological basis of intelligence. For example, some researchers have utilised electroencephalography (EEG) and evoked potentials (EPs) to investigate the relationship between brain electrical activity and intelligence, while other researchers have used metabolic imaging techniques such as positron emission tomography (PET) and functional magnetic resonance imaging (fMRI). See the chapter in this book by Puce for an excellent review of these imaging modalities and techniques. Research investigating the neurobiology of intelligence has generally involved several types of experimental paradigms which we can broadly classify: (1) The first involves correlating brain electrical or metabolic activity while in a resting state with a psychometric measure of intelligence. Typical of these paradigms have been the use of the electroencephalograph (EEG) measures. (2) The second involves investigating the changes in brain electrical or metabolic activity while performing an intelligence test, or a cognitive test related to intelligence. Typically measures such as Event Related

Neurobiology of Intelligence

75

Potentials (ERPs), functional EEG such as the Steady State Visual Evoked Potential (SSVEP) or Positron Emission Tomography (PET) have been employed. (3) The third type assesses the relationship between brain structure usually measured by Magnetic Resonance Imaging (MRI) and intelligence or functional changes using functional Magnetic Resonance Imaging (fMRI) during an intelligence or cognitive task. We will review and discuss all of these areas except for relationships between structural parameters of the brain and IQ such as brain size and white and grey volumetric rations and intelligence. Therefore our review will mainly focus on functional aspects of brain-intelligence relationships. Additionally a review of the emerging literature on pharmacological approaches within intelligence research is also beyond the scope of this chapter. However Stough, Thompson, Bates and Nathan (2001) have provided a recent review on experimental studies and models in this area.

SPONTANEOUS EEG AND INTELLIGENCE Since Hans Berger (1929) reported that during mental arithmetic the amplitude of the alpha rhythm decreased, EEG has been used to investigate the relationship between brain electrical activity and higher mental functions (Gevins & Cutillo, 1995). The majority of studies investigating the relationship between spontaneous EEG and intelligence have focussed their attention on measures of the alpha rhythm (AR), and to a lesser extent the slower theta and delta rhythms. The alpha rhythm has a frequency range of approximately 8–13 Hz, and is the dominant frequency in the EEG recorded from the scalp of adult humans (Pilgreen, 1995; Anokhin & Vogel, 1996; Klimesch, 1999). It is maximal over the occipital lobe when relaxed, but awake and alert, with eyes closed (Steriade, Gloor, Llin´as, Lopes da Silva, & Mesulam, 1990; Pilgreen, 1995). There is multiple evidence that AR recorded from the scalp represents neurophysiological mechanisms directly related to individual differences in information processing in the human brain (Klimesch, Schimke& Pfurtscheller, 1993; Lebedev, 1990). The delta (1.5–4.5 Hz) and theta (4.5–7.0 Hz) rhythms are sometimes referred to as ‘slow wave’ activity, and research has indicated that these slow wave frequencies covary with maturation (decreasing with age), neuropathology, and functional disorders such as learning disabilities and hyperkinesis (Hughes, 1976; John et al., 1977; Stamm & Kreder, 1979). Much of the early research investigating the neurobiology of intelligence attempted to correlate measures of psychometric intelligence with EEG parameters such as the amplitude of alpha activity, alpha index, and the frequency of the alpha and slow wave rhythms. Many of these early studies recorded the EEG from only

76

cindy van rooy et al.

a few sites on the scalp, from only a small sample of participants (many of whom were patients), and visually analysed the frequencies of the spontaneous EEG (Vogel & Broverman, 1964; Ellingson, 1966). The results of this early research was inconsistent and inconclusive, with methodological problems (such as those listed above), plaguing many of the studies. Both Vogel and Broverman (1964) and Ellingson (1966) reviewed this early literature. Vogel and Broverman concluded that the relationship between resting EEG and psychometric intelligence was most noticeable in individuals whose intellectual function is either underdeveloped, such as children, or deteriorated, such as individuals suffering from brain damage or dementia, but the evidence was inconclusive in adults. However, Ellingson concluded that the literature was contradictory and inconclusive in children or people suffering from brain damage or dementia, and non-existent in adults. As the equipment used to record and analyse the EEG became more advanced and sophisticated, so did the EEG variables that were investigated. Advances in technology have also enabled researchers to record EEG from a larger number of sites on the scalp and from a larger sample of subjects. This has enabled more recent research to minimise some of the methodological problems associated with the earlier studies and to investigate topographical differences in spontaneous EEG that may be related to intelligence, as well as EEG coherence and intelligence.

EEG AMPLITUDE AND INTELLIGENCE A few studies have investigated the amplitude of the EEG alpha rhythm to examine if that measure of the human EEG was related to intelligence. Generally, the researchers found no evidence of a relationship between intelligence and alpha amplitude (Giannitrapani, 1969). In a later study, Bosaeus, Matouˇsek, and Peters´en (1977) also found no evidence for a relationship between alpha amplitude while resting and psychometric intelligence in a large sample of five to sixteen year old neurologically normal children. Bosaeus et al. also reported finding no relationship between intelligence scores and EEG amplitude in any of the other five EEG bandwidths they investigated, the delta, theta, beta1 (12.5–17.5 Hz), and beta2 (17.5–25.0 Hz) bands.

ALPHA INDEX AND INTELLIGENCE Some early studies investigated the relationship between psychometric intelligence and a measure of the EEG alpha rhythm known as the alpha index. The alpha index is the proportion of time alpha activity is present in the EEG (Stern, Ray, & Davis, 1980). The results of these studies have been mixed and inconclusive. For example, in a sample of eight year old children, Knott, Friedman, & Bardsley (1942)

Neurobiology of Intelligence

77

reported finding no relationship between intelligence and alpha index while resting with eyes closed. In contrast, in samples of adults studies have reported a low to moderate strength positive relationship between the alpha index and Full Scale IQ (Giannitrapani, 1969) and Verbal IQ scores (Mundy-Castle, 1958; Giannitrapani, 1969). Mundy-Castle also reported finding a moderate strength negative relationship between the alpha index and the Picture Arrangement and Picture Completion subtests of the South African version of the Wechsler-Bellevue Intelligence Scale. Giannitrapani (1969) reported that the positive correlations between the alpha index and the full IQ and verbal IQ were marginally stronger when the EEG was recorded during mental arithmetic. Giannitrapani also reported that during mental multiplication, a moderate to strong positive relationship between performance IQ scale of the Wechsler Adult Intelligence Scale and the alpha index when the EEG was found. During Giannitrapani’s study each subject was given a mental multiplication problem that was adjusted for each subject so as to take a little over five seconds to solve. Giannitrapani noted that even though the tasks were objectively more difficult for the higher IQ subjects, mental multiplication produced less desynchronization of the alpha rhythm in individuals of higher IQ.

EEG FREQUENCIES AND INTELLIGENCE Alpha frequency (AF) was one of the first measures of the EEG that has been investigated in the search for the neurobiological basis of intelligence. Many of the early studies that visually analysed the EEG traces, calculated the AF by counting the number of alpha waves in small samples of the EEG trace and calculating the mean AF. Once again, the results of these early studies were mixed. Knott, Friedman, & Bardsley (1942) reported finding a positive relationship (r = +0.5) between intelligence and alpha frequency while eyes closed in eight year old children, but no relationship in twelve year old children. Ellingson & Lathrop (1973) reported finding a positive, but not significant, relationship in a small sample of psychiatric patients. Although they found no evidence of a relationship in a sample of Down’s syndrome patients (aged 13–42 years; Ellingson & Lathrop, 1973). Shagass (1946) also reported finding no relationship between the occipital alpha frequency when resting with eyes closed and scores on a group ability test in a large sample of aircrew candidates. In contrast, Mundy-Castle (1958) found a series of moderate strength positive correlations between mean alpha frequency when relaxed with eyes closed and the general IQ, verbal IQ, and practical IQ scales on the South African version of the Wechsler-Bellevue Intelligence Scale in a sample of adults. More recent studies have used computers to analyse the EEG activity. These later studies used a slightly different measure of alpha frequency known as the peak AF, which can be defined as the peak frequency of the alpha rhythm, or

78

cindy van rooy et al.

the frequency in the traditional alpha rhythm range (i.e. 8–13 Hz) that shows the largest power estimate. (Klimesch, 1999). AF has the advantage of being not as susceptible to the influence of extracerebral factors such as skull thickness or conductance as the alpha amplitude and the alpha power measures of EEG, and therefore its variation seems to result directly from the variation in brain function (Anokhin & Vogel, 1996). Gasser, Von Lucadou-M¨uller, Verleger and B¨acher (1983) investigated the relationship between intelligence scores and EEG parameters in a group of normal and a group of mildly mentally retarded 10–13 year old children. They reported a low to moderate strength positive correlation between IQ scores and the peak alpha frequency over the parietal and occipital areas for the mentally retarded children, but not the normal children. Anokhin and Vogel (1996) have also reported that higher IQ scores were associated with an increased AF, but this time in a sample of normal, healthy adults. They hypothesised that the pattern of the correlations they found implied that non-verbal inductive reasoning abilities as assessed by Ravens Standard Progressive Matrices (SPM) may be related to neurophysiological properties of the frontal areas of the brain, whereas, the significant correlations between AF and the verbal tasks showed a more diffuse topographical distribution. Neubauer, Sange, and Pfurtscheller (1999) also found that higher IQ scores were associated with a higher individual peak alpha frequency, although the correlations did not reach significance. Studies have also investigated the relationship between intelligence and frequency measures in other EEG bands, both while relaxing with eyes closed or open, or while performing a cognitive task. For example, Giannitrapani (1969) investigated the average EEG frequency in a sample of normal, healthy adults while resting with their eyes closed and while performing mental arithmetic. They reported finding interactions between the average EEG frequency, IQ group and hemisphere, with the average EEG frequency in the left frontal area greater than in the right frontal area for both the average and high IQ groups. In the parietal area, the high IQ showed greater average EEG frequency in the left, while the average group displayed greater average EEG frequency in the right. In the occipital area, the opposite was found with the high IQ group showing greater average EEG frequency in the right occipital area, and the average IQ group showing greater average EEG frequency in the left occipital area. Giannitrapani also reported finding a moderate to strong positive correlation between IQ and the average EEG frequency during mental arithmetic (a positive relationship was also seen during the resting condition, but it was not significant). The study also found that participants with a higher IQ tended to show the smallest difference between the average EEG frequency during the resting and thinking states. Fischer, Hunt and Randhawa (1982) investigated EEG parameters while resting in groups of children. They also found that higher scores on reasoning and mathematical ability tests were

Neurobiology of Intelligence

79

associated with relatively a higher resting EEG frequency in the left hemisphere than in the right hemisphere, but only in a sample of academically talented children, not in the academically handicapped group. Fischer et al. also investigated the ratio of the average alpha frequency to the average overall frequency. While they found no relationship between the alpha ratio measure and performance on various cognitive tests for the academically talented group, there were significant correlations in the academically handicapped group. The findings indicated that for the academically handicapped group, a decrease in the average alpha frequency in relation to the average overall frequency was associated with better scores on a reasoning test and lower scores on the mathematics and reading tests. Marosi et al. (1999) took a different approach and investigated the mean frequency in four EEG bands, delta, theta, alpha and beta, while the subjects were resting with their eyes open. The strength and direction of the relationships depended on the location the EEG was recorded from, the EEG bandwidth, and the scale on the Wechsler Adult Intelligence Scale (WAIS). They found that generally, as IQ scores increased, the mean delta and theta frequencies decreased and the mean alpha and beta frequencies increased. Marosi et al. further concluded that broad band measurements are not an adequate tool in the study of certain abilities as broad bands dilute the effect of IQ when it occurs in a narrower frequency range. At present, only a hypothetical explanation of the AF-intelligence relationship can be proposed based on the possible role of the alpha rhythm for information processing in cortical networks (Klimesch, Schimke & Pfurtscheller, 1993; Lebdev, 1990) and on individual differences in cortical arousal level (Golubeva, 1980; Robinson, 1993). Alpha rhythm emerges as a result of synchronous oscillations of synaptic potentials in large populations of neurons (primarily pyramidal cells) spread throughout the cortex. Although the exact mechanisms of alpha rhythm generation and its functional meaning are not understood completely so far, there is increasing evidence that synchronized oscillatory activity in the cerebral cortex is essential for spatiotemporal coordination and integration of activity of anatomically distributed but functionally related neural elements. Recent experimental and simulation studies of information processing in neuronal networks suggest that synchronized oscillatory activity in cell assemblies plays a key role in encoding, storage, and retrieval of information in the brain. Whereas information seems to be encoded by the temporal sequence of action potentials, synchronized periodic fluctuations of membrane excitability enable temporally and spatially structured co-activation of cells in an assembly (Birmbaumer, Elbert, Canavan & Rockstrih, 1990; Buzasaki & Chrobak, 1995; Lidman & Idiart, 1995). As a consequence, the duration of the cycle of the dominant cerebral rhythm may limit the capacity for storage, transfer, and retrieval of info in individual brains. There are also some indications that AF is related to the level of cortical arousal in both state and trait aspect. AF increases with mental activity compared to rest. Increasing cognitive task difficulty leads to right hemispheric as well as bilateral

80

cindy van rooy et al.

alpha acceleration (Earle, 1988); a decrease in individual AF is always related to a drop in performance on memory tasks (Klimesch, Schimke & Pfurtscheller, 1993). Resting AF is higher in persons with a higher level of tonic cortical arousal regarded as a stable individual train and assessed using either EEG or EP measures. Moreover, it correlates positively with indices of mental activity level, academic performance in high school students, as well as performance on some memory tasks (Golubeva, 1980). It may also be hypothesised that individuals with a higher level of cortical arousal would also show a better performance on intelligence tests. Regarding hypothetical neuroanatomical features underlying stable individual differences in AF, the degree of myelination could play an important role. AR results from cyclic excitations in cortico-cortical and thalamo-cortical circuits involving certain numbers of interneurons. It can be argued that the duration of a cycle would be shorter with greater axonal and dendritic conduction velocity (given the same or a similar number of interneurons) and, hence, the freq of the resulting rhythm would be higher. R. Miller (1994) provided evidence that axonal conduction delay in cortico-cortical connections, rather than synaptic delay, is the major factor limiting EEG propagation velocity. In turn, conduction velocity in cortico-cortical connections is mainly determined by the degree of axonal myelination. This interpretation of the AF-intelligence relationship also seems to be consistent with the brain myelination hypothesis of intelligence (E.M. Miller, 1994). The findings of several experiments suggest that alpha frequency is an indicator of the speed of cognition and memory performance in particular. Early findings reported by Surwillo (1961, 1963a, 1963b, 1964a, 2964b, 1971) indicate that alpha frequency is significantly correlated with the speed of information processing as measured by reaction times (RT). Subjects with high alpha frequency show fast reaction times (RTs), whereas slow subjects have low alpha frequency (see Klimasch et al, 1996). These findings are in good agreement with the results from a variety of experiments from the laboratory of Klimasch which revealed that the alpha frequency of good memory performers is about 1 Hz higher than that of age-matched samples of bad performers (Klimesch, 1996; Klimesch, 1997; Klimesch et al, 1990a; Klimesch et al, 1990b; Klimesch et al, 1993a, Klimesch et al, 1993b). Because good performers are faster in retrieving information from memory than bad performers (Klimesch, 1994), these data indicate that alpha frequency is related to the speed of information processing or RT. These results also suggest that alpha frequency should be related to individual differences in intelligence which is an assertion that is supported by the data in this area of research (Anokhin & Vogel, 1996). All of these findings are based on inter-individual differences in alpha frequency. In, contrast, intra-individual differences or task related shifts in alpha frequency appear not to be related to the speed of information processing (Klimesch et al, 1996) because as asymmetric desynchronization in the broad alpha band

Neurobiology of Intelligence

81

(favouring the lower or upper band) will lead to a shift in power and thus to a distorted estimate of alpha frequency. In summarizing, the reported findings suggest that alpha frequency is an indicator of cognition and memory performance. This conclusion is also supported by the fact that alpha frequency increases from early childhood to adulthood and then decreases with age over the remaining life span in a similar way as brain volume and general cognitive performance (e.g. Bigler et al, 1995; Willerman et al, 1991)

EEG POWER AND INTELLIGENCE Another parameter of the EEG that has been investigated as a possible predictor of intellectual ability is known as power. EEG power refers to the size or magnitude of the signal in that particular frequency range. EEG power measures can be further divided in to relative EEG power and absolute EEG power. Bosaeus et al (1977) investigated the relationship between EEG parameters and intelligence in a large sample of neurologically normal children (aged between five and sixteen years). The reported that the only significant relationship between IQ scores and EEG power when age was included in the analysis as a covariate, was a weak negative correlation between IQ and theta power in the parieto-occipital area. They found no relationship between IQ and delta, alpha, or beta power. In a similar study, Gasser, Von Lucadou-M¨uller, Verleger and B¨acher (1983) investigated the relationship between intelligence scores and EEG parameters in a group of normal and a group of mildly mentally retarded 10–13 year old children. Gasser et al. standardised the EEG parameters for age, and therefore proposed that large EEG parameter values were associated with a relative advancement of brain function as measured by the EEG. Like Bosaeus et al. (1977), they also reported a relationship between IQ scores and age-standardised absolute theta power. They also found a series of correlations between IQ scores and the age standardized absolute alpha1 (7.5–9.5 Hz) power, and age standardised relative beta1 (12.5–17.5 Hz) and beta2 (17.5–25.0 Hz) power. These correlations were found to be higher for the mentally retarded children than for the normal children, and were reduced when age was not taken into consideration. The vast majority of the significant correlations were positive, which Gasser et al. interpreted as indicating that higher IQ scores were associated with a maturational advancement in brain functions. Corning, Steffy and Chaprin (1982) also found evidence for a relationship between the slow wave frequencies and intelligence. They ranked and sorted a group of normal children and children suffering from attentional problems, hyperactivity, and learning and cognitive disorders according to the relative power of their EEG slow wave indices (delta, delta and theta, and the theta/alpha ratio), into either the diffuse slow frequency (DSF) group or the normal frequency (NF)

82

cindy van rooy et al.

group. Discriminant function analyses of the Revised Weschler Intelligence Scale for Children (WISC-R) profiles significantly separated the two EEG groupings. Corning et al. reported that by using the WISC-R subtest scaled scores, and the verbal and performance IQ, they could correctly classify 88% of the children into their EEG group. The subjects clustered into group 1 had lower than average intelligence scores, 95% had at least one subtest score that was six standard points or lower (half of these were on the information subtest), 90% belonged to the DSF group and none of the DSF group were normal subjects. In contrast, group 2 generally had higher intelligence scores, less than 24% had a subtest score of six or lower, 55% were from the NF group, and 14 of the 19 normal subjects were included in this group. Corning et al. proposed that the EEG features and the WISC-R patterns combine to suggest that the diffuse slow frequency profile may reflect a “maturational lag”, which is consistent with Gasser et al.’s (1983) interpretation of their correlational data. Giannitrapani (1988) also reported a series of positive correlations between EEG power and IQ scores in a sample of children, but these correlations were not in the slow wave frequencies. Giannitrapani studied 16 EEG frequency bands (1–32 Hz in 2 Hz bands), and found that the power in the 13 Hz band (which included activity from 12 to 14 Hz), showed the greatest correlations with IQ scores. Giannitrapani pointed out that the 13 Hz band is not to be confused with EEGdominant activity which was investigated in previous EEG-intellectual function studies. The 13 Hz activity has an amplitude many times smaller than dominant alpha activity which for this group was represented in the power of the 11 Hz band. The study did indicate that intellectual function was related to the dominant EEG activity, as there were a series of low-moderate strength positive correlations found between power in the 11 Hz band and IQ scores, they were just not as strong as those with the 13 Hz band. Giannitrapani (1988) proposed that alpha activity relates to the portion of verbal activity which is dependant on old funds of knowledge Whereas, the strongest correlations in the 13 Hz band was with the comprehension subtest of the WISC, which may indicate that it plays a role in facilitating conceptual ability. Marosi et al.’s (1999) study of EEG parameters in a sample of male adults also investigated topographical differences. Marosi et al. investigated the relationship between the absolute and relative power in six broad band parameters, delta, theta, alpha and beta when relaxed with eyes open, and scores on the WAIS scales. The few significant correlations they found for the absolute power measure included a positive correlations between delta power in the right frontal area and the digit span and picture completion subtests, and between alpha power in the parietal and occipital area and the coding and block design subtests. They also found a negative relationship between alpha power in the left frontal area and picture arrangement. There were also a limited number of correlations between the relative EEG power parameters and the WAIS scales. For example, Marosi et al. reported positive correlations between theta power and coding in the right central area, and alpha

Neurobiology of Intelligence

83

power in the left temporal area and the arithmetic subscale. Negative relationships were found between theta power in the left temporal area and performance IQ, theta power in the right occipital area and picture completion, and alpha power in the right central area and digit span. Marosi et al. (1999) point out that while they only found a small number of significant correlations, their data were suggestive of relatively high correlations between the variables. Most of the correlations with absolute EEG power were observed in the frontal regions. The association cortices are involved in many aspect of higher functions, including cognition, emotional behaviour, memory and language, though all these functions require the integrated action of neurons in different regions. Several studies have also investigated if there are differences in the EEG alpha power evinced in intellectually gifted and average intelligence individuals. Alexander, O’Boyle, and Benbow (1996) reported than gifted young adolescents displayed less alpha power when focussed on a fixation point than average young adolescents. In addition, the overall alpha power shown by the gifted adolescents was of a similar level to that displayed by a group of young adults (college students), although, the pattern of alpha power varied between the two groups. This pattern of differences was such that college and gifted subjects had the same right hemisphere alpha power superiority at the frontal and occipital locations. However, at temporal and parietal recording sites the college students exhibited greater left hemisphere alpha power compared to gifted students who evinced greater right hemisphere alpha power. Alexander et al. proposed their results may indicate that gifted subjects may have some form of developmentally advanced levels of alpha power. The lack of difference between the alpha power in the frontal and occipital lobes of the gifted adolescents and college age students may indicate that in the two groups have a similar level of brain maturation for these regions. It is not unreasonable to suggest that gifted adolescents may be more physiologically advanced than average ability adolescents in either brain organization, development, or utilization of brain resources. Jauˇsovec (1996, 1988) also investigated alpha power and giftedness, but the results of these studies were not consistent with those of Alexander et al. (1996). In fact, in a sample of young adults, Jauˇsovec (1996) reported finding no difference between the gifted and average group while the subjects were relaxed with eyes closed whereas when the subjects were relaxed with their eyes open, the gifted group displayed overall higher alpha activity than the average individuals. Jauˇsovec (1996) also reported that this pattern also existed when the subjects were solving problems (but not reading and planning how to solve the problem), and memorizing lists of words or pictures. Jauˇsovec (1998) obtained similar results during short-term memory tasks and tasks requiring complex processes such as arithmetic and proportional reasoning. Both studies also reported that when the subjects were relaxed with their eyes open the differences between the gifted and average groups was more diffuse and did not involve any particular brain area. However, the differences between the two groups during the problem solving and memorizing tasks were concentrated in the frontal and parieto-occipital areas

84

cindy van rooy et al.

(Jauˇsovec, 1996; 1998). Alexander et al. also found evidence for the importance of the frontal and posterior areas in intellectual ability. Jauˇsovec hypothesised that the results indicated that average individuals worked harder than the gifted group when solving the problems (lower alpha power), and that the gifted individuals used specific task-relevant brain areas, while the average individuals activated brain areas irrelevant for superior task performance. A series of recent studies have used a relatively new technique for investigating changes in the EEG associated a cognitive task. This technique is known as eventrelated desynchronization (ERD), and it is based on the well-known phenomenon of a blocking or desynchronization of EEG-background activity within the alphaband. Pfurtscheller (1986) provided a method for quantifying and displaying the ERD in the form of time courses and maps. The technique involves investigating the alpha power during a task and calculating how much it decreases, or increases, in comparison to a baseline level. Neubauer, Freudenthaler, and Pfurtscheller, (1995) investigated the patterns of ERD associated with a sentence verification task (SVT) in a sample of adult males which requires the subject to read a sentence and then decide if the picture they are shown is consistent with the sentence. Neubauer et al. investigated changes in the EEG power in the alpha1 (8–10 Hz) and alpha2 (10–12 Hz) bands during different stages of the SVT, in comparison to the reference interval of the task. The results revealed that the subjects who scored in the average range of the Ravens Advanced Progressive Matrices (APM) showed a continuously increasing overall ERD in the alpha2 band (increasing cortical activation) from the start of the task right through to when they had to decide if the sentence was correct. Whereas, the overall ERD values of the high APM group stayed relatively constant across the task. Comparing the topographical differences in the alpha2 band between the two groups shed more light on the results, with the average group developing strong cortical desynchronization in the posterior areas of the cortex. [Figure 2]. This trend starts with the presentation of the sentence where we have the strongest activation at parietal derivations. Toward the end of the sentence phase, this activation begins to spread to occipito-temporal and central areas. This strong activation of the whole posterior and central cortex remains throughout the blank interval and when the picture is presented. In the frontal cortex, the low IQ participants, show an increase in activation until the end of a trial, indicating that these individuals display a relatively unspecific increase in activation over the whole cortex. This conclusion is especially corroborated by a comparison with the spatiotemporal patterns of high IQ individuals: This group also displays an activation of the posterior areas increasing with time on the sentence trial, but it can be seen that even in these areas there is less ERD compared to the low IQ group. The most remarkable finding for the high IQ group, however, is the deactivation of the frontal cortex with increasing time on the task. This frontal region, which is presumably scarcely involved in SVT performance, therefore seems to provide the best distinction between low and high IQ participants.

Neurobiology of Intelligence

85

IQ low

IQ high

PSI

CRT

BLANK

−36

VRT

72 % Figure 2. ERD and intelligence.

The alpha1 band revealed a similar pattern of results, although none of the differences were significant. This agrees well with a recent interpretation of the functional meaning of the 2 alpha-bands (e.g. Pfurtscheller & Klimesch, 1991): The ERD in the alpha-2-band is a better indicator of task-specific cortical activation, whereas the ERD in the alpha-1-band is more diffuse and widespread. Similar results were obtained in a later study by Neubauer, Sange, & Pfurtscheller, (1999), who used a letter matching task rather than a SVT. Once again the differences between the IQ groups was mainly apparent in the alpha2 band, and lower scores on the APM were associated with greater ERD values, although only in the more difficult semantic letter matching task. Again, the topographical distribution of the ERD changes during the letter matching task revealed that the average IQ group displayed a rather unspecific ERD increase from the easier (physical appearance letter matching) to the more difficult (semantic letter matching) task over the whole cortex. The high IQ group, on the other hand, showed a relatively uniform activation of the whole cortex in the easier task (with a focus at posterior-especially at the occipital sites), exhibited a clear-cut (and somewhat stronger) activation center at parietal and occipital sites in the more difficult test and—at the same time—they showed decreased activation in the frontal area. The main finding from both of the above studies (Neubauer et al., 1995; Neubauer et al, 1999), was an inverse relationship between psychometric

86

cindy van rooy et al.

intelligence and cortical activation during performance of an elementary cognitive task. Although, this was only apparent during tasks that were cognitively demanding in nature (Neubauer et al., 1999). The authors, (Neubauer et al., 1995; Neubauer et al, 1999), hypothesized that their finding of a decrease in cortical activity in the frontal areas for the high IQ groups could give a clue to a possible extension of the concept of neural efficiency. The higher neural efficiency of brighter individuals seems to be characterized not only by their presumed use of fewer neurons or neuronal circuits within those areas required for task performance, but also the ability to focus the cortical resources on those cortical areas that are required for task performance and- at the same time-to deactivate or inhibit other areas. The most important conclusion is that the amount of EEG power in the theta and alpha frequency range is related to cognitive performance and memory performance in particular, if a double dissociation between absolute and evert-related changes in alpha and theta power is taken into account. This double dissociation is characterized by the fact that during a resting state: 1. small theta power but large alpha power (particularly in the freq range of the upper alpha band) indicates good performance, whereas the opposite holds true for event-related changes, where 2. a large increase in theta power (synchronization) but a large decrease in alpha power (desynchronization) reflect good cognitive and memory performance in particular

EEG COHERENCE AND INTELLIGENCE The previously discussed studies have mainly focussed on the EEG parameters that index local cortical activation. Although, an increasing body of evidence from behavioural and cognitive neuroscience suggests that complex goal-directed behaviour, including cognitive test performance, is underlied by a dynamic integration of anatomically distributed, but functionally related neuronal groups. This spatiotemporal ‘binding’ of activities in specialized cortical areas is achieved through coherent oscillatory processes in neural networks (e.g. Livanov, 1977; Singer, 1994; Birbaumer et al., 1995; Klimesch et al., 1997; Tononi & Edelman, 1998; Anokhin, Lutzenberger, & Birbaumer, 1999. Therefore, more recent research has investigated measures of EEG coherence and measures indicating deterministic chaos (e.g. dimensional complexity or Kolmogorov entropy) in an attempt to understand the way in which the EEG patterns differ between regions. EEG coherence is a quantitative measure of the correlation in frequency domain between EEG signals recorded from different cortical sites (Anokhin, Lutzenberger, & Birbaumer, 1999). Past studies have provided evidence that EEG coherence can serve as an index of functional integration of specialized cortical

Neurobiology of Intelligence

87

areas (Anokhin et al., 1999). While both the dimensional complexity (DCx) or Kolmogorov entropy (K2 ) measures are derived from the theory of non-linear dynamics and provide an index of the ‘chaoticness’ of the EEG (Jauˇsvec, 1998; Anokhin et al., 1999). Research has suggested that dimensional complexity can indicate the overall complexity of brain dynamics in terms of the relative number of concurrently activated and competitively interacting neuronal assemblies, i.e. differentiation of neural activity (Anokhin et al., 1999). One of the advantages of using the dimensional complexity measure is that because it is derived from theories of deterministic chaos, it can capture unpredictable, but nonetheless regular recurrent EEG frequency patterns not detectable by traditional EEG analyses based on frequency bands in the power spectrum (Lutzenberger, Birbaumer, Flor, Rockstroh, & Elbert, 1992). Recent studies have provided evidence that coherent oscillations in different frequency bands, most notably theta and gamma, play a crucial role in the dynamic functional integration of brain structures involved in ongoing mental activity. Cognitive tasks typically produce desynchronization of alpha activity and enhancement of theta activity (e.g. Rugg & Dickens, 1982; Gr¨unwald et al., 1999). More specifically, EEG synchrony in the lower frequency band (theta) appears to facilitate functional connections in tasks with high demand for WM and focussed attention (Anokhin et al., 1999). Anokhin et al. hypothesized that individuals who exhibit greater large-scale spatial EEG synchronization during cognitive tasks possess a greater capacity for establishing functional connections between specialized cortical regions involved in mental activity and would therefore show a better performance on intelligence tests. Lutzenberger et al (1992) found adults with higher IQ scores showed greater DCx values when resting with their eyes open than those with lower IQ scores, although this difference disappeared during visual imagery. Lutzenberger at al. hypothesized that during resting periods persons with lower levels of intelligence may be characterized by a relative reduction in the variability of cortical activity with less dynamic alterations in the activity of neuronal call assemblies, and thus lower DCx governing the EEG oscillations. In contrast, Anokhin et al (1999) and Jauˇsvec, (1998) did not find any evidence to support Lutzenberger et al. Anokhin et al found little evidence for a relationship between the DCx measures while resting and intelligence scores in a group of young adolescents and Jauˇsvec found no difference between a group of gifted young adults and a group of average intelligence young adults on the K2 measure of deterministic chaos. In contrast, both Jauˇsovec (1998) and Anokhin et al (1999) found a negative relationship between the deterministic chaos measure during cognitive tasks and intelligence scores. Jauˇsovec found that the gifted group appeared to use processes which displayed a similar complexity of neural mess activity while solving the 10 cognitive tasks, while the average group displayed rather different complexity patterns while solving the 10 tasks. Jauˇsovec proposed that the results indicated

88

cindy van rooy et al.

that average individuals involved brain areas irrelevant for good task performance. Anokhin et al reported that higher intelligence scores in a sample of young adolescents were associated with lower DCx scores during a verbal task and a visuospatial task. They also found that most of the significant correlations were evident in the parietal and occipital areas. Anokhin et al (1999) also found that during the verbal and visuo-spatial tasks measures if EEG coherence in the theta band was positively correlated with intelligence scores. There were also a small number of correlations with the alpha band. The highest correlations with intelligence scores were observed for the frontooccipital connections in the right hemisphere for the verbal and spatial conditions. Anohkin et al. (1999) suggested that relationship between the deterministic chaos and coherence measures of EEG may serve as indicators of a single underlying dimension of individual differences in brain function, e.g. the overall order-to-chaos ratio, or ‘orderliness’, of task-related brain dynamics. Subjects with higher level of cognitive development may be characterized by a more ‘organized’, spatially and temporarily co-ordinated electrocortical activity wile performing a mental task. The conclusion from the studies reviewed so far is that the mature brain at or beyond an age of about 16 is characterized by an increase in absolute power in the upper alpha band and a decrease in theta and delta power as compared to a less developed brain in younger children. A series of other studies support this view and show further that children with poor education (Harmony et al, 1990), with reading, writing/disabilities (Byring et al, 1991), with spelling disabilities (Byring et al, 1991) or with other types of neurological disorders (Schmid et al, 1997) show significantly more delta and theta but less alpha power (for a review see also Schmid et al, 1997). The frequency of the dominant background (or alpha) rhythm gradually increases until it attains its maximum value at about age 10 years, with a mean frequency of 10 Hz. Slow activity gradually decreases until ages 25 to 30 years, when the ‘normal’ adult EEG pattern is established. This age coincides with the final state of myelination (Courchesne, 1990).

SUMMARY OF STUDIES FINDINGS Generally, the early studies have reported that a low intelligence quotient (IQ) was related to slow alpha frequencies and the existence of delta and theta rhythms, while higher levels of intelligence was related to faster alpha frequencies and a lack of delta and theta rhythms (Vogel & Broverman, 1964), but later studies indicate that the relationship is more complex. If the EEG of the mature brain of young healthy adults is compared either with the developing brain, the aging brain or the brain which is affected by neurological diseases of various kinds, the conclusion is that:

Neurobiology of Intelligence

89

r alpha frequency is positively related to cognitive performance, and r large power in the range of the upper alpha band but small power in the theta frequency range indicate good cognitive peformance. These conclusions are based on findings which show that: r alpha frequency increases from early childhood to adulthood but then decrease with increasing age or age related neurological diseases r alpha frequency is lowered in demented subjects (as well as in patients with other types of neurological disorders), r alpha frequency is significantly higher in subjects with good memory performance as compared to age matched controls with bad memory performance r alpha frequency is positively correlated with speed of processing information r theta power decreases and alpha power increases from early childhood to adulthood, r theta power increases and upper alpha power decreases during the late part of the lifespan r theta power is enhanced and alpha power lowered in subjects with a variety of different neurological disorders as compared to age matched controls.

EVOKED POTENTIALS AND INTELLIGENCE Evoked potentials (EPs) are a variation of spontaneous EEG. They involve averaging several brief EEG segments time registered to the stimulus and or response in a simple cognitive task. Averaging enhances signals that are temporally consistent with respect to the event, and it suppresses unrelated background activity. An averaged EP waveform consists of a series of positive and negative waves. A substantial change in latency, amplitude, or duration of an EP peak or wave between experimental conditions that differ in one specific cognitive factor is assumed to reflect the mass neural activity associated with that cognitive factor (Gevins & Cutillo, 1995). We now review some of the research examining the relationships between Event related potentials or EPs, the Steady State Visual Evoked Potentials or SSVEPs and intelligence respectively.

EVOKED POTENTIALS AND INTELLIGENCE The first systematic attempt to relate ERP parameters to IQ was by Ertl and his co-investigators (i.e. Barry & Ertl, 1966; Chalke & Ertl, 1965; Ertl, 1971; 1973; Ertl & Schafer, 1969). Employing crude estimation techniques in the identification

90

cindy van rooy et al.

of visual ERP components, Barry and Ertl (1966) reported correlations of −.88 and −.76 between component latency and IQ in college students. Chalke and Ertl (1965) confirmed that shorter latencies were associated with higher intelligence in a more representative sample of IQ scores. Ertl (1966) reported correlations of −.70 between latency of the third negative peak and IQ in 100 subjects with IQs ranging from 77 to 136. A subsequent replication (Ertl, 1969) obtained a correlation of −.51 between the latency of the third negative peak and IQ in 300 children. In the largest study to date, Ertl and Schafer (1969) tested 573 primary school children (317 males, 256 females; Grades 2 through 8) with the Wechsler Intelligence Scale for Children (WISC), the Primary Mental Abilities test (PMA) and the OTIS test of mental abilities, as well as a visual ERP task in which the subject was required to attend but not to respond to brief photic stimuli. Highest correlations were obtained with the latency of the third and fourth peak and the WISC, PMA, and OTIS scores respectively (−.35 and −.33; −.34 and −.32; −.35 and −.35). Although these three tests of intelligence each pose very different problems for the subject, their correlations with the latency of the third and fourth peaks were nearly identical, indicating that the latency of the third and fourth components may tap some general process common to all tests. Ertl and Schafer (1969) claimed that this finding reflects the role of speed of information processing within the brain and this is a concept further developed by Hendrickson and Hendrickson (1980) in their biological model of intelligence. Other researchers have made similar suggestions when attempting to explain the processes underlying IT and RT and their relationship to IQ (Brand & Deary, 1982; Eysenck, 1987; Eysenck & Barrett, 1985). Other studies have provided evidence for an ERP latency-IQ relationship. In a notable experiment, Schucard and Horn (1972) reported correlations between latency measures of the ERP and a battery of psychological tests designed to measure crystallised and fluid abilities, as well as factors relating to speed and level measurements of fluid ability (see Furneaux 1952, 1961). One hundred and eight subjects (60 males, 48 females, with ages ranging from 16 to 68) attended three ERP conditions in which the level of attention and arousal was systematically manipulated from a requirement to; firstly to respond (high arousal); secondly not to respond but to count (medium arousal); and, thirdly, not to respond or count but to attend (low arousal). Nearly all the correlations between the IQ tests and latency were negative, ranging from −.15 to −.32, with shorter latency associated with higher IQs. Most significant correlations with IQ were obtained with latencies of the later ERP components in the low arousal condition, a finding consistent with the results from similar methodologies employed by Ertl and his co-workers. In fact, some researchers have warned that significant correlations can only be obtained between ERP measures and IQ if the attentional/arousal level of subjects is held at a relatively low level (Eysenck & Barrett, 1985; Schucard & Horn, 1972). Callaway (1973) has also reported significant negative correlations between P3 latency and

Neurobiology of Intelligence

91

IQ in 191 Naval recruits (although much of the detail concerning this experiment was not specified). Rhodes, Dustman and Beck (1969) examined the relationship between ERP measures (latency and amplitude) and IQ. From a sample of 800 children, two groups of 10 and 11-year-old children were selected, a high IQ group (with WISC scores ranging from 120 to 140) and the second with WAIS IQ scores in the borderline to normal range (70–90). One hundred visual evoked potentials (VEPs) were elicited on two occasions (separated by 1 month) at different light intensities (three levels of intensity). Only one of the latency measures (peak delay), recorded at the occipital site, resulted in significant differences between the two groups, with the N4 component occurring later in the low IQ group. Importantly, consistent differences were found between the two groups with respect to amplitude, the high IQ group displaying greater responses from both occipital and centrally elicited VEPs at all intensities. A subsequent manipulation of the ages of the subjects into two groups, reported by Dustman et al. (1976), so that each group contained children varying in ages from 4 to 15 years (Group 1: mean IQ = 110; Group 2: mean IQ = 88) resulted in few significant correlations between IQ and these ERP parameters. The authors concluded that intelligence is only weakly correlated to visual ERP amplitude, and only observable when age-related amplitude changes are controlled. Alternatively, differences in amplitude between low and high subjects may merely reflect different levels of attention. A subsequent experiment, reported by the same authors, showed that greater focussed attention was associated with increases in the amplitude of visual ERPs, indicating that amplitude-IQ differences may reflect attentional differences, with high IQ subjects generally more alert than low IQ subjects. This finding was therefore consistent with more recent work by Haier, Robinson, Braden, and Williams (1983), in which larger N1-P2 peak amplitude excursion was related to higher IQ. Overall, the results of these studies suggest that both ERP latency and amplitude are related to IQ under different stimulus conditions, and that the attentional demands and arousability of the stimulus has played some part in most of these studies reporting a significant ERP-IQ correlation. Overall shorter ERP latency appears to be related to higher IQ, a finding that may overlap with the consistent relationship reported by many researchers between Inspection Time (IT) which measures speed of processing a stimulus and intelligence test scores (see Deary & Stough, 1996 for a review). Hendrickson and Hendrickson (1980) reinterpreted Ertl’s results relating amplitude and latency to IQ and developed a model of neural functioning that relates to individual differences in intelligence. According to these authors the ERP waveform represents electrical activity measured on the scalp and this measurement was regarded by A.E. Hendrickson as reflecting the initial pulse activity and its subsequent propagation through the cortex. Thus, the ERP waveform was regarded as a description of the individual pulse trains that are set off by the stimulus. Pulse trains

92

cindy van rooy et al.

are transmitted from one neuron to the next through synaptic transmission, with errors in propagation of the initial activity resulting in the ERP waveform showing degradation, from one of complexity to simplicity. According to the Hendricksons, it is possible to obtain a measure of the complexity of the ERP which reflects the accuracy of sensory transmission through the cortex by measuring the length of the ERP’s contour, which they termed the string-length. It is this measure that they have related to IQ test performance. Those subjects with complicated ERP waveforms are held to have the ability to process information within the brain more accurately (and quickly) and propagate information within the brain with less error, than those subjects exhibiting less complicated ERP waveforms. To examine the relationship between the string length and IQ, D. E. Hendrickson (1982) re-analysed data presented by Ertl and Schafer (1969). Results are shown in Figure 7. Among 20 children (10 high WISC IQ and 10 low WISC IQ) the string length measure correlated +.77 with IQ, accounting for more than half the IQ variance. Blinkhorn and Hendrickson (1982) also reported significant correlations between string length and scores on the APM among students. ERPs from 100 tones (1–8 second pseudo-random inter-stimulus interval (ISI), 85 dB auditory tone pips) were elicited, although only 90, 64, and 32 trials were used for later analyses. Correlations between the APM and the string length measured +.50, +.36, and +.45 for 90, 64 and 32 trials (averaged) respectively. Other verbal tests (Verbal Concepts and Verbal Critical Reasoning and various divergent thinking tests) were also administered, but did not significantly correlate with the string length. Correcting the correlations for restriction in attenuation of APM scores increased the string length-IQ correlations to a range (.7 to .85) that is characteristic of the internal reliabilities of many IQ tests (Eysenck & Barrett, 1985). D. E. Hendrickson (1982) has reported the most convincing support for the string length measure to date. In a sample of 219 school children (age; Mean age = 15.6 years, SD = 1.13; FSIQ Mean = 108, no SD published) she reported a Pearson correlation between the string length and WISC FSIQ of +.72. Two other ERP measures were also used a variance measure (a measure of the electrophysiological variability of each trial) and a composite measure (variance minus the string length). These measures correlated −.72 and −.83 with FSIQ respectively. The WISC subtests correlated between .3 and .8 with the ERP measures, with the overall result providing further support for the preliminary study reported by Blinkhorn and Hendrickson (1982). Since the Hendrickson (1982) and Blinkhorn and Hendrickson (1982) studies, there have been only a few studies attempting to replicate and extend the string length-IQ relationship. This is surprising, given the potential importance of the string length measure for intelligence theory. Shagass, Roemer, Straumanis and Josiassen (1981) found no significant relationship between auditory, visual and somatosensory string length measures and IQ (Raven’s Progressive Matrices) in 20 subjects ranging in age from 18 to 49. Although significant correlations were

Neurobiology of Intelligence

93

obtained between the string length measure and APM scores, the authors reported larger associations between amplitude measures, especially the peak N1-P2 amplitude excursion (up to .69; p < .0005). They argued that the highest IQ-Amplitude correlations are seen at optimal stimulus intensities. Although the N1-P2 peak amplitude excursion measure was highly correlated with the string measure, ranging from .74 to .80 for different string epochs, when the N140-P200 measure was partialled out of the string length-IQ correlations, this association became small and negligible. This finding would then suggest that the string length parameter is only an “epiphenomenon” of the N140-P200 amplitude excursion differences. In a similar study, Robinson et al. (1984) replicated the results of the Haier et al. (1983) study, but only when 12 subjects (out of 27) were excluded from the analysis because of age, sex and measurement error. Importantly, significant correlations between the string length and IQ were only obtained for female subjects. It should be pointed out that 22 of the 23 subjects tested by Haier et al. (1983) were also females. Caryl and Fraser (1985) have obtained a correlation of .8 between the string length and APM in 10 subjects (IQs ranging from 105 to 140) using stimulus parameters and procedures faithful to the Hendricksons’ experiments. There have been a few other studies examining the string length-IQ relationship (such as Stough, Nettelbeck & Cooper, 1991; Widaman, Carlson, Saetermore and Galbraith, 1993) but on the whole, there is little conclusive evidence that has been systematically replicated linking the string length with IQ. Overall, ERP measures have at best provided some reliable but modest correlations linking shorter component amplitudes with higher psychometric intelligence (particularly the earlier components).

STEADY-STATE VISUALLY EVOKED POTENTIALS AND INTELLIGENCE A technique known as steady-state probe topography (SSPT) has also provided some evidence that the patterns of cortical activity elicited during cognitive tasks are different for individuals of differing levels of intelligence. SSPT is a brain imaging technique that involves using a probe ERP to investigate changes in the steady-state visually evoked potential (SSVEP), which have been shown to be related to cognition (Silberstein et al., 1990). SSVEPs are cortical responses to rapidly repetitive stimuli that consist of individual sinusoidal components at both the stimulus frequency and multiples of it, whose amplitude and phase should remain relatively consistent over time (Regan, 1989). The rationale for the probe ERP technique involves both the probe stimulus and the cognitive processes vying for the cortical resources that are available (Papanicolaou & Johnstone, 1984). Silberstein (1995) has proposed that changes in the amplitude and latency of the SSVEP may occur as a result of neuromodulatory differences in

94

EasyTrials

cindy van rooy et al.

MediumTrials

DifficultTrials

+0.5 +0.3 ( p=0.05)

0.0 - 0.3 ( p=0.05)

-0.5 Figure 3. Correlation between SSVEP amplitude and WAIS-R full scale IQ: Raven Advanced Progressive Matrices n = 31

cortico-cortical re-entrant loops. These cortico-cortical re-entrant loops have been proposed to play a vital role in regulating the frequency of EEG rhythms in the alpha bandwidth (Silberstein, 1995). In the first study to examine SSPT parameters and intelligence Stough, Dennison, Celi & Silberstein (2003) recorded the SSVEP during the Raven Advanced Progressive Matrices (APM) in 31 participants. The Wechsler Adult Intelligence Scale Revised (WAIS-R) was administered separately to the SSVEP recording and used to correlate with the brain activity. Interestingly the correlational brain maps (see Figure 3) indicated two areas of significant correlations between the WAIS-R scores and SSVEP latency measures during the APM which is a predominantly a measure of general intelligence and spatial analysis. These two areas provided opposite relationships with statistically significant negative latency-IQ correlations observed frontally and statistically significant positive latency-IQ correlations observed parietally. This pattern of results may indicate that frontal and parietal networks may be important during APM performance and that individual differences in IQ may relate to the preferential activation of different areas of the brain. Song, Stough and Silberstein (2000) recorded the SSVEP elicited by a 13 Hz visual flicker while participants performed a computerised version of the Ravens Standard Progressive Matrices (SPM). It was found that while performing the SPM, participants showed a decrease in the SSVEP amplitude in the occipito-parietal area that was greater in the more difficult SPM problems than the easier SPM problems. Song et al. also reported that this decrease in SSVEP amplitude was greater for those participants who scored in the average range on the 3rd edition of the Wechsler Adult Intelligence Scale (WAIS-3), than those who performed better on the WAIS-3. Song et al. also noted an increased SSVEP latency over the posterior areas during the more difficult problems, and an increased SSVEP latency in the frontal regions in the high IQ group (Figure 4). Reductions in SSVEP amplitude during a cognitive task have been associated with increase task-related

Neurobiology of Intelligence

95

SSVEP Latency during Stimulus Presentation LOW IQ GROUP 1

2

3

HIGH IQ GROUP 1

4

SPMA

SPMA

SPMB

SPMB

SPMC

SPMC

SPMD

SPMD

SPME

SPME

Warmercolours colours Warmer Latency decrease

-6.1

+6.1 Latency Difference

1=0.5 seconds pre-presentation 2=0.5 seconds post-presentation 3=1.5 seconds post-presentation 4=2.5 seconds post-presentation

2

3

4

Cooler colours Latency increase

Figure 4. SSVEP maps during the Raven Progressive Matrices.

cortical activity (Farrow et al., 1996; Silberstein et al., 1990; 1995). Song et al. proposed that the results were consistent with Haier et al (1992) ‘neural efficiency’ hypothesis, that suggests that it is not how hard the brain works that determines how intelligent the brain is, but how efficiently it works. Van Rooy et al. (2000, 2001) has also found evidence that the brains of individuals who have average scores on an intelligence test have to ‘work harder’ to solve problems than the brains of individuals with higher intelligence scores. Van Rooy et al. (2001) found that rehearsing spatial information in working memory task, was associated with an increased in SSVEP latency in the frontal areas, and an increase in SSVEP amplitude and decreased SSVEP latency in the parietal and occipital areas. These changes in the SSVEP were found to be greater in magnitude for those participants who scored higher on the WAIS-3. Van Rooy et al. (2001) speculated that high intelligence may be characterised by the quality of the frontal executive processes and the ability to store and rehearse spatial information in working memory in the posterior areas of the brain.

96

cindy van rooy et al.

Further support for the importance of the frontal lobes was found in a study that investigated changes in the SSVEP associated with a spatial and verbal N-back working memory task (Van Rooy et al., 2000a, 2001). The study reported that while manipulating and rehearsing spatial information, participants showed increased SSVEP amplitude in the prefrontal and occipito-parietal areas and a decreased SSVEP latency in the frontal area (Van Rooy et al., 2000a). The SSVEP changes in the frontal area were found to be greater during the higher demand 3-back task and for the average IQ group. Silberstein proposed that a decreased SSVEP latency in the prefrontal area may be indicative of an increase in the neural information processing speed and an increase in excitatory processes. Therefore, the greater decrease in SSVEP latency over the frontal area shown by the average IQ group may indicate greater usage of excitatory processes, which in turn, suggests that the average IQ group area ‘working harder’ to retain and manipulate the spatial information in their working memory (Van Rooy et al., 2000). A similar pattern of SSVEP amplitude changes were found when the participants were manipulating and rehearsing verbal information in working memory, with an increased SSVEP amplitude in the prefrontal and occipital areas, with the frontal SSVEP changes being greater in magnitude during the higher demand task and for the average IQ group. The SSVEP latency differences were not quite as consistent with the average IQ group displaying an increased SSVEP latency in the frontal area, while the high IQ group showed a decreased SSVEP latency in the posterior area, which extended over the frontal areas during the higher demand task.

METABOLIC IMAGING AND INTELLIGENCE There have not been a large number of studies conducted investigating intelligence using metabolic imaging techniques such as positron emission tomography (PET) or functional magnetic resonance imaging (fMRI). Unfortunately the large sample required to encompass a broad range of intelligence quotients makes these techniques very costly, and the invasive nature of the PET technique makes it difficult to obtain a representative sample. Metabolic imaging techniques such as PET and fMRI have the advantage of excellent spatial resolution. This enables researchers to investigate if the biological basis of intelligence manifests itself in the amount of cortical activation during a task or at rest, or the areas of the brain recruited to complete a task, or a combination of both. Chase et al. (1984) used PET to investigate the resting metabolic rate of a sample of Alzheimer’s Disease patients and a small selection of normal control subjects. They reported moderately strong positive correlations between the full score, verbal and performance IQ on the WAIS-R and cortical glucose metabolic

Neurobiology of Intelligence

97

rate (GMR). It was also reported that the verbal subtests correlated more highly with glucose use in the left parasylvian area, while performance subtests correlated more strongly with glucose use in the right posterior parietal area. In contrast to Chase et al. (1984) study, the majority of the studies investigating the metabolic correlates of intelligence have investigated the topography of changes in metabolic activity during an intelligence, reasoning or ability tests. Generally, these studies have reported a negative relationship between performance and overall cortical GMR during the Raven’s Advanced Progressive Matrices (APM) (Haier et al., 1988), a visuo-spatial motor task (the tetris game) (Haier et al., 1992a; 1992b), and a verbal fluency test (Parks et al., 1988). Haier et al. (1988) reported a moderate to strong negative correlation between performance on the APM and the absolute cortical GMR, suggesting that those subjects who scored lower on the APM required more cortical activity to perform the task than those who scored higher on the APM. Haier et al. (1988) hypothesised that this may have been because the lower IQ subjects use inefficient neural circuitry to solve the problem, either because they do not have the correct circuits or they cannot or do not access the correct circuits. In contrast, those subjects who perform better on the APM can access the most efficient circuits or my have more efficient circuits generally, and therefore use less energy (see Figure 5). A similar results was obtained by Parks et al. (1988) during a PET study of verbal fluency. Parks et al. reported a moderate strength negative relationship between performance on a verbal fluency test and overall cortical GMR in the frontal, temporal and parietal regions during the verbal fluency task. In a similar vein to Haier et al. (1988), they hypothesised that it was possible that those who performed well on the verbal fluency task used more efficient strategies than those who found the verbal fluency task more difficult. Therefore, those who found the task difficult had to work harder and therefore showed greater activation. These hypotheses gained further support from a series of studies by Haier et al. (1992a, 1992b), who reported that the cortical GMR during a visuo-spatial motor task (the tetris game) decreased after practice. The greatest decrease in cortical GMR was seen in the subjects who improved the most after practice on the task (Haier et al, 1992) and those with the highest scores on the Ravens APM (Haier et al., 1992). Larson et al. (1995) pointed out that in standard intelligence tests everyone receives the same items in the same order. Typically, some low-aptitude participants will experience difficulty with most of the items, whereas many of the high-aptitude participants will excel. From this, it is quite logical to assume that low-aptitude participants may be required to expend more effort (p269). Therefore, Larson et al. (1995) conducted a study where the cognitive task (the backward digit span) was adapted for each subject so as to have the subject performing at the 90% (easy) or 75% (hard) accuracy. It was proposed that this would remove the confounding factor of the low aptitude subject expending more mental effort than the high aptitude subject. In contrast to the earlier studies, Larson et al. reported that subjects

98

cindy van rooy et al.

Figure 5. Cortical GMR using Positron Emission Tomography during the Raven Advanced Progressive Matrices (APM). Glucose uptake in a low APM (left) and high APM (right) participants. From Haier et al. (1998).

with a higher APM score tended to show higher cortical GMR during the backward digit span task than the subjects with the lower APM scores. It was also reported that for the average APM group, cortical GMR decreased during the more difficult task, while the high APM group showed an increased cortical GMR during the harder task. Larson et al. suggested that the differences in cortical GMR during the harder task between the two groups may be the result of group differences in the problem solving strategies utilised.

TOPOGRAPHICAL CHANGES An early study by Risberg (1973) investigated the regional cerebral blood flow (rCBF) during cognitive tasks in a sample of patients who were neurologically normal, but were undergoing treatment for chronic alcoholism. Risberg, (1973) reported that during the reasoning tasks, the largest increases in rCBF were in the occipito-temporal, parietal, frontal and prerolandic areas.

Neurobiology of Intelligence

99

Prabhakaran (1997) used functional magnetic resonance imaging (fMRI) to investigate cortical activity during the performance of the Standard and Advanced Progressive Matrices (SPM and APM) in order to assess the cortical activation of fluid intelligence and reasoning. The participants attempted to solve three types of problems from the SPM and APM, one requiring analytic reasoning, another requiring figural or visuo-spatial reasoning and the third type involving simple pattern matching. When comparing the analytic and figural reasoning problems Prabhakaran (1997) found activation in both the right and left frontal areas and a predominantly left hemispheric activation in the parietal, temporal and occipital areas, which was associated with analytical reasoning. When comparing the figural and the matching problems, figural reasoning was associated with an increase in activation mainly in the right hemisphere in the frontal, parietal and temporal areas that was proposed to be associated with the spatial and abject working memory networks. Comparing the analytical and the matching problems indicated an increase in activity bilaterally in the frontal, parietal, temporal and occipital areas, as well as in the left parietal area, that was associated with fluid intelligence and reasoning. These findings were consistent with those found by Haier et al. (1988, 1992) in earlier PET studies, who also found activation in the parieto-temporal, occipital and frontal areas while performing the APM. The importance of the role of the frontal lobes in intelligence has also been supported by Duncan’s (1995) research into attention, the frontal lobes and intelligence. Prabhakaran (1997) also proposed that the findings suggested that the spatial, object and verbal working memory systems control fluid reasoning ability. In contrast there is a general consensus that the right prefrontal cortex plays an executive role in working memory, performing functions such as holding data and coordinating and integrating the sensory regions (Raichle, 1993; Wickelgren, 1997). This is compatible with previous research that has linked the prefrontal cortex to memory, learning and other higher functions (Wickelgren, 1997a). For example, Duncan (1995) proposed that general intelligence (Spearman’s g), attention and frontal lobe impairment, especially in the dorsolateral prefrontal cortex and anterior cingulate, are all closely related. Haier et al (1988) found negative correlations between the GMR and performance on the APM in the parieto-temporal and occipital areas and to a lesser extent in the frontal lobes. These latter two studies provide some discrepancy in terms of an agreement for a g based frontal lobe theory of intellectual functioning.

SUMMARY AND CONCLUSIONS Initial PET studies of intelligence, however, have produced somewhat counterintuitive evidence concerning localisation and direction of relationships. With regard to localisation, one would expect that the likely brain sites for problem

100

cindy van rooy et al.

solving (based on existing neuropsychological data) would also be the likely sites for intelligence-metabolism correlations. For example, research indicates that certain frontal and prefrontal brain areas are involved in directing attention, planning, holding stimuli in memory, and performing complex stimulus transformations (e.g. Goldman-Rakic, 1987; Owen, Downes, Sahakian, Polkey, & Robbins, 1990; Posner & Peterson, 1990; Roland & Friberg, 1985)-processes required to solve items on many common intelligence tests. Yet, despite the “logical” tie-in between certain specific brain regions and intelligence, PET results to date suggest more diffuse “whole-brain” relationships. To summarize, aptitude for complex problem solving seems to be associated with the whole brain rather than theoretically “appropriate” brain structures, and aptitude level is negatively correlated with brain glucose utilization (Larson, 1995). All these studies (resting during uptake, many using patients) suggest a stronger resting metabolism of high IQ individual, perhaps because they are “more mentally alert and spontaneously engaged in more cog activity when ‘at rest’ than the lower scoring subjects” (Vernon, 1993, p. 175). This hypothesis, however, has not remained uncontradicted: As the positive correlations reported above have been obtained mainly in clinical samples (partially combined with healthy controls), they could only ‘reflect. . . the extent of brain damage and not necessarily the relationship between quality of performance and brain work in the physiological normal brain” (Haier, 1993, p. 321). An interesting question in this area of research reflects the mechanisms by which an efficiency model could work. How do brains that actually use less energy function better? One possibility is that they are faster and less error prone because of more myelination. Because the myelin is chemically inactive (serving roughly the same function as the insulator on a cable), it uses very little glucose. Most energy is used in the movement of ions in and out of axons. Thus, lower energy use in the more intelligence could be merely the result of more relatively inert myelin. Alternatively, more intelligent brains are more ‘efficient’ and somehow activate fewer neurons for any given problem. These results appear strongest in non-diseased individuals (e.g., Haier et al). Studies that have included diseased individuals often show a positive correlation between energy use and intelligence, which is usually attributed to cell death or damage reducing both- energy use and intelligence (see Miller, 1994 for a discussion on this topic). More recent studies that have employed technologies that more directly assess neural processes whilst a cognitive test is being undertaken tend to suggest the importance of frontal and parietal areas. The majority of activation tasks have involved the Raven Progressive Matrices, both a high g loaded task and a task requiring detailed spatial analysis. It is probably not surprising that the results of studies using the Raven tests (SSVEP, fMRI etc) indicate the importance of frontal working memory circuits as well and parietal spatial analysis circuits. The synchrony of these circuits during intelligence tests is worthy of further research

Neurobiology of Intelligence

101

as well as studies that assess brain activation during auditory intelligence test performance. Although the whole brain and efficiency hypotheses cannot be ruled out, evidence is emerging for the importance of frontal and parietal networks for intelligence.

REFERENCES Alexander, J.E., O’Boyle, M.W., & Benbow, C.P. (1996). Developmentally advanced EEG alpha power in gifted male and female adolescents. International Journal of Psycholphysiology, 23, 25–31. Anokhin, A.P., Lutzenberger, W., & Birbaumer, N. (1999). Spatiotemporal organization of brain dynamics and intelligence: and EEG study in adolescents. International Journal of Psychophysiology, 33, 259–273. Anokhin, A. & Vogel, F. (1996). EEG alpha rhythm frequency and intelligence in normal adults. Intelligence, 23, 1–14. Barry, W. M., & Ertl, J. P. (1966). Brain waves and human intelligence. In F. B. Davis (Ed.), Modern Educational Developments: Another Look. New York: Educational Records Bureau. Blinkhorn, S. F., & Hendrickson, D. E. (1982). Averaged evoked responses in psychometric intelligence. Nature, 295, 596–597. Bosaeus, E., Matouˇsek, M., & Peters´en, I. (1977). Correlation between paedopsychiatric findings and EEG-variables in well-functioning children of ages 5–16 years. Scandinavian Journal of Psychology, 18, 140–147. Caryl, P. G., & Fraser, I. C. (1985). The Hendrickson string length and intelligence-a replication. Paper presented at the Psychophysiology Society Scottish Conference, September. Ceci, S.J. (1990). On the relation between microlevel processing efficiency and macrolevel measures of intelligence: some arguments against current reductionism. Intelligence, 14, 141–150. Chen, A.C.N. & Buckley, K. (1988). Neural perspectives of cerebral correlates of giftedness. Inernational Journal of Neuroscience, 41, 115–125. Corning, W.C., Steffy, R.A., & Chaprin, I.C. (1982). EEG slow frequency and WISC-R correlates. Journal of Abnormal Child Psychology, 10(4), 511–530. Deary, I. J., & Stough, C. (1996). Intelligence and Inspection Time: Achievements, Prospects, and Problems, American Psychologist, 51, 599–608. Ellingson, R.J. (1966). Relationship between EEG and test intelligence: a commentary. Psychological Bulletin, 65(2), 91–98. Ellingson, R.J. & Lathrop, G.H. (1973). Intelligence and frequency of the alpha rhythm. American Journal of Mental Deficiency, 78(3), 334–338. Ellingson, R.J., Wilcott, R.C., Sineps, J.G., & Dudek, F.J. (1957). EEG frequency pattern variation and intelligence: a correlational study. Electroencephalography and Clinical Neurophysiology, 9, 657–660. Ellingson, R. J. (1966). Relationship between EEG and test intelligence: A commentary. Psychological Bulletin, 65, 91–98. Ertl, J. P. (1966). Evoked potentials and intelligence. Revue Uni Ottawa, 36, 599–607. Ertl, J. P. (1969). Neural efficiency and human intelligence. Final report, US Office of Education, Project No. 9-0105. Ertl, J. P. (1971). Fourier analysis of evoked potential and human intelligence. Nature, 230, 525–526. Ertl, J. P. (1973). IQ, evoked responses and Fourier analysis. Nature, 241, 209–210. Ertl, J. P., & Schafer, E. W. P. (1969). Brain response correlates of psychometric intelligence. Nature, 223, 421–422.

102

cindy van rooy et al.

Everhart, J.D., China, C.L., & Auger, R.A. (1974). Measures of EEG and verbal intelligence: an inverse relationship. Psychological Psychology, 2, 374–378. Eysenck, H. J. (1982). A Model for intelligence. New York: Springer. Eysenck, H.J. (1986). Toward a new model of intelligence. Personality and Individual Differences, 7(5), 731–736. Eysenck, H.J. (1988). The concept of “intelligence”: useful or useless? Intelligence, 12, 1–16. Eysenck, H. J., & Barrett, P. (1985). Psychophysiology and the measurement of intelligence. In C.R. Reynolds, and V.L. Willson, (Eds.) Methodological and Statistical Advances in the Study of Individual Differences. New York: Plenum Press. Fischer, D.G., Hunt, D., & Randhawa, B.S. (1982). Spontaneous EEG correlates of intellectual functioning in talented and handicapped adolescents. Perceptual and Motor Skills, 54, 751–762. Gasser, Th., Von Lucadou-M¨uller, I., Verleger, R., & B¨acher, P. (1983). Correlating EEG and IQ: a new look at an old problem using computerized EEG parameters. Electroencephalography and Clinical Neurophysiology, 55, 493–504. Gevins, A.S. & Cutillo, B.A. (1995). Neuroelectric measures of mind. In P. Nunez (Ed.) Neocortical dynamics and human EEG rhythms (pp. 304–338). New York: Oxford University Press. Giannitrapani, D. (1969). EEG average frequency and intelligence. Electroencephalography and Clinical Neurophysiology, 27, 480–486. Giannitrapani, D. (1988). The role of 13-Hz activity in mentation. In D. Giannitrapani & L Murri (Eds.) The EEG of mental activities (pp. 149–152). Basel, Switzerland: Karger. Haier, R. J., Robinson, D. L., Braden, W., & Williams, D. (1983). Electrical potentials of the cerebral cortex and psychometric intelligence. Personality and Individual Differences, 4, 591–529. Haier, R. J., Siegel, B.V., Nuechterlein, K. N., Hazlett, E., Wu, J. C., Paek, J., Browning, H., & Buchsbaum, M. S. (1988) Cortical Glucose metabolic rate correlates of abstract reasoning and attention studied with positron emission topography. Intelligence. 12, 199–217. Hendrickson, A. E. (1982a). The biological basis of intelligence: Part II Measurement. In H.J. Eysenck, (Ed.), A Model for Intelligence. New York: Springer-Verlag. Hendrickson, D. E. (1982b). The biological basis of intelligence: Part 1 Theory. In H.J. Eysenck, (Ed.), A Model for Intelligence. New York: Springer-Verlag. Hendrickson, D. E., and Hendrickson, A. E. (1980). The biological basis for individual differences in intelligence. Personality and Individual Differences, 1, 3–33. Jauˇsovec, N. (1996). Are gifted individuals less chaotic thinkers? Personality and Individual Differences, 25, 253–267. Jauˇsovec, N. (1998). Differences in EEG alpha activity related to giftedness. Intelligence, 23, 159–173. Klimesch, W. (1999). EEG alpha and theta oscillations reflect cognitive and memory performance: a review and analysis. Brain Research Reviews, 29, 169–195. Knott, J.R., Friedman, H., & Bardsley, R. (1942). Some electroencephalographic correlates of intelligence in eight-year- and twelve-year-old children. Journal of Experimental Psychology, 30, 300–391. Lutzenberger, W., Birbaumer, N., Flor, H., Rockstroh, B., & Elbert, T. (1992). Dimensional analysis of the human EEG and intelligence. Neuroscience Letters, 143, 10–14. Marosi, E., Rodr`ıguez, H., Harmony, T., Ya˜nez, G., Rodr`ıguez, M., Bernal, J., Fern`andez, T., Silva, J., Reyes, A., & Guerrero, V. (1999). Broad band spectral EEG parameters correlated with different IQ measurements. International Journal of Neuroscience, 97, 17–27. Miller, E.M. (1994). Intelligence and brain myelination: a hypothesis. Personality and Individual Differences, 17(6), 803–832. Mundy-Castle, A.C. (1958). Electrophysiological correlates of intelligence. Journal of Personality, 26, 184–199. Neubauer, A., Freudenthaler, H.H., & Pfurtscheller, G. (1995). Intelligence and spatiotemporal patterns of event-related desynchronization (ERD). Intelligence, 20, 249–266.

Neurobiology of Intelligence

103

Neubauer, A. C., Sange, G., & Pfurtscheller, G. (1999). Psychometric intelligence and event-related desynchronization during performance of a letter matching task. Handbook of Electroencephalography and Clinical Neurophysiology, 6, 219–231. Pfurtscheller, G. (1986). Event-related desynchronization mapping: visualization of cortical activation patterns. In F.H. Duffy (Ed.) Topographical mapping of brain electrical activity. (pp. 99–112). Boston: Butterworth Publishers. Pilgreen, K.L. (1995). Physiologic, medical, and cognitive correlates of electroencephalography. In P. Nunez (Ed.) Neocortical dynamics and human EEG rhythms (pp. 195–248). New York: Oxford University Press. Robinson, D. L., Haier, R. J., Braden, W., & Krengel, M. (1984). Psychometric intelligence and visual evoked potentials: a replication. Personality and Individual Differences, 5, 487–489. Shaw, J.C. (1984). Correlation and coherence analysis of the EEG: a slective tutorial review. International Journal of Psychophysiology, 1, 255–266. Schucard, D., & Horn, J. (1972). Evoked cortical potentials and measurement of human abilities. Psychophysiology, 10, 445–452. Shagass, C., Roemer, R. A., Straumannis, J. J. & Josiassen, R. C. (1981). Intelligence as a factor in evoked potential studies of psychopathology—I. Comparison of low and high IQ subjects. Biological Psychiatry, 16, 1007–1030. Steriade, M., Gloor, P., Llin´as, R.R., Lopes da Silva, F.H., & Mesulam, M.-M. (1990). Basic mechanisms of cerebral rhythmic activities. Electroencephalography and Clinical Neurophysiology, 76, 481– 508. Stern, R.M., Ray, W.J., & Davis, C.M. (1980). Psychophysiological recording. New York: Oxford University Press. Stough, C.K.K., Nettelbeck, T., & Cooper, C. (1990). Evoked brain potentials, string length and intelligence. Personality and Individual Differences, 11(4), 401–406. Stough, C., Thompson, J.C., Bates, T.C. & Nathan, P. J. (2001). Examining Neurochemical Determinants of Inspection Time: Development of a Biological Model. Intelligence, 29, 511–522. Shagass, C. (1946). An attempt to correlate the occipital alpha frequency of the electroencephalogram with performance on a mental ability test. Journal of Experimental Psychology, 36, 88–92. Vogel, W. & Broverman, D.M. (1964). Relationship between EEG and test intelligence: a critical review. Psychological Bulletin, 62(2), 132–144.

5 Neurobiology of Antisociality Lisa J. Cohen

This chapter will explore the current literature on the neurobiology of antisociality. The term “antisociality” will be used in distinction to “criminality”, which can refer to behavior defined purely in terms of its legal status. Antisociality and the related terms, psychopathy and sociopathy, refer to the personality traits which are characteristically associated with antisocial behavior. Antisocial behavior will be defined as that which is intended to serve the short-term interest of the actor but is clearly harmful or destructive to the public at large and expressly transgresses against cultural norms and frequently legal strictures. Further, the individual who performs this behavior is fully cognizant of its implications. Antisocial traits have been divided into two components (Hare, 1991), antisocial attitudes and antisocial behavior. Antisocial behavior is often associated with behavioral dyscontrol and impulsivity, while antisocial attitudes are characterized by a dearth of empathy, guilt, and sense of responsibility towards others. The construct validity of these two components has been consistently confirmed by factor analyses (Hare, 1991; Soderstrom, 2002). Although these factors are highly correlated, (with a correlation coefficient of about 0.5 according to Hare, 1991) they are nonetheless, dissociable. In fact the crimes of psychopathic murderers were found to be far less impulsive and far more instrumental than those of non-psychopathic murderers (Woodward, Porter, 2002). In light of this distinction, James & Blair (2003) propose a definition of psychopathy that focuses solely on attitudinal/emotional deficits, such as lack of guilt and empathy, omitting consideration of behavioral dyscontrol. In this chapter, however, I will address both behavioral and attitudinal components, in keeping with the diagnostic traditions of Lisa J. Cohen

•

Beth Israel Medical Center/Albert Einstein College of Medicine.

107

108

lisa j. cohen

DSM-IV (APA, 1994), ICD-10 (WHO, 1992) and the Hare Psychopathy ChecklistRevised (Hare, 1991). I will also propose that the attitudinal aspect of antisociality may be further subdivided into the moral and affective/interpersonal aspects. The former refers to the failure of an individual to regulate his/her behavior according to a belief system, which puts value on the partial subordination of individual interests in the interest of the public good. This might otherwise be termed a system of values, or a knowledge of “right” and “wrong.” The basis of such a value system develops within the first decade of a child’s life and is centrally dependent on cognitive development as well as significant environmental input (Kohlberg et al, 1983; Gilligan, 1982; Eisenberg, 2000). The study of moral development has received considerable attention within the field of developmental and social psychology (Eisenberg, 2000). What I would term the affective/interpersonal aspects of antisociality are less strictly conceptual, less tied to fairly advanced cognitive development and more dependent on the capacity for emotional empathy. Such affective capacity is heavily dependent on early attachment experiences (Schore, 1994). In contrast, the behavioral component of antisociality has been largely linked to the concept of impulsivity. Impulsivity can be defined as the tendency to act towards short-term, pleasurable goals with insufficient consideration of the long term negative consequences (Cohen et al, 1997). There is a fairly large literature on the neurobiology of impulsivity and behavioral dyscontrol (Coccarro et al, 1989; Hollander et al, 1994; Kavoussi & Coccarro, 1996). In this chapter, I will first discuss the neurobiological literature on impulsivity, then consider the neurobiological research on the attitudinal aspects of antisociality, which is less developed, and finally the interaction between genes and environment in the etiology of antisociality.

PSYCHOBIOLOGY OF IMPULSIVITY As above, impulsivity is defined as acting without thinking, or as behaving recklessly without regard to consequences. A robust literature suggests that deficits in impulse control are associated with abnormalities in neuropsychological, neuroanatomical (primarily frontal), and neurotransmitter function. Research into impulsivity is supplemented by the literature on impulse control disorders. Impulsive disorders are characterized by risk seeking behavior, a defect in harm avoidance, and little anticipatory anxiety. These disorders may include DSM-IV disorders of impulse control, such as intermittent explosive disorder, pyromania, kleptomania, and pathological gambling. Such disorders are characterized by pleasure producing behaviors, although the consequences of such behavior may be painful. This research is relevant to the study of antisociality, as high trait levels of impulsivity

Neurobiology of Antisociality

109

and/or impulsive-aggression may dispose people to perform the behaviors associated with specific impulse control disorders. Likewise, high scores on the motor impulsiveness factor on the Barratt Impulsivity Scale have been associated with a greater number of impulsive acts in inmates (Barratt et al, 1997). Research into the neurobiology of impulse control is further complemented by studies conducted within the impulsive/compulsive conceptual framework (Hollander, & Cohen 1996). Compulsive disorders, to some degree the reverse of impulsive disorders, are characterized phenomenologically by an increased sense of harm avoidance, risk aversiveness, and anticipatory anxiety. Such disorders include obsessive-compulsive disorder (OCD), body dysmorphic disorder, and anorexia nervosa. Moreover, the neurobiology of compulsive disorders is largely the inverse of the impulsive findings, providing additional evidence of the neurobiological substrates of impulsivity.

NEUROPSYCHOLOGICAL STUDIES Neuropsychological studies have shown impaired decision making strategies and other indications of executive dysfunction in impulsive subjects. Neurocognitive measures of impulsive decision-making have shown robust deficits in antisocial subjects, supporting the relevance of impulsivity in the study of antisociality. Antisocial substance abusers favored immediate monetary rewards over larger delayed monetary rewards (discounting of delayed rewards) faster than substance abusers without significant antisocial psychopathology. Both groups discounted delayed rewards at a greater rate than controls and discounting of delayed rewards correlated with a trait measure of impulsivity (Petry, 2002). Similar findings were demonstrated when comparing psychopathic subjects, defined by the Psychopathy Checklist-Revised—PCL-R (Hare, 1991), to incarcerated controls on the gambling task (Bechara et al, 1994), another measure of risk vs. reward decision making (Mitchell et al, 2002), which has been associated with orbital-frontal function (Bechara et al, 1994). Moreover, additional studies have replicated the finding that antisocial subjects favor larger immediate rewards despite long term losses (Van Honk et al, 2002; Mazas et al, 2000). There are also robust findings of decreased performance in other areas of executive functions associated with impulsivity. Pathological gamblers demonstrated higher order attentional impairment and may have elevated rates of childhood ADHD (Rugle & Melamed, 1993). Adolescents without conduct problems showed a significant task condition effect on P300 event-related potentials and left prefrontal activation, suggesting greater brain activity with the harder task. This task effect was reduced or absent in subjects with conduct problems, suggesting difficulty mobilizing the frontal resources in the performance of executive functions (Bauer, Hesselbrock, 2001). In another study, (Barratt et al 1997) assessed

110

lisa j. cohen

executive function in a group of impulsive-aggressive felons compared to those who exhibited non-impulsive aggression. All subjects met DSM-III-R criteria for anti-social personality disorder. Impulsive-aggressive felons showed lower verbal symbol decoding and peak P300 amplitudes relative to non-impulsive aggressive felons. Moreover, impulsiveness and verbal skills were inversely correlated. This study highlights how impulsivity and antisociality, although frequently comorbid, are nonetheless dissociable. The findings are somewhat less consistent regarding executive dysfunction in explicitly antisocial groups. The lack of consistency in this area may be due to the heterogeneity in antisocial samples, such that executive dysfunction relates most to impulsivity but not antisociality per se (i.e., antisocial attitudes vs. behavior) (Morgan, Lillienfeld, 2000). Nonetheless, in a meta-analytic study of 39 reports of executive dysfunction in psychopaths, psychopaths demonstrated a difference of .62 standard deviations below controls (Morgan, Lillienfeld, 2000) but the authors note that inconsistent definitions of antisociality, e.g., confounding antisociality with criminality, complicates interpretation of the finding. Likewise, a study of violent offenders compared with normal and marginally mentally retarded controls, showed deficits in attentional set-shifting but not spatial or figurative working memory or planning (Bergvall et al, 2001). However, in another study, centro-frontal event-related potentials (N350) were stronger in psychopaths vs. nonpsychopaths during three verbal tasks, suggesting abnormal frontal processing of verbal tasks. Moreover, psychopaths showed more difficulty identifying abstract words (Kiehl, Hare et al, 1999).

Neuroanatomical Studies Deficits in impulse control and executive functions have been robustly linked to frontal lobe abnormalities. As above, the gambling task, a measure of impulsive decision making, is significantly impaired in patients with orbito-frontal lesions (Bechara et al, 1994). Impulsive patients, such as borderline personality disorder patients, have decreased frontal glucose metabolic rates, and those with greater aggression have lower frontal activity (Goyer et al, 1991). Further, electrophysiological studies demonstrate a link between impulsivity and impaired frontal function (Kiehl, Hare et al, 1999; Barratt, Stanford et al. 1997). In contrast, hyper-frontality has been consistently documented in compulsive disorders such as OCD (Insel, 1992). On positron emission tomography (PET) several studies demonstrated increased glucose metabolism in orbitofrontal cortex (Baxter, Phelps, Mazziotta et al, 1987; Baxter, Schwartz, Mazziotta et al. 1988; Nordahl, Benkelfat, Semple et al. 1989; Swedo, Schapiro, Brady et al. 1989) although one study found decreased glucose metabolism (Martinot, Allilaire, Mazoyer et al. 1990). Additionally, single photon emission computed tomography (SPECT) has documented increased blood flow (HMPAO uptake) in the frontal

Neurobiology of Antisociality

111

cortex (Rubin, Villaneuva-Meyer, Anath et al. 1992; Machlin, Harris, Pearlson et al, 1991). Moreover, frontal abnormalities on PET have been related to OCD severity (Swedo et al, 1989). In one study with 137 -Xe rCBF, the partial 5-HT agonist mCPP increased OCD symptoms and cortical blood flow, especially in frontal areas (Hollander, Prohovnik, Stein et al. 1995). Of interest, normalization of functional abnormalities in the caudate and orbital frontal areas have been demonstrated following successful treatment of OCD (Hoehn-Saric et al, 1991; Baxter, Schwartz, Bergman et al. 1992; Swedo, Pietrini, Leonard et al. 1992). Neurosurgical procedures such as anterior capsulotomy and cingulotomy decrease frontal lobe input to the limbic system and are effective in severely refractory OCD patients (Jenike et al, 1991).

NEUROTRANSMITTER FUNCTION Considerable evidence implicates serotonergic dysfunction in the neurobiology of impulsivity. While the complexity of neurotransmitter systems demands cautious interpretation, there is evidence of decreased serotonergic tone in impulsive disorders in contrast to increased serotonergic tone found in compulsive disorders. Serotonergic (5-HT) function may be measured by cerebrospinal fluid (CSF) metabolites of 5-HT (5-hydroxyindoleacetic acid: 5-HIAA), by responses to serotonergic probes (m-CPP and others), and by treatment outcome to serotonin reuptake blockers (fluoxetine, clomipramine, fluvoxamine and others). Patients with impulsive aggressive (Linnoila, Virkkunen, Scheinen et al. 1983) and violent suicidal (Asberg, Traskin, Thoren, 1976) behavior have decreased levels of cerebrospinal fluid metabolites of 5-HT (CSF 5HIAA). Patients successfully completing violent suicide also have decreased 5-HT receptors in the frontal cortex (Arora, Meltzer, 1989). In 22 violent offenders, psychopathic features on the PCLR were associated with low CSF-5HIAA (Soderstrom et al, 2001). On the other hand, elevated CSF 5-HIAA has been demonstrated in a subgroup of OCD patients (Insel, Mueller, Alterman et al. 1985) as well as corresponding decreases following successful treatment with clomipramine (Thoren, Asberg, Bertilsson et al. 1980). Other compulsive disorders such as anorexia nervosa (Kaye, Gwirstman, George et al. 1991) demonstrate increased 5-HIAA overall or in subgroups of patients responsive to 5-HT reuptake blockers. Acute challenges with serotonergic agents have provided further evidence of 5-HT involvement in impulsivity. Moreover, there is preliminary evidence that impulsive and compulsive disorders may demonstrate opposing behavioral responses to serotonergic challenge. In response to 5-HT agonists such as mCPP, patients with impulsive disorders, such as impulsive personality disorders (Hollander, Stein, DeCaria et al. 1994) and pathological gambling (DeCaria, Stein, Cohen et al. 1993) for the most part do not show a dysphoric response, but often

112

lisa j. cohen

have a euphoric or a “high” response to m-CPP. On the other hand, compulsive disorder such as OCD (Hollander et al, 1992a; Zohar et al, 1987) and restrictive eating disorders (Buttinger, Holllander, Walsh, 1990) show increased negative affect and increased obsessional thoughts and compulsive urges in response to challenge with 5-HT agonists. In response to 5-HT reuptake blockers, compulsive disorders such as OCD, anorexia nervosa, and body dysmorphic disorder, show clear cut improvement with chronic treatment. In fact, the hyperfrontality demonstrated in OCD has been shown to be reduced following chronic 5-HT reuptake blocker treatment and this is associated with decreased harm avoidance (Hoehn-Saric et al, 1991). As serotonin reuptake inhibitors function to stimulate 5-HT activity, symptoms may initially worsen following acute administration with high doses (Hollander et al, 1992b). Chronic treatment with these agents, however, may work to desensitize or downregulate 5-HT receptors over time (Zohar et al, 1988; Hollander et al, 1991a). Open pilot work in impulsive personality disorders shows some improvement early on (Coccaro et al, 1990) but this effect may wear off with time and long term follow-up studies are needed. It has been postulated that acute administration of 5-HT reuptake blockers might worsen compulsive but improve impulsive disorders, whereas chronic administration of these agents might improve compulsive but ultimately worsen impulsive disorders (Cohen et al, 1997).

THE MESOLIMBIC DOPAMINERGIC REWARD CIRCUITRY Research into impulsivity generally focuses on inhibitory failures. A new area of neurobiological research investigates the motivational component of impulsive behavior. As such, a reward system, subserved by the meso-limbic dopaminergic pathways, is gaining increasing attention. This reward circuitry appears to be geared towards appetitive states, i.e., eager anticipation of reward, rather than consummatory states, or the pleasure associated with obtaining the reward, which may comprise a distinct system (Panksepp, 1998). Moreover, this system appears to comprise a generalized approach to reward stimuli and is not specific to any class of reward, e.g., food, sex, or psychoactive substances (Panskepp, 1998). The initial theories of the reward circuitry derive from animal models (Panskepp, 1998; Koob, 2000; Becerra, 2001), which implicated several subcortical areas, including parts of the amygdala and the nucleus accumbens (Becerra, 2001; Koob, 2000; Panksepp, 1998). These areas are richly innervated by meso-limbic dopaminergic tracts originating in the ventral tegmental area (Becerra, 2001; Koob, 2000) and coursing through the medial forebrain bundle of the lateral hypothalamus (Panksepp, 1998). In humans, brain imaging allows the closest analogue of in vivo animal studies. The notion of anticipatory, appetitive states is particularly suggestive of drug

Neurobiology of Antisociality

113

craving; likewise, the impetus for investigating the reward circuitry in.humans has largely come out of addiction research. There are a few imaging studies following direct infusion of cocaine or methelphenidine (Volkow et al, 1999; Becerra, 2001). Most studies utilize videotapes of drug use or drug paraphernalia as visual stimuli to induce drug craving, which was seen as a rough equivalent of rewardmotivated states (Wexler et al, 2001, Volkow et al, 1999). Technical limitations often preclude sensitive imaging of small subcortical regions, but several studies have demonstrated changes in the nucleus accumbens (Becerra et al, 2001, Wexler et al, 2001, Grant et al, 1996) and the amygdala (Wexler et al, 2001, Childress et al, 1999, Grant et al, 1996) following either direct infusion of a drug of abuse or activation with cue-induced craving. A few studies have noted involvement of the caudate and other striatal areas (Garavan et al, 2000, Childress et al, 1999; Grant et al, 1996). As would be expected in humans, there was also extensive cortical involvement. Most studies showed orbito (Volkow, 1999; Grant et al, 1996), medial (Garavan et al, 2000, Grant et al, 1996) or dorsolateral prefrontal activation (Garavan et al, 2000, Maas et al, 1998, Wexler et al, 2001, Grant et al, 1996) with cue induced craving or infusion (Volkow et al, 1999), suggesting significant cognitive elaboration of emotional/ motivational states in humans. Inferior and anterior parietal regions have also been identified (Garavan et al, 2000, Grant et al, 1996) as have temporal areas, (Garavan et al, 2000, Childress et al, 1999, Becerra et al, 2001), which may be related to memory demands. An area that has received particular attention is the anterior cingulate gyrus (Garavan et al, 2000, Maas et al, 1998, Wexler et al, 2001, Childress et al, 1999, Volkow et al, 1999). The anterior cingulate may subserve a complex array of psychological functions, with affective/motivational, cognitive and sensory components (Liotti et al, 2000, Shin et al, 2000). There are to date, no investigations specifically into the reward circuitry in antisocial populations but we could postulate that such groups would show functional abnormalities in this system. Soderstrom et al (2001) found that elevated levels of CSF homovanillic acid (HVA), an index of dopaminergic function, predicted to high scores on the PCL-R. Furthermore, robust evidence of the high comorbidity of substance abuse and antisocial personality disorder adds support to this hypothesis (Fu et al, 2002).

NOVELTY-SEEKING Elevated novelty-seeking or sensation-seeking has also been proposed in antisocial individuals. Antisocial groups have scored higher on Cloninger’s Novelty-Seeking scale (Cloninger, 1996). Novelty seeking has been associated with noradrenergic function. (Roy et al, 1988). In pathological gamblers, noradrenergic hyperactivation, consistent with elevated novelty-seeking, is suggested

114

lisa j. cohen

by increased rates of CSF 3-methoxy-4-hydroxyphenylglycol (MHPG), a metabolite of norepinephrine (Roy et al, 1988) and increased GH response to clonidine challenge (DeCaria et al, 1993). There are some studies, however, that suggest sensation-seeking (or novelty seeking) may be less central to antisociality than impulsivity (Vitacco, Rogers, 2001).

CENTRALITY OF IMPULSIVITY IN ANTISOCIALITY It must be noted that most antisocial samples studied are identified through criminal behavior, such that the majority of subjects are incarcerated or in courtmandated treatment. Thus most research is conducted on subjects who are identified through their illegal behavior and through their apprehension for such behavior. Impulsivity in these cases is likely to be high. Such studies therefore, may not account for antisocial individuals who suppress their antisocial behavior, engage in antisocial behavior without being apprehended, or do so within the confines of the legal system. In fact world history is replete with genocidal murderers whose heinous acts are state sanctioned. In such cases, antisocial individuals succeed in their antisocial goals through careful planning and disciplined suppression of impulsivity. In light of this consideration, Ishikawa et al, (2001) compared successful vs. unsuccessful psychopaths on measures of cognitive function and other psychological traits. They found that successful psychopaths demonstrated less impulsivity and better conceptual skills than their unsuccessful counterparts. Moreover, even in the prison and forensic samples studied by Hare (1991), the behavioral and attitudinal factors correlated at about 0.5. Factor 2, the behavioral factor, correlated inversely with age, which was not the case with Factor 1, the attitudinal factor. Thus it appears that even in “unsuccessful” psychopaths, antisocial behavior decreases across the lifespan, while attitudes do not change with age. Further, Factor 2 showed a relationship between education and social class, which was not the case with Factor 1. Given the evidence of dissociability between behavioral dyscontrol and antisocial attitudes, many questions remain about the essential relationship between both components of antisociality.

ATTITUDINAL ASPECTS OF ANTISOCIALITY As the attitudinal and behavioral aspects of antisociality appear quite distinct, their respective neurobiological substrates do as well. Although the neurobiological research into the attitudinal aspects is less developed than that of the behavioral aspects, there are nonetheless several promising lines of inquiry. As above, we have chosen to subdivide the attitudinal aspects into moral and affective/interpersonal domains. We will first address the affective/interpersonal domain.

Neurobiology of Antisociality

115

Affective/Interpersonal Aspects of Antisociality In DSM-IV and the Hare Psychopathy Checklist, a core feature of the diagnosis of antisocial personality disorder is the lack of empathy and the absence of committed, intimate relationships. Relationships are shallow, opportunistic, and exploitative. Central to this is the notion of altered emotional processing and there is consistent evidence of abnormalities in this area in antisocial groups. A number of studies have demonstrated reduced ability to recognize and discriminate emotional facial expressions (Blair et al, 2001), and vocal tone (Blair et al, 2002). Moreover, there are robust findings of lowered somatic markers of CNS arousal in response to emotional stimuli. For example, subjects with ASPD show reduced electro dermal skin response, event related potentials, and other physiological response in response to aversive conditioning (Flor et al, 2002). Although psychopaths show reduced reactivity to many emotions, they are particularly insensitive to distress and, most specifically, fear (Blair et al, 2002). In an fMRI study of ASPD subjects’ response to fear-inducing visual stimuli, there was lowered activation of the limbic prefrontal circuit, including the orbitofrontal cortex, insula, amygdale and anterior cingulate cortex relative to a control group, consistent with a blunted fear response to emotionally salient perceptual stimuli (Veit et al, 2002). Similar findings in criminal psychopaths showed decreased affect-related activity in the amygdala, hippocampus, parahippocampal gyrus, ventral striatum, and the anterior and posterior cingulate (Kiehl et al, 2001). The amygdala is centrally involved in the activation of emotional responses to perceptual stimuli and is particularly sensitive to fearful stimuli. The hippocampus is involved in memory and rich interconnections with the amygdala make it particularly salient to the processing of emotional memory. This reduced processing of and reactivity to emotions in general, and fear in particular, may underlie the lack of empathy characteristic of antisocial individuals. The reduced fear response, however, is also consistent with the lowered harm avoidance associated with impulsivity. Thus impaired emotional processing may link impulsivity and lack of empathy in psychopathic individuals. There is also growing interest in the neurobiology of attachment, which is clearly deficient in antisocial individuals, although to our knowledge, this line of research has not yet extended to antisociality. Animal studies have demonstrated the central role of two neuropeptides, vasopressin and oxytocin, in the attachment system (Panksepp, 1998; Insel, 1997). Based on these findings, a number of studies have investigated the role of oxytocin in autism, another psychiatric disorder characterized by attachment deficits (Green et al, 2001). Oxytocin and vasopressin, nine-amino acid peptides (nonapeptides) synthesized in the hypothalamus, differ only by 2 amino acids. Receptors for both peptides are distributed throughout the limbic system and the brain stem. Both are part of a family of nonapeptides that date back to invertebrates. Earlier versions of nonapeptides have been associated

116

lisa j. cohen

with non-mammalian reproductive behaviors such as nest-building. Unique to this family of nonapeptides, oxytocin and vasopressin are found only in mammals and each differ from their putative evolutionary precursor, arginine vasotocin, by only one amino acid. Both peptides are also implicated in characteristic mammalian functions, such as uterine contractions during labor, and milk ejection during nursing. Of interest to the study of human attachment, these nonapeptides are involved in sexual activity and associated adult pair bonding, parental behavior and infant attachment. More specifically, in the highly monogamous prairie vole, but not in the non-monogamous and less generally affiliative montane vole, oxytocin receptors are found in the nucleus accumbens and prelimbic areas associated with the dopaminergic reward circuitry. Two measures of pair bonding found in the monogamous prairie vole but not the montane vole include time spent preferentially in proximity to the mate (partner preference) and, in males, aggressiveness towards third parties after mating (mate guarding). In females oxytocin, but not vasopressin, administered into the lateral ventricle facilitates development of partner preference, even in the absence of mating. Furthermore, an oxytocin antagonist given prior to mating blocks partner preference but does not interfere with mating. In males, partner preference and mate guarding was facilitated selectively by vasopressin and blocked with a vasopressin antagonist. Neither mating behaviors nor non-mating aggression were affected. Of note, both oxytocin and vasopressin are released during human sexual behavior (Insel, 1997; Panksepp, 1998). Further, in rats, onset of maternal behavior is facilitated by administration of oxytocin and blocked following administration of oxytocin antagonists. Vasopressin in male prairie voles is also associated with parenting behavior (See Insel, 1997 for a review). Oxytocin may have an effect on dopamine function from the ventral tegmental area and on opioid release. As noted above, there may be two distinct components of the reward system, appetitive and consummatory pleasure, which are mediated, respectively, by dopaminergic and beta endorphin systems (Panskepp, 1998). Activating “appetitive” pleasure is subserved by the meso-limbic dopaminergic reward circuitry, and is characterized by a kind of eager excitement associated with the anticipation of reward. “Consummatory” pleasure, associated with beta endorphin release at the mu opiate receptors, particularly in the septal area, is characterized by a pleasurable release of tension after obtaining a reward. Thus oxytocin (and possibly vasopressin) may serve to link attachment related behaviors to both the appetitive and consummatory reward circuitry. If vasopressin and oxytocin are deficient in antisocial individuals, it could explain their excessive need to obtain gratification via non-attachment related pleasures, such as illicit substances and impersonal sex. In fact, drugs of abuse such as cocaine directly stimulate the dopaminergic reward circuitry, while heroin and other opiates act upon the endogenous opioid system. Future research could investigate the relationship between oxytocin, vasopressin and the reward circuitry in antisocial individuals.

Neurobiology of Antisociality

117

MORAL ASPECTS OF ANTISOCIALITY Another critical feature of antisociality is the absence or deficiency of any moral code. As noted above, moral function involves cognitive structures about codes of social behavior. More specifically, morality involves a system of beliefs to guide choices where there is conflict between one’s immediate selfinterest and the good of the larger social group. Moral function has been extensively studied within the realm of developmental and social psychology (Kohlberg et al, 1983; Gilligan, 1982; Eisenberg, 2000). Although a full review is beyond the scope of this chapter, there is wide consensus that moral development rests on the interaction between cognitive development and pro-social environmental input. Without sufficient cognitive development truly moral reasoning is not possible, as a conceptual grasp of “the public good” may be beyond the abilities of the individual. In other words, a child or cognitively impaired adult may not understand an abstract system of “right and wrong” and may regulate behavior according to immediate behavioral contingencies (rewards or punishment) or affective cues (Kohlberg’s most primitive stage of moral development; Kohlberg et al, 1983). Likewise, as children mature, their moral belief system grows more abstract and less egocentric, with consequent gains in complexity and sophistication (Piaget, 1954; Kohlberg et al, 1983; Werner & Kaplan, 1963). Such conceptual development during childhood is coincident with tremendous growth in the frontal lobe, including myelinization, synaptic growth and pruning, etc. (Schore, 1994). The dependence of moral functioning on higher order cognitive development (abstraction and decentering) explains why lower frontal function is a significant risk factor for impulsive, destructive behavior (Barratt et al, 1997). On the other hand, and perhaps of more interest, are those “true” psychopaths with intact cognitive function whose interpersonal belief systems are not informed by empathy and attachment. In such cases, we might posit that a deficit in the limbic circuitry underlying attachment and related affect precluded integration of such circuits with the pre-frontal networks subserving higher level cognition. The linking of the frontally-mediated higher order cognitions with more limbic-driven affective states is influenced by early affective experience (Schore, 1994); such that subtle differences in environmental input, specifically parental empathic responses, have far reaching influence on the development of interpersonal concepts. This notion is robustly supported in the large attachment literature (Main et al, 1985; Slade & Aber, 1992). Whether the creation of pro-social interpersonal concepts rests on the integration of prefrontal and limbic circuitry separately subserving affect and cognition or whether there is a specific localization of such interpersonal templates is not known. Watt (1990), suggested a preferential role in the right prefrontal region for interpersonal constructs, based on the right lateralization of several cortical aspects of affective processing (e.g., facial recognition, decoding of facial expressions, emotional prosody). A more recent fMRI study has

118

lisa j. cohen

linked cognitively generated affect, or “affective scripts” to the medial prefrontal cortex. Of note, antisocial individuals are also known for their lack of guilt or remorse. In the literature on the “moral emotions”, such as guilt and shame, these “self conscious emotions” are seen to be integrally linked with higher level cognition, specifically concepts of the self (Eisenberg, 2000; Lewis, 1997). As above, such cognitive elaboration would implicate the dorsal lateral prefrontal cortex while the strictly emotional component would point to limbic involvement. In fact, one PET study of brain activation during guilt-inducing memory scripts implicated the inferior and orbito-frontal, anterior cingulate, insular and temporal cortices in healthy controls (Shin et al, 2000).

SUMMARY To summarize, antisociality is comprised of two related but dissociable components, antisocial attitudes and behavioral dyscontrol. Behavioral dyscontrol is associated with decreased function in the prefrontal cortex. Reduced function in the ventro-medial and orbito-frontal cortex may underlie impaired behavioral inhibition in the context of punishment. Decreased higher order executive functions, such as verbal processing, abstraction, and concept formation, reflecting reduced function in the dorso-lateral prefrontal cortex, are also associated with impulsivity. Further, serotonin appears to play a central role in behavioral inhibition with reduced serotonergic tone associated with impaired impulse control. Although neurobiological research into the attitudinal components of antisociality is less developed, there is evidence of reduced responsivity to affect in general and to fear in particular, which may be associated with decreased amygdala function. Animal research and studies of healthy humans also point to potential lines of investigation, specifically into the relationship between oxytocin, vasopressin and the reward circuitry as well as the integration of cortico-limbic circuitry in antisocial individuals.

GENES AND ENVIRONMENT: THE QUESTION OF ETIOLOGY Multiple twin studies and family studies assess the relevant contribution of genes and environment to antisociality, with estimates of genetic contribution ranging from about 33% (Jacobson et al, 2000; Rhee, Waldman, 2002) up to to 69% (Fu et al, 2002). Raine (2002), however, suggests that the complex interaction of genetic and environmental factors is critical to predicting the development of antisociality. When both deleterious environmental and genetic factors are present, there is an exponentially higher likelihood of an antisocial outcome. Further in

Neurobiology of Antisociality

119

antisocial groups, those individuals with benign environmental backgrounds have the strongest rates of genetic risk factors. Nonetheless, there is clearly a strong genetic component to antisocial behavior, leading to research into the genetic markers of antisociality. Most studies of genetic markers have looked at polymorphisms at specific loci associated with components of the serotonergic system. Such genetic markers have largely been associated with Type II alcoholism, characterized by early age of onset and antisocial behavior, and are seen to predispose the individual for impulsivity and poor behavioral control, as opposed to alcoholism or antisociality per se. The most promising findings involve the genes encoding the neurotransmitter-metabolizing enzyme, monoamine oxydase A (Hill et al, 2002; Caspi et al, 2002; Parsian, Cloninger, 2001). A genotype conferring high levels of expression of monoamine oxydase A activity may protect against the development of impulsive aggressive tendencies, even in the context of a deleterious environment. In fact, Caspi et al, (2002) studied a large sample of maltreated children and found that those who developed antisocial behavior were more likely to have genotypes associated with low MOA-A activity. It must be noted however, that elucidation of the neurobiological substrates of both behavioral and attitudinal components of antisociality should not be mistaken for an assumption of genetic etiology. There are undoubtedly genetic contributions to the components of antisociality, such as impulsivity, novelty seeking, and, most likely, aspects of affective processing. Moreover, genetic research may lead to fruitful psychopharmacological interventions. Nonetheless, there is ample evidence of the contributions of sociological, cultural, and familial factors, as well as individual experiences, such as trauma, to the development of antisociality (Raine, 2002; Hare, 1991; Eisenberg, 2000). A superficial look at most prison populations will show the vastly disproportionate representation of members of lower socioeconomic status groups, ethnic minority groups, and those with lower education. In light of this, the challenge for neurobiological research is to elucidate how environmental inputs influence neurodevelopment and then how to facilitate optimal outcomes and to remedy pathological ones.

REFERENCES American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, 4th ed. Washington. DC, 1994. Arora RC, Meltzer HY: Serotonergic measures in the brains of suicide victims: 5-HT2 binding sites in the frontal cortex of suicide victims and control subjects. Am J Psychiatry 146:730–736, 1989. Asberg M, Traskin L, Thoren P: 5 HIAAA in the cerebrospinal fluid: A biochemical suicide predictor? Arch Gen Psychiatry 33:1193–1197, 1976. Barratt ES, Stanford MS, Kent TA: Felthous A.Neuropsychological and cognitive psychophysiological substrates of impulsive aggression. Biological Psychiatry. 41(10):1045–61, 1997 May 15.

120

lisa j. cohen

Bauer LO, Hesselbrock VM: CSD/BEM localization of P300 sources in adolescents “at-risk”: evidence of frontal cortex dysfunction in conduct disorder. Biological Psychiatry. 50(8):600–8, 2001. Baxter LR, Phelps ME, Mazziotta JC, Guze BH, Schwartz JM, Seline CE: Local cerebral glucose metabolic rates in obsessive-compulsive disorder. A comparison with rates in unipolar depression and in normal controls. Arch Gen Psychiat 1987; 44:211–218. Baxter LR, Schwartz JM, Mazziotta JC et al: Cerebral glucose metabolic rates in non-depressed patients with obsessive compulsive disorder. Am J Psychiatry 1988, 145:1560–3. Baxter LR Jr., Schwartz JM, Bergman KS, Szuba MP, Guze BH, Mazziotta JC, Alazraki Selin CE, Ferng H-K, Munford P, Phelps ME: Caudate glucose metabolic rate changes with both drug and behavior therapy for obsessive-compulsive disorder. Arch Gen Psychiatry. 1992;49:681–689. Becerra L, Breiter HC, Wise R, Gonzalez RG, Borsook D: Reward circuitry activation by noxious thermal stimuli.Neuron, 32:927–946, 2001. Beccera LR, Breiter HC, Stojanovic M, Fishman S, Edwards A, Comite AR, Gonzalez RG, Borsook: Human Brain activation under controlled thermal stimulation and habituation to noxious heat: An fMRI study. Magnetic Resonance in Medicine 1999; 41: 1044–1057. Bechara A, Damasio A, Damasio H, Anderson SW. Insensitivity to future consequences following damage to human prefrontal cortex. Cognition 1994;50:7–15. Bergvall AH, Wessely H, Forsman A, Hansen S.A: deficit in attentional set-shifting of violent offenders. Psychological Medicine. 31(6):1095–105, 2001. Blair RJ, College E, Murray L, Mitchell DG: A selective impairment in the processing of sad and fearful expressions in children with psychopathic tendencies. Journal of Abnormal Child Psychology. 29 (6):491–8, 2001. Blair RJ, Mitchell DG, Richell RA, Kelly S, Leonard A, Newman C, Scott SK: Turning a deaf ear to fear: impaired recognition of vocal affect in psychopathic individuals. Journal of Abnormal Psychology. 111(4):682–6, 2002. Buttinger K, Hollander E, Walsh BT: M-CPP challenges in eating disorder patients. Proceedings of the American Psychiatric Association Annual Meeting. New York, NY, 1990. Caspi A, McClay J, Moffitt TE, Mill J, Martin J, Craig IW, Taylor A, Poulton R: Role of genotype in the cycle of violence in maltreated children.[comment]. Comment in: Science. 297(5582):851–4, 2002. Childress AR, Mozley PD, McElgin W, Fitzgerald J, Reivich M, O’Brien CP: Limbic activation during cue- induced craving. Am J Psychaitry 1999;156:11–18. Cloninger RL: Assessment of the impulsive-compulsive spectrum of behavior by the seven factor model of temperament and character. In: Impulsivity and Compulsivity, (Eds.) Oldham J, Hollander E, Skodol A. Washington DC: American Psychiatric Press. 59–96. Coccaro EF, Siever LJ, Klar HM, Maurer G, Cochrane K, Cooper TB, Mohs RC, Davis KL: Serotonergic studies in affective and personality disorder patients: Correlations with behavioral aggression and impulsivity. Arch Gen Psychiatry 46:587–599, 1989. Cohen LJ, Stein DJ, Simeon D, Hollander E: Obsessive Compulsive Spectrum Disorders. In (eds.) E. Hollander, D.J. Stein. Obsessive Compulsive Disorder-Etiology-Diagnosis-Treatment. Marcel Dekker, 1997. DeCaria CM, Stein DJ, Cohen L, Simeon D, Hwang M, Hollander E: Psychobiology of pathological gambling. Presented at First International Obsessive-Compulsive Disorder Congress, Capri, Italy, Mar 9–12, 1993. Eisenberg N: Emotion, regulation and moral development. Annual Review of Psychology. 51:665–697, 2000. Flor H, Birbaumer N, Hermann C, Ziegler S, Patrick CJ: Aversive Pavlovian conditioning in psychopaths: peripheral and central correlates.Psychophysiology. 39(4):505–18, 2002. Fu Q, Heath AC, Bucholz KK, Nelson E, Goldberg J, Lyons MJ, True WR, Jacob T, Tsuang MT, Eisen SA: Shared genetic risk of major depression, alcohol dependence, and marijuana

Neurobiology of Antisociality

121

dependence: contribution of antisocial personality disorder in men. Archives of General Psychiatry. 59(12):1125–32, 2002. Garavan H, Pankiewicz J, Bloom A, Cho J, Sperry L, Ross TJ, Salmeron BJ, Risinger R, Kelley D, Stein EA: Cue-induced cocaine craving: Neuroanatomical specificity for drug users and drug stimuli. Am J Psychiatry 2000;157:1789–1798. Gilligan, C: In a Different Voice: Psychological Theory and Women’s Development. Cambridge, MA: Harvard University Press, 1982. Goyer PF, Andreason PJ, Semple WE, Clayton AH, King AC, Schultz SC, Cohen RM: Pet and Personality Disorders, Biol Psychiatry 29;43A–185A, 1991. Grant S, London ED, Newlin DB, Villemagne VL, Liu X, Contoreggi C, Phillips RL, Kimes AS, Margolin A: Activation of memory circuits during cue-elicited cocaine craving. Porc Natl Acad Scie 1996;93: 12040–12045. Green L, Fein D, Modahl C, Feinstein C, Waterhouse L, Morris M: Oxytocin and autistic disorder: alterations in peptide forms. Biological Psychiatry. 50:609–13, 2001. Hare RD: The Hare Psychopathy Checklist—Revised: Manual. Northe Tonowanda, NY: Multi-Health Systems, Inc., 1991. Hill EM, Stoltenberg SF, Bullard KH, Li S, Zucker RA, Burmeister M.: Antisocial alcoholism and serotonin-related polymorphisms: association tests. Psychiatric Genetics. 12(3):143–53, 2002. Hoehn-Saric R, Pearlson GD, Harris CJ, Machlin SR, Camargo EE: Effects of fluoxetine on regional cerebral blood flow in obsessive-compulsive patients. Am J Psychiatry 1991;48:1243–1245. Hollander E, Cohen LJ: (1996) The psychobiology and psychopharmacology of compulsive spectrum disorders. In: Impulsivity and Compulsivity, Edited by Oldham J, Hollander E, Skodol A. Washington DC: American Psychiatric Press, Inc., pp. 143–166. Hollander E, Cohen LJ, DeCaria C, Stein DJ, Trungold-Apter S, Islam N: Fluoxetine and depersonalization syndrome (letter), Psychosomatics 33:361, 1992b?. Hollander E, DeCaria C, Gully R, Nitescu A, Suckow RF, Gorman JM, Klein DF, Liebowitz MR: Effects of chronic fluoxetine treatment on behavioral and neuroendocrine response to meta-chlorophenylpiperazine in obsessive-compulsive disorder. Psychiatry Research 36:1–17, 1991a?. Hollander E, DeCaria CM, Nitescu A, et al: Serotonergic function in obsessive-compulsive disorder: behavioral and neuroendocrine responses to oral m-chlorylpiperazine and fenfluramine in patients and healthy volunteers. Arch Gen Psychiatr 1992a;49:21–28. Hollander E, Prohovnik I, Stein DJ: Effects of m-CPP induced obsessions on cortical blood flow.J Neuropsychiatry Clin Neurosci 1995; 7:485–490. Hollander E, Stein DJ, DeCaria CM, Cohen L, Saoud J, B, Skodol AE, Kellman D, Rosnick L, Oldham J: Serotonergic sensitivity in borderline personality disorder: Preliminary findings. Am J Psychiatry February, 1994. Insel TR: Toward a neuroanatomy of obsessive-compulsive disorder. Arch Gen Psychiatry 1992; 49:739–744. Insel TR: A Neurobiological basis of social attachment. American Journal of Psychiatry, 1997; 154:726–735. Insel TR, Mueller EA, Alterman I, Linnoila M, Murphy DL: Obsessive-compulsive disorder and serotonin: Is there a connection? Biol Psychiatry 20:1174–1188, 1985. Ishikawa SS, Raine A, Lencz T, Bihrle S, Lacasse L.: Autonomic stress reactivity andexecutive functions in successful and unsuccessful criminal psychopaths from the community. Journal of Abnormal Psychology. 110(3):423–32, 2001. Jacobson KC, Prescott CA, Kendler KS: Genetic and environmental influences on juvenile antisocial behaviour assessed on two occasions. Psychological Medicine. 30(6):1315–25, 2000. James R, Blair RJ: Neurobiological basis of psychopathy. British J of Psychiatry. 182;5–7: 2003.

122

lisa j. cohen

Jenike MA, Baer L, Ballantine T, Martuza RL, Tynes S, Giriunas I, Buttolph L, Cassem NH: Cingulotomy for refractory obsessive-compulsive disorder: A long term follow up of 33 patients. Arch Gen Psychiatry 48:548–555, 1991. Kavoussi RJ, Coccarro EF: Biology and pharmacological treatment of impulse control disorders. In: Impulsivity and Compulsivity, (Eds.) Oldham J, Hollander E, Skodol A. Washington DC: American Psychiatric Press. 1996:59–96. Kaye WH, Gwirstman HE, George DT, Ebert MH: Altered serotonin activity in anorexia nervosa after long-term weight restoration: Does elevated cerebrospinal fluid 5-hydroxyindoleacitic acid level correlate with rigid and obsessive behavior? Arch Gen Psychiatry 48:556–562, 1991. Kiehl KA, Hare RD, McDonald JJ, Brink J: Semantic and affective processing in psychopaths: an event-related potential (ERP) study. Psychophysiology. 36(6):765–74, 1999. Kiehl KA, Smith AM, Hare RD, Mendrek A, Forster BB, Brink J, Liddle PF: Limbic abnormalities in affective processing by criminal psychopaths as revealed by functional magnetic resonance imaging. Biological Psychiatry. 50(9):677–84, 2001. Kohlberg L, Levine C, Hewer A: Moral stages: A current formulation and a response to critics. In JA Meacham (ed.) Contributions to Human Development, Basil, Switzerland: Karger: 1983. Koob GF. Neurobiology of addiction: Toward the development of new therapies. Annals of the NY Academy of Sciences 2000; 909: 170–185. Lewis M: The self in self-conscious emotions. In (eds) JG Snodgrass, RL Thompson, The Self Across Psychology: Self Recognition, Self-Awareness, and the Self-Concept. Annals of the New York Academy of Sciences; 1997;818: 119–142. Linnoila M, Virkkunen M, Scheinen M, Nuutila A, Rimon R, Goodwin FK: Low cerebrospinal fluid 5-hydroxyindoleacetic acid concentration differentiates impulsive from nonimpulsive violent behavior. Life Sci 33:2609–2514, 1983. Liotti M, Mayberg HS, Brannan SK, McGinnis S, Jerabek P, Fox PT: Differential limbic—cortical correlates of sadness and anxiety in healthy subjects: implications for affective disorders. Biological Psychiatry 2000; 48:30–42. Maas LC, Lukas SE, Kaufman MJ, Weiss RD, Daniels SL, Rogers VW, Kukes TJ, Renshaw PF: Functional magnetic resonance imaging of human brain activation during cue-induced cocaine craving. Am J Psychiatry 1998; 155:124–126. Machlin SR, Harris GJ, Pearlson GD, Hoehn-Saric R, Jeffrey P, Camargo EE: Elevated medialfrontal cerebral blood flow in obsessive-compulsive patients: a SPECT study. Am J Psychiatry 1991. Main M, Cassidy J, Kaplan N: Security of attachment in childhood, adolescence, and adulthood: A move to the level of representation. In Bretherton I & Waters E (eds.) Growing points of attachment theory and research, Monographs of the Society for Research in Child Development. 50;1985. Martinot JL, Allilaire JF, Mazoyer BM, Hantouche E, Huret JD, Legaut-Demare F, Deslauriers AG, Hardy P, Pappata S, Baron JC, Syrota A: Obsessive-compulsive disorder. A clinical, neuropsychological and positron emissiontomography study. Acta Psychiatr Scand 1990; 82:233–242. Mazas CA, Finn PR, Steinmetz JE: Decision-making biases, antisocial personality, and early-onset alcoholism. Alcoholism: Clinical & Experimental Research. 24(7):1036–40, 2000. Mitchell DG, Colledge E, Leonard A, Blair RJ: Risky decisions and response reversal: is there evidence of orbitofrontal cortex dysfunction in psychopathic individuals?. Neuropsychologia. 40(12):2013– 22, 2002. Morgan AB, Lilienfeld SO: A meta-analytic review of the relation between antisocial behavior and neuropsychological measures of executive function. Clinical Psychology Review. 20(1):113–36, 2000. Nordahl TE, Benkelfat C, Semple WE, Bross M, King AC, Cohen RM: Cerebral glucose metabolic rates in obsessive- compulsive disorder. Neuropsychopharm 1989;2:23–8.

Neurobiology of Antisociality

123

Panksepp J: Affective Neuroscience: The Foundations of Human and Animal Emotions. New York: Oxford University Press, 1998. Parsian A, Cloninger CR: Serotonergic pathway genes and subtypes of alcoholism: association studies.Source Psychiatric Genetics. 11(2):89–94, 2001. Petry NM: Discounting of delayed rewards in substance abusers: relationship to antisocial personality disorder.Psychopharmacology. 162(4):425–32, 2002. Piaget J: The Construction of Reality in the Child. New York, Basic Books, 1954/1986. Raine A: Biosocial studies of antisocial and violent behavior in children and adults: a review. Journal of Abnormal Child Psychology. 30(4):311–26, 2002. Rhee SH, Waldman ID: Genetic and environmental influences on antisocial behavior: a meta-analysis of twin and adoption studies. Psychological Bulletin. 128(3):490–529, 2002. Roy A, Adinoff B, Roehrich L, et al: Pathological gambling: A psychobiological study. Arch Gen Psychiatry 45:369–373, 1988. Rubin RT, Villaneuva-Meyer J, Anath J, Trajmar PG, Mena I: Regional xenon 133 cerebral blood flow and cerebral technetium Tc 99m-HMPAO uptake in unmedicated patients with obsessivecompulsive disorder and matched normal control subjects: determination by high-resolution single-photon emission computed tomography. Arch Gen Psychiatry 1992;49:695–702. Rugle L, Melamed L. Neuropsychological assessment of attention problems in pathological gamblers. J Nerv Ment Dis 1993;181:107–112. Schore AN: Affective Regulation and the Origin of the Self: The Neurobiology of Emotional Development. Hillsdale, NJ: Lawrence Earlbaum, 1994. Shin LM, Dougherty DD, Orr SP, Pitman RK, Lasko M, Macklin ML, Alpert NM, Fischman AJ, Rauch SL: Activation of anterior paralimbic structures during guilt-related script-driven imagery. Biological Psychiatry 2000;48:43–50. Slade A, Aber L: Attachments, drives and development: conflicts and convergences in theory. In Interface of Psychoanalysis and Psychology. Edited by Barron J, Eagle M, Wolitsky D. Washington, DC, American Psychological Association, 1992. Soderstrom H, Hultin L, Tullberg M, Wikkelso C, Ekholm S, Forsman A: Reduced frontotemporal perfusion in psychopathic personality.Psych Res. 114(2):81–94, 2002. Soderstrom H, Blennow K, Manhem A, Forsman A: CSF studies in violent offenders. I. 5-HIAA as a negative and HVA as a positive predictor of psychopathy. Journal of Neural Transmission. 108(7):869–78, 2001. Swedo SE, Pietrini P, Leonard HL, Schapiro MB, Rettew DC, Goldberger El, Rapoport SI, Rapoport JL, Grady CL: Cerebral glucose metabolism in childhood-onset obsessive-compulsive disorder: revisualization during pharmacotherapy. Arch Gen Psychiatry 1992;49:690–694. Swedo SE, Schapiro MB, Brady CL et al: Cerebral glucose metabolism in childhood-onset obsessive compulsive disorder. Arch Gen Psychiatry 1989;46:518–523. Thoren R, Asberg M, Bertilsson L, Mellstrom B, Syoquist F, Trachman L: Clomipramine treatment of obsessive-compulsive disorder. II. Biochemical aspects. Arch Gen Psychiatry 37:1289–1294, 1980. Van Honk J, Hermans EJ, Putman P, Montagne B, Schutter DJ: Defective somatic markers in subclinical psychopathy. Neuroreport. 13:1025–7, 2002. Veit R, Flor H, Erb M, Hermann C, Lotze M. Grodd W, Birbaumer N: Brain circuits involved in emotional learning in antisocial behavior and social phobia in humans. Neuroscience Letters. 328(3):233–6, 2002. Vitacco MJ, Rogers R: Predictors of adolescent psychopathy: the role of impulsivity, hyperactivity, and sensation seeking. J Am Acad Psychiatry Law. 2001;29(4):383–6. Volkow ND, Wang GJ, Fowler JS, Hitzemann R, Angrist B, Gatley SJ, Logan J, Ding YS, Pappas N: Association of methylphenidate-induced craving with changes in right striato-orbitofrontal metabolism in cocaine abusers: implications in addiction. Am J Psychiatry 1999;156:19–26.

124

lisa j. cohen

Watt DF: Higher cortical functions and the ego: explorations of the boundary between behavioral neurology, neuropsychology, and psychoanalysis. Psychoanalytic Psychology. 1990; 7:487–527. Werner H, Kaplan B: Symbol Formation. Hillsdale, NJ, Erlbaum, 1963/1984. Wexler BE, Gottschalk CH, Fulbright RK, Prohovnik I, Lacadie CM, Rounsaville BJ, Gore JC: Functional magnetic resonsnace imaging of cocaine craving. Am J Psychiatry 2001;158:86–95. World Health Organization, 1992, Section V-Mental Disorders. International Statistical Classification of Diseases, Injuries, and Causes of Death, 10th Revision. Geneva: World Health Organization. Woodworth M, Porter S: In cold blood: characteristics of criminal homicides as a function of psychopathy. Journal of Abnormal Psychology. 111(3):436–45, 2002. Zohar J, Mueller EA, Insel TR, Zohar-Kadouch RC, Murphy DL: Serotonergic responsivity in obsessive-compulsive disorder: Comparison of patients and healthy controls. Arch Gen Psychiatry 44: 946–951, 1987. Zohar J, Insel TR, Berman KF, Foa EB, Hill JL, Weinberger DR: Anxiety and cerebral blood flow during behavioral challenge: Dissociation of central from peripheral and subjective measures. Arch Gen Psychiatry 45:167–172, 1988.

6 Neurobiology of Autism, Mental Retardation, and Down Syndrome What Can We Learn about Intelligence? Christopher J. Lawrence, Ira Lott, and Richard J. Haier

Structural and functional brain studies of autism, Down syndrome, and idiopathic mental retardation are reviewed to help identify brain areas and parameters that may be related to intelligence. Six brain factors appear common among people with autism, Down syndrome, and idiopathic mental retardation: cerebellum size, brain stem size, hippocampus size, abnormal dendritic development, whole brain size, and whole brain metabolism. How these areas may relate to intelligence is discussed.

WHAT MAKES THE BRAIN INTELLIGENT? What can we learn about the neurobiological basis of intelligence by studying people with aberrations of intelligence? Autism, mental retardation, and Down syndrome manifest a variety of intellectual deviations from low intelligence quotient (IQ) to aspects of genius. Most research on these disorders has focused Christopher J. Lawrence • Department of Neurobiology and Behavior, University of California, Irvine Ira Lott, and Richard J. Haier • Department of Pediatrics, University of California, Irvine

125

126

christopher j. lawrence et al.

on the underlying etiology. Independently, intelligence researchers have sought to understand the neurobiology of intelligence. We will review brain research in autism, mental retardation, and Down syndrome, and try to relate relevant findings to intelligence research. Neuropathological data allow an in vitro look at the brain. The advent of brain imaging technologies such as Positron Emission Tomography (PET) and functional Magnetic Resonance Imaging (fMRI) allow an in vivo look at the working brain. After a brief review of some animal work and intelligence research, we will review studies on autism, mental retardation, and Down syndrome, especially those that include an analysis of high and low-functioning individuals.

Animal Work and Intelligence Research In a comprehensive study of rats, Thompson and colleagues (1990) performed lesions to understand what parts of the brain underlie learning and problem solving. They discovered eight regions that significantly impaired performance on all of tests studied. These areas impairing general performance were the venterolateral thalamus, pontine reticular formation, dorsal caudatoputamen, globus pallidus, substantia nigra, ventral tegmental area, median raphe, and superior colliculus. There were also lesioned regions that impaired performance on some specific tests and not on others. These were the superior colliculus, posterior cingulate cortex, dorsal hippocampus, posterolateral hypothalamus, parietal cortex, and occipitotemporal cortex. Haier and colleagues (1993) compared these areas to the results of human PET studies of learning and problem solving. In one study (Haier et al., 1988), they used the Raven’s Advanced Progressive Matrices (a nonverbal test of abstract reasoning highly correlated with intelligence) and found brain areas related to abstract reasoning. In another study (Haier et al., 1992), they reported regional glucose metabolic rate (GMR) changes after subjects learned a complex task. All these areas from both human studies (putamen, superior colliculus, hippocampus, frontal lobe regions, posterior cerebral cortex, and cingulate gyrus ) were compared to the areas identified in the rat studies of problem solving from Thomson et al. (1990). The only direct overlap between these areas and the rat areas were the superior colliculus and hippocampus. This tentative comparison suggested that looking for brain areas commonly implicated in studies of learning and reasoning may help identify areas related to intelligence. This chapter will extend this kind of analysis by reviewing brain areas associated with other disorders where intelligence is abnormal—autism, Down syndrome, and mental retardation.

How are Autistic Brains Different? Autism occurs in anywhere from 4/10,000 births (Klauck et al., 1998) to 1/1000 births (Bailey et al., 1996). It is a clinical syndrome defined by deficiencies

Neurobiology of Autism, Mental Retardation and Down Syndrome

127

in emotion, communication, some motor coordination, social interaction, and social awareness (CAN Consensus Group, 1998). There is no stipulation on intelligence in the definition of autism; however, autism is frequently noted in the most extreme cases of intelligence—anywhere from severe retardation to genius. While many autistic individuals are quite intelligent, 50–75% are classed as at least mildly mentally retarded (Bailey et al., 1996). Ten percent of the autistic population is classed as autistic savant (Rimland & Fein, 1988), as seen in the movie “Rain Man.” Some may perform complex calculations typically only done with computers, and some may reproduce on piano a song heard only once (Rimland & Fein, 1988). The nature of intelligence defined in these individuals remains a mystery. Many inconsistencies are apparent across studies, possibly due to difficulties in diagnosing autism. Autism is diagnosed by a psychological profile determined by behavioral observations of both parents and their doctors. This is not a completely objective diagnosis; there is no conclusive test, such as a genetic test, that identifies autism. Sometimes the diagnosis is confused with other syndromes. Sometimes the confusion roots within the clinical field itself. For example, there is still some disagreement among physicians if autism and Asperger syndrome are part of the same neurologic disorder or are two separate ones. Asperger syndrome is sometimes described as “high-level autism.” Heterogeneity of autism within research samples, therefore, continues to confuse research findings. Differences Seen from Autopsy Post mortem studies yielded the first evidence of structural abnormalities in the autistic brain. For example, Bailey et al. (1993) reported that brain weight and volume are larger than controls, which yields a larger head circumference. Although macrocephaly [head circumference more than 2 SD above the mean] is not usually apparent at birth (Lainhart et al., 1997), it is frequently reported in adults. Three separate studies totalling over 200 individuals with autism reported macrocephaly in anywhere from 14%–42% (Bailey et al., 1993; Davidovitch et al., 1996; Lainhart et al., 1997). An analysis of subcortical structures and the limbic system reveals many irregularities. There is some evidence of increased neuronal packing density with shrunken cell size in the amygdala, anterior cingulate cortex, entorhinal cortex, hippocampus, mammilary body, septum, and subiculum (Bauman & Kemper, 1994). In addition, the hippocampus appears smaller than in “normal” brains (Rimland & Fein, 1988). This has not been replicated; however, Raymond and colleagues (1996) did find that hippocampal neurons were smaller with a decrease in dendritic branching. Autistic individuals also appear to have changes in the cerebellum. Ritvo and colleagues (1986) reported a reduction in Purkinjie cells. Bauman and Kemper (1994) also reported a reduction in Purkinjie cells in addition to a reduction of granular cells throughout the cerebellum.

128

christopher j. lawrence et al.

Structural Differences Seen from Brain Imaging Structural brain images have been obtained using magnetic resonance imaging (MRI) and computerized tomography (CT) scanning. Most studies have used the older CT scanning. Unfortunately, the resolution of CT is poorer than MRI, which limits detailed analysis. Gross anatomical differences, however, are easily discernable using either imaging technique. With CT scanning, two studies have reported a small number of cases of hemispheric asymmetry within the autistic brain (Hier et al., 1979; Damasio et al., 1980). However, in a review of imaging studies, Minshew and Dombrowski (1994) reported that studies overall have not supported this finding. Consistent with pathological data, a few studies have reported macrocephaly with MRI (Filipek et al., 1992; Piven et al., 1995; Piven et al., 1996). Piven et al. (1995) reported that the supratentorial volume of the autistic brain is larger than that of controls. Piven and colleagues (1996) followed that study with an analysis of lobe sizes. Significant enlargement was reported in the temporal, parietal, and occipital lobes, but not the frontal lobe (Piven et al., 1996). However, in an analysis of the parietal lobes, Courchesne and colleagues (1993) reported a loss of cortical volume in nine (43%) of 21 high-functioning patients. The ventricular system has been analyzed in several studies. Two studies using CT scanning found an enlargement of the lateral ventricles in 15–25 percent of individuals (Campbell et al., 1982; Rosenbloom et al., 1984). Using MRI, Piven and colleagues reported an increase in lateral ventricle volume as well. Hoshino and colleagues (1984) reported a tendency toward enlargement of the lateral ventricles as the patients got older when divided into age subgroups. Gaffney and Tsai (1987) selected a group of relatively high-functioning autistics (IQ = 60–135) and reported enlarged lateral ventricles as well. Enlargement of the lateral ventricles has not been entirely consistent, and this may be due to the screening techniques used and resolution of CT scanning. Two studies found that some individuals show an increase in size of the third ventricle (Campbell et al., 1982; Jacobson et al., 1988). However, this finding has not been consistent either. There is little evidence to support a difference in size of the fourth ventricle (Minshew & Dombrowski, 1994; Piven et al., 1992; Kleiman et al., 1992). However, three studies did find a larger fourth ventricle when analyzing high-functioning autistics (IQ range 60– 135) (Gaffney et al., 1987a; Gaffney et al., 1987b; Gaffney & Tsai, 1987). None of the evidence on the ventricular system has been conclusive for two main reasons— other neurologic diseases exist within the autistic population, and the resolution of CT scanning lacks sufficient detail. MRI may solve the problem of detail, but the problem of heterogeneity and comorbidity also need attention. The corpus callosum has also been studied in the autistic population. Courchesne and colleagues reported a thinning in the size of the corpus callosum in 2 (10%) of 21 high-functioning patients. In the related Asperger’s syndrome, Berthier

Neurobiology of Autism, Mental Retardation and Down Syndrome

129

(1994) reported a similar finding in three (16%) of 19 high-functioning patients. Filipek and colleagues (1992) reported no difference in size of the corpus callosum, but they included both low and high-functioning individuals. Piven and colleagues (1997) followed these studies also using MRI. When controlling sex, age, IQ, and for the increased brain size of 35 individuals with autism, they reported a smaller relative size of the corpus callosum (Piven et al., 1997). The cerebellum has yielded some conflicting results. Many studies have reported a reduced size and/or a hypoplasia [a reduced rate of synaptic density] of the cerebellum (Courchesne et al., 1987; Gaffney et al., 1987a; Gaffney et al., 1987b; Courchesne et al., 1988; Murakami et al., 1989; Hsu et al., 1991; Piven et al., 1992; Hashimoto et al., 1995). More specifically, five of these studies focused on reduced size and/or hypoplasia of the neocerebellum (Courchesne et al., 1987; Courchesne et al., 1988; Murakami et al., 1989; Hsu et al., 1991; Piven et al., 1992). One study reported no difference in the size of the cerebellum (Kleiman et al., 1992) and one study reported an increase in cerebellar volume (Piven et al., 1997). The brain stem is the final region in imaging analysis and contains the same disagreements as the prior regions. Consistent with Rodier et al. (1996), two studies reported a smaller brain stem than controls (Gaffney et al., 1988; Hashimoto et al., 1995). However, three studies contradict those findings and report no difference between autistic and controls (Hsu et al., 1991; Kleiman et al., 1992; Piven et al., 1992). Thus, there have been many conflicting reports on the structure of the autistic brain. However, consistencies have appeared in some brain structures. Macrocephaly is a common finding and abnormalities of the following have been reported: cerebellum, ventricles, corpus callosum, brain stem, and neuronal development. The relationship of there findings to intelligence is unclear. Functional Differences Seen from Brain Imaging Functional information has been obtained using imaging techniques such as positron emission tomography (PET) and functional MRI (fMRI). One of the first interesting functional aberrations reported in the autistic brain is a cerebral hypermetabolism with PET (Rapoport et al., 1983). Rumsey and colleagues (1985) reported a 20% global metabolic increase when compared to controls. This was measured using no cognitive task and having subjects lie on a bed with eyes and ears closed. Rumsey and colleagues (1985) did report a relative metabolic decrease in the right superior frontal and precentral gyri. A few studies have reported on metabolic hemispheric asymmetries (Rumsey et al., 1985; Horowitz et al., 1988, Buchsbaum et al., 1992; Baron-Cohen et al., 1994; Goel et al., 1996; Happe et al., 1996). This may coincide with the structural hemispheric asymmetries seen by Hier et al. (1979) and Damasio et al. (1980). However, Prior and colleagues (1984) reported that no cognitive asymmetries exist. Two

130

christopher j. lawrence et al.

different laboratories reported data of a temporal lobe hypometabolism (Chugani et al., 1996; Boddaert et al., 1998). In addition, Buchsbaum and colleagues (1992) reported that the right posterior frontal lobe had a reduced GMR. Implications of Structural and Functional Differences on Intelligence A review of the above information yields some interesting observations of the autistic brain and possibly some insight into the “normal” brain. Cortical abnormalities have been identified in all areas except for the occipital lobe. Subcortical and limbic system abnormalities are even more pervasive. The corpus callosum, midbrain, brain stem, and cerebellum have all been implicated in autism. Which of these abnormalities are causing the autistic disorder and which are causing the mental retardation frequently associated with it? Analyses comparing low and high-functioning individuals with autism may help to answer this question. Three studies have separated autistic individuals into high and lowfunctioning groups. Filipek and colleagues (1992) separated autistic individuals by an IQ of 80. Individuals with an IQ > 80 (n = 9) were placed in the highfunctioning group and those with an IQ < 80 (n = 13) were placed in the low functioning group. Filipek et al. found that a larger brain positively correlates with higher functioning. Conflicting reports of cerebellar size and hypo or hyperplasia [an increased rate of synaptic density] led Courchesne and colleagues (1994) to divide the autistic population into two groups. Courchesne and colleagues (1994) divided the groups by hypoplasia (n = 43) and hyperplasia (n = 6) of the cerebellar vermal lobules. After dividing into two groups, all individuals with hyperplasia were low-functioning (IQ < 70), and the only individuals who were high-functioning (IQ > 70) had hypoplasia (Courchesne et al., 1994). In addition, the higher functioning individuals had larger lobules than the lower functioning individuals with hypoplasia. Minshew (1994) also separated individuals with autism by IQ and reported a correlation with brain metabolism. Using magnetic resonance spectroscopy, she reported that the metabolic rate of autistic brains increased as the IQ’s decreased. This may imply that hypoplasia is correlative with a higher IQ and a brain of lesser IQ may actually have to work harder. As discussed below, using PET Haier et al. (1988; 1995) have reported an inverse correlation between brain metabolism and IQ.

Down syndrome Differences in the Brain Down syndrome occurs in one out of every 800 to 1000 live births (National Down syndrome Society, 1999). It can usually be diagnosed early due to many phenotypic differences and is conclusive with a blood test. Most individuals with

Neurobiology of Autism, Mental Retardation and Down Syndrome

131

Down syndrome have an IQ that falls within the mild (IQ = 50 to 70) to moderate range (IQ = 35 to 50) of mental retardation (National Down syndrome Society, 1999). Differences Seen from Autopsy An initial pathological inspection of the Down syndrome brain appears to reveal a reduced brain weight (Benda, 1969; Benda, 1971; Friede, 1975; Urich, 1976; Whalley, 1982; Wisniewski et al., 1985; Coyle et al., 1986). Benda (1971) also reported a compression of the occipital lobes. A few studies have reported a narrowed superior temporal gyrus in approximately 50% of cases (Friede, 1975; Urich, 1976; Zellweger, 1977; Kemper, 1988). Two separate studies have also reported a hypoplasia of the frontal lobes (Benda, 1971; Crome & Stern, 1972). Other cortical abnormalities are reported on neural density. Many studies have reported a decrease in neural density (Apert, 1914; Davidoff, 1928; Colon, 1972; Crome & Stern, 1972; Urich, 1976; Zellweger, 1977; Ross et al., 1984; Wisniewski et al., 1984; Wisniewski et al., 1986; Coyle et al., 1986; Becker et al., 1991) while only one has reported an increase (Norman, 1938). Two studies have reported an increase in synaptic density (Huttenlocher, 1974; Cragg, 1975). Synaptic density was later reported to change with age (Huttenlocher, 1979). Cortical pyramidal neurons have been reported to display a dysgenesis [abnormal development] of dendritic spines and impaired growth of basal dendrites (Marin-Padilla, 1976; Takashima et al., 1981; Huttenlocher, 1991). Subcortical abnormalities have also been reported in the Down syndrome brain. The hippocampus has been reported to have a decreased number of neurons (Sylvester, 1983). Also in the hippocampus, Suetsugu and Mehraein (1980) reported a decrease in the number of dendritic spines, and Coyle and colleagues (1986) reported abnormal dendritic spines. The brain stem and cerebellum have also been reported to be reduced in size (Davidoff, 1928; Crome et al., 1966; Wolstenholme, 1967; Friede, 1975; Urich, 1976; Gandolfi et al., 1981). Coyle and colleagues (1986) reported abnormalities of dendritic spines in the cerebellum as well. Structural Differences Seen from Brain Imaging MRI has been used to identify structural changes in the Down syndrome brain. In agreement with the pathological data, two studies have reported a whole brain volume of about 80% the size of controls (Haier et al., 1995 & Pulsifer, 1995). Raz et al. (1995) reported that Down syndrome brains have smaller cerebral and cerebellar hemispheres. Kesslak and colleagues (1994) reported a smaller frontal cortex. Wisniewski and colleagues (1986) reported shrinkage of the temporal lobe. However, the medial temporal gyrus has been reported to be larger than controls (Kesslak et al., 1994).

132

christopher j. lawrence et al.

Subcortical structures have also shown some differences with MRI technology. Raz and colleagues (1995) reported a smaller ventral pons, mammilary bodies, and hippocampal formations than controls. Kesslak et al. (1994) reported a smaller cerebellum and hippocampus. Larger brain areas were found in the Down syndrome group in the parahippocampal gyrus (Kesslak et al., 1994; Raz et al., 1995). Functional Differences Seen from Brain Imaging All functional studies reviewed here used [18F]-fluorodeoxyglucose (FDG) with PET and measured glucose metabolic rate. Three separate small sample studies have reported that Down syndrome shows a higher global GMR than controls (Schwartz et al., 1983; Cutler, 1986; Haier et al., 1995). More specifically, a higher GMR is evidenced in the left and right medial frontal, left and right cingulate, left and right caudate, left thalamus, left putamen, right medial temporal lobe, left uncus, left anterior cerebellum, and left occipital lobe (Haier et al., 1995). One study did report a decrease in global GMR, but this difference may be due to a lack of a cognitive task; the subjects sat idle with eyes open (Shapiro et al., 1990). Haier et al. (1998) also reported the inter-correlations among GMR in three areas: frontal lobe, caudate nucleus, and putamen. Whereas these areas were correlated to each other in controls and individuals with mild mental retardation, there was a lack of correlation between the caudate and frontal lobe in Down syndrome, suggesting a functional disconnection. Implications on Intelligence of Structural and Functional Differences Pathological study reveals a hypoplasia and perhaps a hyperplasia in Down syndrome. Like autism, there may be two subtypes of Down syndrome that fill these two categories, possibly separating high and low-functioning individuals. No Down syndrome studies have compared low and high-functioning subgroups. When compared to controls, similarities with autism include the following shrunken structures: hippocampus, cerebellum, and ventral pons (Rimland & Fein, 1988; Courchesne et al., 1994; Hsu et al., 1991; Kesslak et al., 1994; Raz et al., 1995). Like autism, Down syndrome may also have a higher GMR than controls (Rumsey et al., 1985; Haier et al., 1995).

What are Possible Causes to the Low IQ of Mental Retardation? Mental retardation is defined as an IQ of less than 70, and occurs in 1–3% of the population (Schaefer et al., 1994 & Pulsifer, 1995). Most mental retardation is classed as mild (IQ from 50 to 70) (Schaefer et al., 1994 & Pulsifer, 1995). Most mild mental retardation is idiopathic [has no identifiable etiology] (Pulsifer, 1995).

Neurobiology of Autism, Mental Retardation and Down Syndrome

133

Pulsifer (1995) reviewed pathological and imaging data of the five major identifiable prenatal causes of mental retardation in order to identify the neuropsychology of mental retardation. After a review of the literature, Pulsifer reported on both neuroanatomical abnormalities and cognitive deficits in each and every disorder. After reviewing fetal alcohol syndrome, Down syndrome, fragile X syndrome, Prader-Willi syndrome, Angelman syndrome and idiopathic mental retardation, Pulsifer found only two neuroanatomical sites with abnormalities common to all disorders: the hippocampus and cerebellum. Cognitive deficits common to all disorders were in attention, sequential information processing, and short-term memory. Differences Seen from Autopsy Yakovlev (1960) performed some of the first pathological research on idiopathic mental retardation. He reported a decreased weight of the forebrain, cerebellum, and brain stem. Benda (1971) also reported a decreased weight of the cerebellum and brain stem. Pathological data provides one major consistent finding regarding dendritic development. Dendritic arborization is sparse; basal dendrites are stunted and secondary branchings are diminished (Huttenlocher, 1974). Dendritic spines, which enhance synaptic communication, also appear to be sparse (Huttenlocher, 1974 & Purpura, 1974). Marin-Padilla (1972) also reported abnormalities of dendritic spines. Polednak looked at gross neuropathological findings of idiopathic mental retardation and separated them by level of retardation. Polednak (1977) reported that the incidence of microgyria [small gyri] and microcephaly [small brain] increased as IQ decreased. Atrophy or agenesis of the cerebellum and brainstem was also reported to be more common in profound mental retardation (Polednak, 1977). Agenesis [absence] or dysgenesis of the corpus callosum has also been reported as more common in the mentally retarded population (Jeret et al., 1986), but still only has a prevalence of 2–3% (Freytag & Lindenberg, 1967; Andermann, 1981). Structural Differences Seen from Brain Imaging Reports from structural imaging studies of idiopathic mental retardation reveal many inconsistencies. An initial computerized tomography (CT) study reported that 72% of patients had normal scans (Lingam et al., 1982). In another CT study one year later, Lingham and Kendall (1983) reported that 75% of scans were abnormal. However, CT scans lacked the current morphometric analysis techniques and resolution may have suffered. Even though CT resolution does not compare to that of MRI, analysis techniques have improved significantly and another study on mental subnormality was published in 1995. Prassopoulos and colleagues (1996) reported the third ventricle enlarged in 77% of cases and also a widening of the subarachnoid spaces in about 30% of cases. As the severity of

134

christopher j. lawrence et al.

the idiopathic mental retardation increased, they reported that all children showed third ventricle enlargement, widening of the subarachnoid spaces, and an enlargement of the lateral ventricles as well (Prassopoulos et al., 1995). Using MRI in 1995, Haier and colleagues reported a 20% decrease in whole brain size in mild mental retardation. Schaefer and colleagues published a recent study on neurogenetic syndromes using MRI in 1996. They revisited the study done by Courchesne and colleagues (1988) regarding hypoplasia of cerebellar vermal lobules found in autism. Schaefer et al. (1996) found this hypoplasia in idiopathic mental retardation but also in other forms of mental retardation. Schaefer & Bodensteiner (1992) also reported a hypoplasia of the corpus callosum. Abnormalities of the corpus callosum appear in many structural studies on mental retardation, but still amount to less than 10% of these individuals (Lingam et al., 1982; Jeret et al., 1986; Schaefer et al., 1991). Functional Differences Seen from Brain Imaging Functional analysis of idiopathic mental retardation is very much in its infancy. Most functional information on idiopathic mental retardation has been obtained using GMR with PET scanning. Chugani and colleagues (1987) performed the first functional study on individuals with mental retardation (n = 27). Like Shapiro et al. (1990) in his report on Down syndrome, Chugani and colleagues (1987) reported a decrease in GMR. Consistent with Shapiro et al. (1990), no cognitive task was used and subjects sat awake with eyes open. Haier and colleagues (1995) performed a study using a continuous performance task (CPT), and reported that mental retardation shows a 30% overall GMR increase over control individuals; this is a similar increase as seen in the Down syndrome individuals (Haier et al., 1995). More specifically, there is an increased GMR in the cingulate, globus pallidus, posterior temporal lobe, frontal white matter, and the cerebellum (Haier et al., 1995). Despite the overall increase, the following areas showed a relative GMR decrease: medial thalamus, anterior cingulate, medial temporal lobe, superior frontal cortex, posterior gyrus rectus, putamen, and uncus (Haier et al., 1995). Implications on Intelligence of Structural and Functional Differences Considering the paucity of information on imaging studies of idiopathic mental retardation, it is difficult to develop reasonable conclusions or theories. Many studies have identified particular brain regions abnormal or missing, but few have controlled for idiopathic mental retardation. However, abnormalities of the hippocampus and cerebellum are frequently mentioned and the brain stem is frequently implicated as well. Dendritic development is consistently mentioned in the literature and may have strong implications on intelligence. Because most idiopathic mental retardation is mild, it is difficult to separate groupings by severity

Neurobiology of Autism, Mental Retardation and Down Syndrome

135

within the class of idiopathic mental retardation. One study that did group by severity found that intelligence suffered as the likelihood of abnormalities in the cerebellum and brain stem increased (Polednak, 1977). Another theme that comes from the functional data is a difference between resting and active GMR. In studies of mentally retarded individuals at rest, a lower GMR has been reported. However, when similar studies required mentally retarded individuals to work at a task, they show a higher GMR than controls.

Discussion Based on this review of autism, Down syndrome, and mental retardation, six factors may be linked to intelligence. These are brain size, brain metabolism, neuronal/dendritic development, cerebellum size, brain stem size, and hippocampus size. It is interesting to note the lack of cortical areas implicated in this review. Higher intelligence is traditionally thought to be located in the frontal lobes. However, the frontal lobes are rarely implicated in the literature. It is important to discern if the abnormality is causing a low intelligence or if the low intelligence is causing the abnormality. It is equally important to keep in mind that these abnormalities may be affecting performance and not intelligence. Anderson (in press) recently completed an analytic review of the brain size/IQ relationship. Anderson reviewed studies of brain size and intelligence and reported positive correlations of anywhere from r = 0.35 to r = 0.69. Autism is frequently associated with macrocephaly or a big brain. Autism has a relatively low incidence of mental retardation when compared to Down syndrome, and some individuals show an exceptionally high IQ. Within autism itself, a larger brain has been reported to have a higher IQ (Filipek, 1992). Down syndrome frequently results in mental retardation, and Down syndrome has been reported to have a decreased brain size (Haier et al., 1995). Idiopathic mental retardation has also been reported to have a decreased brain size (Haier et al., 1995). While there is a definite link between brain size and IQ, the nature of why larger brains go with higher IQ is not apparent. Brain metabolism appears to have significant correlation with intelligence. Both Down syndrome and idiopathic mental retardation have been reported to have a decreased GMR in a resting state (Chugani et al., 1987; Shapiro et al., 1990). However, both Down syndrome and idiopathic mental retardation were reported to have an increased GMR when doing an active task (Haier et al., 1995). Both autistic and “normal” brains have been reported to have an inverse correlation with metabolic rate and IQ (Haier et al., 1988; Minshew, 1994). A brain that is less intelligent appears to need to work harder; perhaps a brain that is less intelligent is less efficient with its use of glucose. Autism was reported to have an increased GMR in the resting state (Rapoport et al., 1983). Autism provides an interesting dichotomy here. Autistic individuals have a larger brain, but they have a lower IQ

136

christopher j. lawrence et al.

Autism

Down Syndrome

Macrocephaly [large brain] Hypoplasia [a reduced rate of synaptic density]

Hypoplasia of frontal lobes

in cerebellum Increased neuronal packing density GMR with no cognitive task

Impaired dendritic development Smaller brain stem, cerebellum, and hippocampus

GMR with cognitive task

Larger lateral ventricles

GMR with no cognitive task Microcephaly [small brain]

Larger third ventricle Smaller forebrain

Idiopathic Mental Retardation

Figure 1. Converging brain areas for Autism, Down Syndrome, and Idiopathic Mental Retardation.

and an increased GMR. Perhaps in autism a larger brain is not more intelligent because it is not entirely efficient. Neuronal/dendritic development yields conflicting reports. Autism has been reported to have an increased neural density (Bauman & Kemper, 1994) while Down syndrome has been reported to have a decreased neural density (Apert, 1914; Davidoff, 1928; Colon, 1972; Crome & Stern, 1972; Urich, 1976; Zellweger, 1977; Ross et al., 1984; Wisniewski et al., 1984, 1986; Coyle et al., 1986; Becker et al., 1991). The wide range in IQ in autism and the typically low IQ in Down syndrome may be explained by neural density. Huttenlocher (1979) reported on synaptic density in the brain over the lifetime of “normal” individuals. It changes throughout life, increasing by 70% from birth to about four years of age, and returning to baseline by the early twenties. A decrease in synaptic density appears in the early sixties and continues until death. Information on hypo/hyperplasia should account for changes in age. Hypoplasia may be due to many things including a decrease in dendritic development or an increase in synaptic pruning. A decrease in dendritic development may be the result of a less stimulating environment. Synaptic pruning may correlate with the decreased GMR seen in high-functioning individuals. Haier (1993) noted the possible role of pruning abnormalities with

Neurobiology of Autism, Mental Retardation and Down Syndrome

137

hypothetical variations on Huttenlocher’s synaptic count data. Haier hypothesized that a lack of synaptic pruning might result in an overactive brain and mental retardation; increased pruning may result in a particularly efficient brain and increased intelligence, but too much pruning may result in damage or a psychiatric disorder. The cerebellum was implicated in autism, Down syndrome, and idiopathic mental retardation. All three disorders are reported to have a smaller cerebellum and abnormalities of the dendrites (Ritvo et al., 1986; Bauman & Kemper, 1994; Benda, 1971; Crome & Stern, 1972; Marin-Padilla, 1976; Takashima et al., 1981; Huttenlocher, 1991). Autism and idiopathic mental retardation are reported to have hypoplasia, and Down syndrome is reported to have abnormal dendritic spines. The cerebellum is traditionally implicated in balance and motor movement, but perhaps it holds higher functions as well. The brain stem has been reported to be reduced in size in autism, Down syndrome, and idiopathic mental retardation (Rodier et al., 1996; Gaffney et al., 1988; Hashimoto et al., 1995; Davidoff, 1928; Crome et al., 1966; Wolstenholme, 1967; Gullotta & Redher, 1974; Friede, 1975; Urich, 1976; Gandolfi et al., 1981; Polednak, 1977). There is little evidence that the brain stem is involved in intelligence, but not much data exists. The last factor of intelligence implicated in this review is the hippocampus. The hippocampus is smaller in size and also contains dendritic abnormalities (Rimland & Fein, 1988; Kesslak et al., 1994; Raz et al., 1995; Pulsifer, 1995). This is also the only brain structure implicated by Haier and colleagues (1993) in their review of human and animal intelligence. The hippocampus has been significantly studied over the years and has been heavily associated with memory—both in forming new memories and remembering old ones. Individual differences in shortterm memory have been proposed as the basic element of general intelligence (Kyllonen & Christal, 1990; Kyllonen, 1996). Pulsifer also implicated a deficit in short-term memory in her review of mental retardation. It seems reasonable that an abnormality in the hippocampus is going to affect learning and possibly cortical development. If the hippocampus is not working correctly, stimulation to the surrounding cortex may be impaired. This may provide an explanation to hypoplasia by a lack of environmental stimulation. If plasticity is dependent on stimulation, and the hippocampus is required for this stimulation, then a dysfunctional hippocampus might hinder development and therefore adversely affect intelligence. It appears that there is no one location in the brain where intelligence is found. Many factors appear to affect the intelligence of the individual—anywhere from gross anatomical differences such as brain size to the microscopic differences such as synaptic branching. However, it seems that some parts of the brain play a key role in influencing intelligence. Of all the different parts of the brain, the following seem to have a large effect on intelligence and deserve further study: brain size, metabolic rate, dendritic development, cerebellum, brain stem, and hippocampus.

138

christopher j. lawrence et al.

REFERENCES Andermann, E. (1981) “Agenesis of the corpus callosum,” in P. J. Vinken & G. W. Bruyn (Eds.), Handbook of clinical neurology. Vol 42. (pp. 6–9). Amsterdam: Elsevier/North-Holland Biomedical Press Apert, E. (1914) “Mongolism.” Le Monde Medicale 24:201. Bailey, A., Phillips, W., & Rutter, M. (1996) “Autism: towards an integration of clinical, genetic, neuropsychological, and neurobiological perspectives.” Journal of Child Psychology and Psychiatry 37:89–126. Bailey, A., Luthert, P., Bolton, P. et al. (1993) “Autism and megencephaly.” Lancet 349 (9067):1744– 1745. Baron-Cohen, S., Ring, H., Moriarty, J., Schmitz, B., Costa, D., Ell, P. (1994) “Recognition of mental state terms. Clinical findings in children with autism and a functional neuroimaging study in normal adults.” British Journal of Psychiatry 165:640–649. Bauman, M. L., and Kemper, T. L. (1994) “Neuroanatomic brain observations of the brain in autism,” in M. L. Bauman & T. L. Kemper (Eds.), The Neurobiology of Autism (pp. 119–145). Baltimore: The John Hopkins University Press. Becker, L. E., Armstrong, D. L. & Chan, F. (1986) “Dendritic atrophy in children with Down’s syndrome.” Annals of Neurology 20:520–526. Benda, C. (1969) Down’s Anomaly, 2nd ed., New York: Grune and Stratton. Benda, C. E. (1971) “Mongolism,” in J. Mencken (Ed.), Pathology of the Nervous System (pp. 1361– 1371). New York: McGraw-Hill. Berthier, M. L. (1994) “Corticocallosal anomalies in Asperger’s syndrome.” American Journal of Roentonology, Letters 162:236–237. Boddaert, N., Zilbovicious, M., Barthelemy, C. et al. (1998) “Individual detection of temporal hypoperfusion with PET in autistic children.” Society for Neuroscience 24:819. Buchsbaum, M. S., Siegal, B. V., Jr., Wu, J. C. et al. (1992) “Brief report: attention performance in autism and regional brain metabolic rate assessed by positron emission tomography.” Journal of Autism and Developmental Disorders 22:115–125. Campbell, M. S., Rosenbloom, S., Perry, R. et al. (1982) “Computerized axial tomography in young autistic children.” American Journal of Psychiatry 139:510–512. Chugani, H. T., Silva, E. D., & Chugani, D. C. (1996) “Infantile spasms: III. Prognostic implications of bitemporal hypometabolism positron emission tomography.” Annals of Neurology 39(5):643–649. Chugani, H. T., Phelps, M. E., Light, R. K. et al. (1987) “Metabolic correlates of mental retardation in children determined with FDG positron emission tomography (PET).” Journal of Cerebral Blood Flow and Metabolism 7(Suppl. 1):533 Colon, E. Y. (1972) “The structure of the cerebral cortex in Down’s syndrome: A quantitative analysis.” Neuropaediatrie 3:362–376. Courchesne, E., Hessenlink, J. R., Hernigan, T. L. et al. (1987) “Abnormal neuroanatomy in a nonretarded person with autism: unusual findings with magnetic resonance imaging.” Archives of Neurology 44:335–341. Courchesne, E., Press, G. A. & Yeung-Courchesne, R. (1993) “Parietal lobe abnormalities detected with MR in patients with infantile autism.” American Journal of Roentgenology 160(2):387–393. Courchesne, E., Saitoh, O., Yeung-Courchesne, R. et al. (1994) “Abnormality in the cerebellar vermian lobules VI and VII in patients with infantile autism: identification of hypoplastic and hyperplastic subgroups with MR imaging.” American Journal of Roentonology 162:123–130. Courchesne, E., Yeung-Courchesne, R., Press, G. A., et al. (1988) “Hypoplasia of cerebellar vermal lobules VI and VII in autism.” New England Journal of Medicine 318: 1349–1354. Coyle, J. T., Oster-Granite, M. L. & Gearhart, J. D. (1986) “The neurobiologic consequences of Down syndrome.” Brain Research Bulletin 16:773–787.

Neurobiology of Autism, Mental Retardation and Down Syndrome

139

Cragg, B. G. (1975) “The density of synapses and neurons in normal, mentally defective ageing human brains.” Brain 98(1):81–90. Crome, L. & Stern, J. (1972) Pathology of Mental Retardation. Baltimore: William & Wilkins. Crome, L., Cowie, V. & Slater, E. (1966) “A statistical note on cerebellar and brain-stem weight in mongolism.” Journal of Mental Deficiency Research 10:69–72. Cure Autism Now (CAN) Consensus Group (1998) “Autism screening and diagnostic evaluation: CAN consensus statement.” CNS Spectrum 3(3):40–49. Cutler, N. (1986) “Cerebral metabolism as measured with PET and FDG: Healthy aging, Alzheimer’s disease and Down’s syndrome.” Progress in Neuro-Psychopharmacology & Biological Psychiatry 10:309–321. Damasio, H., Maurer, R. G., Damsio, A. R. et al. (1980) “Computerized tomographic scan findings in patients with autistic behavior.” Archives of Neurology 37:504–510. Davidoff, L. M. (1928) “The brain of mongolian idiocy.” Archives of Neurological Psychiatry 20:1229– 1257. Davidovitch, M., Patterson, B. & Gartside, P. (1996) “Head circumference measurements in children with autism.” Journal of Child Neurology 11(5):389–93. Filapek, P.A. (1992) “Morphometric analysis of the brain in developmental language disorders and autism.” Annals of Neurology 32:475. Freytag, E. & Lindenberg, R. (1967) “Neuropathologic findings in patients of a hospital for the mentally deficient. A survey of 359 cases.” Johns Hopkins Medical Journal 121:379–392. Friede, R. L. (1975) Developmental Neuropathology Berlin: Springer-Verlag. Gaffney, G. R., Kuperman, S., Tsai, L. Y. et al. (1988) “Morphological evidence of brainstem involvement in infantile autism.” Biological Psychiatry 24:578–586. Gaffney, G. R., & Tsai, L. Y. (1987) “Brief report: magnetic resonance imaging of high level autism.” Journal of Autism and Developmental Disorders 17:433–438. Gaffney, G. R., Kuperman, S., Tsai, L. Y. et al. (1987a) “Midsagittal magnetic resonance imaging of autism.” Bristish Journal of Psychiatry 151:831–833. Gaffney, G. R, Tsai, L. Y., Kuperman, S. et al. (1987b) “Cerebellar structure in autism.” American Journal of Diseases of Children 141:1330–1332. Gandolfi, A., Horoupian, D. S. & DeTeresa, R. M. (1981) “Pathology of the auditory system in autosomal trisomies with morphometric and quantitative study of the central cochlear nucleus.” Journal of Neurological Science 51:43–50. Goel, V., Grafman, J., Sadato, N., and Hallett, M. (1996) “Modeling other minds.” Neuroreport 6(13):1741–1746. Haier, R. J., Chueh, D., Touchette, P. et al. (1995) “Brain size and cerebral glucose metabolic rate in nonspecific mental retardation and Down syndrome.” Intelligence 20:191–210. Haier, R. J.; Siegel, B. V. Jr.; Crinella, F. M. et al. (1993) “Biological and psychometric intelligence: Testing an animal model in humans with positron emission tomography” in: D. K. Detterman et al. (Ed), Individual differences and cognition. (pp. 157–170). Norwood: Ablex Publishing Corp. Haier, R. J., Siegel, B. V., MacLachlan, A. et al. (1992) “Regional glucose metabolic changes after learning a complex visuospatial/motor task: a positron emission tomographic study.” Brain Research 570(1–2):134–43. Haier, R. J., Siegel, B. V., Nuechterlein, K. H. et al. (1988) “Cortical glucose metabolic rate correlates of abstract reasoning and attention studied with positron emission tomography.” Intelligence 12(2):199–217. Happe, F., Ehlers, S., Fletcher, P. et al. (1996) “Theory of mind’ in the brain. Evidence from a PET scan study of Asperger syndrome.” Neuroreport 8(1):197–201. Hashimoto, T., Tayama, M., Murakawa, K. et al. (1995) “Development of the brainstem and cerebellum in autistic patients.” Journal of Autism and Developmental Disorders 25:1–18.

140

christopher j. lawrence et al.

Hier, D. B., LeMay, M., & Rosenberger, P. B. (1979) “Autism and unfavorable left-right asymmetries of the brain.” Journal of Autism and Developmental Disorders 9:153–159. Horowitz, B., Rumsey, J. M., Grady, C. L., and Rappoport, S. I. (1988) “The cerebral metabolic landscape in autism. Intercorrelations of regional glucose utilization.” Archives of Neurology 45:749–755. Hoshino, U., Manome, T., Kaneko, M. et al. (1984) “Computed tomography of the brain in children with early infantile autism.” Folia Psychiatrica et Neurologica Japonica 38:33–34. Hsu, M., Yeung-Courchesne, R., Courchesne, E., and Press, G. A. (1991) “Absence of magnetic resonance imaging evidence of pontine abnormality in infantile autism.” Archives of Neurology 48:1160–1163. Huttenlocher, P. R. (1974) “Dendritic development in neocortex of children with mental defect and infantile spasms.” Neurology 24(3):203–10. Huttenlocher, P. R. (1978) “Synaptic density in human frontal cortex—developmental changes and effects of aging.” Brain Research 163:195–205. Huttenlocher, P. R. (1991) “Dendritic and Synaptic pathology in mental retardation.” Pediatric Neurology 7(2):79–85. Jacobson, R., LeCouteur, A., Howlin, P. et al. (1988) “Selective subcortical abnormalities in autism.” Psychological Medicine 18:39–48. Jeret, J. S., Serur, D., Wisniewski, K. et al. (1986) “Frequency of agenesis of the corpus callosum in the developmentally disabled population as determined by computerized tomography.” Pediatric Neuroscience 12:101–103. Kemper, T. L. (1988) “Neuropathology of Down syndrome,” in L. Nadel et al. (Ed.), The psychobiology of Down syndrome (pp. 269–289). Cambridge: Mit Press. Kesslak, J. P., Nagata, S. F., Nalcioglu, O. et al. (1994) “Magnetic resonance imaging analysis of age related changes in the brains of individuals with Down syndrome.” Neurology 44:1039– 1045. Klauk, S. M. (1998) “Searching for susceptibility genes for infantile autism.” American Journal of Medical Genetics 81(6):481. Kleiman, M. D., Neff, S., & Rosman, P. (1992) “The brain in infantile autism: are the posterior fossa structures abnormal?” Neurology 42:753–760. Kyllonen, P. C. (1996) “Is working memory capacity Spearman’s g?” in I. Dennis & P. Tapsfield et al. (Eds.), Human abilities: Their nature and measurement (pp. 49–75). Mahwah: Lawrence Erlbaum Assoc. Kyllonen, P. C. & Christal, R. E. (1990) “Reasoning ability is (little more than) working-memory capacity?” Intelligence 14(4):389–433 Lainhart, J. E., Piven, J., Wzorek, M. et al. (1997) “Macrocephaly in Children and Adults With Autism” in The Journal of the American Academy of Child Adolescence and Psychiatry 36(2): 282–290. Lingam, S., Read, S., Holland, I. M. et al. (1982) “Value of computerised tomography in children with non-specific mental subnormality.” Archives of Disease in Childhood 57:381–383. Marin-Padilla, M. (1972) “Structural abnormalities of the cerebral cortex in human chromosomal aberrations: A Golgi study.” Brain Research 44:625–629. Marin-Padilla, M. (1976) “Pyramidal cell abnormalities in the motor cortex of a child with Down’s syndrome: A Golgi study.” Journal of Comparative Neurology 167:63–82. Minshew, N. J. (1994) “In vivo brain chemistry of autism 31 P magnetic resonance spectroscopy studies,” in (Eds.) M. L. Bauman, MD & T. L. Kemper The Neurobiology of Autism (pp. 86–101). Baltimore: The John Hopkins University Press. Minshew, N. J., and Dombrowski, S. M. (1994) “In vivo neuroanatomy of autism: neuroimaging studies,” in M. L. Bauman & T. L. Kemper (Eds.), The Neurobiology of Autism (pp. 66–85). Baltimore: The John Hopkins University Press.

Neurobiology of Autism, Mental Retardation and Down Syndrome

141

Murakami, J.W., Courchesne, E., Press, G. A. et al. (1989) “Reduced cerebellar hemisphere size and its relationship to vermal hypoplasia in autism.” Archives of Neurology 46:689–694. Norman, R. M. (1938) “Some observations on the depth and nerve cell content of the supragranular cortex in normal and mentally defective persons.” Journal of Neurological Psychiatry 1:198–210. Piven, J., Arndt, S. A., Bailey, J. et al. (1995) “An MRI study of brain size in autism.” American Journal of Psychiatry 152(8):1145–1149. Piven, J., Arndt, S., Bailey, J. et al., (1996) “Regional brain enlargement in autism: a magnetic resonance imaging study.” The Journal of the American Academy of Child Adolescence and Psychiatry 35(4):530–536. Piven, J., Bailey, J., Ranson, B. J. et al. (1997) “An MRI study of the corpus callosum in autism.” American Journal of Psychiatry 154(8):1051–1056. Piven, J., Nehme, E., Simon, J. et al. (1992) “Magnetic resonance imaging in autism: measurement of the cerebellum, pons, and fourth ventricle.” Biological Psychiatry 31:491–504. Polednak, A. P. (1977) “Post-mortem neuropathology in a mentally retarded population.” The Lancet 1(8009):492–3. Prassopoulos, P., Cavouras, D., Ioannidou, M. et al. (1996) “Study of subarachnoid spaces in children with idiopathic mental retardation.” Journal of Child Neurology 11(3):197–200. Prior, M. R., Tress, B., Hoffman, M. L. et al. (1984) “Computed tomographic study of children with classic autism.” Archives of Neurology 41:482–484. Pulsifer, M. (1996) “The neuropsychology of mental retardation.” Journal of the International Neuropsychological Society 2:159–176. Purpura, D. P. (1973) “Normal and aberrant development of synaptic pathways in human hippocampus.” Transactions of the American Neurological Association 98:224–226. Purpura, D.P. (1974) “Dendritic spine ‘dysgenesis’ and mental retardation. Science 186(4169):1126– 1128. Rapoport. J. L., Rumsey, J., Duara, R. et al. (1983) “Cerebral metabolic rate for glucose in adult autism as measured with positron emission tomography (PET).” Journal of Cerebral Blood Flow and Metabolism 3(Suppl. 1):264–265. Raymond, G. V., Bauman, M. L. & Kemper, T. L. (1996) “Hippocampus in autism: a Golgi analysis.” Acta Neuropathology 91:117–119. Raz, N., Torres, I. J., Briggs, S. D. et al. (1995) “Selective neuroanatomic abnormalities in Down’s syndrome and their cognitive correlates: evidence from MRI morphometry.” Neurology 45(2): 356–66. Rimland, B. & Fein, D. (1988). Special Talents of Autistic Savants. In L.K. Obler & D. Fein (Eds.), The Exceptional Brain: Neuropsychology of Talent and Special Abilities (pp. 474–492). New York: The Guilford Press. Ritvo, E. R., Freeman, B. J., Scheibal, A. B. et al. (1986) “Lower Purkinjie cell counts in the cerebella of four autistic subjectsL intial findings of the UCLA-NSAC autopsy research report.” American Journal of Psychiatry 146:862–866. Rodier, P. M., Ingram, J. L., Tisdale, B. et al. (1996) “Embryological origin for autism: developmental anomalies of the cranial nerve motor nuclei.” The Journal of Comparative Neurology 370:247– 261. Rosenbloom, S., Campbell, M. & George, A. E. (1984) “High resolution CT scanning in infantile autism: A quantitative approach.” Journal of the American Academy of Child Psychiatry 23:72– 77. Ross, H. M., Galaburda, A. M. & Kemper, T. L. (1984) “Down’s syndrome: Is there a decreased population of neurons?” Neurology 34:909–916. Rumsey, J. M., Duara, R., Grady, C. et al. (1985) “Brain metabolism in autism: resting cerebral glucose utilization as measured with positron emission tomography (PET).” Archives of General Psychiatry 15:448–457.

142

christopher j. lawrence et al.

Schaefer, G.B. & Bodensteiner, J.B. (1992) “Evaluation of the child with idiopathic mental retardation.” Pediatric Clinics of North America 39(4):929–43. Schaefer, G. B., Bodensteiner, J. B., Thompson, J. N. et al. (1991) “Clinical and morphometric analysis of the hypoplastic corpus callosum.” Archives of Neurology 48:933–36. Schaefer, G. B., Sheth, R. D. & Bodensteiner, J. B. (1994) “Cerebral dysgenesis: an overview.” Pediatric Neurogenetics 12(4):773–788. Schaefer, G. B., Thompson, J. N., Bodensteiner, J. B. et al. (1996) “Hypoplasia of the cerebellar vermis in neurogenetic syndromes.” Annals of Neurology 39(3):382–385. Schwartz, M., Duara, R., Haxby, J. et al. (1983) “Down’s syndrome in adults: Brain metabolism.” Science 221:781–783. Shapiro, M. B., Grady, A., Kumar, A. et al. (1990) “Regional glucose metabolism is normal in young adults with Down syndrome.” Journal of Cerebral Blood Flow and Metabolism 10:199–206. Suetsugu, M. & Mehraein, P. (1980) “Spine distribution along the apical dendrites of the pyramindal neurons in Down’s syndrome.” Acta Neuropathology 50:207–210. Sylvester, P. E. (1983) “The hippocampus in Down’s syndrome.” Journal of Mental Deficiency Research 27:227–236. Takashima, S., Becker, L. E., Armstrong, D. L. et al. (1981) “Abnormal neuronal development in the visual cortex of the human fetus and infant with Down’s syndrome: A quantitative and qualitative Golgi study.” Brain Research 225:1–21. Thompson, R. F., Crinella, F.M. & Yu, J. (1990) Brain mechanisms in problem solving and intelligence. New York and London: Plenum Press. Urich, H. (1976) “Malformations of the nervous system, perinatal damage and related conditions early in life,” in W. Blackwood (Ed.), Greenfield’s Neuropathology (pp. 361–496). Chicago: Year Book Medical Whalley, L. J. (1982) “The dementia of Down’s syndrome and its relevance to aetiological studies of Alzheimer’s disease.” Annals of New York Academy of Sciences 396:39–54. Wisniewski, H. M. & Rabe, A. (1986) “Discrepancy between Alzheimer-type neuropathology and dementia in persons with Down’s syndrome.” Annals of New York Academy of Sciences 477:247– 259. Wisniewski, K. E., Laure-Kamionowska, M. & Wisniewski, H. M. (1984) “Evidence of arrest of neurogenesis and synaptogenesis in brains of patients with Down’s syndrome.” New England Journal of Medicine 311:1187–1188. Wisniewski, K. E., Wisniewski, H. M. & Wen, G. Y. (1985) “Occurrence of neuropathological changes and dementia of Alzheimer’s disease in Down’s syndrome.” Annals of Neurology 17:278–282. Wolstenholme, G. E. W. (1967) “Mongolism.” CIBA Foundation Study Group 25:1–99. Yakovlev, P. (1960) “Anatomy of the human brain and the problem of mental retardation,” in P. W. Bowan & H. V. Maotner (Eds.), Proceedings of the First International Conference in Mental Retardation (pp. 1–43). New York: Grune & Stratton. Zellweger, H., (1977) “Down syndrome,” in P. J. Vinken & G. W. Buryn (Eds.), Handbook of Clinical Neurology Volume 31, Part II (pp. 367–469). Amsterdam: North-Holland.

7 Neurobiology of ADHD Maree Farrow, Florence Levy, and Richard Silberstein

INTRODUCTION Attention-deficit/hyperactivity disorder (ADHD) is a childhood psychiatric disorder, estimated to affect 3 to 5% of school-aged children (American Psychiatric Association, 1994; Barkley, 1997). Children diagnosed with ADHD vary widely in the type and severity of symptoms that they demonstrate, but the disorder is generally characterized by developmentally inappropriate levels of inattention, impulsivity, and hyperactivity (American Psychiatric Association, 1987, 1994; Barkley, 1997; Cantwell, 1996). The diagnostic criteria for ADHD, specified in the fourth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV: American Psychiatric Association, 1994), divides symptoms into three domains of inattention, hyperactivity, and impulsivity. Inattention symptoms include difficulty concentrating and distractibility, impulsivity symptoms include acting without thinking and taking risks, and hyperactivity symptoms include being constantly “on the go” and excessive restlessness or fidgeting. The DSM-IV describes three subtypes of ADHD—predominantly inattentive type, predominantly hyperactive-impulsive type, and combined type, which describes children who exhibit features of all three symptom domains (American Psychiatric Association, 1994). Maree Farrow • School of Psychology, Psychiatry and Psychological Medicine, Monash University, Victoria, Australia. Florence Levy • School of Psychiatry, University of New South Wales, New South Wales, Australia. Richard Silberstein • Brain Sciences Institute, Swinburne University, Melbourne, Australia.

143

144

maree farrow et al.

These diagnostic criteria result in a heterogeneous group of children being diagnosed with ADHD. Between 50 and 80% of these children also meet criteria for other psychiatric disorders or learning disabilities (Biederman et al., 1991; Cantwell, 1996; Tannock, 1998). Comorbid diagnoses commonly seen in children with ADHD include conduct disorder, oppositional defiant disorder, learning disabilities, affective disorders, and anxiety disorders. Social difficulties, low selfesteem, and aggression are also commonly associated problems (Barkley, 1997; Whalen, 1989; Wood, 1995). The problem behaviours associated with ADHD usually appear early in a child’s development and are sustained over a long period of time, continuing into adolescence in 50 to 80% of cases and into adulthood in 30 to 50% of these cases (Barkley, 1997). Severe childhood ADHD can be associated with behavioural, social and academic problems that lead to difficulties coping with everyday life. Adolescents and adults with a history of ADHD are at greater risk for antisocial problems, alcohol and drug abuse, criminal behaviour, and academic and employment difficulties (Barkley, 1997; Castellanos, 1997; Fischer et al., 1990). The most common treatment for ADHD is the prescription of the stimulant drugs methylphenidate (Ritalin) and dexamphetamine (Dexedrine). The efficacy of these drugs in alleviating the symptoms of ADHD in around 80% of cases has been well documented (Barkley, 1998; Campbell et al., 1989; Erickson, 1987; Jarman, 1996; Schachar, 1991; Solanto, 1998). They improve attention, concentration and self-control, while reducing impulsive behaviour, restlessness, motor overactivity and aggression (Barkley, 1995, 1998; Campbell et al., 1989; Jarman, 1996; Wood, 1995). Stimulants have a short half-life, producing therapeutic effects within 20 to 60 minutes after ingestion, which then dissipate within 3 to 7 hours (Dinklage & Barkley, 1992; Jarman, 1996). These drugs are thought to enhance the activity of dopamine and noradrenaline, principally by blocking their reuptake (Bradley, 1989; Cooper et al., 1986; Solanto, 1998). It is believed that these short-term effects on neurotransmitter systems are responsible for the improved attention, reduced motor activity and reduced impulsivity that is seen in children with ADHD treated with these drugs. Non-pharmacological treatments for ADHD include behavioural and cognitive therapies such as positive reinforcement for appropriate behaviour, behaviour modification training, social-skills training, and special education programs (Cantwell, 1996; Dinklage & Barkley, 1992; Gelfand et al., 1988; Jarman, 1996; Levine, 1984; Wood, 1995). It is generally recognised that ADHD is best managed using a multiple-modality approach, combining psychosocial and pharmacological interventions (Cantwell, 1996; Dinklage & Barkley, 1992; Swanson et al., 1998b). Despite extensive research, the underlying neurobiological mechanisms involved in ADHD are not well understood and its cause is not known (Barkley, 1998; Cantwell, 1996; Zametkin, 1995), although several theories to explain the disorder have been suggested. Brain damage, frontal lobe dysfunction, genetic factors,

Neurobiology of ADHD

145

developmental lag, neurotransmitter imbalance, dysfunction of the behavioural inhibition system, psychosocial influences and food allergies have all been implicated as possible causes of or contributing factors in the development of ADHD (Barkley, 1990; Gelfand et al., 1988; Fadely & Hosler, 1992; Whalen, 1989). As ADHD is a heterogeneous disorder there is likely to be multiple etiologies which might combine genetic predisposition, brain dysfunction and psychosocial factors (Cantwell, 1996; Levine, 1984; Tannock, 1998; Wood, 1995). Most recently, disinhibition has come to be seen as a core deficit in ADHD (Barkley, 1996, 1997; Pennington & Ozonoff, 1996; Quay, 1997; Schachar, 1991; Tannock, 1998), but there is disagreement about the precise nature of this deficit and several different inhibitory dysfunction models of ADHD have been proposed (Quay, 1988, 1997; Schachar et al., 1995; Schachar & Logan, 1990; Barkley, 1996, 1997; Sonuga-Barke, 1994; Sonuga-Barke et al., 1992a, 1992b, 1996; Van der Meere, 1996). In each of these models, a primary deficit in inhibition is hypothesised to be responsible for the myriad of attentional and behavioural problems associated with ADHD.

NEUROCHEMISTRY The positive response of children with ADHD to treatment with stimulant medications, which have dopaminergic and noradrenergic agonist activity, suggests catecholamine abnormalities in ADHD (Cantwell, 1996; Pliszka et al., 1996). As direct measurement of catecholamine concentrations in the brain is not possible, evidence for differences in ADHD has been sought from measurements of catecholamine metabolite concentrations in cerebrospinal fluid (CSF), blood and urine. The results of these studies are inconsistent (Cantwell, 1996; Mason, 1984; Pliszka et al., 1996; Raskin et al., 1984; Weizman et al., 1990), thus the precise nature of catecholamine anomalies in ADHD remains unknown and several different theories have been suggested.

Dopamine The efficacy of stimulant drugs in treating ADHD suggests involvement of the dopamine system in this disorder. Methylphenidate and dexamphetamine block the reuptake of dopamine and facilitate its release from presynaptic terminals (Bradley, 1989; Malone et al., 1994; Solanto, 1998). This might suggest that dopamine may be deficient at prefrontal synapses, leading to the deficits in inhibitory control, working memory and executive functions that are common in ADHD (Pliszka et al., 1996). However, this simple hypothesis is not supported by findings that dopamine agonists such as L-dopa have proven inefficient in treating ADHD, while antidepressants, which have virtually no direct effect on the dopamine system, can be an effective treatment (Pliszka et al., 1996). There are also contradictory

146

maree farrow et al.

findings regarding the effects of dopamine antagonists, with some studies showing deleterious effects and others not (Levy, 1991). As the activity of dopamine neurons is modulated by autoreceptors and feedback loops (Cooper et al., 1986), the enhancement of synaptic dopamine activity may not be enough to explain the effects of stimulant drugs on this complex system. Dopamine autoreceptors are more sensitive to dopamine and to low doses of stimulants than post-synaptic receptors, suggesting that the inhibition of synthesis and release of dopamine due to autoreceptor feedback mechanisms may play a role in the therapeutic effect of stimulants, and therefore that elevated pre-synaptic dopaminergic activity could be associated with ADHD (Solanto, 1998). Studies of metabolite concentrations have failed to find conclusive direct evidence for a dopamine deficiency in ADHD. The best measure of CNS dopamine levels is thought to be the concentration of homovanillic acid (HVA) in CSF (Castellanos, 1997; Raskin et al., 1984). Some early studies found reduced HVA levels in CSF or urine in children with ADHD compared with controls (Shaywitz et al., 1977), while other studies found no group differences (Shekim et al., 1977, 1979; Shetty & Chase, 1976). More recent studies have attempted to determine the relationship between CSF HVA concentrations, symptom severity and drug response, and findings have not supported the theory of deficient dopamine in ADHD. A significant positive correlation between HVA levels and measures of hyperactivity was found by Castellanos et al. (1994), and subsequently replicated (Castellanos et al., 1996). The later study also found that baseline CSF HVA concentration was a predictor of stimulant drug response, with higher HVA levels being associated with better drug response (Castellanos et al., 1996). These results are consistent with findings that CSF HVA levels decrease after treatment with stimulants (Castellanos, 1997; Raskin et al., 1984; Weizman et al., 1990), and also decrease with age (Castellanos, 1997). These findings and the assumption based on animal studies that the majority of CSF HVA originates in the striatum, led Castellanos (1997) to conclude that the motor hyperactivity in ADHD may be associated with increased HVA concentrations in the caudate, as would be found in younger children. Interest in dopamine system abnormalities in ADHD has also arisen from findings of brain imaging studies which implicate brain structures with rich dopamine innervation such as fronto-striatal circuits (Tannock, 1998). Genetic or environmental factors may affect the development of frontal lobe—basal ganglia neural networks and the dopamine systems that modulate activity in these networks (Swanson et al., 1998b). However, given the complexity of and the discrepancies amongst the findings discussed above, it seems unlikely that ADHD is related to a simple overall hypofunctioning of dopamine systems (Pliszka et al., 1996). This can be explained in part by the fact that at least five different dopamine receptors have been identified, with different anatomical distributions, different functional characteristics, and different drug responses (Levy, 1991; Pliszka et al., 1996).

Neurobiology of ADHD

147

Noradrenaline The best indicator of central noradrenaline levels is thought to be urinary concentrations of 3-methoxy-4-hydroxyphenylglycol (MHPG), the principal metabolite of noradrenaline. As with dopamine metabolite levels, there are conflicting findings for MHPG levels, with some studies finding reduced urinary MHPG in children with ADHD (Shekim et al., 1977, 1979, 1983, 1987), others finding increased levels (Khan & Dekirmenjian, 1981; Oades et al., 1998), and others finding no difference relative to controls (Rapoport et al., 1978). Some of the discrepancies among findings may be explained by subjects’ prior use of stimulants, as urinary MHPG concentrations have been found to decrease after treatment with dexamphetamine, but not after treatment with methylphenidate (Raskin et al., 1984; Weizman et al., 1990). MHPG concentrations in plasma and in CSF have also been measured in just a few studies. Castellanos et al. (1994) found that CSF MHPG levels were positively correlated with measures of aggressive and disruptive behaviour in children with ADHD. However, they found no relationship between plasma MHPG levels and behavioural measures, a finding consistent with those of two studies by Halperin and colleagues (Halperin et al., 1993, 1997). As with dopamine, the strongest evidence for noradrenaline dysfunction playing a role in ADHD comes from pharmacological studies. Stimulants facilitate the release and inhibit the reuptake of noradrenaline (Mason, 1984). This is thought to inhibit the locus coeruleus via feedback loops (Cooper et al., 1986; Malone et al., 1994; Solanto, 1998). Clonidine, which has also been found to be effective in treating ADHD, acts on presynaptic α2 noradrenergic receptors and inhibits noradrenaline release and locus coeruleus activity (Pliszka et al., 1996; Weizman et al., 1990). Thus excess noradrenergic activity in the locus coeruleus, where neural activity is dampened by drugs effective in the treatment of ADHD, has been postulated to play a role in this disorder (Malone et al., 1994; Pliszka et al., 1996; Solanto, 1998). This excessive noradrenergic activity might lead to increased inhibition of cortical activity and a reduced ability to enhance the brain’s signal-to-noise ratio appropriately for different stimuli (Pliszka et al., 1996). As the posterior attention system receives noradrenergic innervation from the locus coeruleus, this noradrenergic dysfunction may contribute to the attentional problems seen in children with ADHD (Pliszka et al., 1996). In contrast Arnsten et al. (1996) argue that diminished brainstem noradrenergic activity disrupts the inhibitory control mechanisms of the prefrontal cortex and produces the deficits in behavioural inhibition characteristic of children with ADHD.

Other Neurotransmitters While most neurochemical research has focussed on dopamine and noradrenaline, other neurotransmitters have been suggested to play a role in ADHD,

148

maree farrow et al.

principally serotonin and adrenaline. Serotonergic involvement in ADHD was suggested because stimulant and antidepressant drugs have serotonergic agonist activity (Mason, 1984; Weizman et al., 1990). However, measures of serotonin metabolites in CSF indicate that serotonergic systems may be normal in children with ADHD (Raskin et al., 1984; Weizman et al., 1990). There is some evidence for a role of adrenergic influences in ADHD (Castellanos, 1997; Pliszka et al., 1996). Deficits in both the central and peripheral adrenergic systems may contribute to excessive locus coeruleus activity in ADHD (Pliszka et al., 1996).

Summary Given the complexity of interactions between neurotransmitters and their pharmacological effects, it seems unlikely that a simple deficiency or excess of any one neurotransmitter can account for the symptoms associated with ADHD or their improvement under drug treatment (Malone et al., 1994; Pliszka et al., 1996). Evidence suggests that dysfunctions of both noradrenergic and dopaminergic systems are involved, and several different theories have been suggested. One hypothesis is that noradrenergic deficits may be related to the cognitive and attentional problems associated with ADHD, while dopaminergic deficits may be related to hyperactivity (Mason, 1984). Another suggests that the central noradrenaline system may be dysregulated in ADHD, leading to inefficient priming of the posterior attention system to external stimuli, while effective mental processing is affected by deficient dopaminergic function in the executive anterior attention system (Pliszka et al., 1996). Malone et al. (1994) suggest that excessive noradrenergic activity in the locus coeruleus and deficient dopaminergic activity in frontal-mesolimbic pathways results in the symptoms of ADHD, as stimulants have a dampening effect on the locus coeruleus and facilitate the release of dopamine from the striatum. However, as stimulants produce a similar behavioural response in individuals with and without ADHD, they may simply provide compensatory effects rather than targeting a specific neurochemical deficit (Solanto, 1998). While the precise nature of the neurochemical deficits in ADHD remains unclear, there is growing evidence from several fields of biological research to suggest that dysfunction of frontal—basal ganglia dopamine pathways plays an important role (Castellanos, 1997; Swanson et al., 1998b). Mutations of dopamine receptor genes within this system that innervates fronto-striatal circuits may reduce dopamine activity and alter the normal development of dopamine systems (Swanson et al., 1998b,c) and may increase susceptibility for ADHD (Tannock, 1998). Children with ADHD may lack reciprocal inhibitory interactions between mesocortical and striatal dopamine neurons due to a deficit or delay in cortical development, resulting in reduced motor inhibition and goal-directed activity and increased environmentally dependant and inappropriate instinctual activity (Levy, 1991).

Neurobiology of ADHD

149

MOLECULAR GENETICS Long standing evidence from family, twin and adoption studies suggests that ADHD is heritable and that genetics play a major role in the aetiology of this disorder (Levy et al., 1997; Tannock, 1998; Thapar et al., 1999). Evidence from pharmacological and brain imaging studies for dopamine system abnormalities in ADHD has focussed recent molecular genetic research on dopaminergic genes. Polymorphisms defined by variable numbers of tandem repeats of two genes in particular have now been found to be associated with ADHD in multiple studies, although conflicting findings have also occurred. Because drugs that inhibit the dopamine transporter (including methylphenidate, dexamphetamine and pemoline) are effective in treating ADHD, the dopamine transporter (DAT1) gene has been examined as a potential candidate gene for ADHD. Studies involving children with ADHD and their parents have found a significant association between a 10-repeat allele of the DAT1 gene and ADHD (Cook et al., 1995; Daly et al., 1999; Gill et al., 1997; Waldman et al., 1998). This allele was found to be preferentially transmitted to ADHD probands. It has been speculated that this finding, along with the efficacy of dopamine transporter inhibiting drugs in treating ADHD, may be associated with an overactive dopamine transporter in ADHD, which would increase the reuptake of dopamine and reduce the time it has to act within the synapse (Barkley, 1998; Swanson & Castellanos, 1998). Waldman et al. (1998) found that the association with the 10-repeat DAT1 allele was stronger for ADHD—Combined Type than for ADHD—Inattentive Type and that the number of high risk alleles was correlated with the severity of hyperactive/impulsive symptoms but not inattentive symptoms. The relationship between ADHD and DAT1 polymorphisms may therefore vary depending on the subtype and severity of the disorder. A recent study by Swanson et al. (2000b) obtained the opposite result to previous studies, finding that the 10-repeat DAT1 allele was more often not transmitted in a sample of children with ADHD— Combine type with no serious comorbidities and a demonstrated positive response to methylphenidate. However, these authors stated that non-replication is expected as ADHD is a complex disorder likely to be associated with multiple genes and so did not discount the earlier findings of an association between ADHD and the DAT1 gene. Other studies have found an association between ADHD and a 7-repeat allele of the dopamine-4 receptor (DRD4) gene (Faraone et al., 1999; LaHoste et al., 1996; Rowe et al., 1998; Smalley et al., 1998; Sunohara et al., 1997; Swanson et al., 1998c). These studies found increased frequency of the 7-repeat DRD4 allele in samples of children and adults with ADHD compared to control groups. The DRD4 gene has been examined because the 7-repeat allele has been reported to be associated with novelty seeking (Benjamin et al., 1996; Ebstein et al., 1996), a trait that shares similar characteristics with ADHD. It has been suggested that

150

maree farrow et al.

the association between ADHD and polymorphism of the DRD4 gene may be related to reduced postsynaptic responsiveness to dopamine (LaHoste et al., 1996; Swanson & Castellanos, 1998; Tannock, 1998). This variation in the DRD4 gene has previously been shown to mediate a blunted intracellular response to dopamine (Ashgari et al., 1995). However, one study of children with ADHD found that this variation in the DRD4 gene occurred equally often in ADHD and healthy control groups (Castellanos et al., 1998). These authors also found that anatomic MRI measures of the brain and behavioural measures that had previously been found to discriminate between ADHD children and controls did not differ between subjects who had and those who lacked the 7-repeat allele. Another recent study evaluated symptom severity and neuropsychological task performance in subgroups of children with ADHD defined by the presence or absence of the 7-repeat DRD4 allele (Swanson et al., 2000). They found that both groups displayed similar symptom severity, but that the group with the 7-repeat allele had normal reaction times, while those without this D4 polymorphism had slower and more variable responses as expected for children with ADHD. That is, children with the 7-repeat allele, previously associated with ADHD, failed to show neuropsychological abnormalities also typically associated with ADHD, a finding opposite the authors’ predictions (Swanson et al., 2000). Rowe et al. (1998) found that the 7-repeat DRD4 allele occurred more frequently in children with ADHD—Inattentive type than in controls, and that a greater number of 7-repeat alleles was associated with higher levels of inattentive symptoms. This is perhaps consistent with the abundance of D4 receptors in the dorsolateral prefrontal cortex, which suggests that this dopaminergic system plays a role in mediating attention (Meador-Woodruff et al., 1994). The high risk allele frequency was also greater for ADHD—Combined type than controls, but there was no relationship between the 7-repeat allele and hyperactive/impulsive symptoms. The findings of this study again suggest that the relationship between ADHD and dopamine gene polymorphisms depends on the subtype and severity of the disorder. Findings of anomalies in dopamine transporter and D4 receptor genes are consistent with models of hypodopaminergic activity and altered development of dopamine systems playing a role in ADHD (Swanson et al., 1998b, 2000b; Tannock, 1998). Brain regions rich in dopamine receptors are thought to be involved in component processes of attention including alerting and executive control (Posner & Raichle, 1994). Genetic variations in these attentional networks may result in subsensitive dopamine receptors or overactive reuptake of dopamine and may contribute to the attentional deficit that characterises ADHD (Swanson et al., 2000b). However, associations between dopamine genes and ADHD suggested by recent molecular genetic studies can only be considered preliminary given that the associations have not been strong, there have been conflicting findings

Neurobiology of ADHD

151

and heterogeneous samples have been involved (Swanson et al., 1998b, 2000b; Tannock, 1998).

NEUROPSYCHOLOGICAL RESEARCH Studies of Attention Attention has been a major focus of neuropsychological research into ADHD and deficits on various tasks designed to measure attentional aspects of performance have been found in children with ADHD. However, task performance in these children has been found to be affected by many factors other than attention and it is generally concluded that poor performance cannot be attributed to a specific deficit in attention (Barkley, 1997; Schachar, 1991; Swanson et al., 1990; Van der Meere, 1996). Instead, the cognitive deficit in ADHD appears to be at the output or motor stage of information processing rather than at input or attentional stages (Barkley, 1997; Van der Meere, 1996). In a series of studies that manipulated task parameters in order to isolate different aspects of information processing Van der Meere, Sergeant and colleagues found no evidence of specific deficits in selective, focussed, sustained or divided attention, encoding, search or decision processes, or acquisition of automatic processing (Sergeant & Scholten, 1983, 1985a,b; Van der Meere & Sergeant, 1987, 1988a,b,c). They did find that children with ADHD consistently made more errors and had slower and more variable reaction times than normal controls. They concluded that this task inefficiency in children with ADHD could not be explained in terms of an information processing or attention deficit, but was due to a deficit in output or motor processes (Van der Meere, 1996; Van der Meere & Sergeant, 1988a). Deficits in motor timing, preparation and control may better explain the poor performance of children with ADHD on attentional tasks. In terms of behavioural symptoms, children with ADHD are consistently reported by parents and teachers to demonstrate inattention (Barkley, 1990, 1997). This behavioural inattention and distractibility may also arise from deficits in inhibition and selfregulation and the effects these deficits have on task persistence and interference control (Barkley, 1997; Schachar et al., 1995; Van der Meere, 1996). Sustained Attention—the Continuous Performance Task The continuous performance task (CPT) was originally developed by Rosvold et al. (1956) as a measure of sustained attention or vigilance. The CPT has been widely used as both a research and a diagnostic instrument in the field of ADHD (Corkum & Siegel, 1993). These tasks require the subject to view sequences of randomly ordered stimuli, such as letters, and to respond to a particular infrequent stimulus, such as the letter X, or to a particular sequence of stimuli, such as X

152

maree farrow et al.

preceded by A. Omission errors (missed targets) are considered a measure of inattention, while commission errors (responses to non-targets or false alarms) are considered a measure of impulsivity (Burke, 1990; Corkum & Siegel, 1993). Reaction time (time between stimulus onset and response) to target stimuli is considered a measure of alertness (Levy, 1980). Reaction time is often not reported in CPT studies, which tend to concentrate on the number and/or type of errors made (Chee et al., 1989). However, children with ADHD have been found to have slower reaction times in response to target stimuli than normal controls in several studies (Chee et al., 1989; Klorman et al., 1979; Overtoom et al., 1998; Schechter & Timmons, 1985; Strandburg et al., 1996; Wood et al., 1999). This finding has been interpreted as suggesting an inability to process and respond to information quickly in children with ADHD (Wood et al., 1999). Children with ADHD also have more variable reaction times than their normal peers (Klorman, 1991; Van Leeuwen et al., 1998). Children with ADHD have been shown to make significantly more errors of omission than normal controls in many studies and this finding has been interpreted as evidence for inattention and deficient arousal in children with ADHD (Corkum & Siegel, 1993; Losier et al., 1996). Children with ADHD also make more errors of commission than their normal peers and this is thought to result from poor inhibition and more impulsive responding (Barkley, 1997; Halperin et al., 1993; Kupietz, 1990; Van Leeuwen et al., 1998). Losier et al. (1996) performed a metaanalytic review of error rates in CPT studies of ADHD and found that across 11 studies children with ADHD made twice as many omission errors and more than twice as many commission errors as normal controls. Both these differences were statistically significant. Increased CPT error rates in children with ADHD is a fairly consistent result, although many studies find significant group differences for only one type of error (omission or commission), and some studies find no significant difference in either type of error between ADHD and control subjects (e.g. Werry, 1987; Wood et al., 1999). While performance of the CPT requires sustained attention, it also taps many other processes including arousal, motivation and inhibition (Corkum & Siegel, 1993; Klorman, 1991). Therefore, poor performance on the CPT does not necessarily reflect a specific sustained attention or vigilance deficit. Several other explanations for CPT performance deficits in children with ADHD have been proposed including momentary concentration problems (Corkum & Siegel, 1993; Oades, 1998), compromised allocation of effort (Dinklage & Barkley, 1992; Van der Meere & Sergeant, 1988), or an inability to modulate activation according to task demands (Van der Meere, 1996). The cognitive processes measured by the CPT and therefore the nature of the deficits revealed by poor performance may also depend on task and external variables, which vary greatly between studies (Corkum & Siegel, 1993; Losier et al., 1996).

Neurobiology of ADHD

153

Performance on the CPT has been shown to be improved by stimulants, which generally reduce reaction times and the number of errors made (Klorman, 1991; Klorman et al., 1979, 1991; Losier et al., 1996; Michael et al., 1981; Rapoport et al., 1980). Losier et al. (1996) performed a meta-analytic review of the effects of methylphenidate on the CPT performance of children with ADHD. They found that across 15 studies omission errors were reduced by 39% after methylphenidate administration, and commission errors were reduced by 29%. Klorman et al. (1991) suggested that this improved accuracy results from more efficient stimulus evaluation. Stimulants have been found to also improve CPT performance in normal children and adults (Klorman, 1991; Rapoport et al., 1980; Strauss et al., 1984), so this effect is not specific to ADHD.

Studies of Frontal Lobe/Executive Functions More recently, neuropsychological studies of ADHD have focused on measures of executive functions, or processes thought to involve the frontal lobes of the brain (Pennington & Ozonoff, 1996; Tannock, 1998). Performance deficits on various measures of executive functions have been found in ADHD subjects, using tasks such as the Tower of Hanoi or Tower of London (Aman et al., 1998; Pennington et al., 1993), the Matching Familiar Figures Test (Pennington et al., 1993; Robins, 1992), the Stroop colour-word interference task (Boucugnani & Jones, 1989; Grodzinsky & Diamond, 1992) and the Wisconsin Card Sort Task (Boucugnani & Jones, 1989; Shue & Douglas, 1992). Adults with prefrontal lobe injuries are also found to perform poorly on these same tasks and brain imaging studies have found frontal activation during performance of these tasks, leading to the suggestion that poor performance on executive function tasks by children with ADHD indicates frontal lobe deficits (Barkley, 1997). In their review of studies of executive functions in ADHD, Pennington and Ozonoff (1996) concluded that executive function deficits in ADHD were found in 15 of 18 studies. Measures of motor inhibition consistently found differences between ADHD and normal control groups, e.g. the Go—No-Go task (Shue & Douglas, 1992) and the Stop task (Aman et al., 1998). The executive function deficit in ADHD does not appear to be exclusively in motor inhibition however, as studies of other executive function tasks also find group differences, e.g. working memory tasks such as Self Ordered Pointing (Shue & Douglas, 1992). Pennington and Ozonoff (1996) concluded that executive function deficits in ADHD have been consistently found in well controlled studies and do not appear to be a factor of IQ, age or comorbidity. Apart from measures of vigilance, non-executive function measures including verbal and visuo-spatial tasks are far less consistent in finding deficits in ADHD (Pennington & Ozonoff, 1996).

154

maree farrow et al.

Inhibition Tasks Deficient performance by children with ADHD on frontal lobe tasks such as the WCST, the Stroop and the MFFT has been interpreted as being a consequence of poor response inhibition in ADHD (Barkley & Grodzinsky, 1994). However, these tasks rely on the effective use of many executive functions in addition to inhibition of prepotent responses, as discussed above. More direct evidence comes from purer measures of motor inhibition, which fairly consistently demonstrate deficient inhibition in children with ADHD (Barkley, 1997; Pennington & Ozonoff, 1996). The Go—No-Go task requires inhibiting a motor response to a “no-go” stimulus. Children with ADHD have been found to have difficulties inhibiting responding and to make more errors on this task, suggesting motor control difficulties similar to those reported for patients with frontal lobe lesions (Iaboni et al., 1995; Shue & Douglas, 1992; Trommer et al., 1988). The Stop task requires stopping an already initiated motor response when a “stop signal” is presented during a choice reaction time task. Children with ADHD are consistently found to demonstrate difficulties inhibiting responses in this task (Aman et al., 1998; Oosterlaan & Sergeant, 1996; Oosterlaan et al., 1998; Rubia et al., 1998; Schachar & Logan, 1990; Schachar et al., 1995). These findings have been interpreted as indicating less efficient inhibitory control in ADHD (Quay, 1997; Schachar & Logan, 1990). The involvement of frontal lobe inhibitory mechanisms in the stop task is suggested by findings of ERP differences in stop trials compared with go trials at frontal recording sites (De Jong et al., 1990, 1995). In a variation of the stop task, the Change task, which requires making an alternative response to the stop signal, children with ADHD also perform more poorly than controls (Oosterlaan & Sergeant, 1998; Schachar et al., 1995). The Change task requires the inhibition of an on-going action and rapid shifting to an alternate action, abilities which are both found to be deficient in children with ADHD, suggesting impairments in response re-engagement in addition to response inhibition (Oosterlaan & Sergeant, 1998; Schachar et al., 1995).

Studies of Parietal Lobe Functions A small number of studies have used neuropsychological tasks to examine parietal lobe function in children with ADHD. The idea of parietal lobe involvement in ADHD arose from observations that children with ADHD and patients with right parietal lobe damage showed similar symptoms of inattention and hypoarousal (Aman et al., 1998; Voeller & Heilman, 1988). Children with ADHD were found to make more errors of omission and more left-sided errors than controls on the Letter Cancellation Task, suggesting deficits similar to those of adults with right hemisphere dysfunction (Voeller & Heilman, 1988). Mental rotation tasks have

Neurobiology of ADHD

155

also shown deficits in children with ADHD (Aman et al., 1998; Snow, 1990), similar to those reported for patients with right parietal lesions (Ditunno & Mann, 1990). The right parietal cortex has been shown to be involved in visuo-spatial attention in PET studies (Corbetta et al., 1993) and in lesion studies (Posner & Raichle, 1994). The pattern of performance exhibited by patients with right parietal lesions on visuo-spatial attention tasks such as the COVAT is an increased invalid cue effect for LVF targets, suggesting difficulties in disengaging attention from invalid cues presented to the RVF (Posner & Raichle, 1994). However, with one exception (Wood et al., 1999) this pattern of COVAT performance has not been found in children with ADHD (Aman et al., 1998; Swanson et al., 1991). Aman et al. (1998) did find deficits in children with ADHD on their two other measures of right parietal function, the Turning Task and Spatial Relations. They suggested that these findings may reflect the presence of right parietal dysfunction in ADHD and that this dysfunction may be too subtle to produce deficits on the COVAT which may be sensitive only to more severe parietal dysfunction. As they found stronger evidence for frontal lobe deficits in ADHD, Aman et al. (1998) also suggested an alternative explanation that their findings may reflect the influence of frontal lobe deficits on parietal lobe function.

Summary While parietal lobe function in ADHD has not been extensively studied, some evidence does exist for possible right parietal dysfunction in children with ADHD (Aman et al., 1998). There is stronger evidence for deficits in frontal lobe function in children with ADHD, who are fairly consistently found to have difficulties with aspects of task performance thought to be mediated by the frontal lobes, including motor control, problem solving, formulating and testing hypotheses, using feedback to modify responding, organizing responses, and adhering to task constraints (Barkley, 1997; Barkley & Grodzinsky, 1994; Boucugnani & Jones, 1989; Gorenstein et al., 1989; Pennington & Ozonoff, 1996; Shue & Douglas, 1992). This combination of deficits is similar to that found for patients with frontal lobe lesions and suggests that ADHD may be associated with comprehensive frontal lobe deficits in planning, hypothesis testing and inhibitory control (Shue & Douglas, 1992). One of the most consistent findings from neuropsychological studies is that of deficits in the executive function of response inhibition in children with ADHD, providing compelling evidence that ADHD involves impaired behavioural inhibition (Barkley, 1997; Barkley & Grodzinsky, 1994). However, inhibitory deficits may not be sufficient to explain the range of executive dysfunction in children with ADHD, who are also found to have difficulties in problem solving, effective

156

maree farrow et al.

use of feedback, and generation and use of strategies (Shue & Douglas, 1992). The integrative function of the frontal lobes may be impaired in ADHD as it is in patients with frontal lobe lesions, and impairment of higher order cognitive processing may result from difficulties in integrating information (Shue & Douglas, 1992). Alternatively, these other deficits may arise from the influence of deficient inhibition on other executive functions as suggested by Barkley (1997). While the majority of studies do find executive function deficits in children with ADHD, there are conflicting findings for most of the frontal lobe tasks that have been used (Pennington & Ozonoff, 1996). Some of the inconsistencies in the results of neuropsychological studies of ADHD may be due to methodological differences such as selection criteria and type of tests used, or to comorbidity and heterogeneity of ADHD subject groups (Barkley & Grodzinsky, 1994; Seidman et al., 1995a). These inconsistencies in methodology and results also plague the literature on studies of attention in ADHD. While many studies find that children with ADHD perform poorly on attentional tasks, it is generally concluded that this poor performance cannot be explained by attention deficits (Barkley, 1997; Schachar, 1991; Swanson et al., 1990; Van der Meere, 1996). Studies that have isolated various aspects of information processing have failed to find deficits in orienting of attention, encoding of information, selective attention or divided attention, but have found deficits in motor processes (Sergeant & Scholten, 1983, 1985a,b; Van der Meere & Sergeant, 1987, 1988c). These findings suggest that the inattention that is characteristic of ADHD may also be related to deficits in inhibition and self-regulation (Barkley, 1997; Van der Meere, 1996). Children with ADHD fairly consistently demonstrate performance deficits on continuous performance or vigilance tasks in terms of slower and more variable reaction times and increased errors of omission and of commission. These findings have been interpreted as reflecting concentration problems (Corkum & Siegel, 1993; Oades, 1998), compromised regulation of effort or activation (Dinklage & Barkley, 1992; Van der Meere, 1996; Van der Meere & Sergeant, 1988), and poor inhibition (Barkley, 1997). So, executive function deficits may also be related to the poor performance of children with ADHD on the CPT. This is further supported by evidence that the frontal lobes are involved in vigilance tasks (Pardo et al., 1991).

BRAIN IMAGING RESEARCH Various brain imaging methodologies have been applied to ADHD research in an attempt to find the underlying neurobiological basis of the disorder. These have included both structural and functional imaging methods, revealing some differences in both brain anatomy and brain function in subjects with ADHD.

Neurobiology of ADHD

157

Structural Imaging Computerized Tomography Early studies which utilized Computerized Tomography (CT) to image brain structure in children and adults with ADHD revealed few significant differences from controls. In a comparison between children with ADHD and a control group who required CT scans for various clinical indications, no group differences were found in lateral ventricular size or in frontal lobe width, but the frontal lobes were found to be more symmetric in the ADHD group (Shaywitz et al., 1983). These authors concluded that if anatomic abnormalities exist in ADHD, they were not distinguishable using CT techniques available at that time. In a study of young adult males, no group differences were found in ventricular or hemispheric areas between an ADHD group and a group who required CT scans for evaluation of head trauma (Nasrallah et al., 1986). This study did find a 58% prevalence of mild to moderate cortical atrophy in the ADHD group, but the authors concluded that this finding may be due as much to coexisting psychiatric diagnoses or alcohol abuse as to the presence of ADHD (Nasrallah et al., 1986). The use of ionizing radiation in CT has limited it’s use in studies involving children, and therefore in studies of ADHD (Filipek et al., 1992; Tannock, 1998). Magnetic Resonance Imaging More recently Magnetic Resonance Imaging (MRI) has been used to study anatomical differences between ADHD and normal control groups. Several brain regions, predominantly in the right hemisphere, have been found to be smaller in children with ADHD than in their normal peers using MRI. ADHD subjects have consistently been found to have a significantly smaller right frontal cortex than normal controls (Castellanos et al., 1996a; Filipek et al., 1997; Hynd et al., 1990). ADHD subjects also failed to demonstrate the normal right > left frontal asymmetry found in control subjects (Castellanos et al., 1996a; Hynd et al., 1990), which is consistent with the greater frontal symmetry in ADHD subjects found using CT (Shaywitz et al., 1983). These findings have been interpreted as suggestive of right prefrontal deficits in ADHD (Castellanos et al., 1996a; Hynd et al., 1990). However, one reviewer suggested that this assumed right-sided dysfunction is actually on the left and that these findings may reflect a lack of age-appropriate synaptic pruning in the left prefrontal cortex in children with ADHD (Oades, 1998). Some researchers have speculated that the observed abnormal frontal symmetry and proposed frontal dysfunction in ADHD might be related to differences in the size of the corpus callosum (Giedd et al., 1994; Hynd et al., 1991). Various regions of the corpus callosum have been reported to be smaller in ADHD subjects. In several studies, anterior regions of the corpus callosum (rostrum, rostral body

158

maree farrow et al.

or genu) were found to be significantly smaller in ADHD subjects compared with normal controls (Baumgardner et al., 1996; Giedd et al., 1994; Hynd et al., 1991). Other studies found that posterior regions (splenium or isthmus) were significantly smaller in ADHD groups (Hynd et al., 1991; Lyoo et al., 1996; Semrud-Clikeman et al., 1994). However, Castellanos et al. (1996a), whose sample included the subjects who participated in the study by Giedd et al. (1994) reported no significant difference between the larger ADHD and control groups in the total midsagittal area of the corpus callosum, or in the area of any of the seven subregions measured. The discrepancies between results for the corpus callosum may be due to differences in the selection of the subregions measured, the MRI procedures used, and the characteristics of the groups studied (Castellanos et al., 1996a; Peterson, 1995; Semrud-Clikeman et al., 1994; Tannock, 1998). The known anatomy of cortical interconnections via the corpus callosum suggests that differences in the anterior regions of this structure may be related to frontal dysfunction in ADHD (Giedd et al., 1994; Hynd et al., 1991). Differences in posterior callosal regions may be related to the visuo-spatial, math and sustained attention deficits seen in ADHD (Hynd et al., 1991; Semrud-Clikeman et al., 1994). Recently Overmeyer et al. (2000) reported no significant differences for any corpus callosum areas between boys with ADHD and healthy male siblings of children with ADHD, and concluded that diagnosis could not be based on callosal abnormalities. Other MRI studies of ADHD have focused on the basal ganglia, due to its connections with frontal cortex and suggestions of fronto-striatal deficits underlying ADHD symptoms. Castellanos and colleagues found a right > left asymmetry of the caudate nucleus in normal control subjects (Castellanos et al., 1994; Castellanos et al., 1996a). This asymmetry was absent in ADHD subjects who had a significantly smaller right caudate nucleus than controls. These studies also found that age-related reductions in caudate volume that occurred in normal controls were diminished in ADHD subjects. This is consistent with a later finding that adolescents with ADHD had a larger right caudate than normal controls, which was suggested to be related to a failure of the normal maturational processes resulting in caudate volume reduction (Mataro et al., 1997). However, other studies have found a left > right caudate asymmetry in controls, which was reversed in ADHD subjects who were found to have a significantly smaller left caudate (Filipek et al., 1997; Hynd et al., 1991). Aylward et al. (1996) found no group differences in caudate asymmetry. Castellanos et al. (1996a) discounted the contrasting earlier finding of left > right normal caudate asymmetry (Hynd et al., 1990), stating that right > left caudate asymmetry had been found in three independent samples of normal adults (Breier et al., 1992; Flaum et al., 1995; Peterson et al., 1993) in addition to their paediatric sample. They also suggested that the strengths of their study included a larger sample size and better characterized subjects. The discrepancies between results for the caudate nucleus may also be due in part to methodological differences in the way this structure was measured, such as whether the head and

Neurobiology of ADHD

159

body were included or just the head, or whether axial or coronal slices were used (Castellanos et al., 1994, 1996a). The globus pallidus, another basal ganglia nucleus, has also been shown to be smaller and to differ from normal asymmetry in ADHD subjects. Singer et al. (1993) found that the left globus pallidus was smaller in children with Tourette’s Syndrome with comorbid ADHD than in normal controls. This finding was replicated by Aylward et al. (1996), who also found reduced total globus pallidus volume, which was predominant on the left, in 10 children with ADHD. In contrast, Castellanos and colleagues (1996a; 1996b) found that the right globus pallidus was smaller and that the normal right > left asymmetry was reversed in ADHD subjects. Two studies have reported a smaller total brain volume in children with ADHD, but this did not account for any of the regional volume differences also found (Castellanos et al., 1994; Castellanos et al., 1996a). Other brain structures found to be smaller in ADHD subjects include the cerebellum (Berquin et al., 1998; Castellanos et al., 1996a) and frontal and parieto-occipital white matter (Filipek et al., 1997). Some attempts have been made to correlate anatomical measures found to be deviant in ADHD with behavioural measures of ADHD symptoms. Areas of anterior regions of the corpus callosum (rostrum and rostral body) found to be smaller in boys with ADHD were negatively correlated with parent and teacher ratings of hyperactivity/impulsivity, suggesting a role for abnormal frontal lobe circuitry in ADHD (Giedd et al., 1994). Casey et al. (1997) found that task performance on response inhibition tasks was significantly correlated with anatomical measures of fronto-striatal circuitry (prefrontal cortex, caudate and globus pallidus) previously reported to be abnormal in children with ADHD (Castellanos et al., 1996a), but not with the size of normal structures (putamen). As the correlations between task performance and measures of prefrontal cortex and caudate nuclei were prominant in the right hemisphere, the authors suggested that these findings indicate the involvement of right fronto-striatal circuitry in response inhibition and in ADHD (Casey et al., 1997). In another study, larger caudate nucleus areas were associated with higher ratings on the Conners Teachers Rating Scale and poorer performance on tests of attention in adolescents, providing further evidence of caudate involvement in the deficits found in ADHD (Mataro et al., 1997). More recently, Semrud-Clikeman et al. (2000) found significant relationships between previously reported anatomical measures (Filipek et al., 1997) and neuropsychological measures. A smaller left caudate was associated with more externalising behaviours, reversed caudate asymmetry (R > L) was associated with poor performance on the Stroop interference task, and smaller right frontal white matter volume was related to poor performance on sustained attention tasks. In contrast to these findings, Castellanos et al. (1994) found no significant correlations between caudate volumes or asymmetry and

160

maree farrow et al.

continuous performance task errors or teacher and parent hyperactivity and conduct ratings.

Summary Structural brain imaging studies, especially MRI studies, have revealed the possibility of anatomical brain differences in children with ADHD. Regions of the frontal lobes and basal ganglia have been found to be about 10% smaller in ADHD groups than in control groups (Swanson et al., 1998a). These findings point to possible fronto-striatal system differences, consistent with biological theories of ADHD that implicate fronto-striatal dopamine pathways in the pathophysiology of ADHD (Castellanos, 1997; Levy, 1991; Swanson et al., 1998b) and with neuropsychological theories that see ADHD as resulting from frontal lobe deficits, in particular deficits in response inhibition (Barkley, 1997; Casey et al., 1997). The majority of studies suggest that right hemisphere abnormalities are predominant in ADHD. This is especially so for the frontal cortex (Castellanos et al., 1996a; Filipek et al., 1997; Hynd et al., 1990), although the laterality of basal ganglia abnormalities is less clear (Castellanos et al., 1994; Hynd et al., 1991). Despite some concordance between results, there are inconsistencies between findings for normal controls as well as for children with ADHD, which may be due to differences between studies in subject selection criteria or in MRI methods and image analysis (Peterson, 1995; Tannock, 1998). In addition, structural MRI studies have failed to identify consistent structural landmarks associated with ADHD, limiting the diagnostic potential of this technique (Vaidya & Gabrieli, 1999).

Functional Imaging Positron Emission Tomography Positron Emission Tomography (PET) studies of ADHD have mostly been restricted to older subjects due to concerns about exposing children to the radioactive isotopes necessary for this technique (Lou, 1992; Zametkin et al., 1990, 1993). The few studies that have been conducted have yielded inconsistent results. Zametkin et al. (1990) found reduced global glucose metabolism and reduced regional metabolism in 30 of the 60 areas examined in adults with a history of childhood ADHD, when compared with normal controls. The greatest decreases in metabolism occurred in premotor and superior prefrontal cortex, areas associated with control of motor activity and attention, suggesting a relationship between reduced frontal metabolic activity and ADHD symptoms (Zametkin et al., 1990). Reduced metabolism in ADHD subjects was also found in the striatum and the thalamus. In a subsequent study of adolescents with ADHD (Zametkin et al., 1993), no significant group differences were found for global or absolute regional cerebral

Neurobiology of ADHD

161

glucose metabolism. However, reduced normalized (regional/global) metabolism in ADHD subjects was found in six regions including the left anterior frontal cortex, where significantly reduced glucose metabolism was also found in adults with ADHD in the previous study (Zametkin et al., 1990). Left anterior frontal metabolism was significantly correlated with ADHD symptom severity in the adolescent sample, providing further evidence of a link between reduced frontal metabolism and deficits in motor and attentional control (Zametkin et al., 1993). In both the adult (Zametkin et al., 1990) and adolescent (Zametkin et al., 1993) studies, a trend toward stronger group differences in metabolism in female subjects than in males was observed (Ernst et al., 1994a; Zametkin et al., 1993). When the adolescent sample was expanded in a subsequent study, no group differences in global or absolute regional metabolism were found across the entire groups (Ernst et al., 1994a). However, global glucose metabolism was 15% lower in girls with ADHD than in normal girls, and significant regional metabolism reductions occurred in premotor, orbito-frontal and temporal cortex in girls with ADHD. No significant differences were found between boys with ADHD and normal boys. A further study which included a larger independent sample of adolescent girls failed to replicate this finding, as no differences in global or regional metabolism between girls with ADHD and normal girls were found (Ernst et al., 1997). Differences in sample characteristics and data analysis techniques were discussed as possible reasons for the conflicting results (Ernst et al., 1997). Lateralization of normalized metabolism was significantly different between groups in the later study, with lower metabolism in the left hemisphere in girls with ADHD and in the right hemisphere in controls (Ernst et al., 1997). Two recent PET studies by Ernst and colleagues used the tracer [flourine18]flourodopa to examine presynaptic dopaminergic function in adults and children with ADHD. In the first study with adults, prefrontal cortex, striatum and midbrain regions were examined, but only the prefrontal cortex showed significantly lower DOPA decarboxylase activity in ADHD adults, with medial and left prefrontal areas showing the largest differences (Ernst et al., 1998). Given this and previous findings of discrepancies between adults and adolescents with ADHD, the authors hypothesized that a prefrontal dopaminergic dysfunction underlies ADHD symptoms in adults and that this dysfunction may be secondary to subcortical dopaminergic deficits and their interactions with maturational processes (Ernst et al., 1998). This hypothesis is supported by their subsequent findings that children with ADHD had higher midbrain [18F]flourodopa accumulation than normal controls and that right midbrain [18F]flourodopa accumulation was correlated with symptom severity (Ernst et al., 1999). There were no differences in prefrontal and striatal measures in these children. A study examining regional cerebral blood flow changes associated with working memory found that rCBF changes in men with ADHD were prominent in occipital regions, while those in healthy controls were prominent in frontal

162

maree farrow et al.

and temporal regions (Schweitzer et al., 2000). The authors concluded that their results suggest compensatory mechanisms in subjects with ADHD in response to disrupted inhibition and internally guided behavior. Neither methylphenidate nor dexamphetamine have been shown to alter global glucose metabolism in adults with ADHD (Ernst et al., 1994b; Matochik et al., 1993, 1994). In addition, both drugs produced inconsistent patterns of increases and decreases in regional metabolism (Matochik et al., 1993, 1994). Single Photon Emission Computed Tomography Studies using Single Photon Emission Computed Tomography (SPECT) have revealed reduced cerebral blood flow in the frontal lobes and in the caudate nucleus in children with ADHD (Amen & Carmichael, 1997; Lou et al., 1984, 1989; Sieg et al., 1995). In the first, small study using Xenon-133 SPECT, cerebral blood flow was reduced in frontal lobe white matter regions and in the caudate nuclei in ADHD subjects (Lou et al., 1984). These findings were replicated in subsequent studies which found hypoperfusion in the striatum, especially on the right, and hyperperfusion in the occipital lobes and left sensori-motor and primary auditory regions (Lou et al., 1989, 1990a). The reduced activity in frontal and striatal regions and increased activity in primary sensory regions were partly reversible with administration of methylphenidate (Lou et al., 1984, 1989). The authors postulated a primary dysfunction of striatal structures in ADHD, which leads to disinhibition and hyperfunction of primary sensory and sensori-motor cotices (Lou et al., 1989). Lower striatal activity was also found in pre-school children compared to older children in a normal developmental study (Lou et al., 1990b), suggesting a neurobiological correlate to the behavioural immaturity seen in children with ADHD (Lou, 1992). These findings of reduced striatal activity in children with ADHD are consistent with the known anatomical connections between the caudate nuclei and prefrontal cortex (O’Tuama & Treves, 1993) and suggest that dysfunction in these pathways may be associated with the symptoms of ADHD (Dinklage & Barkley, 1992). One problem with the Lou et al. (1984, 1989) studies is that many of their subjects suffered from some type of early neurological insult such as hypoxia or encephalitis, raising the question of whether the findings are applicable to more typical groups of children with ADHD whose development was free of potentially damaging traumas (O’Tuama & Treves, 1993). In a more recent study using N-Isopropyl I-123 IMP SPECT, Sieg et al. (1995) found greater hemispheric I-123 IMP uptake asymmetry in ADHD subjects, with reduced blood flow in left frontal and left parietal regions, in comparison to psychiatric controls. The authors suggested their results, in addition to other PET and SPECT findings, might reflect a maturational lag resulting from delayed myelinization in ADHD, especially in the frontal lobes (Sieg et al., 1995). Reduced frontal activation was also found by Amen and Carmichael (1997) using high-resolution

Neurobiology of ADHD

163

SPECT in children and adolescents with ADHD under resting and “intellectual stress” conditions. Under intellectual stress, i.e. when performing a concentration task, 65% of ADHD subjects showed reduced perfusion compared to the resting condition in the prefrontal cortex. Only 5% of control subjects showed the same reduction in prefrontal activation. Functional Magnetic Resonance Imaging There have been only a few preliminary studies of ADHD using functional magnetic resonance imaging (fMRI) to date. A study of adolescents with ADHD found that, compared with normal peers, they had reduced activation in right hemisphere cortical regions including the anterior cingulate, pre- and post-central gyrus and posterior parietal cortex during a visual stop signal task (Rubia et al., 1999). However the ADHD group had increased activation in subcortical areas including the right and left insula and left caudate. The authors concluded that ADHD is associated with dysfunction of right hemisphere inferior frontal and striatal regions during motor inhibition (Rubia et al., 1999). Another study of 10 ADHD subjects aged from 14 to 51 years found that activation was predominant in the right middle frontal gyrus during a visual vigilance task (Sunshine et al., 1997). No comparison group was included in this study, but the authors reported that similar areas of activation were found in a previous study of normal subjects using the same task (Lewin et al., 1996). Some additional areas of activation not seen in normal subjects were found in the ADHD group in right and left frontal, left precentral and left parietal regions. The authors concluded that this result might represent true regions of abnormality in the ADHD subjects during visual vigilance, perhaps related to attempted compensation for their disorder, or alternatively may be due to artifacts (Sunshine et al., 1997). A recent fMRI study examined fronto-striatal function in children with ADHD and matched controls during the performance of go/no-go tasks (Vaidya et al., 1998). Compared to the controls, the ADHD group had significantly reduced striatal activation during a stimulus-controlled (more difficult) task and significantly increased frontal activation during a response-controlled (easier) task. The authors suggested that the finding of increased frontal activation, which differs from previous reports of frontal hypometabolism in ADHD (Sieg et al., 1995; Zametkin et al., 1990), might reflect greater inhibitory effort. The finding of reduced striatal activation is consistent with other functional imaging studies (Lou et al., 1984, 1989) and with structural imaging studies that have reported associations between anatomical abnormalities of the striatum and poor performance on inhibitory tasks in ADHD subjects (Casey et al., 1997; Mataro et al., 1997). The authors concluded that activation in children with ADHD may be abnormally high or low depending on the specific demands of inhibitory control imposed by the task (Vaidya et al., 1998).

164

maree farrow et al.

Summary In agreement with many of the findings of anatomical differences in children with ADHD, functional brain imaging studies also suggest that dysfunction of fronto-striatal networks may be involved in ADHD. However, the findings of functional brain imaging studies of ADHD have been inconsistent in terms of the precise nature of this fronto-striatal system dysfunction. These inconsistencies may be due in part to differences in the age groups and methodologies employed. Many of the studies discussed above used small, heterogeneous samples with wide age ranges, some included subjects with comorbid developmental learning disabilities or early neurological insult (Lou et al., 1984, 1989), some used children with other psychiatric disorders or siblings of ADHD subjects as controls (Lou et al., 1984, 1989; Sieg et al., 1995), and within each imaging modality a variety of imaging and analysis methodologies have been used. These limitations and methodological differences mean that future studies using larger samples, more specific subject selection criteria, and comparable imaging and analysis techniques are needed to verify the nature of functional abnormalities in ADHD, how they are affected by developmental changes, and their specificity to ADHD (Tannock, 1998). This will not be a simple task given that these imaging studies are limited by the difficulties associated with getting young children, especially hyperactive children, to comply with the demands of the scanning procedures.

Electrophysiological Research EEG Studies EEG studies have tended to find increased slowing of cortical electrical activity in children with ADHD, although there have been conflicting findings. Increased theta (4–7 Hz) activity in ADHD subjects compared with normal controls has been found in several studies, particularly in frontal regions (Clarke et al., 1998; Mann et al., 1992; Matsuura et al., 1993; Chabot & Serfontein, 1996). Increased delta (1–3 Hz) activity, especially in posterior regions, has also been observed (Clarke et al., 1998; Matousek et al., 1984). Mann et al. (1992) found that frontal theta power was increased in children with ADHD compared to normal controls during a resting condition and increased further during cognitive tasks. As theta activity decreases with age (Gasser et al., 1988), the findings for children with ADHD resembled those of younger children, suggesting a maturational delay in children with ADHD (Mann et al., 1992). Matsuura et al. (1993) also related their findings to brain immaturity in children diagnosed with ADHD. Chabot & Serontein (1996) also found increased theta in children with ADHD, which was greatest in frontal regions, but concluded that this represented a deviation from normal development rather than a maturational lag. Clarke et al. (1998) found increased theta power

Neurobiology of ADHD

165

in children with ADHD in all regions including midline frontal. Frontal theta was greater for the ADHD-Combined type group than for the ADHD-Inattentive type group. The authors suggested this finding may be related to an association between frontal dysfunction and the overt behavioural problems exhibited by children with ADHD-Combined type (Clarke et al., 1998). Reduced or slower alpha (8–12 Hz) activity has been found in children with ADHD (Callaway et al., 1983; Chabot & Serfontein, 1996; Clarke et al., 1998; Dykman et al., 1982; Matsuura et al., 1993; Shetty, 1971), as has reduced or slower beta (13–21 Hz) activity (Callaway et al., 1983; Caresia et al., 1984; Chabot & Serfontein, 1996; Clarke et al., 1998; Dykman et al., 1982; Mann et al., 1992; Oades, 1987). Mann et al. (1992) found reduced beta activity in children with ADHD in posterior and temporal regions during cognitive tasks, and related this finding to maturational delays in brain systems involved in attention. Clarke et al. (1998) also found less beta activity in posterior regions in children with ADHD, and in addition found decreased alpha in all regions with the greatest difference in posterior regions. As fast frequency activity increases with age (Gasser et al., 1988), they suggested that their findings support the idea of a maturational lag in ADHD. These findings of increased slow EEG activity (delta and theta) and decreased fast EEG activity (alpha and beta) suggest EEG slowing in children with ADHD and have been interpreted in terms of cortical underarousal and less active information processing (Ackerman et al., 1994; Chabot & Serfontein, 1996; Lubar, 1991; Mann et al., 1992; Oades, 1998; Tannock, 1998). This is supported by findings of increased deficits when the EEG is recorded during reading or drawing (Lubar, 1991; Mann et al., 1992). Some researchers have suggested that increased EEG slowing reflects delayed brain maturation in children with ADHD (Clarke et al., 1998; Matsuura et al., 1993; Ucles & Lorente, 1996; Tannock, 1998), while others disagree or suggest that development is deviant from normal rather than delayed (Callaway et al., 1983; Chabot & Serfontein, 1996). This characteristic EEG slowing in children with ADHD has led to the use of Neurometrics or Quantitative EEG (QEEG) techniques, which have been claimed to be useful in diagnosing ADHD and in distinguishing between subtypes of the disorder. There are reports of these techniques being able to correctly classify between 75% and 95% of subjects as either normal or ADHD (Chabot et al., 1996; Chabot & Serfontein, 1996; Lubar, 1991; Mann et al., 1992; Monastra et al., 1999). Lubar (1991) suggests that the ratio of theta to beta is the best measure to distinguish those with ADHD from controls, although this is based on studies of children with attention deficit disorder without hyperactivity (Mann et al., 1992) and with learning disabilities (LD) with attention deficits (Lubar et al., 1985). Chabot et al. (1996) suggest that discriminant functions using combinations of QEEG features best discriminate children with ADHD from those with LD and from normal controls. However, Neurometrics is not generally viewed as a valid diagnostic tool

166

maree farrow et al.

for ADHD due to the lack of evidence supporting its clinical usefulness (Levy & Ward, 1995; Zametkin et al., 1998). While increased slow EEG activity is found in most studies, there have been some conflicting findings. Kuperman et al. (1996) found increased beta activity in children with ADHD compared with normal controls, and suggested that this finding indicated elevated mental activity and overarousal in children with ADHD, which might contribute to sustained attention difficulties. Their subjects were drawn from a community rather than a clinical sample, but met DSM-III-R criteria for ADHD according to teacher reports. Chabot and Serfontein (1996) found increased beta activity, especially in frontal regions, in a subgroup of their ADHD subjects (13%), suggesting hyperarousal of fronto-striatal systems in these children, in contrast to the hypoarousal found in the majority of ADHD subjects. Clarke et al. (1998) excluded four ADHD subjects from their analyses because they had much higher levels of beta activity (>3 SD above mean) than the rest of the group. They concluded that this subset of subjects in their study and in the study by Chabot and Serfontein (1996) suggested there may be a subtype of ADHD characterised by increased beta activity. Another issue which raises questions about the assumption that EEG slowing characterises ADHD is that EEG slowing has been found for other clinical populations, especially LD (Ackerman et al., 1994; Lubar et al., 1985). Event Related Potential (ERP) Studies P3 in ADHD In ERP studies of ADHD, the most commonly examined component is the P3 (also labelled P300 or P3b), which is a late positive wave with a latency of 300 to 800 ms (Klorman, 1991). The amplitude of the P3 is influenced by stimulus probability, relevancy or novelty (Johnson, 1988; Klorman, 1991; Levy & Ward, 1995; Pritchard, 1981), and the latency of the P3 is influenced by cognitive, perceptual or memory load (Ford et al., 1982; Klorman, 1991). Because of these effects on its amplitude and latency, the P3 component is thought to reflect allocation of attention and to mark the end of stimulus evaluation processes which precede response selection and execution (Klorman, 1991). It has also been related to updating of internal representations and to working memory (Tannock, 1998). The majority of studies examining the P3 have found that the amplitude of this component is smaller in children with ADHD than in normal controls (Frank et al., 1994; Holcomb et al., 1985, 1986; Jonkman et al., 1997; Kemner et al., 1996; Klorman et al., 1979, 1983; Loiselle et al., 1980; Michael et al., 1981; Novak et al., 1995; Overtoom et al., 1998; Robaey et al., 1992; Satterfield et al., 1990, 1994; Strandburg et al., 1996; Van Leeuwen et al., 1988; Verbaten et al., 1994; Winsberg et al., 1993). There are contradictory findings however, as some studies have reported no significant group differences in P3 amplitude (Frank et al., 1998; Satterfield et al., 1988; Taylor et al., 1993; Winsberg et al., 1997).

Neurobiology of ADHD

167

Smaller P3 amplitude in ADHD subjects has been found in both auditory and visual modalities and in response to both target and non-target stimuli and has been interpreted in different ways. It may reflect a general underarousal or underreactivity to task relevant stimuli (Satterfield et al., 1990; Tannock, 1998), cognitive and information processing difficulties (Frank et al., 1994; Klorman, 1991; Satterfield et al., 1994; Winsberg et al., 1993), or deficits in selective or sustained attention (Holcomb et al., 1985; Loiselle et al., 1980; Michael et al., 1981; Overtoom et al., 1998). The reduction in P3 amplitude in children with ADHD is generally associated with poorer performance on the task used to elicit the ERP, leading to the suggestion that it is likely to reflect deficient cognitive processing rather than a generalised neurophysiologic deficit (Klorman, 1991; Levy & Ward, 1995; Tannock, 1998). The specific cognitive deficits associated with smaller P3 amplitude in children with ADHD are dependent on the tasks used to elicit the ERP. Visual continuous performance and oddball tasks fairly consistently elicit smaller P3 to target stimuli in children with ADHD than in normal controls and this result has been interpreted as reflecting attention deficits (Klorman et al., 1979; Michael et al., 1981; Overtoom et al., 1998) or inappropriate allocation of attentional resources (Holcomb et al., 1985). Auditory oddball or tone discrimination tasks have elicited smaller P3 amplitude to target tones in children with ADHD in some studies (Frank et al., 1994; Holcomb et al., 1986; Kemner et al., 1996), but not in others (Frank et al., 1998; Lazzaro et al., 1997; Winsberg et al., 1997). Kemner et al. (1996) found that reduced P3 to deviant auditory stimuli occurred irrespective of the task relevance of the stimulus and concluded that smaller P3 amplitude in ADHD is due to abnormal processing of deviant stimuli. Frank et al. (1994) suggested that the reduced P3 amplitude in their ADHD with LD group reflects cognitive and processing difficulties rather than an attention deficit. Smaller P3 amplitude has also been found using selective attention tasks, which simultaneously present visual and auditory oddball paradigms and require the subject to attend to one modality or the other, or require subjects to attend to tones presented to one ear or the other. These results have been interpreted as suggesting that children with ADHD have a deficit in the activation of P3 processes (Jonkman et al., 1997), a selective attention dysfunction (Loiselle et al., 1980), a deficit in preferential processing of attended stimuli (Satterfield et al., 1994), or insufficient locus coeruleus activity that is normally triggered by attended, task-relevant stimuli (Satterfield et al., 1990). Other visual tasks employed in ERP studies that have resulted in findings of smaller P3 amplitude in children with ADHD include categorization (Robaey et al., 1992), spatial orienting (Novak et al., 1995) and delayed-go tasks (Brandeis et al., 1998). Smaller P3 amplitude in children with ADHD has been found in a variety of different task conditions and modalities, however, the P3 to target stimuli consistently shows the greatest reduction in amplitude in children with ADHD compared to normal controls, in

168

maree farrow et al.

particular when task performance is also deficient (Brandeis et al., 1998; Klorman, 1991). This common finding is therefore likely to reflect deficits in cognitive processing of task-relevant stimuli in children with ADHD. Many studies have used only a few electrodes to record the ERP and have discussed overall P3 amplitude reductions in terms of cognitive processes but not in terms of topography. The P3 to visual targets is generally found to have a maximum amplitude over parietal regions (Klorman, 1991; Satterfield et al., 1988). Differences in topography in children with ADHD may be as important as overall differences in amplitude. Some recent studies have used larger numbers of electrodes and have examined the topography of the P3 in children with ADHD. Using 17 electrodes and an auditory tone discrimination task, Johnstone & Barry (1996) found that P3 amplitudes to target stimuli were smaller in the posterior brain region but larger in the frontal region for children with ADHD compared with normal controls. They suggested that the ADHD group utilized an additional frontally distributed cognitive process when processing task-relevant stimuli, which might reflect an attentional compensation mechanism (Johnstone & Barry, 1996). The laterality of the P3 component may also differ in children with ADHD. Oades et al. (1996) used 19 electrodes and an auditory tone discrimination task and found a right-biased P3 asymmetry in normal controls that was absent in their ADHD group, suggesting a right hemisphere impairment in terms of stimulus processing in children with ADHD. Other researchers have used ERP microstates and source localization techniques to examine the topography of the ERP in children with ADHD and controls (Brandeis et al., 1998; Van Leeuwen et al., 1998). Microstates are successive ERP segments with stable topographies that vary in duration and are related to different stages of information processing (Brandeis & Lehmann, 1986). The global field power (GFP), defined as “the spatial standard deviation over all voltages in a map” (Van Leeuwen et al., 1998, p. 100) can be calculated for each microstate and is similar to an amplitude measure. Van Leeuwen et al. (1998) recorded the ERP at 30 electrode sites while children performed the A-X version of the CPT. They found no significant group differences in the topography of ERP microstates, but found reduced GFP in a CNV/P3 microstate to cues (the A) in the ADHD group. The topography of this microstate was defined by a posterior positivity in both the ADHD and control groups. Source localization analysis using low resolution electromagnetic tomography (LORETA: Pascual-Marqui et al., 1994) identified posterior sources for this microstate that were less right biased in the ADHD group. The authors concluded that their results suggest impaired orienting to cues in children with ADHD, possibly involving the posterior attention system (Van Leeuwen et al., 1998). Using a delayed-go task, Brandeis et al. (1998) also found reduced GFP in late P3 type microstates in their ADHD group and related this finding to less efficient posterior orienting mechanisms. These results are consistent with findings of reduced parietal P3 amplitude in children with ADHD and suggest that possible deficits in parietal brain mechanisms in ADHD and their relationship

Neurobiology of ADHD

169

to well established frontal deficits should not be overlooked (Brandeis et al., 1998; Van Leeuwen et al., 1998). Reduced P3 amplitude has also been found in several other clinical populations including children with autism, learning disabilities and schizophrenia, so this effect does not seem to be specific to ADHD (Klorman, 1991; Levy & Ward, 1995; Oades, 1998). A recent study addressed the issue of the specificity of abnormal ERPs to ADHD using auditory and visual oddball tasks, and found that only the parietal P3 amplitude to deviant auditory stimuli was smaller in children with ADHD than in groups of autistic and dyslexic children (Kemner et al., 1998). In an earlier study, Kemner et al. (1994) found that visual P3 amplitude in autistic children did not differ from that of children with ADHD or dyslexia. Frank et al. (1994) found that auditory P3 amplitude was smaller in children with learning disabilities (LD) and in children with LD and ADHD than in normal controls. They suggested that smaller P3 amplitude in children with LD and/or ADHD is due to cognitive processing difficulties rather than an attention deficit. In a later study (Frank et al., 1998), they found significantly smaller auditory P3 amplitude than normal controls in a LD group and a LD + ADHD group, but not in a pure ADHD group. They again suggested that P3 abnormalities in children with learning and attentional problems reflect processing rather than attentional deficits. Abnormalities in P3 latency in children with ADHD have also been reported. Some studies that examined P3 latency to visual stimuli found it to be longer in ADHD subjects than in normal controls (Holcomb et al., 1985; Strandburg et al., 1996; Sunohara et al., 1997; Taylor et al., 1993). This finding has been interpreted as suggesting that stimulus evaluation and attentional processes are slower and more difficult for children with ADHD (Holcomb et al., 1985; Klorman, 1991; Tannock, 1998; Taylor et al., 1993). Holcomb et al. (1985) also found that P3 latency in their ADHD group increased across blocks of their visual target detection task, suggesting a deterioration of these processes over time. In contrast, shorter P3 latencies in children with ADHD have been reported for a visual categorization task (Robaey et al., 1992) and an auditory selective attention task (Loiselle et al., 1980), while other studies have found no significant group differences in P3 latency (Holcomb et al., 1986; Lazzaro et al., 1997; Michael et al., 1981; Satterfield et al., 1988, 1994). Some of the discrepancies in findings for P3 latency may be due to the different tasks employed. Visual target detection tasks may produce longer P3 latencies in children with ADHD due to slowed evaluation processes (Holcomb et al., 1985; Strandburg et al., 1996; Sunohara et al., 1997; Taylor et al., 1993), while tasks requiring selective attention or categorization may produce shorter P3 latencies due to less efficient modulation of processing speed according to task demands (Loiselle et al., 1980; Robaey et al., 1992). N1 and N2 in ADHD While the majority of studies find that P3 amplitude, especially to target stimuli, is smaller in children with ADHD, it is less clear

170

maree farrow et al.

whether this is preceded by abnormalities in earlier ERP components and thus earlier stages of information processing (Tannock, 1998). The other most commonly examined ERP components in studies of ADHD are earlier negative waves, the N1 and N2. The N1, occurring at a latency of around 100ms, is generally larger to attended than to non-attended stimuli and is thought to reflect an attentive division between concurrent stimulus channels (Loiselle et al., 1980). The N2 occurs at a latency of around 200 ms, is generally larger to novel than to frequent stimuli, and is thought to reflect automatic orienting to deviant stimuli (Robaey et al., 1992; Satterfield et al., 1988). N2 has also been related to stimulus comparison and categorization (Oades, 1998; Robaey et al., 1992), and to inhibition (Overtoom et al., 1998; Yong-Liang et al., 2000). The increased negativity to rare versus frequent stimuli is termed mismatch negativity (MMN), while the difference between attended and non-attended stimulus ERPs is termed processing negativity (PN). These early negative components of the ERP have predominantly been studied in ADHD using selective attention tasks. The N1 to attended auditory targets in a selective attention task using simultaneously presented visual and auditory oddball paradigms was found to be significantly smaller in children with ADHD than in normal controls (Satterfield et al., 1994). The 6 year old ADHD subjects in this study also showed a smaller difference in N1 amplitude between attended and non-attended stimuli than controls. A similar finding has been obtained for older ADHD subjects (12 to 14 years old) using an auditory selective attention task (Loiselle et al., 1980; Zambelli et al., 1977). These results have been interpreted as reflecting a selective attention dysfunction in children with ADHD (Klorman, 1991; Loiselle et al., 1980). In an auditory oddball task that did not require selective attention, no group differences in N1 amplitude were found (Winsberg et al., 1997). A smaller N2 amplitude to auditory targets in children with ADHD has been found in some studies (Satterfield & Braley, 1977; Satterfield & Schell, 1984; Satterfield et al., 1988, 1994; Winsberg et al., 1993). In addition, children with ADHD have been found to have a smaller difference in N2 amplitude between attended and non-attended stimuli (Satterfield et al., 1994) and between target and non-target stimuli (Satterfield et al., 1988). These findings of reduced N2 amplitude have been suggested to reflect deficiencies in children with ADHD in preferential processing of attended stimuli and in orienting to target or novel stimuli (Satterfield et al., 1988, 1994). In contrast, using visual categorization tasks Robaey et al. (1992) found that ADHD boys had a larger N2 amplitude than normal controls, as did Prichep et al. (1976) using an auditory guessing paradigm. Robaey et al. (1992) suggested that the parieto-occipital N2 was related to stimulus classification and was larger in ADHD subjects due to enhanced automatic processes. Prichep et al. (1976) related their finding of larger N2 amplitude to low arousal levels in children with ADHD, as N2 amplitude was reduced after administration of methylphenidate. In other studies, no differences in N2 amplitude

Neurobiology of ADHD

171

were found between ADHD subjects and normal controls using an auditory oddball task (Winsberg et al., 1997), visual feature detection tasks (Holcomb et al., 1985; Taylor et al., 1993) or the continuous performance task (Overtoom et al., 1998). Overtoom et al. (1998) did find however, that N2 amplitude was smaller in a subgroup of children with ADHD and comorbid ODD. They suggested that the fronto-central N2 component to non-targets in the AX-CPT (A followed by not X) was related to inhibitory processes and that deficiencies in these processes or increased impulsivity may be restricted to the comorbid group. Yong-Liang et al. (2000) found that frontal N2 amplitude was larger for no-go stimuli in a go/no-go task and suggested that it reflected inhibition of responding. This N2 amplitude was smaller in ADHD subjects than in controls, but only when the no-go task was performed second, suggesting an inhibitory regulation problem in ADHD (YongLiang et al., 2000). A related finding is that of Johnstone & Barry (1996), who found that frontal N2 amplitude to non-targets in a tone discrimination task was smaller for children with ADHD compared to normal controls. Johnstone & Barry (1996) also found that N2 amplitude was larger in the posterior region in children with ADHD, perhaps consistent with the findings of Robaey et al. (1992). Mismatch negativity (MMN), an enhancement of early negativity in the ERP to infrequent target stimuli compared to that to frequent standard stimuli, is thought to be related to automatic orienting to novel stimuli and to be a process which is not under voluntary control (Satterfield et al., 1988). MMN was found to be smaller in children with ADHD in a selective attention task (Satterfield et al., 1988), but was found to be normal using an auditory oddball task (Winsberg et al., 1997). Children with ADHD are often said to be less responsive to target stimuli, but this is more frequently linked with smaller P3 than with smaller MMN. Oades et al. (1996) found that MMN was left lateralized in children with ADHD but right lateralized in normal controls, which in conjunction with a similar finding for P3 laterality was interpreted as suggesting right hemisphere impairment in ADHD. Processing negativity (PN) is an enhancement of early negativity in the ERP to attended compared to non-attended stimuli and is thought to reflect attentional processes that are under voluntary control (Satterfield et al., 1988). PN has been found to be smaller in children with ADHD in several studies using selective attention tasks (Jonkman et al., 1997; Satterfield et al., 1988, 1990, 1994). These findings have been interpreted as suggesting poor discrimination and poor preferential processing of attended stimuli and are consistent with deficits in selective attention (Klorman, 1991; Satterfield et al., 1988, 1994). Reduced PN in the frontal region in children with ADHD may also be consistent with findings of reduced frontal blood flow (Lou et al., 1984, 1989) and reduced frontal metabolism (Zametkin et al., 1990) in ADHD (Satterfield et al., 1988). As the above discussion indicates, the results for N1, N2, MMN and PN vary from study to study. Methodological differences between studies make it difficult to directly compare these results. Some of the discrepancies between results for N1

172

maree farrow et al.

and N2 may be due to differences in the tasks and modalities used to elicit the ERP (Klorman, 1991), to differences in the age groups studied and developmental effects (Levy & Ward, 1995; Oades, 1998; Satterfield et al., 1990), to heterogeneity of subject groups and comorbidity in ADHD subjects, or to differences in the electrode sites used to record the ERP. Posterior/anterior differences (Johnstone & Barry, 1996) and laterality differences (Oades et al., 1996) suggest that topography may be important for the early negative ERP peaks, as for the P3. Smaller negative ERP components in children with ADHD are most consistently found in the auditory rather than the visual modality (Tannock, 1998), and when selective attention tasks are used that include a set of stimuli to be ignored and place greater demand for selective attention (Klorman, 1991). Latencies for the early negative ERP peaks are not often reported. Using visual feature detection tasks Sunohara et al. (1997) found that N2 latency was longer in children with ADHD than normal controls, while Taylor et al. (1993) found no group difference in N2 latency. The latency of the auditory N1 component was found to be shorter in children with ADHD than in controls in two studies (Oades et al., 1996; Satterfield et al., 1994). This finding may suggest that children with ADHD process perceptual information faster than their normal peers (Oades, 1998; Oades et al., 1996). In contrast, Loiselle et al. (1980) found no group difference for auditory N1 latency. ERP Studies of the CPT in ADHD The continuous performance task (CPT) has been used in several studies to examine attentional processes and the associated visual ERP. Klorman et al. (1979) found reduced P3 amplitude at Cz in hyperactive children compared with normal controls when they performed the X version of the CPT. P3 amplitude to both targets and non-targets was reduced in the hyperactive group, and their task performance was significantly worse. These results were replicated in a follow-up study (Michael et al., 1981), which in addition found reduced P3 amplitude at both Cz and Pz and deficient task performance in hyperactive children during a B-X version of the CPT (similar to CPT-AX, using B as the cue rather than A). The P3 to both targets and non-targets was significantly smaller in the hyperactive group for the CPT-X, while only the P3 to targets was reduced for the CPT-BX. These findings of reduced P3 amplitude in hyperactive children during the CPT were concluded to reflect deficits in sustained attention (Klorman et al., 1979; Michael et al., 1981). More recently, Overtoom et al. (1998) recorded the ERP at Fz, Cz, Pz and Oz while children with ADHD and normal controls performed the A-X version of the CPT. They suggested that the parietal P3 to targets (X preceded by A) could be used as a measure of attentional processes, while the fronto-central N2 to non-targets (not-X preceded by A) could be used as a measure of inhibitory processes. The ADHD group had a smaller parietal P3 amplitude to targets, indicating attention deficits. But there were no group differences in fronto-central N2 to non-targets,

Neurobiology of ADHD

173

indicating a lack of expected inhibition deficits in the ADHD group. These results were consistent with task performance results, as the ADHD group performed significantly worse on an inattention score but not on an impulsivity score. The authors concluded that deficient processing in the ADHD group was related to attention rather than to response inhibition (Overtoom et al., 1998). Similar results for early negative and late positive ERP components were obtained by Strandberg et al. (1996) for two versions of the CPT. These authors concluded that their findings of reduced P3 amplitude and longer P3 latency reflect processing problems in children with ADHD that occur in later rather than early stages, as ERP components related to earlier stages of processing were normal. In a recent ERP microstate study of the CPT-AX, Van Leeuwen et al. (1998) found reduced GFP in a CNV/P3 microstate (277–605 ms) to the cue (A) in children with ADHD, but not to the target (X). The authors concluded that impaired orienting to cues involving a posterior attention system, rather than impaired target processing involving frontal executive processes, was involved in ADHD children’s deficient performance of the CPT-AX (Van Leeuwen et al., 1998). Steady-State Visually Evoked Potential In our research we have examined differences in the steady-state visually evoked potential (SSVEP) between children with ADHD and healthy controls. Using the technique known as steady-state probe topography (SSPT) enables examination of disturbances in the spatial distribution and the dynamics of brain electrical activity in children with ADHD. Seventeen boys with ADHD (mean age = 10 years 9 months, SD = 2 years) and seventeen healthy male controls (mean age = 11 years, SD = 1 year 7 months) performed computerised tasks while their brain electrical activity was recorded from 64 scalp electrodes. A 13 Hz sinusoidal flicker was presented simultaneously to evoke the SSVEP. The subject characteristics and methods used are described in more detail in Silberstein et al. (1998). Subjects performed a low demand visual vigilance task (the reference task) and the AX version of the continuous performance task (CPT-AX). We found that, compared to the mean amplitude during the reference task, control subjects demonstrated SSVEP amplitude reductions during the A–X interval. Transient amplitude reductions occurred in frontal regions, while right parietal and occipital amplitude reductions were sustained throughout the 3.5 second A–X interval. Reductions in SSVEP amplitude during cognitive tasks have previously been associated with increased task related cortical activation (Farrow et al., 1996; Silberstein et al., 1990, 1995). In contrast to the control group results, ADHD subjects demonstrated much smaller frontal amplitude reductions and increased parieto-occipital amplitude, suggesting they failed to increase regional cortical activation according to task demands. Group differences in SSVEP amplitude were prominent in the right parietal region where sustained increased activation was

174

maree farrow et al.

Normalized Amplitude

0.4

ADHD Control

0.2

0.0

−0.2

A on

−0.4 1

A off

2

3

X on

4

5

6

Time (s) Figure 1. SSVEP amplitude during the CPT-AX target interval for ADHD and control groups at electrode 57 (right parietal). The dashed horizontal line represents the mean normalized amplitude for the reference task and is set to zero for both groups. CPT-AX related amplitude changes are therefore expressed as differences from the mean reference task amplitude. The vertical lines indicate the times of onset and offset of the cue A and onset of the target X.

seen in the control group. Figure 1 illustrates the amplitude differences at electrode 57, a right parietal site. In this region the control group show a large amplitude reduction, which follows the appearance of the A and is sustained until after the appearance of the X. In the ADHD group there is a relative amplitude reduction at the disappearance of the A, but the amplitude is increased compared with the reference level throughout the A–X interval. Control subjects also demonstrated significant SSVEP latency reductions in the right prefrontal region at the appearance and disappearance of the A and the appearance of the X. Reductions in SSVEP latency are interpreted as reflecting increased efficiency of coupling between neural networks and faster information processing (Silberstein et al., 1996; 1998). Right prefrontal latency reductions were much smaller and occurred later in the ADHD group, who predominantly demonstrated latency increases or slower processing at frontal and temporal sites throughout the A–X interval. Figure 2 illustrates the SSVEP latency at electrode 4, a right prefrontal site. In controls, the disappearance of the A coincides with a large reduction in SSVEP latency and smaller latency reductions occur at the appearances of the A and the X. However, in the ADHD group only very small latency reductions compared with the reference level are evident at the disappearance of the A and the appearance of the X.

Neurobiology of ADHD

175

10 ADHD Control

Latency Difference (ms)

5

0 −5 −10 −15

A on

1

X on

A off

2

3

4

5

6

Time (s) Figure 2. SSVEP latency during the CPT-AX target interval for ADHD and control groups at electrode 4 (right prefrontal). The dashed horizontal line represents the mean latency for the reference task and is set to zero ms for both groups. CPT-AX related latency changes are therefore expressed as differences from the mean reference task latency. The vertical lines indicate the times of onset and offset of the cue A and onset of the target X.

The largest group differences occurred at the disappearance of the A (figure 3a). At this time, there is extensive activation in the controls, predominant in the parieto-occipital region. In contrast, in ADHD subjects increased activation is restricted to the prefrontal region and there is reduced activation in the parieto-occipital region. The disappearance of the A also coincides with large latency reductions in the controls at frontal and temporal sites, moreso in the right hemisphere. While in the ADHD group a latency increase is predominant in these regions. As the Hotelling’s T maps show, the SSVEP differences in the control group at the disappearance of the A are highly significant for this single comparison in most regions. In another study we examined changes in the SSVEP following methylphenidate administration in 60 boys with ADHD (mean age = 10 years 1 month, SD = 1 year 10 months), who were recently diagnosed according to DSM-IV criteria and had never previously been treated with stimulants. We compared SSVEP amplitude and latency during the A–X interval before methylphenidate with the SSVEP 90 minutes after administration of a 0.3 mg/kg dose of methylphenidate. We found regional increases in activation and reductions in latency that partly coincided with regions most active in the control group in

176

maree farrow et al.

(A) Control

ADHD

+0.5

−0.5 +15ms

Amplitude Difference

−15ms 0.0

Latency Difference

6.0

Hotelling’s T

(B) Methyl. Effects

+0.3

−0.3

Amplitude Difference

+5ms

−5ms

Latency Difference

0.0

4.0

Hotelling’s T

Figure 3. (A) Topography of reference task – CPT-AX differences for ADHD and control groups at the disappearance of the cue A. For SSVEP amplitude (normalized units) and latency (ms), warmer colours (pink/red) indicate reductions at this time in the CPT-AX relative to the reference task, cooler colours (blue) indicate increases. Hotelling’s T maps indicate the statistical strength of these differences, warmer colours indicate higher T values. Iso-T contours represent uncorrected p values of 0.01 and 0.001. (B) Topography of pre-methylphenidate CPT-AX – post-methylphenidate CPT-AX differences for ADHD boys at the disappearance of the cue A. For SSVEP amplitude (normalized units) and latency (ms), warmer colours (pink/red) indicate reductions at this time post-methylphenidate relative to pre-methylphenidate, cooler colours (blue) indicate increases. Hotelling’s T maps indicate the statistical strength of these differences, warmer colours indicate higher T values. Iso-T contours represent uncorrected p values of 0.01 and 0.001.

our earlier study (Farrow et al., 1999a, 2000a). At the disappearance of the A (figure 3b) there is increased activation after methylphenidate compared with before, predominantly in frontal and occipital regions, and reduced parieto-occipital latency but increased frontal latency. Methylphenidate produced other transient increases in frontal activation and reductions in frontal latency, particularly after the disappearance of the A. Methylphenidate also produced increased activation

Neurobiology of ADHD

177

and reduced latency in parieto-occipital regions. The most sustained and most significant changes occurred in the right parietal region. We also found that there were differences in methylphenidate effects on the SSVEP between boys with ADHD-Combined type and boys with ADHDPredominantly Inattentive type (Farrow et al., 1999b, 2000b). The Combined type subjects demonstrated more right frontal and right parieto-occipital amplitude and latency reductions following methylphenidate, a pattern similar to the activity seen in controls subjects during the same task. The Inattentive type subjects demonstrated more central and left parieto-occipital activation and left temporal latency reductions, suggesting methylphenidate has quite different effects on brain activity in children with ADHD-Inattentive type, which may be related to different underlying deficits. The greater changes in left hemisphere activation may be related to the higher incidence of learning difficulties in children with ADHD-Predominantly Inattentive type. These results suggest that right prefrontal and right parietal processes are involved in performance of the CPT-AX in healthy control children, are deficient in ADHD children, and are enhanced by methylphenidate in ADHD children. The findings of reduced and slower brain activity in children with ADHD are consistent with other findings of reduced cortical activation in ADHD subjects, indexed by increased slow wave EEG (eg. Chabot & Serfontein, 1996; Clarke et al., 1998), reduced P3 amplitude (Klorman, 1991), or reduced metabolic activity (eg. Lou et al., 1984, 1989; Rubia et al., 1999). They may also be consistent with findings of poorer performance on the CPT by ADHD subjects (Corkum & Siegel, 1993; Losier et al., 1996). The topography of the prominent SSVEP differences between ADHD and control groups and changes after methylphenidate (ie. right frontal and right parietal) is consistent with involvement of these regions in vigilance task performance (Pardo et al., 1991; Posner & Raichle, 1994). The diminished activity in these regions in children with ADHD and its enhancement by methylphenidate suggest the possible involvement of both the anterior and posterior attention systems in this disorder (Farrow et al., 1999a, 2000a). Deficits in the anterior or executive attention network, responsible for target detection and executive control of attention and involving frontal cortical regions, are consistent with current theories of ADHD that see frontal deficits as a core component (Barkley, 1997). This is supported by evidence from functional and structural brain imaging studies, including our own, that consistently find frontal cortical regions to be underactive and smaller in children with ADHD (eg. Castellanos et al., 1996; Farrow et al., 1996; Lou et al., 1989; Silberstein et al., 1998). Further evidence comes from neuropsychological studies that find poor performance by children with ADHD on tasks designed to assess frontal or executive function (Pennington & Ozonoff, 1996). Deficits in the posterior or orienting network, responsible for orienting attention to salient stimuli and involving the posterior parietal cortex, in ADHD are less well studied. Reduced right parietal activation in children with ADHD has been

178

maree farrow et al.

found in a few imaging studies including our own (Brandeis et al., 1998; Farrow et al., 1996; Rubia et al., 1999; Van Leeuwen et al., 1998), and there is some neuropsychological evidence for parietal deficits (Aman et al., 1998; Voeller & Heilman, 1988).

Summary Findings from electrophysiological research are in agreement with those from other functional brain imaging studies and suggest that children with ADHD generally have underactive brains compared with their healthy peers. EEG studies mostly find increased slow EEG activity (delta and theta) and decreased fast EEG activity (alpha and beta), suggesting EEG slowing in children with ADHD, although there are some conflicting findings. EEG slowing in ADHD has been interpreted in terms of cortical underarousal and less active information processing (Ackerman et al., 1994; Chabot & Serfontein, 1996; Lubar, 1991; Mann et al., 1992). However, EEG slowing is not specific to ADHD as it is also found in other clinical populations, including children with learning disabilities (Ackerman et al., 1994; Lubar et al., 1985). Several differences in cognitive ERPs between children with ADHD and normal controls have been found. The most consistent of these is reduced amplitude of the P3 component to attended target stimuli recorded from the parietal region. This is generally associated with poorer task performance. This finding suggests that children with ADHD are under-reactive to task-relevant stimuli and may have deficits in allocation of attention and later stages of stimulus processing (Klorman, 1991; Tannock, 1998). However, the finding of reduced P3 amplitude in children with ADHD may have limited value in explaining the specific deficits associated with ADHD as it is also commonly found in other clinical populations (Brandeis et al., 1998). Findings for earlier ERP components associated with the initial orienting of attention are less consistent. Findings of reduced amplitude of N2 and MMN in children with ADHD have been related to deficits in a basic orienting response and in the locus coeruleus noradrenergic system which enhances responsiveness to important signals (Satterfield et al., 1988, 1994). Findings of reduced frontal PN suggest deficits in preferential processing of attended stimuli and are thought to be consistent with reduced frontal metabolism in ADHD (Satterfield et al., 1988, 1994). These findings also suggest deficits in selective attention (Klorman, 1991). However, there are conflicting findings for these early negative ERP components and behavioural studies of selective attention have failed to demonstrate a specific deficit in this aspect of cognition (Pearson & Lane, 1990; Van der Meere & Sergeant, 1988c). ERP studies suggest that deviant processing in children with ADHD appears to be most pronounced for relevant target stimuli and to be associated with later, controlled stages of processing, indexed by reduced P3 amplitude (Brandeis et al.,

Neurobiology of ADHD

179

1998; Klorman, 1991). This is consistent with neuropsychological research that suggests deficits in response selection and organization and motor processes in children with ADHD (Klorman, 1991; Schachar et al., 1993; Tannock, 1998; Van der Meere, 1996). In concordance with EEG and ERP findings, our studies of the SSVEP in children with ADHD also suggest reduced and slower brain activity in these children (Farrow et al., 1996; Silberstein et al., 1998). In particular, our findings suggest deficits in both anterior and posterior attention systems associated with CPT performance. Increased and speeded activation was found to occur in the cortical regions associated with these attention systems after administration of methylphenidate, providing further evidence of right hemisphere frontal and parietal involvement in ADHD and in the effects of stimulants on attentional processes (Farrow et al., 1999a, 2000a). However, these results may depend on the diagnostic characteristics of the subjects, as differences in the effects of methylphenidate on the SSVEP were found between boys meeting criteria for ADHD-Combined type and those diagnosed with ADHD-Predominantly Inattentive type (Farrow et al., 1999b, 2000b). Differences in subject selection criteria and the heterogeneous nature of ADHD are likely to have contributed to many of the conflicting findings in electrophysiological and other brain imaging research, as evidenced by these results for different ADHD subtypes.

CONCLUSIONS The findings of neuropsychological, neurochemical, genetic and neuroimaging research reviewed in this chapter clearly indicate that ADHD does have a neurobiological basis, although its precise nature remains speculative. Given the complexity of interactions between brain regions and neurotransmitter systems and the heterogeneity of ADHD, it is likely that multiple aetiologies and neurobiological deficits can result in the cognitive and behavioural problems we currently define as ADHD. However, despite conflicting theories and research findings, some consistent characteristics do seem to be emerging. Neuropsychological studies implicate deficits in response inhibition as being a core component of ADHD (Barkley, 1997; Pennington & Ozonoff, 1996; Tannock, 1998). Disinhibition and subsequent executive function deficits are thought to produce the impulsive and disorganised behaviour and inattention that characterise ADHD (Barkley, 1997). These inhibition deficits are thought to be associated with abnormalities of the prefrontal cortex and its connections with the basal ganglia, implicated by brain imaging studies (Barkley, 1997; Castellanos, 1997; Tannock, 1998). Fronto-striatal networks are known to be involved in regulation of motor function and behaviour. The prefrontal cortex exerts inhibitory control over motor function via its connections with the basal ganglia and the basal ganglia in turn provides feedback to the prefrontal and premotor cortices.

180

maree farrow et al.

Prefrontal and caudate regions have been found to be around 10% smaller in children with ADHD (Swanson & Castellanos, 1998), suggesting possible reduced fronto-striatal connectivity. Smaller anterior corpus callosum regions have also been found, suggesting reduced interhemispheric connectivity in this system (Baumgardner et al., 1996; Giedd et al., 1994; Hynd et al., 1991). Functional imaging and electrophysiological studies implicate reduced and slower activity in frontal and striatal regions as being associated with executive function deficits in ADHD (Amen & Carmichael, 1997; Lou et al., 1984, 1989; Rubia et al., 1999; Schweitzer et al., 2000; Silberstein et al., 1998; Zametkin et al., 1990). Frontal lobe and basal ganglia abnormalities in ADHD have also been related to networks involved in attention. While neuropsychological findings do not point to a specific deficit in any of the components of attention in ADHD (Van der Meere, 1996), inattentive symptoms are a core feature of the disorder and children with ADHD are found to perform poorly on neuropsychological tasks requiring attention, particularly the CPT (Corkum & Siegel, 1993; Losier et al., 1996). ADHD symptoms involving sustained attention may be related to deficits in the alerting attention network and abnormalities in right frontal regions (Posner & Raichle, 1994; Swanson & Castellanos, 1998; Swanson et al., 1998). The anterior or executive attention network also involves prefrontal cortical regions found to be smaller and less active in children with ADHD (Posner & Raichle, 1994; Swanson & Castellanos, 1998; Swanson et al., 1998). While frontal deficits have been extensively studied and for some time have been concluded to play a major role in ADHD, deficits in posterior brain systems are also suggested by some research findings. The parietal cortex provides the prefrontal cortex with higher order sensory input and the prefrontal cortex in turn inhibits processing of irrelevant stimuli by the parietal cortex, thereby protecting important tasks from interference. Findings of smaller parietal white matter and posterior corpus callosum regions suggest that these reciprocal connections might be reduced in ADHD (Filipek et al., 1997; Hynd et al., 1991; Semrud-Clikeman et al., 1994). Reduced right parietal activation in children with ADHD has been found in brain imaging studies including our own (Brandeis et al., 1998; Farrow et al., 1996; Rubia et al., 1999; Van Leeuwen et al., 1998), and there is some neuropsychological evidence for parietal deficits (Aman et al., 1998; Voeller & Heilman, 1988). The involvement of the parietal cortex, especially the right parietal region, in the orienting or posterior attention network suggests that right parietal abnormalities in ADHD may be associated with symptoms of poor selective attention and vigilance (Posner & Raichle, 1994; Swanson et al., 1998). Recently, some reviewers have attempted to relate the findings of anterior and posterior attention system deficits in ADHD to abnormalities in the associated neurotransmitter systems. Involvement of dopamine and noradrenaline in ADHD is implicated by the efficacy of stimulant drugs in treating the disorder. The parietal cortex receives extensive noradrenergic innervation from the locus coeruleus. This

Neurobiology of ADHD

181

input is thought to inhibit spontaneous parietal activity, thus priming the posterior attention system to orient to and engage novel or salient stimuli. Disruption to this noradrenergic system in ADHD may lead to reduced selective processing of salient stimuli and attention deficits (Pliszka et al., 1996). The prefrontal cortex also receives noradrenergic input from the locus coeruleus, which primes the prefrontal cortex to process task relevant stimuli, suppress task irrelevant stimuli, and inhibit behaviour. Disruption to this system in ADHD may lead to inhibitory control and executive function deficits and behavioural disinhibition (Arnsten, 2000; Arnsten et al., 1996). Noradrenergic input to the prefrontal cortex also modulates dopaminergic activity in this region, and frontal deficits in ADHD are more often related to dopamine dysfunction. Abnormalities in fronto-striatal regions and in dopamine gene polymorphisms suggest an important role for dopamine in ADHD. The response of the prefrontal cortex to input from other regions is modulated by dopaminergic innervation from the ventral tegmental area. This mesocortical dopamine system is involved in selectively gating excitatory inputs, thereby reducing irrelevant activity and improving the signal-to-noise ratio for important stimuli, and is important for the anterior attention system and executive functions. Reduced dopamine in this system may lead to an inability to gate inputs to the anterior attention system and to executive function deficits (Castellanos, 1997; Pliszka et al., 1996). Neurons within the mesocortical dopamine system have few inhibitory autoreceptors, so stimulant drugs are thought to increase postsynaptic dopaminergic effects in the prefrontal cortex and promote integration of input from other cortical regions and enhance executive functions (Castellanos, 1997). Striatal regions receive dopminergic input from the substantia nigra. Activity in this nigrostriatal dopamine system is tightly regulated by inhibitory autoreceptors and feedback from the cortex, so stimulant drugs may produce a net inhibition of dopamine transmission in this system leading to reduced motor activity (Castellanos, 1997). Hence, Castellanos (1997) suggests that executive function deficits in ADHD are associated with reduced activity of the mesocortical dopamine system, while hyperactive and impulsive symptoms are associated with over activity of the nigrostriatal dopamine system. Theories of altered dopaminergic and noradrenergic activity in ADHD are consistent with research findings of altered brain structure and brain function in regions innervated by these neurotransmitter systems, in particular the frontal cortex, the basal ganglia and the parietal cortex. Neurochemical and neurofunctional deficits in these regions may give rise to the disinhibition, hyperactivity and inattention that are the core features of ADHD. However, further research is needed to confirm and clarify the precise nature of these deficits and their role in the symptoms of ADHD. Improved understanding of the neurobiology of ADHD should lead to improved diagnosis and treatment, and improved quality of life for those who suffer from this disorder.

182

maree farrow et al.

REFERENCES Ackerman, P.T., Dykman, R.A., Oglesby, D.M. & Newton, J.E. EEG power spectra of children with dyslexia, slow learners, and normally reading children with ADD during verbal processing. Journal of Learning Disabilities, 1994, 27: 619–630. Alberts, E. & Van der Meere, J. Observations of hyperactive behaviour during vigilance. Journal of Child Psychology and Psychiatry, 1992, 33: 1355–1364. Aman, C.J., Roberts, R.J. & Pennington, B.F. A neuropsychological examination of the underlying deficit in attention deficit hyperactivity disorder: frontal lobe versus right parietal lobe theories. Developmental Psychology, 1998, 34: 956–969. Amen, D.G. & Carmichael, B.D. High-resolution brain SPECT imaging in ADHD. Annals of Clinical Psychiatry, 1997, 9: 81–86. Amen, D.G., Paldi, J.H. & Thisted, R.A. Brain SPECT imaging. Journal of the American Academy of Child Psychiatry, 1993, 32: 1080–1081. American Psychiatric Association. Diagnostic and statistical manual of mental disorders (2nd ed.). Washington DC: Author. 1968. American Psychiatric Association. Diagnostic and statistical manual of mental disorders (3rd ed.). Washington DC: Author. 1980. American Psychiatric Association. Diagnostic and statistical manual of mental disorders (3rd ed., revised). Washington DC: Author. 1987. American Psychiatric Association. Diagnostic and statistical manual of mental disorders (4th ed.). Washington DC: Author. 1994. Arnett, P.A., Rao, S.M., Bernadin, M.S., Grafman, J., Yetkin, F.Z. & Lobeck, L. Relationship between frontal lobe lesions and Wisconsin card sorting test performance in patients with multiple sclerosis. Neurology, 1994, 44: 420–425. Arnsten, A.F., Streere, J.C & Hunt, R.D. The contribution of alpha α2 -noradrenergic mechanisms of prefrontal cortical function. Potential significance for attention-deficit hyperactivity disorder. Archives of General Psychiatry, 1996, 53: 448–455. Ashgari, V., Sanyal, S., Buchwaldt, S., Paterson, A., Jovanovic, V. & Van Tol, H.H.M. Modulation of intracellular cyclic AMP levels by different human dopamine D4 receptor variants. Journal of Neurochemistry, 1995, 65: 1157–1165. August, G.J. & Garfinkel, B.D. Behavioural and cognitive subtypes of ADHD. Journal of the American Academy of Child and Adolescent Psychiatry, 1989, 28: 739–748. Aylward, E.H., Reiss, A.L., Reader, M.J., Singer, H.S., Brown J.E. & Denckla, M.B. Basal ganglia volumes in children with attention-deficit hyperactivity disorder. Journal of Child Neurology, 1996, 11: 112–115. Barkley, R.A. Attention deficit hyperactivity disorder: A handbook for diagnosis and treatment. New York: Guilford Press. 1990. Barkley, R.A. Workshop Manual: Attention Deficit Hyperactivity Disorder in Children and Adolescents. Worcester: Author. 1995. Barkley, R.A. The North American perspective on attention deficit hyperactivity disorder. The Australian Educational and Developmental Psychologist, 1996, 13: 2–23. Barkley, R.A. Behavioural inhibition, sustained attention and executive functions: constructing a unifying theory of ADHD. Psychological Bulletin, 1997, 121: 65–94. Barkley, R.A. Attention-deficit hyperactivity disorder. Scientific American, 1998 (September): 44–49. Barkley, R.A. & Grodzinsky, G.M. Are tests of frontal lobe functions useful in the diagnosis of attention deficit disorders? The Clinical Neuropsychologist, 1994, 8: 121–139. Barkley, R.A., Grodzinsky, G.M. & DuPaul, G.J. Frontal lobe functions in attention deficit disorder with and without hyperactivity: a review and research report. Journal of Abnormal Child Psychology, 1992, 20: 163–188.

Neurobiology of ADHD

183

Baumgardner, T.L., Singer H.S., Denckla, M.B., Rubin, M.A., Abrams, M.T., Colli, M.J. & Reiss, A.L. Corpus callosum morphology in children with Tourette syndrome and attention deficit hyperactivity disorder. Neurology, 1996, 47: 477–482. Bench, C.J., Frith, C.D., Grasby, P.M., Friston, K.J., Paulesu, E., Frackowiak, R.S.J. & Dolan, R.J. Investigations of the functional anatomy of attention using the Stroop test. Neuropsychologia, 1993, 31: 907–922. Benjamin, J., Li, L., Patterson, C., Greenberg, B.D., Murphy, D.L. & Hamer, D.H. Population and familial association between the D4 dopamine receptor gene and measures of novelty seeking. Nature Genetics, 1996, 12: 81–84. Berman, K.F., Ostrem, J.L., Randolph, C., Gold, J., Goldberg, T.E., Coppola, R., Carson, R.E., Herscovitch, P. & Weinberger, D.R. Physiological activation of a cortical network during performance of the Wisconsin card sorting test: a positron emission tomography study. Neuropsychologia, 1995, 33: 1027–1046. Berquin, P.C., Giedd, J.N., Jacobsen, L.K., Hamburger, S.D., Krain, A.L., Rapoport, J.L. & Castellanos, F.X. Cerebellum in attention-deficit hyperactivity disorder: a morphometric MRI study. Neurology, 1998, 50: 1087–1093. Biederman, J., Newcorn, J. and Sprich, S. Comorbidity of attention deficit hyperactivity disorder with conduct, depressive, anxiety and other disorders. American Journal of Psychiatry, 1991, 148: 564–577. Boucugnani, L.L. & Jones, R.W. Behaviours analogous to frontal lobe dysfunction in children with attention deficit hyperactivity disorder. Archives of Clinical Neuropsychology, 1989, 4: 161–173. Bradley, C. The behaviour of children receiving Benzedrine. American Journal of Psychiatry, 1937, 94: 577–585. Bradley, P.B. Introduction to neuropharmacology. London: Wright. 1989. Breen, M.J. Cognitive and behavioural differences in ADHD boys and girls. Journal of Child Psychology and Psychiatry, 1989, 30: 711–716. Burke, R.S. A cognitive-developmental approach to studying attention deficits. In J.T. Enns (Ed.) The development of attention: Research and theory. New York: North Holland. 1990, pp. 365–381. Callaway, E., Halliday, R. & Naylor, H. Hyperactive children’s event-related potentials fail to support underarousal and maturational-lag theories. Archives of General Psychiatry, 1983, 40: 1243–1248. Campbell, M., Cohen, I.L., Perry, R. & Small A.M. Psychopharmacological Treatment. In T.H. Ollendick & M. Hersen (Eds.). Handbook of child psychopathology. (2nd ed.). New York: Plenum Press. 1989, pp. ???? Cantwell, D.P. Attention deficit disorder: a review of the past 10 years. Journal of the American Academy of Child and Adolescent Psychiatry, 1996, 35: 978–987. Caresia, L., Pugnetti, L., Besana, R., Barteselli, F., Cazullo, A.G., Musetti, L. & Scarone, S. EEG and clinical findings during pemoline treatment in children and adults with attention deficit disorder. Neuropsychobiology, 1984, 11: 158–167. Carter, C.S., Krener, P., Chaderjian, M., Northcutt, C. & Wolfe, V. Asymmetrical visual-spatial attentional performance in ADHD: evidence for a right hemispheric deficit. Biological Psychiatry, 1995, 37: 789–798. Casey, B.J., Castellanos, F.X., Giedd, J.N., Marsh, W.L., Hamburger, S.D., Schubert, A.B., Vauss, Y.C., Vaituzis, A.C., Dickstein, D.P., Sarfatti, S.E. & Rapoport, J.L. Implication of right frontostriatal circuitry in response inhibition and Attention-Deficit/Hyperactivity Disorder. Journal of the American Academy of Child and Adolescent Psychiatry, 1997, 36: 374–383. Casey, B.J., Trainor, R.J., Orendi, J.L., Schubert, A.B., Nystrom, L.E., Giedd, J.N., Castellanos, F.X., Haxby, J.V., Noll, D.C., Cohen, J.D., Forman, S.D., Dahl, R.E. & Rapoport, J.L. A developmental functional MRI study of prefrontal activation during performance of go-no-go task. Journal of Cognitive Neuroscience, 1997, 9: 835–847.

184

maree farrow et al.

Castellanos, F.X. Toward a pathophysiology of attention-deficit/hyperactivity disorder. Clinical Pediatrics, 1997, 36: 381–393. Castellanos, F.X., Elia, J., Kruesi, M.J., Gulotta, C.S., Mefford, I.N., Potter, W.Z., Ritchie, G.F. & Rapoport, J.L. Cerebrospinal fluid monoamine metabolites in boys with attention-deficit hyperactivity disorder. Psychiatry Research, 1994, 52: 305–316. Castellanos, F.X., Elia, J., Kruesi, M.J., Marsh, W.L., Gulotta, C.S., Potter, W.Z., Ritchie, G.F., Hamburger, S.D. & Rapoport, J.L. Cerebrospinal fluid homovanillic acid predicts behavioural response to stimulants in 45 boys with attention deficit/hyperactivity disorder. Neuropsychopharmacology, 1996, 14: 125–137. Castellanos, F.X., Giedd, J.N., Eckburg, P., Marsh, W.L., Vaituzis, C., Kaysen, D., Hamburger, S.D. & Rapoport, J.L. Quantitative morphology of the caudate nucleus in attention deficit hyperactivity disorder. American Journal of Psychiatry, 1994, 151: 1791–1796. Castellanos, F.X., Giedd, J.N., Marsh, W.L., Hamburger, S.D., Vaituzis, C., Dickstein, D.P., Sarfatti, S., Vauss, Y.C., Snell, J.W., Lange, N., Kaysen, D., Krain, A.L., Ritchie, G.F., Rajapakse, J.C. & Rapoport, J.L. Quantitative brain magnetic resonance imaging in attention-deficit/hyperactivity disorder. Archives of General Psychiatry, 1996a, 53: 607–616. Castellanos, F.X., Giedd, J.N., Hamburger, S.D., Marsh, W.L. & Rapoport, J.L. Brain morphometry in Tourette’s syndrome: the influence of comorbid attention-deficit/hyperactivity disorder. Neurology, 1996b, 47: 1581–1583. Castellanos, F.X., Lau, E., Tayebi, N., Lee, P., Long, R.E., Giedd, J.N., Sharp, W., Marsh, W.L., Walter, J.M., Hamburger, S.D., Ginns, E.I., Rapoport, J.L. & Sidransky, E. Lack of an association between a dopamine-4 receptor polymorphism and attention-deficit/hyperactivity disorder: genetic and brain morphometric analyses. Molecular Psychiatry, 1998, 3: 431–434. Chabot, R.J., Merkin, H., Wood, L.M., Davenport, T.L. & Serfontein, G. Sensitivity and specificity of QEEG in children with attention deficit or specific developmental learning disorders. Clinical Electroencephalography, 1996, 27: 26–34. Chabot, R.J. & Serfontein, G. Quantitative electroencephalographic profiles of children with attention deficit disorder. Biological Psychiatry, 1996, 40: 951–963. Chee, P., Logan, G., Schacher, R., Lindsay, P. & Wachsmuth, R. Effects of event rate and display time on sustained attention in hyperactive, normal, and control children. Journal of Abnormal Child Psychology, 1989, 17: 371–391. Chelune, G.J., Ferguson, W., Koon, R. & Dickey, T.O. Frontal lobe disinhibition in attention deficit disorder. Child Psychiatry and Human Development, 1986, 16: 221–234. Clarke, A.R., Barry, R.J., McCarthy, R. & Selikowitz, M. EEG analysis in attention-deficit/hyperactivity disorder: a comparative study of two subtypes. Psychiatry Research, 1998, 81:19–29. Cook, E.H., Stein, M.A., Krasowski, M.D., Cox, N.J., Olkon, D.M., Kieffer, J.E. & Leventhal, B.L. Association of attention-deficit disorder and the dopamine transporter gene. American Journal of Human Genetics, 1995, 56: 993–998. Cooper, J.R., Bloom, F.E. & Roth, R.H. The biochemical basis of neuropharmacology (5th ed.). New York: Oxford University Press. 1986. Corkum, P.V. & Siegel, L.S. Is the continuous performance task a valuable research tool for use with children with attention-deficit-hyperactivity disorder? Journal of Child Psychology and Psychiatry, 1993, 34: 1217–1239. Corkum, P.V. & Siegel, L.S. Debate and argument: Reply to Dr Koelega: Is the continuous performance task useful in research with ADHD children? Comments on a review. Journal of Child Psychology and Psychiatry, 1995, 36: 1487–1493. Daly, G., Hawi, Z., Fitzgerald, M. & Gill M. Mapping susceptibility loci in attention deficit hyperactivity disorder: preferential transmission of parental alleles at DAT1, DBH and DRD5 to affected children. Molecular Psychiatry, 1999, 4: 2192–2196. De Hass, P.A. Attentional style and peer relationships of hyperactive and normal boys and girls. Journal of Abnormal Child Psychology, 1986, 14: 457–467.

Neurobiology of ADHD

185

Delis, D.C., Squire, L.R., Bihrle, A. & Massman, P. Componential analysis of problem-solving ability: performance of patients with frontal lobe damage and amnesic patients on a new sorting test. Neuropsychologia, 1992, 30: 683–697. Dinklage, D. & Barkley, R.A. Disorders of attention in children. In S.J. Segalowitz & I. Rapin (Eds.) Handbook of neuropsychology. Volume 7: Child neuropsychology. Amsterdam: Elsevier. 1992, pp. 279–307. Douglas, V.I. Attentional and cognitive problems. In M. Rutter (Ed.) Developmental Neuropsychiatry. New York: Guilford Press. 1983, pp. 280–328. Douglas, V.I. Cognitive deficits in children with attention deficit disorder with hyperactivity. In L.M. Bloomingdale & J.A. Sergeant (Eds.) Attention deficit disorder: Criteria, cognition, intervention. Oxford: Pergamon Press. 1988, pp. 65–83. Douglas, V.I. & Benezra, E. Supraspan verbal memory in attention deficit disorder with hyperactivity, normal, and reading disabled boys. Journal of Abnormal Child Psychology, 1990, 18: 617–638. Dykman, R.A., Holcomb, P.J., Oglesby, D.M. & Ackerman, P.T. Electrocortical frequencies in hyperactive, learning disabled, mixed, and normal children. Biological Psychiatry, 1982, 17: 675–685. Ebstein, R.P., Novick, O., Umansky, R., Priel, B., Osher, Y., Blaine, D., Bennett, E.R., Nemenov, L., Katz, M. & Belmaker, R.H. Dopamine D4 receptor (D4DR) exon III polymophism associated with the human personality trait of novelty seeking. Nature Genetics, 1996, 12: 78–80. Erickson, M.T. Behaviour disorders of children and adolescents. Englewood Cliffs: Prentice Hall. 1987. Ernst, M., Cohen, R.M., Liebenauer, L.L., Jons, P.H. & Zametkin, A.J. Cerebral glucose metabolism in adolescent girls with attention-deficit/hyperactivity disorder. Journal of the American Academy of Child and Adolescent Psychiatry, 1997, 36: 1399–1406. Ernst, M., Liebenauer, L.L., King, A.C., Fitzgerald, G.A., Cohen, R.M. & Zametkin, A.J. Reduced brain metabolism in hyperactive girls. Journal of the American Academy of Child and Adolescent Psychiatry, 1994a, 33: 858–868. Ernst, M., Zametkin, A.J., Matochik, J.A., Jons, P.H. & Cohen, R.M. DOPA decarboxylase activity in attention deficit hyperactivity disorder adults. A [flourine-18] flourodopa positron emission tomographic study. Journal of Neuroscience, 1998, 18: 5901–5907. Ernst, M., Zametkin, A.J., Matochik, J.A., Liebenauer, L.L., Fitzgerald, G.A. & Cohen, R.M. Effects of intravenous dextroamphetamine on brain metabolism in adults with attention-deficit hyperactivity disorder (ADHD). Preliminary findings. Psychopharmacology Bulletin, 1994b, 30: 219–225. Ernst, M., Zametkin, A.J., Matochik, J.A., Pascualvaca, D., Jons, P.H. & Cohen, R.M. High midbrain [18F]DOPA accumulation in children with attention deficit hyperactivity disorder. American Journal of Psychiatry, 1999, 156: 1209–1215. Fadely, J.L. & Hosler, V.N. Attentional deficit disorder in children and adolescents. Springfield: Charles C. Thomas. 1992. Faraone, S.V., Biederman, J., Weiffenbach, B., Keith, T., Chu, M.P., Weaver, A., Spencer, T.J., Wilens, T.E., Frazier, J., Cleves, M. & Sakai, J. Dopamine D4 gene 7-repeat allele and attention deficit hyperactivity disorder. American Journal of Psychiatry, 1999, 156: 5768–5770. Farrow, M., Silberstein, R.B., Levy, F., Pipingas, A., Wood, K., Hay, D.A. & Jarman, F.C. Prefrontal and parietal deficits in ADHD suggested by brain electrical activity mapping during children’s performance of the AX CPT. The Australian Educational and Developmental Psychologist, 1996, 13: 59–68. Feingold, B.F. Why your child is hyperactive. New York: Random House. 1975. Filipek, P.A., Kennedy, D.N. & Caviness, V.S. Neuroimaging in child neuropsychology. Ch. 11 in I. Rapin & S.J. Segalowitz (Eds.) Handbook of Neuropsychology, Volume 6: Child Neuropsychology. Elsevier, Amsterdam, 1992, pp. 301–329. Filipek, P.A., Semrud-Clikeman, M., Steingard, R.J., Renshaw, P.F., Kennedy, D.N. & Biederman, J. Volumetric MRI analysis comparing subjects having attention-deficit hyperactivity disorder with normal controls. Neurology, 1997, 48: 589–601.

186

maree farrow et al.

Fischer, M., Barkley, R.A., Edelbrock, C.S. & Smallish, L. The adolescent outcome of hyperactive children diagnosed by research criteria: II. Academic, attentional, and neuropsychological status. Journal of Consulting and Clinical Psychology, 1990, 58: 580–588. Fischer, M., Newby, R.F. & Gordon, M. Who are the false negatives on continuous performance tests? Journal of Clinical Child Psychology, 1995, 24: 427–433. Frank, Y., Seiden, J.A. & Napolitano, B. Event-related potentials to an “oddball” auditory paradigm in children with learning disabilities with or without attention deficit hyperactivity disorder. Clinical Electroencephalography, 1994, 25: 136–141. Frank, Y., Seiden, J.A. & Napolitano, B. Electrophysiological changes in children with learning and attentional abnormalities as a function of age: event-related potentials to an “oddball” paradigm. Clinical Electroencephalography, 1998, 29: 188–193. Fuster, J.M. The prefrontal cortex. New York: Raven Press. 1989. Gasser, T., Verleger, R., Bacher, P. & Sroka, L. Development of the EEG of school-age children and adolescents. I. Analysis of band power. Electroencephalography and Clinical Neurophysiology, 1988, 69: 91–99. Gelfand, D.M., Jenson, W.R. & Drew, C.J. Understanding child behaviour disorders (2nd ed.). New York: Holt, Rinehart and Winston Inc. 1988. Giedd, J.N., Castellanos, F.X., Casey, B.J., Kozuch, P., King, A.C., Hamburger, S.D. & Rapoport, J.L. Quantitative morphology of the corpus callosum in attention deficit hyperactivity disorder. American Journal of Psychiatry, 1994, 151: 665–669. Gill, M., Daly, G., Heron, S., Hawi, Z. & Fitzgerald, M. Confirmation of association between attention deficit hyperactivity disorder and a dopamine transporter polymorphism. Biological Psychiatry, 1997, 2: 311–313. Goel, V. & Grafman, J. Are the frontal lobes implicated in “planning” functions? Interpreting data from the Tower of Hanoi. Neuropsychologia, 1995, 33: 623–642. Gorenstein, E.E., Mammato, C.A. & Sandy, J.M. Performance of inattentive-overactive children on selected measures of prefrontal-type function. Journal of Clinical Psychology, 1989, 45: 619–632. Grodzinsky, G.M. & Diamond, R. Frontal lobe functioning in boys with attention deficit hyperactivity disorder. Developmental Neuropsychology, 1992, 8: 427–445. Halperin, J.M., Matier, K., Bedi, G., Sharma, V. & Newcorn, J.H. Specificity of inattention, impulsivity, and hyperactivity to the diagnosis of attention-deficit hyperactivity disorder. Journal of the American Academy of Child and Adolescent Psychiatry, 1992, 31: 190–196. Halperin, J.M., Newcorn, J.H., Koda, V.H., Pick, L., McKay, K.E. & Knott, P. Noradrenergic mechanisms in ADHD children with and without reading disabilities: a replication and extension. Journal of the American Academy of Child and Adolescent Psychiatry, 1997, 36: 1688–1697. Halperin, J.M., Newcorn, J.H., Matier, K., Sharma, V., McKay, K.E. & Schwartz, S. Discriminant validity of attention-deficit hyperactivity disorder. Journal of the American Academy of Child and Adolescent Psychiatry, 1993, 32: 1038–1043. Halperin, J.M., Newcorn, J.H., Schwartz, S.T., McKay, K.E., Bedi, G. & Sharma, V. Plasma catecholamine levels in ADHD boys with and without reading disabilities. Journal of Clinical Child Psychology, 1993, 22: 219–225. Halperin, J.M., O’Brien, J.D., Newcorn, J.H., Healy, J.M., Pascualvaca, D.M., Wolf, L.E. & Young, J.G. Validation of hyperactive, aggressive, and mixed hyperactive/aggressive childhood disorders: a research note. Journal of Child Psychology and Psychiatry, 1990, 31: 455–459. Harper, G.W. & Ottinger, D.R. The performance of hyperactive and control preschoolers on a new computerized measure of visual vigilance: the Preschool Vigilance Task. Journal of Child Psychology and Psychiatry, 1992, 33: 1365–1372. Holcomb, P.J., Ackerman, P.T. & Dykman, R.A. Cognitive event-related brain potentials in children with attention and reading deficits. Psychophysiology, 1985, 22: 656–667. Hooks, K., Milich, R. & Lorch, E. Sustained and selective attention in boys with attention deficit hyperactivity disorder. Journal of Clinical Child Psychology, 1994, 23: 69–77.

Neurobiology of ADHD

187

Horn, W.F., Wagner, A.E. & Ialongo, N. Sex differences in school aged children with pervasive attention deficit hyperactivity disorder. Journal of Abnormal Child Psychology, 1989, 17: 109–125. Hynd, G.W., Semrud-Clikeman, M., Lorys, A.R., Novey, E.S. & Eliopulos, D. Brain morphology in developmental dyslexia and attention deficit disorder/hyperactivity. Archives of Neurology, 1990, 47: 919–926. Hynd, G.W., Semrud-Clikeman, M., Lorys, A.R., Novey, E.S., Eliopulos, D. & Lyytinen, H. Corpus callosum morphology in attention-deficit hyperactivity disorder: morphometric analysis of MRI. Journal of Learning Disabilities, 1991, 24: 141–146. Hynd, G.W., Hern, K.L, Novey, E.S., Eliopulos, D., Marshall, R., Gonzalez, J.J. & Voeller, K.K.S. Attention deficit hyperactivity disorder and asymmetry of the caudate nucleus. Journal of Child Neurology, 1993, 8: 339–347. Iaboni, F., Douglas, V.I. & Baker, A.G. Effects of reward and response costs on inhibition in ADHD children. Journal of Abnormal Psychology, 1995, 104: 232–240. Jarman, F.C. Current approaches to management of attention deficit hyperactivity disorder. The Australian Educational and Developmental Psychologist, 1996, 13: 46–55. Jensen, P.S., Martin, B.A. & Cantwell, D.P. Comorbidity in ADHD: Implications for research, practice, and DSM-V. Journal of the American Academy of Child and Adolescent Psychiatry, 1997, 36: 1065–1079. Johnstone, S.J. & Barry, R.J. Auditory event-related potentials to a two-tone discrimination paradigm in attention deficit hyperactivity disorder. Psychiatry Research, 1996, 64: 179–192. Jonkman, L.M., Kemner, C., Verbaten, N.M., Koelega, H.A., Camfferman, G., van der Gaag, R.J., Buietelaar, J.K. & van Engeland, H. Event-related potentials and performance of attention-deficit hyperactivity disorder children and normal controls in auditory and visual selective attention tasks. Biological Psychiatry, 1997, 41: 595–611. Kagan, J. Reflection and impulsivity and reading ability in primary grade children. Child Development, 1965, 37: 583–594. Kemner, C., Verbaten, M.N., Cuperus, J.M., Camfferman, G. & Van Engeland, H. Visual and somatosensory event-related brain potentials in autistic children and three different control groups. Electroencephalography and Clinical Neurophysiology, 1994, 92: 225–237. Kemner, C., Verbaten, M.N., Koelega, H.S., Buitelaar, J.K., van der Gaag, R.J., Camfferman, G. & van Engeland, H. Event-related brain potentials in children with attention-deficit and hyperactivity disorder: effects of stimulus deviancy and task relevance in the visual and auditory modality. Biological Psychiatry, 1996, 40: 522–534. Kemner, C., Verbaten, M.N., Koelega, H.S., Camfferman, G. & van Engeland, H. Are abnormal eventrelated potentials specific to children with ADHD? A comparison with two clinical groups. Perceptual Motor Skills, 1998, 87: 1083–1090. Kesner, R.P., Hopkins, R.O. & Fineman, B. Item and order dissociation in humans with prefrontal cortex damage. Neuropsychologia, 1994, 32: 881–891. Khan, A.U. & Dekirmenjian, H. Urinary excretion of catecholamine metabolites in hyperkinetic child syndrome. American Journal of Psychiatry, 1981, 138: 108–112. Klorman, R. Cognitive event-related potentials in attention deficit disorder. Journal of Learning Disabilities, 1991, 24: 130–140. Klorman, R., Brumaghim, J.T., Fitzpatrick, P.A. & Borgstedt, A.D. Methylphenidate speeds evaluation processes of attention deficit disorder adolescents during a continuous performance test. Journal of Abnormal Child Psychology, 1991, 19: 263–283. Klorman, R., Salzman, L.F. & Borgstedt, A.D. Brain event-related potentials in evaluation of cognitive deficits in attention deficit disorder and outcome of stimulant therapy. In L.M. Bloomingdale (Ed.). Attention deficit disorder. Volume 3. Oxford: Pergamon Press. 1988, pp. ???? Klorman, R., Salzman, L.F., Pass, H.L., Borgstedt, A.D. & Dainer, K.B. Effects of methylphenidate on hyperactive children’s evoked responses during passive and active attention. Psychophysiology, 1979, 16: 23–29.

188

maree farrow et al.

Koelega, H.S. Is the continuous performance task useful in research with ADHD children? Comments on a review. Journal of Child Psychology and Psychiatry, 1995, 36: 1477–1485. Kuperman, S., Johnson, B., Arndt, S., Lindgren, S. & Wolraich, M. Quantitative EEG differences in a nonclinical sample of children with ADHD and undifferentiated ADD. Journal of the American Academy of Child and Adolescent Psychiatry, 1996, 35: 1009–1017. Kupietz, S.S. Sustained attention in normal and in reading-disabled youngsters with and without ADDH. Journal of Abnormal Child Psychology, 1990, 18: 357–372. LaHoste, G.J., Swanson, J.M., Wigal, S.B., Glabe, C., Wigal, T., King, N. & Kennedy, J.L. Dopamine D4 receptor gene polymorphism is associated with attention deficit hyperactivity disorder. Molecular Psychiatry, 1996, 1: 121–124. Landau, S., Lorch, E.P. & Milich, R. Visual attention to and comprehension of television in attentiondeficit hyperactivity disordered and normal boys. Child Development, 1992, 63: 928–937. Leung, P.W. & Connolly, K.J. Attentional difficulties in hyperactive and conduct-disordered children: a processing deficit. Journal of Child Psychology and Psychiatry, 1994, 35: 1229–1245. Leung, P.W. & Connolly, K.J. Distractibility in hyperactive and conduct-disordered children. Journal of Child Psychology and Psychiatry, 1996, 37: 305–312. Leung, P.W. & Luk, S.L. Differences in attention control between ‘clinic-observable’ and ‘reported’ hyperactivity: a preliminary report. Child Care, Health and Development, 1988, 14: 199–211. Levine, M.D. Persistent inattention and unintention in middle childhood. In M.D. Levine & P. Satz (Eds.) Middle childhood: Development and dysfunction. Baltimore: University Park Press. 1984, pp. ???? Levy, F. The development of sustained attention (vigilance) and inhibition in children: some normative data. Journal of Child Psychology and Psychiatry, 1980, 21: 77–84. Levy, F. The dopamine theory of attention deficit hyperactivity disorder (ADHD). Australian and New Zealand Journal of Psychiatry, 1991, 25: 277–283. Levy, F., Hay, D.A., McStephen, M. & Wood, C. Attention-deficit hyperactivity disorder: a category or a continuum? Genetic analysis of a large-scale twin study. Journal of the American Academy of Child Psychiatry, 1997, 36: 737–744. Levy, F. & Hobbes, G. The diagnosis of attention deficit disorder (hyperkinesis) in children. Journal of the American Academy of Child Psychiatry, 1981, 20: 376–384. Levy, F. & Ward, P.B. Neurometrics, dynamic brain imaging and attention deficit hyperactivity disorder. Journal of Paediatrics and Child Health, 1995, 31: 279–283. Lezak, M.D. Neuropsychological assessment (2nd ed.). New York: Oxford University Press. 1983. Loiselle, D.L., Stamm, J.S., Maitinsky, S. & Whipple, S.C. Evoked potential and behavioural signs of attentive dysfunctions in hyperactive boys. Psychophysiology, 1980, 17: 193–201. Losier, B.J., McGrath, P.J. & Klein, R.M. Error patterns on the continuous performance test in nonmedicated and medicated samples of children with and without ADHD: a meta-analytic review. Journal of Child Psychology and Psychiatry, 1996, 37: 971–987. Lou, H.C. Cerebral single photon emission tomography (SPECT) and positron emission tomography (PET) during development and in learning disorders. Ch. 12 in I. Rapin & S.J. Segalowitz (Eds.) Handbook of Neuropsychology, Volume 6: Child Neuropsychology. Elsevier, Amsterdam, 1992, pp. 331–338. Lou, H.C. Etiology and pathogenesis of attention-deficit hyperactivity disorder (ADHD): significance of prematurity and perinatal hypoxic-haemodynamic encephalopathy. Acta Paediatrica, 1996, 85: 1266–1271. Lou, H.C., Henriksen, L. & Bruhn, P. Focal cerebral hypoperfusion in children with dysphasia and/or attention deficit disorder. Archives of Neurology, 1984, 41: 825–829. Lou, H.C., Henriksen, L. & Bruhn, P. Focal cerebral dysfunction in developmental learning disabilities. Lancet, 1990a, 335: 8–11.

Neurobiology of ADHD

189

Lou, H.C., Henriksen, L., Bruhn, P., Borner, H. & Nielsen, J.B. Striatal dysfunction in attention deficit and hyperkinetic disorder. Archives of Neurology, 1989, 46: 48–52. Lou, H.C., Henriksen, L., Greisen, G. & Schneider, S. Redistribution of cerebral activity during childhood. Brain Development, 1990b, 12: 301–305. Lubar, J.F. Discourse on the development of EEG diagnostics and biofeedback for attentiondeficit/hyperactivity disorders. Biofeedback and Self-Regulation, 1991, 16: 201–225. Lubar, J.F., Bianchini, B.A., Calhoun, W.H., Lambert, E.W., Brody, Z.H. & Shabsin, H.S. Spectral analysis of EEG differences between children with and without learning disabilities. Journal of Learning Disabilities, 1985, 18: 403–408. Lyoo, K., Noam, G.G., Lee, C.K., Lee, H.K., Kennedy, B.P. & Renshaw, P.F. The corpus callosum and lateral ventricles in children with attention-deficit hyperactivity disorder: A brain magnetic resonance imaging study. Biological Psychiatry, 1996, 40: 1060–1063. Malone, M.A., Kershner, J.R. & Swanson, J.M. Hemispheric processing and methylphenidate effects in attention-deficit hyperactivity disorder. Journal of Child Neurology, 1994, 9: 181–189. Mann, C., Lubar, J.F., Zimmerman, A.W., Miller, C.A. & Muenchen, R.A. Quantitative analysis of EEG in boys with attention-deficit-hyperactivity disorder: Controlled study with clinical implications. Paediatric Neurology, 1992, 8: 30–36. Mason, S.T. Catecholamines and behaviour. Cambridge: Cambridge University Press. 1984. Mataro, M., Garcia-Sanchez, C., Junque, C., Estevez-Gonzalez, A. & Pujol, J. Magnetic resonance imaging measurement of the caudate nucleus in adolescents with attention-deficit hyperactivity disorder and its relationships with neuropsychological and behavioural measures. Archives of Neurology, 1997, 54: 963–968. Matochik, J.A., Liebenauer, L.L., King, A.C., Szymanski, H.V., Cohen, R.M. & Zametkin, A.J. Cerebral glucose metabolism in adults with attention deficit hyperactivity disorder after chronic stimulant treatment. American Journal of Psychiatry, 1994, 151: 658–664. Matochik, J.A., Nordahl, T.E., Gross, M., Semple, W.E., King, A.C., Cohen, R.M. & Zametkin, A.J. Effects of acute stimulant medication on cerebral metabolism in adults with hyperactivity. Neuropsychopharmacology, 1993, 8: 377–386. Matsuura, M., Okubo, Y., Toru, M., Kojima, T., He, Y., Hou, Y., Shen, Y. & Lee, C.K. A cross-national EEG study of children with emotional and behavioural problems: a WHO collaborative study in the western Pacific region. Biological Psychiatry, 1993, 34: 59–65. Meador-Woodruff, J.H., Grandy, D.K., Van Tol, H.H., Damask, S.P., Little, K.Y., Civelli, O. & Watson, S.J. Jr. Dopamine receptor gene expression in the human medial temporal lobe. Neuropsychopharmacology, 1994, 10: 239–248. Michael, R.L., Klorman, R., Salzman, L.F., Borgstedt, A.D. & Dainer, K.B. Normalizing effects of methylphenidate on hyperactive children’s vigilance performance and evoked potentials. Psychophysiology, 1981, 18: 665–677. Monastra, V.J., Lubar, J.F., Linden, M., VanDeusen, P., Green, G., Wing, W., Phillips, A. & Fenger, T.N. Assessing attention deficit hyperactivity disorder via quantitative electroencephalography: an initial validation study. Neuropsychology, 1999, 13: 424–433. Morris, R.G., Ahmed, S., Syed, G.M. & Toone, B.K. Neural correlates of planning ability: frontal lobe activation during the tower of London test. Neuropsychologia, 1993, 31: 1367–1378. Newcorn, J.H., Halperin, J.M., Healy, J.M., O’Brien, J.D., Pascualvaca, D.M., Wolf, L.E., Morganstein, A., Sharma, V. & Young, J.G. Are ADDH and ADHD the same or different? Journal of the American Academy of Child and Adolescent Psychiatry, 1989, 28: 734–738. Newton, J.E.O., Oglesby, D.M., Ackerman, P.T. & Dykman, R.A. Visual slow brain potentials in children with attention deficit disorder. Integrative Physiological and Behavioural Science, 1994, 29: 39–54. Oades, R.D. Attention deficit disorder with hyperactivity (ADDH): the contribution of catecholaminergic activity. Progress in Neurobiology, 1987, 29: 365–391.

190

maree farrow et al.

Oades, R.D. Frontal, temporal and lateralized brain function in children with attention-deficit hyperactivity disorder: a psychophysiological and neuropsychological viewpoint on development. Behavioural Brain Research, 1998, 94: 83–95. Oades, R.D., Daniels, R. & Rascher, W. Plasma neuropeptide-Y levels, monoamine metabolism, electrolyte excretion and drinking behaviour in children with attention-deficit hyperactivity disorder. Psychiatry Research, 1998, 80: 177–186. Oosterlaan, J., Logan, G.D. & Sergeant, J.A. Response inhibition in AD/HD, CD, comobid AD/HD + CD, anxious, and control children: a meta-analysis of studies with the stop task. Journal of Child Psychology and Psychiatry, 1998, 39: 411–425. Oosterlaan, J. & Sergeant, J.A. Inhibition in ADHD, aggressive, and anxious children: a biologically based model of child psychopathology. Journal of Abnormal Child Psychology, 1996, 24:19–36. Oosterlaan, J. & Sergeant, J.A. Response inhibition and response re-engagement in attentiondeficit/hyperactivity disorder, disruptive, anxious, and normal children. Behavioural Brain Research, 1998, 94: 33–43. O’Tuama, L.A. & Treves, S.T. Brain single-photon emission computed tomography for behaviour disorders in children. Seminars in Nuclear Medicine, 1993, 23: 255–264. Overmeyer, S., Simmons, A., Santosh, J., Andrew, C., Williams, S.C., Taylor, A., Chen, W. & Taylor, E. Corpus callosum may be similar in children with ADHD and siblings of children with ADHD. Developmental Medicine and Child Neurology, 2000, 42: 8–13. Overtoom, C.C.E., Verbaten, M.N., Kemner, C., Kenemans, J.L., van Engeland, H., Buitelaar, J.K., Camfferman, G. & Koelega, H.S. Associations between event-related potentials and measures of attention and inhibition in the continuous performance task in children with ADHD and normal controls. Journal of the American Academy of Child and Adolescent Psychiatry, 1998, 37: 977– 985. Pardo, J.V., Fox, P.T. & Raichle, M.E. Localization of a human system for sustained attention by positron emission tomography. Nature, 1991, 349: 61–64. Pardo, J.V., Pardo, P.J., Janer, K.W. & Raichle, M.E. The anterior cingulate cortex mediates processing selection in the Stroop attentional conflict paradigm. Proceedings of the National Academy of Science, 1990, 87: 256–259. Pascual-Marqui, R.D., Michel, C.M. & Lehmann, D. Low resolution electromagnetic tomography: a new method for localizing electrical activity in the brain. International Journal of Psychophysiology, 1994, 18: 49–65. Pearson, D.A. & Lane, D.M. Reorientation in hyperactive and non-hyperactive children: Evidence for developmentally immature attention? In J.T. Enns (Ed.) The development of attention: Research and theory. Amsterdam: North-Holland. 1990, pp. 345–363. Pennington, B.F., Groisser, D. & Welsh, M.C. Contrasting deficits in attention deficit hyperactivity disorder versus reading disability. Developmental Psychology, 1993, 29: 511–523. Pennington, P.F. & Ozonoff, S. Executive functions and developmental psychopathology. Journal of Child Psychology and Psychiatry, 1996, 37: 51–87. Pliszka, S.R., McCracken, J.T. & Maas, J.W. Catecholamines in attention-deficit hyperactivity disorder: current perspectives. Journal of the American Academy of Child and Adolescent Psychiatry, 1996, 35: 264–272. Posner, M.I. Orienting of attention. Quarterly Journal of Experimental Psychology, 1980, 32: 3–25. Posner, M.I. & Petersen, S. The attention system of the human brain. Annual Review of Neuroscience, 1990, 13:25–42. Posner, M.I. & Raichle, M.E. Images of Mind. New York: Scientific American Library. Power, T.J. Contextual factors in vigilance testing of children with ADHD. Journal of Abnormal Child Psychology, 1992, 20: 579–593. Prendergast, M., Taylor, E., Rapoport, J.L., Bartko, J., Donnelly, M., Zametkin, A., Ahearn, M.B., Dunn, G. & Wieselberg, H.M. The diagnosis of childhood hyperactivity: a U.S.—U.K.

Neurobiology of ADHD

191

cross-national study of DSM-III and ICD-9. Journal of Child Psychology and Psychiatry, 1988, 29: 289–300. Pribram, K.H. & McGuiness, D. Arousal, activation and effort in the control of attention. Psychological Review, 1975, 2: 116–149. Prichep, L.S., Sutton, S. & Hakerem, G. Evoked potentials in hyperkinetic and normal children under certainty and uncertainty: a placebo and methylphenidate study. Psychophysiology, 1976, 13: 419–428. Prior, M. Implications of ADHD for learning. The Australian Educational and Developmental Psychologist, 1996, 13: 24–28. Quay, H.C. Attention deficit disorder and the behavioural inhibition system: The relevance of the neuropsychological theory of Jeffrey A. Gray. In L.M. Bloomingdale & J.A. Sergeant (Eds.) Attention deficit disorder: Criteria, cognition, intervention. Oxford: Pergamon Press. 1988, pp. 117–125. Quay, H.C. Inhibition and attention deficit hyperactivity disorder. Journal of Abnormal Child Psychology, 1997, 25: 7–13. Rapoport, J.L., Buchsbaum, M.S., Weingartner, H., Zahn, T.P., Ludlow, C. & Mikkelsen, E.J. Dextroamphetamine. Its cognitive and behavioural effects in normal and hyperactive boys and normal men. Archives of General Psychiatry, 1980, 37: 933–943. Rapoport, J.L., Mikkelsen, E.J., Ebert, M.H., Brown, G.L., Weise, V.L. & Kopin, I.J. Urinary catecholamine and amphetamine excretion in hyperactive and normal boys. Journal of Nervous and Mental Disorders, 1978, 66: 731–737. Rapport, M.D., Tucker, S.B., DuPaul, G.J., Merlo, M. & Stoner, G. Hyperactivity and frustration: the influence of control over size of rewards in delay of gratification. Journal of Abnormal Child Psychology, 1986, 14: 191–204. Raskin, L.A., Shaywitz, S.E., Shaywitz, B.A., Anderson, G.M. & Cohen, D.J. Neurochemical correlates of attention deficit disorder. Pediatric Clinics of North America, 1984, 31: 387–396. Riccio, C.A., Hynd, G.W., Cohen, M.J. & Gonzalez, J.J. Neurological basis of attention deficit hyperactivity disorder. Exceptional Children, 1993, 60: 118–124. Robaey, P., Breton, F., Dugas, M. & Renault, B. An event-related potential study of controlled and automatic processes in 6–8-year-old boys with attention deficit hyperactivity disorder. Electroencephalography and Clinical Neurophysiology, 1992, 82: 330–340. Robaey, P., Cansino, S., Dugas, M. & Renault, B. A comparative study of ERP correlates of psychometric and Piagetian intelligence measures in normal and hyperactive children. Electroencephalography and Clinical Neurophysiology, 1995, 96: 56–75. Robins, P.M. A comparison of behavioural and attentional functioning in children diagnosed as hyperactive or learning-disabled. Journal of Abnormal Child Psychology, 1992, 20: 65–82. Rosenbaum, M. & Baker, E. Self control behaviour in hyperactive and non-hyperactive children. Journal of Abnormal Child Psychology, 1984, 12: 303–318. Rosenthal, R.H. & Allen, T.W. An examination of attention, arousal, and learning dysfunctions of hyperkinetic children. Psychological Bulletin, 1978, 85: 689–715. Rosenthal, R.H. & Allen, T.W. Intratask distractibility in hyperkinetic and nonhyperkinetic children. Journal of Abnormal Child Psychology, 1980, 8: 175–187. Rosvold, H.E., Mirsky, A.F., Sarason, I., Bransome, E.D. & Beck, L.H. A continuous performance test of brain damage. Journal of Consulting Psychology, 1956, 20: 343–350. Rowe, D.C., Stever, C., Giedinghagen, L.N., Gard, J.M., Cleveland, H.H., Terris, S.T., Mohr, J.H., Sherman, S., Abramowitz, A. & Waldman, I.D. Dopamine DRD4 receptor polymorphism and attention deficit hyperactivity disorder. Molecular Psychiatry, 1998, 3: 419–426. Rubia, K., Overmeyer, S., Taylor, E., Bulimore, E., Brammer, M., Williams, S., Simmons, A. & Andrew, C. Right fronto-striatal abnormalities in hyperactive adolescents during motor response inhibition in functional magnetic resonance imaging (fMRI).

192

maree farrow et al.

Sanders, A.F. Toward a model of stress and human performance. Acta Psychologica, 1983, 53: 61–97. Satterfield, J.H., Schell, A.M. & Nicholas, T. Preferential neural processing of attended stimuli in attention-deficit hyperactivity disorder and normal boys. Psychophysiology, 1994, 31: 1–10. Satterfield, J.H., Schell, A.M., Nicholas, T. & Backs, R.W. Topographic study of auditory eventrelated potentials in normal boys and boys with attention deficit disorder with hyperactivity. Psychophysiology, 1988, 25: 591–606. Satterfield, J.H., Schell, A.M., Nicholas, T.W., Satterfield, B.T., Freese, T.E. Ontogeny of selective attention effects on event-related potentials in attention-deficit hyperactivity disorder and normal boys. Biological Psychiatry, 1990, 28: 879–903. Schachar, R. Childhood hyperactivity. Journal of Child Psychology and Psychiatry, 1991, 32: 155–191. Schachar, R. & Logan, G.D. Are hyperactive children deficient in attentional capacity? Journal of Abnormal Child Psychology, 1990a, 18: 493–513. Schachar, R. & Logan, G.D. Impulsivity and inhibitory control in normal development and childhood psychopathology. Developmental Psychology, 1990b, 26: 710–720. Schachar, R., Logan, G., Wachsmuth, R. & Chajczyk, D. Attaining and maintaining preparation: a comparison of attention in hyperactive, normal, and disturbed control children. Journal of Abnormal Child Psychology, 1988, 16: 361–378. Schachar, R., Tannock, R., Marriott, M. & Logan, D. Deficient inhibitory control in attention deficit hyperactivity disorder. Journal of Abnormal Child Psychology, 1995, 23: 411–437. Schechter, M.D. & Timmons, G.D. Objectively measured hyperactivity—I. Comparison with normal controls. Journal of Clinical Pharmacology, 1985, 25: 269–275. Schweitzer, J.B., Faber, T.L., Grafton, S.T., Tune, L.E., Hoffman, J.M. & Kilts, C.D. Alterations in the functional anatomy of working memory in adult attention deficit hyperactivity disorder. American Journal of Psychiatry, 2000, 157: 278–280. Seidel, W.T. & Joschko, M. Evidence of difficulties in sustained attention in children with ADDH. Journal of Abnormal Child Psychology, 1990, 18: 217–229. Seidman, L.J., Benedict, K.B., Biederman, J., Bernstein, J.H., Seiverd, K., Milberger, S., Norman, D., Mick, E. & Faraone, S.V. Performance of children with ADHD on the Rey-Osterrieth complex figure: a pilot neuropsychological study. Journal of Child Psychology and Psychiatry, 1995a, 36: 1459–1473. Seidman, L.J., Biederman, J., Faraone, S.V., Milberger, S., Norman, D., Seiverd, K., Benedict, K., Guite, J., Mick, E. & Kiely, K. Effects of family history and comorbidity on the neuropsychological performance of children with ADHD: preliminary findings. Journal of the American Academy of Child and Adolescent Psychiatry, 1995b, 34: 1015–1024. Semrud-Clikeman, M., Filipek, P.A., Biederman J., Steingard, R., Kennedy, D., Renshaw, P. & Bekken, K. Attention-deficit hyperactivity disorder: magnetic resonance imaging morphometric analysis of the corpus callosum. Journal of the American Academy of Child and Adolescent Psychiatry, 1994, 33: 875–871. Semrud-Clikeman, M., Steingard, R.J., Filipek, P., Biederman J., Bekken, K. & Renshaw, P.F. Using MRI to examine brain-behavior relationships in males with attention deficit disorder with hyperactivity. Journal of the American Academy of Child and Adolescent Psychiatry, 2000, 39: 477–484. Sergeant, J.A. & Scholten, C.A. A stages-of-information approach to hyperactivity. Journal of Child Psychology and Psychiatry, 1983, 24: 49–60. Sergeant, J.A. & Scholten, C.A. On data limitations in hyperactivity. Journal of Child Psychology and Psychiatry, 1985a, 26: 111–124. Sergeant, J.A. & Scholten, C.A. On resource strategy limitations in hyperactivity: cognitive impulsivity reconsidered. Journal of Child Psychology and Psychiatry, 1985b, 26: 97–109.

Neurobiology of ADHD

193

Sergeant, J.A. & Van der Meere, J.J. What happens when the hyperactive child commits an error? Psychiatry Research, 1988, 24: 157–164. Shapiro, S.K. & Garfinkel, B.D. The occurrence of behaviour disorders in children: the interdependence of attention deficit disorder and conduct disorder. Journal of the American Academy of Child Psychiatry, 1986, 25: 809–819. Shaywitz, B.A., Cohen, D.J. & Bowers, M.B. CSF amine metabolites in children with minimal brain dysfunction: evidence for alteration of brain dopamine—a preliminary report. Journal of Pediatrics, 1977, 90: 67–71. Shaywitz, B.A., Shaywitz, S.E., Byrne, T., Cohen, D.J. & Rothman, S. Attention deficit disorder: quantitative analysis of CT. Neurology, 1983, 33: 1500–1503. Shekim, W.O., Dekirmenjian, H. & Chapel, J.L. Urinary catecholamine metabolites in hyperactive boys treated with d-amphetamine. American Journal of Psychiatry, 1977, 134: 1276–1279. Shekim, W.O., Dekirmenjian, H. & Chapel, J.L. Urinary MHPG in minimal brain dysfunction and its modification by d-amphetamine. American Journal of Psychiatry, 1979, 136: 667–671. Shekim, W.O., Javaid, J., Dans, M. & Bylund, D.B. Urinary MHPG and HVA excretion in boys with attention deficit disorder and hyperactivity treated with d-amphetamine. Biological Psychiatry, 1983, 18: 707–714. Shekim, W.O., Sinclair, E., Glaser, R., Horwitz, E., Javaid, J. & Bylund, D.B. Norepinephrine and dopamine metabolites and educational variables in boys with attention deficit disorder and hyperactivity. Journal of Child Neurology, 1987, 2: 50–56. Shetty, T. Alpha rhythms in the hyperkinetic child. Nature, 1971, 234: 476–478. Shetty, T. & Chase, T.N. Central monoamines and hyperkinesis of childhood. Neurology, 1976, 26: 1000–1006. Shue, K.L. & Douglas, V.I. Attention deficit hyperactivity disorder and the frontal lobe syndrome. Brain and Cognition, 1992, 20: 104–124. Sieg, K.G., Gaffney, G.R., Preston, D.F. & Hellings, J.A. SPET brain imaging abnormalities in attention deficit hyperactivity disorder. Clinical Nuclear Medicine, 1995, 20: 55–60. Silberstein, R.B., Ciorciari, J. & Pipingas, A. Steady-state visually evoked potential topography during the Wisconsin card sorting test. Electroencephalography and Clinical Neurophysiology, 1995, 96: 24–35. Silberstein, R.B., Farrow, M., Levy, F., Pipingas, A., Hay, D.A. & Jarman, F.C. Functional brain electrical activity mapping in boys with attention-deficit/hyperactivity disorder. Archives of General Psychiatry, 1998, 55: 1105–1112. Singer, H.S., Reiss, A.L., Brown, J.E., Aylward, E.H., Shih, B., Chee, E., Harris, E.L., Reader, M.J., Chase, G.A., Bryan, R.N. & Denckla, M.B. Volumetric MRI changes in basal ganglia of children with Tourette’s syndrome. Neurology, 1993, 43: 950–956. Smalley, S.L., Bailey, J.N., Palmer, C.G., Cantwell, D.P., McGough, J.J., Del’Homme, M.A., Asarnow, J.R., Woodward, J.A., Ramsey, C. & Nelson, S.F. Evidence that the dopamine D4 receptor is a susceptibility gene in attention deficit hyperactivity disorder. Molecular Psychiatry, 1998, 3: 427–430. Solanto, M.V. Neuropsychopharmacological mechanisms of stimulant drug action in attention-deficit hyperactivity disorder: a review and integration. Behavioural Brain Research, 1998, 94: 127–152. Sonuga-Barke, E.J.S. Annotation: On dysfunction and function in psychological theories of childhood disorder. Journal of Child Psychology and Psychiatry, 1994, 35: 801–815. Sonuga-Barke, E.J.S., Houlberg, K. & Hall, M. When is impulsiveness not impulsiveness? The case of children’s cognitive style. Journal of Child Psychology and Psychiatry, 1994, 35: 1247–1253. Sonuga-Barke, E.J.S., Taylor, E. & Heptinstall, E. Hyperactivity and delay aversion—II. The effect of self versus externally imposed stimulus presentation periods on memory. Journal of Child Psychology and Psychiatry, 1992a, 33: 399–409.

194

maree farrow et al.

Sonuga-Barke, E.J.S., Taylor, E., Sembi, S. & Smith, J. Hyperactivity and delay aversion—I. The effect of delay on choice. Journal of Child Psychology and Psychiatry, 1992b, 33: 387–398. Sonuga-Barke, E.J.S., Williams, E., Hall, M. & Saxton, T. Hyperactivity and delay aversion—III. The effect on cognitive style of imposing delay after errors. Journal of Child Psychology and Psychiatry, 1996, 37: 189–194. Still, G.F. Some abnormal psychical conditions in children. Lancet, 1902, 1: 1008–1012, 1077–1082, 1163–1168. Strandberg, R.J., Marsh, J.T., Brown, W.S., Asarnow, R.F., Higa, J., Harper, R. & Guthrie, D. Continuous-processing-related event-related potentials in children with attention deficit hyperactivity disorder. Biological Psychiatry, 1996, 40: 964–980. Strauss, J., Lewis, J.L., Klorman, R., Peloquin, L.J., Perimutter, R.A. & Salzman, L.F. Effects of methylphenidate on young adults’ performance and event-related potentials in a vigilance and a paired-associates learning test. Psychophysiology, 1984, 21, 609–621. Sunohara, G., Barr, C., Jain, U., Schachar, R., Roberts, W., Tannock, R., Malone, M. & Kennedy, J. Association of the D4 receptor gene in individuals with ADHD: 1. A family-based control study of children with ADHD; 2. A case control study of adults with ADHD. The American Journal of Human Genetics, 1997, 61 (suppl.): A296. Sunohara, G.A., Voros, J.G., Malone, M.A. & Taylor, M.J. Effects of methylphenidate in children with attention deficit hyperactivity disorder: a comparison of event-related potentials between medication responders and non-responders. International Journal of Psychophysiology, 1997, 27: 9–14. Sunshine, J.L., Lewin, J.S., Wu, D.H., Miller, D.A., Findling, R.L., Manos, M.J. & Schwartz, M.A. Functional MR to localize sustained visual attention activation in patients with attention deficit hyperactivity disorder: a pilot study. American Journal of Neuroradiology, 1997, 18: 633–637. Swanson, J.M. & Castellanos, F.X. Biological bases of attention deficit hyperactivity disorder: neuroanatomy, genetics, and pathophysiology. Proceedings of the NIH Consensus Development Conference on Diagnosis and Treatment of Attention Deficit Hyperactivity Disorder. 1998 (November), Bethesda, USA, pp. 37–42. Swanson, J.M., Flodman, P., Kennedy, J., Spence, M.A., Moyzis, R., Schuck, S., Murias, M., Moriarity, J., Barr, C., Smith, M. & Posner, M. Dopamine genes and ADHD. Neuroscience and Biobehavioral Reviews, 2000, 24: 21–25. Swanson, J.M., McBurnett, K., Wigal, T., Pfiffner, L.J., Lerner, M.A., Williams, L., Christian, D.L., Tamm, L., Willcutt, E., Crowley, K., Clevenger, W., Khouzam, N., Woo, C., Crinella, F.M. & Fisher, T.D. Effect of stimulant medication on children with attention deficit disorder: a “review of reviews”. Exceptional Children, 1993, 60: 154–162. Swanson, J., Oosterlaan, J., Murias, M., Schuck, S., Flodman, P., Spence, M.A., Wasdell, M., Ding, Y., Chi, H.C., Smith, M., Mann, M., Carlson, C., Kennedy, J.L., Sergeant, J.A., Leung, P., Zhang, Y.P., Sadeh, A., Chen, C., Whalen, C.K., Babb, K.A., Moyzis, R. & Posner, M.I. Attention deficit/hyperactivity disorder children with a 7–repeat allele of the dopamine receptor D4 gene have extreme behavior but normal performance on critical neuropsychological tests of attention. Proceedings of the National Academy of Science USA, 2000, 97: 4754–4759. Swanson, J.M., Posner, M., Potkin, S., Bonforte, S., Youpa, D., Fiore, C., Cantwell, D. & Crinella, F. Activating tasks for the study of visual-spatial attention in ADHD children. A cognitive anatomic approach. Journal of Child Neurology, 1991, 6 (Suppl.): S119–S127. Swanson, J.M., Sergeant, J.A., Taylor, E., Sonuga-Barke, E.J.S., Jensen, P.S. & Cantwell, D.P. Attention-deficit hyperactivity disorder and hyperkinetic disorder. Lancet, 1998a, 351: 429–433. Swanson, J.M., Shea, C., McBurnett, K., Potkin, S.G., Fiore, C. & Crinella, F. Attention and Hyperactivity. In Enns, J.T. (Ed) The development of attention: Research and theory. Amsterdam: North-Holland. 1990, pp. 383–403.

Neurobiology of ADHD

195

Swanson, J.M., Sunohara, G.A., Kennedy, J.L., Regino, R., Fineberg, E., Wigal, T., Lerner, M., Williams, L., LaHoste, G.J. & Wigal, S. Association of the dopamine receptor D4 (DRD4) gene with a refined phenotype of attention deficit hyperactivity disorder (ADHD): a family-based approach. Molecular Psychiatry, 1998b, 3: 38–41. Sykes, D.H., Douglas, V.I. & Morgenstern, G. Sustained attention in hyperactive children. Journal of Child Psychology and Psychiatry, 1973, 14: 213–220. Sykes, D.H., Douglas, V.I., Weiss, G. & Minde, K.K. Attention in hyperactive children and the effect of Ritalin. Journal of Child Psychology and Psychiatry, 1971, 12: 129–139. Tannock, R. Attention deficit hyperactivity disorder: advances in cognitive, neurobiological, and genetic research. Journal of Child Psychology and Psychiatry, 1998, 39: 65–99. Tannock, R., Ickowicz, A. & Schachar, R. Differential effects of methylphenidate on working memory in ADHD children with and without comorbid anxiety. Journal of the American Academy of Child and Adolescent Psychiatry, 1995a, 34: 886–896. Tannock, R., Schachar, R. & Logan, G. Methylphenidate and cognitive flexibility: dissociated dose effects in hyperactive children. Journal of Abnormal Child Psychology, 1995b, 23: 235–266. Taylor, E. Medications and their effects. In M.D. Levine & P. Satz (Eds.). Middle childhood: Development and dysfunction. Baltimore: University Park Press. 1984, pp. ???? Taylor, M.J., Voros, J.G., Logan, W.J. & Malone, M.A. Changes in event-related potentials with stimulant medication in children with attention deficit hyperactivity disorder. Biological Psychology, 1993, 36: 139–156. Thapar, A., Holmes, J., Poulton, K. & Harrington, R. Genetic basis of attention deficit and hyperactivity. British Journal of Psychiatry, 1999, 174: 105–111. Trommer, B.L., Hoeppner, J.B., Lorber, R. & Armstrong, K. The go-no-go paradigm in attention deficit disorder. Annals of Neurology, 1988, 24: 610–614. Tucker, D.M. & Williamson, P.A. Asymmetric neural control systems in human self-regulation. Psychological Review, 1984, 91: 185–215. Ucles, P. & Lorente, S. Electrophysiologic measures of delayed maturation in attention-deficit hyperactivity disorder. Journal of Child Neurology, 1996, 11: 155–156. Vaidya, C.J., Austin, G., Kirkorian, G., Ridlehuber, H.W., Desmond, J.E., Glover, G.H. & Gabrielli, J.D.E. Selective effects of methylphenidate in attention deficit hyperactivity disorder: a functional magnetic resonance study. Proceedings of the National Academy of Science, 1998, 95: 14494– 14499. Vaidya, C.J. & Gabrielli, J.D.E. Searching for a neurobiological signature of attention deficit hyperactivity disorder. Molecular Psychiatry, 1999, 4: 206–208. Vance, A., Luk, S.L., Tonge, B.J. & Pantelis, C. A preliminary investigation of an ’anxious subtype’ of attention deficit hyperactivity disorder. Paper presented at the 11th Annual Conference of The Faculty of Child and Adolescent Psychiatry, The Royal Australian and New Zealand College of Psychiatrists. 1998 (October), Sydney, Australia. Van der Meere, J.J. The role of attention. In S.T. Sandberg (Ed.) Monographs in child and adolescent psychiatry. Hyperactivity disorders of childhood. Cambridge: Cambridge University Press. 1996, pp. 111–148. Van der Meere, J.J. & Sergeant, J.A. A divided attention experiment with pervasively hyperactive children. Journal of Abnormal Child Psychology, 1987, 15: 379–391. Van der Meere, J.J. & Sergeant, J.A. Acquisition of attention skill in pervasively hyperactive children. Journal of Child Psychology and Psychiatry, 1988a, 29: 301–310. Van der Meere, J.J. & Sergeant, J.A. Controlled processing and vigilance in hyperactivity: time will tell. Journal of Abnormal Child Psychology, 1988b, 16: 641–655. Van der Meere, J.J. & Sergeant, J.A. Focused attention in pervasively hyperactive children. Journal of Abnormal Child Psychology, 1988c, 16: 627–639.

196

maree farrow et al.

Van der Meere, J.J., Shalev, R., Borger, N. & Gross-Tsur, V. Sustained attention, activation and MPH in ADHD: a research note. Journal of Child Psychology and Psychiatry, 1995a, 36: 697– 703. Van der Meere, J.J., Stemerdink, N. & Gunning, B. Effects of presentation rate of stimuli on response inhibition in ADHD children with and without tics. Perceptual and Motor Skills, 1995b, 81: 259–262. Van der Meere, J., Van Baal, M. & Sergeant, J. The additive factor method: a differential diagnostic tool in hyperactivity and learning disability. Journal of Abnormal Child Psychology, 1989, 17: 409–422. Van der Meere, J.J., Vreeling, H.J. & Sergeant, J.A. A motor presetting experiment in hyperactive, learning disabled and control children. Journal of Child Psychology and Psychiatry, 1992, 33: 1347–1354. Van der Meere, J.J., Wekking, E. & Sergeant, J.A. Sustained attention and pervasive hyperactivity. Journal of Child Psychology and Psychiatry, 1991, 32: 275–284. Van Leeuwen, T.H., Steinhausen, H.-Ch., Overtoom, C.C.E., Pascual-Marqui, R.D., van’t Klooster, B., Rothenberger, A., Sergeant, J.A. & Brandeis, D. The continuous performance test revisited with neuroelectric mapping: impaired orienting in children with attention deficits. Behavioural Brain Research, 1998, 94: 97–110. Vendrell, P., Junque, C., Pujol, J., Jurado, M.A., Molet, J. & Grafman, J. The role of prefrontal regions in the Stroop task. Neuropsychologia, 1995, 33: 341–352. Verin, M., Partiot, A., Pillon, B., Malapani, C., Agid, Y. & Dubois, B. Delayed response tasks and prefrontal lesions in man–evidence for self generated patterns of behaviour with poor environmental modulation. Neuropsychologia, 1993, 31: 1379–1396. Waldman, I.D., Rowe, D.C., Abramowitz, A., Kozel, S.T., Mohr, J.H., Sherman, S.L., Cleveland, H.H., Sanders, M.L., Gard, J.M. & Stever, C. Association and linkage of the dopamine transporter gene and attention-deficit hyperactivity disorder in children: heterogeneity owing to diagnostic subtype and severity. American Journal of Human Genetics, 1998, 63: 1767–1776. Weizman, R., Weizman, A. & Deutsch, S.I. Biological studies of attention-deficit disorder. In Deutsch, S.I., Weizman, A. & Weizman, R. (Eds.). Application of basic neuroscience to child psychiatry. New York: Plenum Publishing Corporation. 1990, pp. 231–239. Werry, J.S., Elkind, G.S. & Reeves, J.C. Attention deficit, conduct, oppositional, and anxiety disorders in children: III. Laboratory differences. Journal of Abnormal Child Psychology, 1987, 15: 409– 428. Whalen, C.K. Attention deficit and hyperactivity disorders. In T.H. Ollendick & M. Hersen (Eds.) Handbook of child psychopathology (2nd ed.). New York: Plenum Press. 1989, pp. 131–169. Winsberg, B.G., Javitt, D.C. & Silipo, G.S. Electrophysiological indices of information processing in methylphenidate responders. Biological Psychiatry, 1997, 42: 434–445. Wood, C., Maruff, P., Levy, F., Farrow, M. & Hay, D. Covert orienting of visual spatial attention in attention deficit hyperactivity disorder: does comorbidity make a difference? Archives of Clinical Neuropsychology, 1999, 14: 179–189. Wood, K. Attention deficit hyperactivity disorder: A guide for professionals. Melbourne: La Trobe University. 1995. World Health Organization. The ICD-9 classification of mental and behavioural disorders. Geneva: Author. 1978. World Health Organization. The ICD-10 classification of mental and behavioural disorders: clinical descriptions and diagnostic guidelines, 1992; diagnostic criteria for research, 1993. Geneva: Author. 1993. Yong-Liang, G., Robaey, P., Karayanidis, F., Bourassa, M., Pelletier, G. & Geoffroy, G. ERPs and behavioral inhibition in a Go/No-go task in children with attention-deficit hyperactivity disorder. Brain and Cognition, 2000, 43: 215–220.

Neurobiology of ADHD

197

Yordanova, J., Dumais-Hubar, C., Rothenberger, A. & Woerner, W. Frontocortical activity in children with comorbidity of tic disorder and attention-deficit hyperactivity disorder. Biological Psychiatry, 1997, 41: 585–594. Zahn, T.P., Kruesi, M.J. & Rapoport, J.L. Reaction time indices of attention deficits in boys with disruptive behaviour disorders. Journal of Abnormal Child Psychology, 1991, 19: 233–252. Zametkin, A.J. Attention-deficit disorder: born to be hyperactive? The Journal of the American Medical Association, 1995, 273: 1871–1874. Zametkin, A.J., Ernst, M. & Silver, R. Laboratory and diagnostic testing in child and adolescent psychiatry: a review of the past 10 years. Journal of the American Academy of Child and Adolescent Psychiatry, 1998, 37: 464–472. Zametkin, A.J., Liebenauer, L.L., Fitzgerald, G.A., King, A.C., Minkunas, D.V., Herscovitch, P., Yamada, E.M. & Cohen, R.M. Brain metabolism in teenagers with attention-deficit hyperactivity disorder. Archives of General Psychiatry, 1993, 50: 333–340. Zametkin, A.J., Nordahl, T.E., Gross, M., King, A.C., Semple, W.E., Rumsey, J., Hamburger, S. & Cohen, R.M. Cerebral glucose metabolism in adults with hyperactivity of childhood onset. New England Journal of Medicine, 1990, 323: 1361–1366. Zametkin, A.J. & Rapoport, J.L. The neuropharmacology of attention deficit disorder with hyperactivity: where have we come in 50 years? Journal of the American Academy of Child and Adolescent Psychiatry, 1987, 26: 676–686. Zentall, S.S. Effects of colour stimulation on performance and activity of hyperactive and nonhyperactive children. Journal of Educational Psychology, 1986, 78: 159–165. Zentall, S.S. & Zentall, T.R. Optimal stimulation: a model of disordered activity and performance in normal and deviant children. Psychological Bulletin, 1983, 94: 446–471.

8 Neurobiology of Savant Syndrome Robyn Young

OVERVIEW Examples of individuals who demonstrate extreme variations between abilities present a challenge to our understanding of brain functioning and in particular a unified view of intelligence. Such individuals often referred to as “savants” are the focus of this chapter. The question as to how brilliance can be achieved in a specific domain despite limited cognitive functioning in all other domains has occupied researchers for more than a century. Several explanations have been advanced which include inheritance, eidetic memory, attention, concrete thinking, sensory deprivation, compensation, reinforcement and intuition yet there remains no clear understanding as to the nature of the mental functioning involved. This chapter presents a detailed account of recent research in this field. Cognitive and neurological explanations of savants have been evaluated to determine processes that may underpin and sustain savant skills. The conclusion proposed is that the existence of savants is consistent with a theory that some skills are based on relatively well-differentiated neurological capacities that have been preserved despite damage to other areas of the brain. The nature of these capacities will also be discussed.

Robyn Young

•

School of Psychology, Flinders University, Adelaide, South Australia, Australia.

199

200

robyn young

SAVANT SYNDROME Savants occur in less than 1% of the intellectually disabled population (Hill, 1978) but their incidence is much higher within the autistic population (approximately 10%; Rimland, 1978a, 1978b) and hence the use of the term “autistic savant”. All savants, however, are not autistic and therefore the term “savant syndrome” is preferred by the present author. Despite the vast array of skills available to humans, those skills developed by savants are limited to a discrete range of abilities; musical precocity, arithmetical and calendrical calculations, verbal representations, highly developed sensory discriminations, artistic ability, mechanical dexterity, mathematical skills, and memory for facts. Savants vary with regard to their level of skill and the terms “prodigious savant” and “talented savant” (Treffert, 1989, 2000) provides a useful distinction. It is likely, however, that a prodigious savant represents an extreme instance within a continuum of skills rather than a discrete category. The term splinter skill is also used to describe a skill that is extraordinary only in comparison to one’s overall functioning. These skills are more common among the autistic population and typically involve memory for facts. Individuals demonstrating only splinter skills do not warrant “savant” classification.

EARLY INTEREST AND RECENT RESEARCH Early reports on savants were largely anecdotal, containing inconsistencies about the level of ability, age at which the skill became apparent, formal instruction, amount of practice, medical history and intellectual ability (e.g., Anastasi & Levee, 1960; Duckett, 1976; Hill, 1975; LaFontaine, 1974; Monty, 1981; O’Connell, 1974; Scheerer, Rothman & Goldstein, 1945; Southall, 1979, 1983; Viscott, 1970; Zsako & Urban, 1938). Recent research in the 1980s and 90s has been more scientific (e.g., Hermelin & O’Connor, 1986, 1990; O’Connor & Hermelin, 1984, 1987a, 1987b, 1989, 1990, 1991a, 1991b; Miller, 1989; Treffert, 1989, 2000, Young, 1995, Young & Nettelbeck, 1994, 1995). Despite many attempts to evaluate the implications that savant skills hold for our understanding of brain functioning (e.g., Goldsmith & Feldman, 1988; Hermelin & O’Connor, 1983; O’Connor, 1987a, 1987b, 1989; O’Connor & Hermelin, 1991a; Rimland & Hill, 1984; Rimland & Fein, 1988; Treffert, 1989, 2000) no consensus has been reached. Howe (1989a, 1989b) cites the existence of savants to support his view that intellectual abilities are multi-faceted. O’Connor and Hermelin (1988, 1990; 1991a; O’Connor, 1987a, 1987b), however, accept the notion of a general ability but suggest that savant abilities are supported by cognitive functions that are independent of general intelligence, even though general

Neurobiology of Savant Syndrome

201

intelligence may determine the manner in which the skills develop. Nettelbeck and Young (1996) suggest that specific cognitive processes can support these skills without transferring to other areas of functioning. It is therefore suggested by them that some skills reflect a modular nature of some cognitive functions, while general intelligence is important to more higher level processes such as thinking creatively and imagination. Kehrer (1992) has offered a general neurological explanation of savant skills, in terms of an abnormal ability to perceive and store information, together with a behaviourist one—the focussed reinforcement by others of the emerging skill under circumstances where the individual shows inadequacies in other abilities. As yet there has been no consistent physiological evidence to support either of these hypotheses. Other studies have focussed on the neurological underpinnings of particular savant skills and these will be addressed in turn.

Music Charness, Clifton and MacDonald (1988) noted that, despite divergent aetiology, there are common characteristics and common areas of cognitive functioning found among musical savants. For example, musical savants have all been reported to have absolute pitch which appears to be an important component underpinning the development of wider musical skills. Similarly, musical savants have also often had physical and/or language impairment with blindness being the most common impairment. More than half of the 18 savants documented by Judd (1988) were blind. This is consistent with speculation that development of musical skills may reflect, among other things, compensation for sensory deprivation (Hill, 1978). Many cases have also involved language disorders, consistent with left hemispheric damage, and this has resulted in speculation that musical competence reflects intact functions in the right hemisphere. Although the quality of savants’ musical performances may be mechanical and rigid, in general they are comparable to those shown by normally skilled musicians (e.g., Miller, 1987a, 1987b; Sloboda, Hermelin & O’Connor, 1985) (Charness, et al, 1988; Judd, 1988; Miller, 1987a; Sloboda et al., 1985). Young and Nettlebeck (1995) also showed that the errors demonstrated by a musical savant were virtually all structure preserving suggesting that his memory for music was well organised, structured and knowledge based, consistent with that of an expert pianist.

Calendar Calculation Calendrical calculation refers to the ability to state correctly the day of the week upon which any given date will or has fallen. The span of dates may vary

202

robyn young

between individuals, ranging from a year to several centuries and is prevalent in savant populations throughout the world including the United States, England, Japan and Australia. From the experiments described in Young and Nettelbeck (1994) and Young (1995) it is apparent that calendrical skills appear inflexible and not readily adapted, manipulated or modified. These abilities might therefore be consistent with what Jensen (1970) described as Level 1 processing (i.e., simple associative learning) requiring little cognitive manipulation of cognitive stimuli. These are qualitatively different from what Jensen (1970) might describe as Level ll processes which require more “cognitive ability”. Generalisations about this ability are difficult because the proficiency of the skill and the level of disability of the individual are often difficult to determine from the literature. It is also possible that the strategies employed by those whose skills appear to be more automatic may be different from those whose responses are more laborious. It may be that controlled processing (Shiffrin & Schneider, 1977) may be involved initially in the acquisition of calendrical calculation, but as savants develop their skills, through familiarity and practice, the processes underpinning these skills become more automatic.

Artistic Talent Although artistic dimensions such as creativity and abstraction have been recognised in the artwork of savants, savants are generally noted for their abilities in reproduction and perspective. Artistic ability can be divided into two kinds which may operate somewhat independently of each other; representational accuracy and artistic merit (Hermelin & O’Connor, 1990; O’Connor & Hermelin, 1990). Representational accuracy involves the ability to draw, paint or sculpt subjects in a way which fully presents the precise details, location and orientation of the subject being reproduced. Artistic merit, however, involves the presence of personal style and composition and therefore reflects abstract qualities. Hermelin and O’Connor’s conclusion is that savants typically demonstrate high levels of representational accuracy and that these may be IQ independent; but that savants lack many of the more creative qualities which define artistic merit and which may depend on general levels of intelligence. Young (1995) supports this claim finding the level of savant skill (including artistic merit) to be correlated with IQ. Other research has focussed on the reason for the development of savant artistic ability, offering explanations in terms of socialisation (White, 1988) and facilitation of development because of deficits in other areas (Stevens & Moffitt, 1988). White’s (1988) discussion of a female artistic savant focussed on the cognitive structures presumably utilised for the development of such skills, concluding that processes usually required by language and long term memory may be involved in the development of artistic skills. O’Connor and Hermelin (1987a, 1987b, 1990) and Rosenblatt and Winner (1988) have also concluded

Neurobiology of Savant Syndrome

203

that memory is involved in the development of the abilities required for artistic expression.

Verbal Skills: Representation without Comprehension This ability involves the recall or reproduction of material presented visually or aurally without understanding the content of the material. Examples of this involve the reproduction of words in any language, perfect spelling and pronunciation, the ability to translate languages easily and reading ability superior to one’s ability to comprehend; commonly known as hyperlexia. Hyperlexia, like other savant skills, is reported to emerge suddenly, with no formal instruction, frequently appearing between the age of 21/2 and 31/2 years. Hyperlexia is almost always associated with a developmental disorder where verbal skills are impaired and language is not communicative, with autism being the most common. Goldberg (1987) has discussed the development of hermetic reading (a term used specifically to describe an isolated word reading ability within a context of other language impairments). He has proposed that, despite deficits in procedural memory, other areas of memory may remain intact and may be influential in the development of the skill. One such function is declarative memory, which incorporates the storage and retrieval of distinct units of verbal and visual-spatial information (e.g., rote-memory, cognitive mapping or list-learning). Because the neuroanatomical substrate for declarative memory is allegedly distinct from other memory structures, this may be possible, although the theory is not yet substantiated.

Mathematical Ability Mathematical skills demonstrated by savants typically involve rapid calculation almost always involving multiplication. Steel, Gorman and Flexman (1984) suggest that the existence of such individuals supports the idea that isolated, highly complex skills can be developed and preserved in an otherwise damaged brain. Stevens and Moffitt (1988) support the idea of an isolated ability, hypothesising that an intact verbal memory function may have been responsible for the mathematical skills demonstrated by their mathematical savant with Asperger’s syndrome.

Mechanical Ability These skills involve preoccupation with mechanical objects and generally include repairing and modifying mechanical and electrical equipment and/or knowledge of technical concepts in mechanics and electronics. Brink (1980) provided a detailed account of a man who, despite receiving brain-damage when shot in the left temple when he was 9 years-old, demonstrated “outstanding” mechanical

204

robyn young

abilities. He was able to modify and manipulate multi-gear bicycles, design a punching bag and had excellent carpentry skills. What is clear from this account is the development of the skills involved motivation, intense practice and reinforcement from family members and did not involve a developmental disability as is typically the case.

Memory The most frequently observed skill in this category is the memorisation of relatively trivial facts such as post-codes, telephone numbers, capital cities, street names and dates. Recent research suggests that certain aspects of memory may remain intact, despite deficits in general intelligence (e.g., Duckett, 1976; Steel et al., 1984; Stevens & Moffitt, 1988; Dorman, 1991; White, 1988). The involvement of preserved memory has often been suspected as essential to the development of other savant skills and will be discussed further in this chapter.

Multiple Skills It is often assumed that most savants develop only one skill. However, results from previous research (Duckett, 1976; Rimland, 1978a, Young, 1995) suggest that the development of more than one skill is widespread. It has been suggested that certain skills often accompany each other to form common groupings, with the combination of musical and memory skills being the most common, followed by mathematical and musical skills (Rimland, 1978b). If skills are found to be consistently grouped, it may be that they require common neurological structures but that these structures are insufficient to support other areas of intellectual functioning. Alternatively it may be that incidents of multiple skills demonstrate that more than one discrete module (highly specific forms of knowledge, Anderson, 1992) may have been preserved. This would not, however, explain why particular module groupings were consistently preserved.

REPRESENTATION OF ABILITIES At this point we need to consider what we know about savant skills so we can hypothesise about what we don’t know. First, we know that these skills develop in neurologically impaired individuals with idiosyncratic and divergent profiles of intellectual ability. We also know that these skills are neither unique to this population, nor do they occur randomly within it. That is, the same skills are universally represented among savants and one might therefore suggest that the processes that underpin such skills are also universal (Nettelbeck, 1999). What remains to be identified then is the nature of the neurological impairment together with the

Neurobiology of Savant Syndrome

205

preserved neurological capacity(ies) that enable a savant to process information in a manner relevant to their skill. We will now discuss possible neurological bases for these skills.

POSSIBLE NEUROBIOLOGICAL BASIS FOR SAVANT SYNDROME Although the mechanisms responsible for savant syndrome remain unknown, deficits in structure and function of a number of brain regions have been suggested as possible factors contributing to the reported mixture of cognitive deficits and striking but isolated skills. There has been little in the way of systematic investigation of pathology in savants. However, there are a small number of studies reporting results of investigations from single cases. Here we will briefly review what has been reported.

HEMISPHERIC LOCALISATION OF DAMAGE IN SAVANTS A number of studies have reported abnormalities in the structure of the brain. These abnormalities are usually confined to, or predominate in, the left hemisphere. This is interesting given that savants typically show impairment of skills thought to be predominantly related to left hemisphere function (eg language function, conceptual or abstract analysis) with preservation, or even exceptional development, of skills and abilities that are predominantly associated with right hemisphere function (eg spatial skills, artistic abilities, mechanical performance) (Rimland 1978a). We will now briefly review the studies reporting abnormalities in brain structure of savants. Hauser and colleagues (1975) employed pneumographic techniques to investigate structural abnormalities in autistic subjects. While not all the subjects investigated were savants, 4 out of 17 participants investigated exhibited some signs of savant abilities. Fifteen cases demonstrated enlargement of the left ventricle, in particular, enlargement of the left temporal horn. These abnormalities represented mild and variable atrophy in the left cerebral hemisphere. Treffert (2000) reports the results of a CAT scan on one prodigiously talented savant. This scan demonstrated a considerable left sided abnormality, especially in the left frontal lobe, which was considered a large area of scarring. There was also damage to the anterior and posterior portions of the left parietal lobes and to a minor degree damage to the left and right occipital lobes. Charness and colleagues (1988) report abnormalities in a blind and retarded savant with exceptional musical abilities, with a CAT scan demonstrating clear left sided abnormalities. These limited reports suggest that often there may be evidence of left hemispheric abnormalities. However, not all savants have demonstrable left sided

206

robyn young

lesions with Steel and colleagues (1984) reporting that in a savant with mathematical skills the CAT scan was normal. However, this subject had an IQ of 91, and so may have been atypical. Therefore, the extremely limited pathophysiological data suggests that some, if not all, savants may have some pathology in the left hemisphere.

DEVELOPMENTAL FACTORS CONTRIBUTING TO LEFT HEMISPHERE DYSFUNCTION One important factor in the development of early left hemisphere damage may be a change in the chemical influences in the intrauterine environment. There are clear sex differences in the prevalence of savant syndrome with male savants being 6 times more common than female savants (Hill, 1978; Treffert, 1989, 2000). It is possible that this male preponderance may be related to developmental factors during the neonatal period. A theory has been proposed by Geschwind and Galaburda (1985) to provide a basis for lateralisation of the brain. This theory has several important implications for development of the pathophysiological changes seen in savants. Traditionally it was thought that genetic factors were largely responsible for the lateralisation of the brain. However, this theory proposes that other intrauterine influences may be at least as important in determining brain development. One of the proposed influences is the level of sex hormones, specifically testosterone, to which the developing brain is exposed. In the intrauterine environment both males and females are exposed to testosterone. However, when the male testes develop the levels of testosterone reach very high levels. The left hemisphere develops more slowly than the right hemisphere during the intrauterine period (Chi, Dooling & Giles, 1977; Geschwind & Galaburda, 1985), resulting in the left hemisphere being more sensitive to developmental influences for a longer period of time than the right hemisphere. Testosterone is known to affect growth and development of various regions of the brain. Therefore, high levels of testosterone (or a period of increased sensitivity to testosterone), which are more likely in males, may preferentially affect development of the left hemisphere. This may further enhance the differential development of the two hemispheres. Delays in the development of regions of the left hemisphere may produce compensatory overdevelopment of regions of the right hemisphere. A factor that may contribute to the over development of the right hemisphere is the finding of an over abundance of neurons in the cerebral hemispheres during early development. Just before birth there is a dramatic reduction in the number of neurons, with an elimination of approximately 40% (Nowakowski, 1987), and this process of elimination can continue in early infancy. However, the survival of neurons is critically dependent upon interaction between the neuron and its target (Hamburger & Levi-Montalcini, 1949). Damage

Neurobiology of Savant Syndrome

207

to any region of the brain during these critical development periods can result in “overdevelopment” of the corresponding region in the other hemisphere, possibly due to a reduction in competition for specific targets. If we consider the motor system, the above reasoning coupled with delayed left hemisphere development (and, therefore, greater risk of left hemisphere damage) in males may well explain why left handedness is more common in males than females and also more commonly reported among savants (Young, 1995). This type of “reorganisation” may also lead to development of right hemisphere abilities. Several studies have provided some evidence to support this hypothesis. For example it has been reported that spatial talents are better developed in male subjects (Buffery & Gray, 1972). Delay in left hemisphere development may lead to a relative deficit in left hemisphere dominant abilities (eg language skills, abstract thought). Therefore this theory provides an explanation for some typical features of savant syndrome such as preservation or enhancement of right hemisphere abilities, impairment of left hemisphere attributes, male preponderance, and increased incidence of left handedness.

DAMAGE TO LEFT HEMISPHERE LATER IN LIFE Damage to the left hemisphere at any stage in life may result in the development of savant like abilities, or skills that underpin savant abilities. Evidence to support this conclusion has been presented recently by Miller and colleagues (Miller et al., 1998) who reported on a group of elderly subjects with dementia. This group reported on 5 patients who presented with frontotemporal dementia (FTD). Four of the 5 exhibited the temporal lobe variant in which there is atrophy of the anterior temporal and basal frontal lobes with dorsolateral frontal areas remaining intact. The remarkable feature of these patients was that they developed new artistic skills in the presence of devastating cognitive decline. The pathology in these subjects was bilateral but with a disproportionate dominant hemisphere involvement. As far as we know this is the only type of dementia in which cognitive decline is reported to be associated with development of anything like savant abilities. How degeneration in the left frontotemporal cortex results in development of exceptional skills remains unclear, but right hemisphere “compensatory” development may have a role to play. Even late in life there appears to be a capacity for the brain to open up existing but “silent” connections. This may be the basis of the recovery of function seen following damage such as stroke. Damage to one hemisphere may act as the trigger for “unmasking” of the silent connections in the undamaged corresponding hemisphere. As suggested by Treffert (1989, 2000), it is likely that overdevelopment of right hemisphere capabilities and concurrent dysfunction of the left hemisphere

208

robyn young

is not sufficient in itself to explain the skills and abilities of prodigious savants. A further, crucial, requirement may be dysfunction of attentional and/or memory systems.

ATTENTION AND MEMORY Savants are capable of intense focussing of their attention on discrete and repetitive tasks and they seem not to suffer from the distractions that the normal population experience when performing stereotyped and repetitive tasks. Savants demonstrate little novelty-seeking behaviour and the resultant over-focussing on a limited array of behaviours may be crucial for the development of savant skills. It has been suggested (Casey, Gordon, Mannheim, Rumsey, 1993) that regions of the brain, such as the thalamus and midbrain, which are implicated in processes such as engaging and shifting of attention, may be dysfunctional. As far as we know there is no pathophysiological data available from savants to support this theory of dysfunction of attentional mechanisms The involvement of prodigious memory has been associated with the development of skills such as music (Hermelin et al., 1989, Young & Nettelbeck, 1995), mathematics (Stevens & Moffitt, 1988), artistic ability (O’Connor & Hermelin, 1987a), calendrical calculation (Hermelin & O’Connor, 1986; Young & Nettelbeck, 1994), and hyperlexia (Goldberg, 1987). Types of memory suggested as being well-developed among savants and therefore involved in their achievements have included: eidetic imagery (e.g., Brink, 1980, LaFontaine, 1974; Roberts, 1945), semantically organised memory schemata (Pring & Hermelin, 1993), retention of information stored in the long-term (Rosen, 1981), auditory verbal memory (Stevens & Moffitt, 1988), and detailed memory for objects or events perceivedprobably visually (Howe & Smith, 1988). Furthermore, many savants have also demonstrated remarkable memories, in addition or as a complement to their other prodigious skill. For example, musical savants have been shown to demonstrate remarkable memories for musical information, or an ability to recite pages of music history and repeat composers’ catalogues of work (Sacks, 1985). The musical savant presented by Viscott (1970) had also memorised pages of the phone book. Despite such claims, the role of memory is often discounted in the development of savant skills because savants typically perform poorly on standardised tests of memory (Howe, 1989a). These poor performances, however, might be expected given the general nature of memory tests compared with the specific nature of savant memory. Because the memory of savants is not typically extended to other domains and no tests of general memory depend on the recall of information such as that involving calendars, music or post codes, it is unlikely that savants will score well on standardised tests of memory. The importance of memory in the development of savant skills should not, however, be underestimated due to poor

Neurobiology of Savant Syndrome

209

performances on such tests. Instead, tests are needed which address the specific processes of memory preserved among such individuals. For example, Valentine and Wilding (1994) reported that two of their savant participants scored on the second and fourth percentile when recalling the details of a story—the type of recall often required in standardised tests of memory—but scored on the hundredth percentile for memory of telephone numbers. While savant memory may be preserved it is not, however, suggested here that it is “bizarre” or peculiar. The argument presented above might equally be extended to individuals demonstrating prodigious skills in the normal population. For example, Ericsson and Faivre (1988) found that chess masters had memorised large numbers of chess games and had acquired high-level concepts relating to the interaction of chess pieces. There was, however, no evidence to suggest that these experts had superior general memory systems, nor that this knowledge could be applied to tasks outside their field. These arguments support the belief that memory is not unitary but rather involves distinct functions, such as recall for visual and verbal information, storage, recognition and whether the information is available for short or long terms. Neurological evidence from investigations into epilepsy (Blaxton, 1992; Samson & Zatorre, 1991), amnesia (Glisky, 1995; Schmidtke, Handschu & Vollmer, 1996), Alzheimers (Deloche, Hannequin, Carlomago, Agniel et al, 1995) and other neurological diseases (Soliveri, Brown, Johanshahi & Marsden, 1992) also support the diversity of memory systems by providing evidence that separate memory capacities remain intact despite damage to other areas of the brain. As indicated above, it is likely that there are a number of memory systems, which can be differentially affected by pathological processes. We will now briefly review different memory systems and their relevance to savant skills.

Eidetic Imagery Eidetic imagery is the ability to retain an intense visual image for at least 40 seconds (e.g., Brink, 1980, LaFontaine, 1974; Roberts, 1945). It is prominent in young children and can become a persistent feature in some forms of chronic brain damage. Although this type of imagery has been demonstrated in a small number of savants (Roberts, 1945), it has not been demonstrated in the majority (Duckett 1976). Therefore it is unlikely that eidetic imagery plays a major role in savant abilities.

Short-term Memory There is some limited evidence of superior short-term memory in savants. Spitz and LaFontaine (1973) tested digit span in a group of 8 savants and a control group of mentally handicapped subjects. Digit span scores of the savants were

210

robyn young

superior to those of the intellectually disabled controls, and matched those of a group of intellectually normal controls. Several other studies have reported digit spans in savants that are within, or exceed, the normal range (Hill, 1975; Ho, Tsang & Ho, 1991; Young & Nettlebeck 1994; Young, 1995).

Declarative Memory Declarative or associative memory is the acquisition, retention and retrieval of information that can be intentionally recollected (Cohen & Squire, 1980). Savants do not have generally superior declarative memory but do show better mnemonic performance for tasks associated with their specific exceptional abilities. For example, savant calender calculators have been shown to have much better performance than control subjects when performing memory tasks involving date retrieval (Heavey, Pring & Hermelin, 1999). The authors suggested that this implied that these savants had access to an extensive knowledge base concerning the calender, and that a database of calender knowledge would facilitate the recall of calender year information. Declarative memory is dependent upon the hippocampal system, which includes a group of medial temporal lobe structures consisting of the hippocampus and underlying entorhinal, perirhinal, and parahippocampal cortices (VarghaKhadem et al., 1997). Modifications in the functioning of these brain regions have not been investigated in savants and would be a worthwhile line of enquiry.

Non-cognitive Memory Another type of memory system is non-cognitive or “habit” memory. This type of memory is more automatic than declarative memory and cognitive input appears to be unnecessary. Sub-cortical basal ganglia structures are implicated in this type of memory (Mishkin, Malamut & Bachevalier, 1984). Savants appear to have superior memory skills in certain, specific and narrow areas and it may be that development of this type of memory system is important in savant syndrome. As far as I am aware this hypothesis has never been tested and no evidence is available to support this theory.

Ancestral Memory Treffert (1989, 2000) has proposed that another type of memory, which he calls “ancestral memory”, is important in regard to Savant Syndrome. This is type of memory is composed of inherited traits and skills Treffert suggests that some of the skills exhibited by savants could not have developed by practise and experience alone. For example some musical savants have skills that seem to require extensive understanding of the rules of music. However, they have received no formal training

Neurobiology of Savant Syndrome

211

and are apparently incapable of reading and comprehending instructional books. Therefore it has been suggested that some form of inherited knowledge must be responsible. The structures and mechanisms involved in this type of memory are unknown.

DO SAVANT ABILITIES RESIDE IN ALL OF US? Recently, an intriguing theory has been proposed suggesting that savant skills may be inherent in all of us, and given the right conditions may become apparent (Snyder & Mitchell, 1999). These authors suggest that the types of skills exhibited by savants are in all of us but are not accessible due to interference from higher cognitive functioning. However, with a lesion or temporary disruption of function in some, critical, cortical region it may be possible to “unmask” these skills. The work of Miller reported above lends some support to this theory. Snyder and Mitchell proposed that it might be possible to test this hypothesis by temporarily disrupting function of the left hemisphere. We have recently investigated this hypothesis by employing repetitive transcranial magnetic stimulation (rTMS) to disrupt function of the frontotemporal cortex in normal subjects (Young et al., in submission). The fronto-temporal cortex was selected as a target for stimulation (and resultant disruption of function), as Miller’s work suggested it as a region whose dysfunction was crucial in the development of savant like abilities. We investigated 17 subjects during three conditions; no stimulation (control), left motor cortex stimulation (stimulation control), and left fronto-temporal cortex stimulation (test condition). We were able to demonstrate evidence of disruption to fronto-temporal cortical function in 5 of these subjects (evidenced by deficits in delayed memory tasks). In these 5 subjects there was striking improvement in skills associated with savant abilities such as artistic representation, calendrical calculation and mathematical abilities. None of the subjects demonstrated skill levels approaching those seen in prodigious savants. However, it must be remembered that subjects were only stimulated for a period of a few minutes, and development of the talents seen in savants may take years. Again, the potential importance of attentional and practise effects is raised. It would be intriguing to ask subjects to practise specific tasks during repeated or prolonged periods of disruption to fronto-temporal cortex function. Although this work is preliminary it does provide some evidence in support of Snyder and Mitchell’s hypothesis that the skills underpinning savant abilities maybe within in us all. In conclusion, there is very little pathophysiological data to help provide an explanation for the neurobiological basis of savant syndrome. The limited data suggest damage to regions of the left cerebral hemisphere, and particularly the fronto-temporal cortex, may be important. Compensatory “overdevelopment” of right hemisphere skills may be central to the development of savant syndrome.

212

robyn young

However, it is not possible to rule out that the development of the extraordinary abilities seen in savants may also require damage or dysfunction to other additional brain regions. Using modern functional imaging techniques such as PET and fMRI it should be possible to investigate the biological basis of savant syndrome more fully. Such investigations would allow for a far more illuminating insight into this fascinating condition.

REFERENCES Anastasi, A., & Levee R. (1960). Intellectual deficit and musical talent: A case report. American Journal of Mental deficiency, 64, 695–703. Anderson, M. (1992). Intelligence and Development: A cognitive Theory. Cambridge. MA. Blaxton, T.A. (1992). Dissociations among memory measures in memory-impaired subjects: Evidence for a processing account of memory. Memory and Cognition., 20: 549–562. Brink, T. L. (1980). Idiot savant with unusual mechanical ability: An organic explanation. American Journal of Psychiatry, 137, 250–251. Buffery AWH, Gray JA. (1972). Sex differences in the development of spatial and linguistic skills. In Ounsted C,. Taylor DC (eds), Gender Differences: Their Ontogeny and Significance. New York, Churchil Livingstone, Inc. Casey BJ, Gordon CT, Mannheim GB, Rumsey JM. (1993). Dysfunctional attention in autistic savants. J Clin Exp Neuropsych 15: 933–946. Charness N, Clifton J, MacDonald L. (1988). A case study of a musical mono-savant: a cognitive psychological focus. In Olber LK and Fein DA (eds), The Exceptional Brain: Neuropsychology of Talent and Special Abilities. New York, Guilford Press. Chi E, Dooling X, Giles F. (1977). Left right asymmetries of the temporal speech areas of the human fetus. Arch. Neurol. 34: 346–348. Cohen NJ, & Squire LR. (1980). Preserved learning and retention of pattern-analysing skill in amnesia: dissociation of knowing how and knowing that. Science 210, 207–210. Deloche G, Hannequin D, Carlomagno S, Agniel A, Dordain M, Pasquier F, Pellat J, Denis P, Desi M, Beauchamp D (1995). Calculation and number processing in mild Alzheimer’s disease. J Clin Exp-Neuropsychol, 17, 634–639. Dorman, C. (1991). Exceptional calendar calculation ability after early left hemispherectomy. Brain Cognition, 15, 26–36. Duckett, J. (1976). Idiot savants; super-specialization in mutually retarded persons. Unpub. Doc. Dissertation. University of Texas, Austin, USA. Ericsson, K. A., & Faivre, I. (1988). What’s exceptional about exceptional abilities? In L. Obler & D.A. Fein (Eds.), The exceptional brain (pp. 436–473). New York: Guilford Press. Geschwind N, & Galaburda AM. (1985). Cerebral lateralisation I. Biological mechanisms, associations, and pathology. Arch. Neurol. 42, 428–459. Glisky, E. L (1995). Acquisition and transfer of word processing skill by an amnesic patient. Neuropsychological Rehabilitation, 5: 299–318. Goldberg, T.E. (1987). On hermetic reading abilities. Journal of Autism and Developmental Disorders, 17, 29–44. Goldsmith, L. T., & Feldman, D. H. (1988). Idiots savants—thinking about remembering: A response to White. New Ideas in Psychology, 6(1), 15–23. Hamburger V, & Levi-Montalcini R. (1949). Proliferation, differentiation and degeneration in the spinal ganglia of the chick embryo under normal and experimental conditions. J Exp Zool 111, 457–501.

Neurobiology of Savant Syndrome

213

Hauser SL, DeLong GR, Rosman NP. (1975). Pneumographic findings in the infantile autism syndrome. A correlation with temporal lobe disease. Brain 98, 667–688. Heavey L, Pring L, & Hermelin B. (1999). A date to remember: the nature of memory in savant calendrical calculators. Psychological Medicine 29, 145–160. Hermelin, B., & O’Connor, N. (1983). The idiot savant: Flawed genius or clever Hans? Psychological Medicine, 13, 479–481. Hermelin, B., & O’Connor, N. (1986). Idiot-savant calendrical calculators: Rules and regularities. Psychological Medicine, 16, 885–893. Hermelin, B., & O’Connor, N. (1990). Art and accuracy: The drawing ability of idiot-savants. Journal of Child Psychology and Psychiatry and Allied Disciplines, 31, 217–228. Hill, A. L. (1975). An investigation of calender calculating by an idiot-savant. American Journal of Psychiatry, 132, 557–560. Hill, A.L. (1978). Savants: Mentally retarded individuals with special skills. In N.R. Ellis (Ed.). International review of research in mental retardation, 9, 277–298. New York: Academic Press. Ho E.D.F, Tsang A.K.T, & Ho D.Y.F. (1991). An investigation of the calender calculation ability of a Chinese calender savant. Journal of Autism and Developmental Disorders 21, 315–327. Howe, M.J. (1989a). Fragments of genius: The strange feats of idiot savants. London: Routledge. Howe, M.J. (1989b). Separate skills or general intelligence: The autonomy of human abilities. British Journal of Educational Psychology, 59, 351–360. Jensen, A. (1970). A theory of primary and secondary familial retardation. In N.Ellis (Ed..) International Review of Research in Metal Retardation (pp. 52–105). Academic Press: New York, London. Judd, T. (1988). The varieties of musical talent. In L. Obler & D.A. Fein (Eds.), The Exceptional Brain (pp. 127–152). New York: Guilford Press. Kehrer, H. E. (1992). Savant capabilities of autistic persons. Acta Paedopsychiatrica International Journal of Child and Adolescent Psychiatry, 55, 151–155. LaFontaine, L. (1974). Divergent abilities in the idiot savant. Unpublished doctoral Dissertation, Boston University. Miller, L. K. (1987a). Determinants of melody span in a developmentally disabled musical savant. Psychology of Music, 15, 76–89. Miller, L. (1987b). Developmentally delayed musical savant’s sensitivity to tonal structure. American Journal of Mental Deficiency, 91, 467–471. Miller, L. (1989). Musical savants: Exceptional skill in the mentally retarded. Hillsdale, New Jersey: Lawrence Erlbaum. Miller BL, Cummings J, Mishkin F, Boone K, Prince F, Ponton M, Cotman C. (1998). Emergence of artistic talent in frontotemporal dementia. Neurology 51, 978–982. Mishkin M, Malamut B, Bachevalier J. Memories and habits: two neural systems. In Lynch G, McGaugh JL, Weinberger NM. (eds), Neurobiology of Learning and Memory. New York, Guilford Press, 1984. Monty, S. (1981). May’s Boy. Nashville, TN. Thomas Nelson. Nettelbeck, T (1999) Savant syndrome—rhyme without reason. In M.Anderson (Ed). The Development of Intelligence. Studies in Developmental Psychology, pp. 247–273. Hove, England: Psychology Press. Nowakowski SG. (1987). Basic concepts of CNS development. Child Dev. 58, 568–595. O’Connell, T. (1974). The musical life of an autistic boy. Journal of autism and Childhood Schizophrenia, 3, 223–229. O’Connor, N. (1987a). Cognitive psychology and mental handicap. Journal of Mental Deficiency Research, 31, 329–336. O’Connor, N. (1987b). Intelligence, handicaps and talents. Mental Retardation and Learning Disability Bulletin, 15, 41–56.

214

robyn young

O’Connor, N. (1989). The performance of the “idiot savant”: Implicit and explicit. British Journal of Disorders of Communication, 24, 1–20. O’Connor, N., & Hermelin, B. (1984). Idiot savant calendrical calculators: Maths or memory? Psychological Medicine, 14, 801–806. O’Connor, N., & Hermelin, B. (1987). Visual and graphic abilities of the idiot savant artist. Psychological Medicine, 17, 79–90. O’Connor, N., & Hermelin, B. (1988). Low intelligence and special abilities. Journal of Child Psychology and Psychiatry and Allied Disciplines, 29, 391–396. O’Connor, N., & Hermelin, B. (1989). The memory structure of autistic idiot savant mnemonists. British Journal of Psychology, 80, 97–111. O’Connor, N., & Hermelin, B. (1990). The recognition failure and graphic success of idiot savant artists. Journal of Child Psychology and Psychiatry and Allied Disciplines, 31, 203–215. O’Connor, N., & Hermelin, B. (1991a). A specific linguistic ability. American Journal on Mental Retardation, 95, 673–680. O’Connor, N., & Hermelin, B. (1991b). Talents and preoccupations in idiot-savants. Psychological Medicine, 21, 959–964. Pring, L., & Hermelin, B. (1993). Bottle, tulip and wineglass: Semantic and structural picture processing by savant artists. Journal of Child Psychology and Psychiatry and Allied Disciplines, 34, 1365– 1385. Rimland, B. (1978a). Inside the mind of the autistic savant. Psychology Today, 68–80. Rimland, B. (1978b). Savant capabilities of autistic children and their cognitive implications. In G. Serban (Ed.). Cognitive Defects in the Development of Mental illness (pp. 43–65). New York: Brunner/Mazel. Rimland, B., & Fein, D. (1988). Special talents of autistic savants. In L. Obler & D.A. Fein (Eds.), The Exceptional Brain (pp. 472–492). New York: Guilford Press. Rimland, B., & Hill, A.L. (1984). Idiot savants. In J. Wortes (Ed.), Mental retardation and developmental disabilities, 13, (pp. 155–169). New York: Plenum Press. Roberts, AD. (1945). Case history of a so-called idiot-savant. J. Gen. Psychology 66, 259–265. Rosen, A.M. (1981). Adult calendar calculators in a psychiatric OPD: A report of two cases and comparative analysis of abilities. Journal of Autism and Developmental Disorders, 11, 285–292. Rosenblatt, E., & Winner, E. (1988). Is superior Visual Memory a Component of superior drawing ability? In L. Obler, & D. Fein (Eds.), The Exceptional Brain (pp. 341–363). New York: Guilford. Sacks, O. (1985). The Twins. The New York Review of Books, 32, 16–20. Samson, S & Zatorre, R (1991). Recognition memory for text and melody of songs after unilateral temporal lobe lesion: Evidence for dual encoding. Journal of Experimental Psychology: Learning, Memory, and Cognition, 17, 793–804. Scheerer, M., Rothmann, E., & Goldstein, K. (1945). A case of “idiot savant:” an experimental study of personality orgaization. Psychology Monograph, 58, 1–63. Schmidtke, K., Handschu, R & Vollmer, H. (1996). Cognitive procedural learning in amnesia. Brain and Cognition, 32, 441–467. Shiffrin, R.M., & Schneider, W. (1977). Controlled and automatic human information processing: II. Perceptual learning, automatic attending and a general theory. Psychological Review, 84, 127–190. Sloboda, J. Hermelin, B., & O’Connor, N. (1985). An exceptional musical memory. Musical Perception, 3, 155–170. Snyder, A.W., & Mitchell, D.J. (1999). Is integer arithmetic fundamental to mental processing? The mind’s secret arithmetic. Proceedings of the Royal Society of London Series B, 266, 587–592. Soliveri, P; Brown, R. Jahanshahi, M; Marsden, C. (1992). Procedural memory and neurological disease. European Journal of Cognitive Psychology, 4. 161–193. Spitz H.H & La Fontaine L. (1973). The digit span of idiot savants. American Journal of Mental Deficiency 77, 757–759.

Neurobiology of Savant Syndrome

215

Steel JG, Gorman R, Flexman JE. (1984). Neuropsychiatric testing in an autistic mathematical idiotsavant: Evidence for nonverbal abstract capacity. Journal of the American Academy of Child Psychiatry, 23, 704–707. Stevens, D.E, & Moffitt, T. E. (1988). Neuropsychological profile of an Asperger’s syndrome case with exceptional calculating ability. Clinical Neuropsychologist, 2, 228–238. Treffert, D.A. (1989, 2000). Extraordinary people. London: Bantam Press. Valentine, E.R. & Wilding, J.M. (1994). Memory Expertise, The Psychologist. Sept, 405–408. Vargha-Khadem F, Gadian DG, Watkins KE, Connelly A, Van Paesschen W, Mishkin M. (1997). Differential effects of early hippocampal pathology on episodic and semantic memory. Science, 277, 376–380. Viscott, D. (1970). A musical idiot savant. Psychiatry, 33, 494–5151. White, P. (1988). The structured representation of information in long term memory: A possible explanation for the accomplishments of “idiot savants”. New Ideas in Psychology, 6, 3–14. Young, R. (1995). Savant syndrome: Processes underlying extraordinary abilities. Unpublished doctoral dissertation. University of Adelaide, South Australia. Young, R., & Nettelbeck, T. (1994). The “intelligence” of calendrical calculators. American Journal on Mental Retardation, 99, 186–200. Young, R., & Nettelbeck, T. (1995). The abilities of a musical savant and his family. Journal of Autism and Developmental Disorders, 25, 229–245.

Index ADHD, 109, 143–181 Alpha frequency, 9, 15, 56, 57, 58, 59, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 88, 89, 165 Amphetamine, 40, 45, 46, 55, 144, 145, 147, 149, 162 Amygdala, 112, 113, 114, 115, 118, 128 Anterior Cingulate, 62, 113, 114, 115, 118, 128, 135, 163 Antisociality, 43, 107–124 Artistic ability, 64 ASPD, 114, 115 Asperger’s Syndrome, 128, 129 Attachment, 114–117 Autism, 125–142 Arithmetic, 78, 82 Beta frequency, 9, 56, 58, 59, 76, 79, 81, 82, 165, 166 Boredom Susceptibility (BS), 32, 34 Brain stem, 115, 130, 134, 137 Brain Weight or Size, 125, 131, 133, 134, 135, 136, 138 Brain volume, 81, 128, 159 Cerebral blood flow, 62, 63, 65, 67, 98 Cocaine, 113 Coherence, 57, 58, 59 Compulsive disorders, 109 Computerised Tomography (CT), 5, 8, 128, 129, 134, 157 Cortisol, 41, 42 Convergent thinking, 59 Corpus Callosum, 129, 134, 157–159, 178, 180

Creativity, 53–72 Criminality, 107 Crystallised ability, 90 Cyngulate gyrus, 127 Delta frequency, 9, 58, 59, 69, 75, 76, 79, 81, 82, 88, 164 Dementia, 63, 76, 96, 207 Disinhibition, 32, 34, 35, 37, 41, 42, 45, 47, 55, 56 Divergent thinking, 54, 65 Dopamine, 43, 44, 46, 55, 60, 116, 144–150 Dopamine antagonist, 46, 55 Dopamine beta hydroxylase (DBH), 38, 44 Dopamine metabolite homovanillic acid (HVA), 38, 113, 146 Dopamine neurons, 39, 41, 112, 146, 148, 161 Dopamine Receptor D4 gene, 36, 150 Dopamine receptors, 36, 146, 150 Down Syndrome, 125–142 EEG Coherence, 11, 13, 67, 68, 86, 88 EEG Power, 13, 67, 69, 78, 80, 81, 82, 86 Electroencephalography (EEG), 9, 10, 11, 12, 13, 16, 24, 37, 53, 56, 57, 58, 59, 64, 65, 66, 67, 68, 69, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 86, 87, 89, 110, 164, 165, 166, 177–179 Empathy, 107, 108, 115 Event Related Desynchronization (ERD), 84, 85 Event Related Potentials (ERPs), 11, 12, 13, 14, 16, 23, 37, 38, 39, 40, 60, 74, 80, 89–93, 109, 169, 170

217

218

Index

Evoked Potentials: see Event Related Potentials (ERPs) Experience Seeking (ES), 32, 34, 45, 47 Extraversion, 43, 44 Eysenck Personality Questionnaire, 41, 45, 56, 69

IQ or Full Scale, IQ, 74–75, 77–79, 81, 82, 84, 85, 87, 89, 90–94, 96–97, 126, 129, 131, 133, 134, 136, 200, 204

Flexible thinking, 69 Fluency: see Verbal fluency Fluid abilities, 90 Frontal cortex, 60, 61, 62, 63, 65, 66, 67, 69, 78, 82, 83, 84, 85, 86, 94, 95, 96, 97, 98–101, 110, 111, 114, 115, 117, 127, 129, 130, 132, 135, 136, 144, 148, 153, 154, 155, 157, 158, 160, 161, 181 Frontal Gyrus, 62, 163 Functional Magnetic Resonance Imaging (fMRI), 21, 22, 23, 24, 62, 69, 74, 75, 96, 99, 100, 115, 117, 126, 130, 163

Magnetic Resonance Imaging (MRI), 5, 6, 7, 8, 14, 17, 18, 20, 23, 24, 53, 63, 75, 128, 129, 134, 150, 157, 158, 160, 210 Magnetoencephalography (MEG), 16, 17 Mathematical ability, 78 Memory, 80 Mental Retardation, 78, 109, 125–142 Mesolimbic Dopaminergic System, 46, 112, 179 Methylphenidate: see Amphetamine Monoamine Oxidase (MAO), 38, 41, 42, 43, 44, 119 Moral development, 108, 117, 118

Gamma frequency, 87 Genetics, 32–36, 43, 108, 118, 119, 127, 144, 145, 146, 149, 150 Giftedness, 83, 84, 87 Glucose Metabolic Rate (GMR), 127, 130, 132, 133, 135, 136, 137

NEO PI-R, 68, 69 Neurobiological techniques, 3–28 Neuroticism, 44 Nicotine, nicotinic receptors or tobacco, 15, 42, 45, 66 Noradrenaline, 113, 144, 145, 147, 148, 180 Norepinephrine, 43, 44, 114 Novelty Seeking, 32, 36, 43, 44, 45, 46, 113, 206 Nucleus Accumbens (NA), 40, 46, 112, 113, 115

Harm avoidance, 44, 108, 109, 115 Hemispheric asymmetry, 53 Hippocampus, 114, 115, 127, 128, 133, 135, 137, 138 Hyperactivity, 143, 146, 148, 159, 160, 165, 181 Hypothalamic-Pituitary-Adrenocortical (HYPAC), 39, 40, 41 Hypothalamus, 39, 40, 112, 127 Impulse Sensation Seeking, 36 Impulsivity, 34, 41, 45, 107, 108, 109, 110, 111, 112, 114, 115, 118, 119, 143, 144, 152, 159, 171, 173 Information Processing or speed of information processing, 15, 80, 151, 156, 165, 167, 168, 170, 174, 178 Inattention, 143, 151, 152 Inhibition, 53, 55, 145–156, 159, 160, 162, 163, 171, 173, 179 Inspection Time, 90, 91 Intelligence, 73–105, 54, 125, 136, 199–202, 204

Limbic system, 46, 111, 114, 115, 117, 118, 128, 130

Obsessive Compulsive Disorder (OCD), 109, 110, 111, 112 Occipital Cortex, 12, 22, 59, 61, 63, 67, 69, 75, 78, 81, 82, 83, 84, 94, 95, 96, 98, 99, 127, 129, 132 Openness to Experience, 53, 68 Originality, 54 Parietal Cortex, 59, 60, 62, 63, 67, 69, 78, 81, 82, 84, 94, 95, 96, 97, 98, 99, 101, 113, 127, 129, 154, 155, 177, 180, 181 PCL-R, 109, 113, 115 Performance IQ, 82, 83, 96 P300, 53, 60, 61, 109, 166–169, 176, 177 Pituitary, 40 Positron Emission Tomography (PET), 17, 18, 19, 20, 22, 47, 53, 61, 62, 66, 67, 68, 69, 74, 75, 96, 97, 99, 110, 111, 118, 126, 127, 130, 131, 132, 135, 155, 160, 161, 162, 163, 212

Index Power Spectral Analysis, 10, 11, 59 Prefrontal Cortex, 46, 60, 63, 64, 96, 99, 100, 109, 112, 113, 114, 115, 117, 118, 150, 153, 159, 160, 161, 163, 179, 205 Psychopathy, 107 Psychoticism: see Eysenck Personality Questionnaire Raven Progressive Matrices (APM or SPM), 78, 84, 92, 93, 94, 97, 98, 99, 100, 127 Reasoning, 78, 127 Reward Dependence or Circuitry, 44, 113 Savant Syndrome, 196–213 Sensation seeking, 31–52, 31, 37, 40, 45, 46, 68, 114 Sensation Seeking Scale (SSS), 32, 33, 34, 38, 41, 42, 44, 45, 47 Serotonin metabolite 5-Hydroxindoleacetic acid (5-HIAA), 38, 111 Serotonin, 43, 44, 46, 55, 56, 61, 66, 111, 118, 148 Serotonin receptors, 45, 112 Serotonin reuptake blockers, 111, 112 Short term memory, 66, 83, 137 Single Photon Emission Computerised Tomography (SPECT), 17, 20, 24, 63, 64, 110, 162–163

219 Sociopathy, 107 Steady-State Probe Topography (SSPT or SSVEP), 14, 15, 16, 74, 93–96, 100, 173–177, 179 Stroop colour naming, 55, 153 Temporal Cortex, 12, 22, 56, 60, 61, 62, 63, 64, 65, 67, 69, 82, 83, 84, 97, 99, 113, 118, 130, 132, 135, 161, 205, 209 Testosterone, 41, 204 Thalamus, 127, 132, 135 Theta frequency, 9, 58, 59, 69, 75, 79, 81, 82, 83, 86, 87, 88, 89 Thrill and Adventure Seeking (TAS), 32, 34 Transcrannial magnetic stimulation (rTMS), 211 Twins, 33, 35 Verbal fluency, 53, 54, 57, 61, 62, 65, 97 Verbal IQ, 77, 82, 96 Visual Cortex, 14 Wechsler Intelligence Scales, 62, 77, 79, 81, 82, 90, 91, 92, 94, 95, 96 Working memory, 60, 61, 64, 87, 95, 96, 99, 100, 109, 145, 161, 166