Serotonergic Neurocircuitry in Mood and Anxiety Disorders

Serotonergicneurocircuitry inmood and anxiety

disorders

Dan J SteinMRC Unit of Anxiety Disorders

University of StellenboschCape Town, South Africa

andDept of Psychiatry

University of FloridaGainesville, USA

LONDON AND NEW YORK

The views expressed in this publication are those of the authorand not necessarily those of Martin Duntiz Ltd.

© 2003Martin Dunitz Ltd, a member of the Taylor & Francisgroup

First published in the United Kingdom in 2003 by Martin DunitzLtd,

Taylor and Francis Group, 11 New Fetter Lane, London EC4P4EE.

Tel.: +44 (0)20 7583 9855Fax.: +44 (0) 20 7842 2298

E-mail: [email protected]: http://www.dunitz.co.uk

This edition published in the Taylor & Francis e-Library, 2005.“To purchase your own copy of this or any of Taylor & Francis or

Routledge’s collection of thousands of eBooks please go towww.eBookstore.tandf.co.uk.”

All rights reserved. No part of this publication may be reproduced,stored in a retrieval system, or transmitted, in any form or by any

means, electronic, mechanical, photocopying, recording, orotherwise, without the prior permission of the publisher or inaccordance with the provisions of the Copyright, Designs andPatents Act 1988 or under the terms of any licence permittinglimited copying issued by the Copyright Licensing Agency, 90

Tottenham Court Road, London W1P 0LP.Although every effort has been made to ensure that all owners ofcopyright material have been acknowledged in this publication, wewould be glad to acknowledge in subsequent reprints or editions

any omissions brought to our attention.This book is an abridged version of Cognitive-Affective

Neuroscience ofDepression and Anxiety Disorders (ISBN 1-84184-100-5) also

publishedby Martin Dunitz Ltd.

A CIP record for this book is available from the British Library.

ISBN 0-203-64073-X Master e-book ISBN

ISBN 0-203-67666-1 (Adobe eReader Format)ISBN 1-84184-303-2 (Print Edition)

Distributed in the USA by Fulfilment Center, Taylor & Francis,10650

Tobben Drive, Independence, KY 41051, USA

Toll Free Tel.: +1 800 634 7064E-mail: [email protected]

Distributed in Canada by Taylor & Francis, 74 Rolark Drive,Scarborough,

Ontario M1R 4G2, CanadaToll Free Tel.: +1 877 226 2237

E-mail: [email protected] in the rest of the world by Thomson Publishing

Services,Cheriton House, North Way, Andover, Hampshire SP10 5BE, UK

Tel.: +44 (0)1264 332424E-mail: [email protected]

iii

Acknowledgements

The support of the Medical Research Council of SouthAfrica is gratefully acknowledged. Permission of theUniversity of Stellenbosch to reprint the illustrations isappreciated, with particular thanks to Carol Lochner forher artistic collaboration. While a range of primary sourceswas used to develop the illustrations, particular gratitudeis owed to the University of Washington for their BrainAtlas, and to Salloway and colleagues for their volume onthe Neuropsychiatry of Limbic and Subcortical Disorders.Dr Andre Joubert played a valuable role by encouragingand advising on the volume.

Contents

1. Introduction 1

2. Major depression 4

3. Generalized anxiety disorder 15

4. Obsessive-compulsive disorder 24

5. Panic disorder 34

6. Post-traumatic stress disorder 43

7. Social phobia 52

8. Conclusion 60

9. References 62

Index 76

1.Introduction

In the past several decades there have been tremendousadvances in understanding the mood and anxietydisorders. The development of a rigorous classificationsystem fostered the study of a range of different conditionsand provided the basis for epidemiological studiesdemonstrating how prevalent, chronic, and disabling theseconditions are. Advances in neurochemistry,neuropsychology, and brain imaging have led to anunderstanding that each of these disorders has specificpsychobiological correlates. And the introduction ofspecific pharmacotherapeutic agents andpsychotherapeutic techniques has significantly improvedthe prognosis of patients with mood and anxiety disorders.

During this time, one of the critical developments in thefield of mood and anxiety disorders has been theintroduction of the selective serotonin reuptake inhibitors(SSRIs), and more recently of other agents that directly orindirectly act on the serotonergic system. The SSRIs areeffective not only for the treatment of depression, but alsofor the major anxiety disorders (generalized anxietydisorder, obsessive-compulsive disorder, panic disorder,post-traumatic stress disorder, and social phobia). Incomparison to the older tricyclic antidepressants andbenzodiazepines these agents are considerably safer andfar better tolerated; this made them a first-linepharmacotherapy for patients with mood and anxietydisorders.

The efficacy of the SSRIs and other serotonergicantidepressants in mood and anxiety disorders raisesmany questions for the researcher. What is the role of the

widely distributed serotonin sys tem (Figure 1.1) inmediating these conditions? Does serotonergic involvementdiffer across these various conditions, or is there aspectrum of dysfunction that underpins their frequentcomorbidity? How does pharmacotherapy with the SSRIsact on the neurocircuits that underpin mood and anxietysymptoms? These questions cut across research innosology, neurochemistry, neuroanatomy, andpharmacotherapy. An integrated understanding of theserotonin system and of response to serotonergicantidepressants in mood and anxiety disorders is animportant goal.

This volume intends to bring this goal of integration alittle closer. The data on serotonergic involvement indepression and each of the major anxiety disorders arereviewed, and for each condition a model of howserotonergic antidepressants are effective is developed. Theconcepts here fall in an area that might be termed‘cognitive-affective neuroscience’; they attempt tosynthesize data across a range of disciplines. Clinical worksimilarly spans a range of different levels (for example, frombiological to psychosocial), and indeed the volume aims tobe relevant to the practicing clinician. An integrated under-standing of the mood and anxiety disorders will hopefullycontribute to improved clinical management of theseimportant conditions.

2

Figure1.1Serotonergic neurons branch widely: neurons originatein the raphe nucleus andbranch extensively to the amygdala,hippocampus, hypothalamus, striatum,cingulate, prefrontal andfrontal cortex.

3

2.Major depression

Symptoms and assessment

Major depression (major depressive disorder) is one of themost prevalent of all psychiatric disorders (Kessler et al,1994), and one of the most disabling of all medicaldisorders (Murray and Lopez, 1996). It is a disorder seen inall age groups, both genders (although it is more commonin women), and throughout the world. The complication ofsuicide remains a crucial worry for clinicians, and animportant rationale for rigorous treatment of depression.

The symptoms of depression are both cognitive andsomatic (Table 2.1). Arguably at the core of depression isthe phenomenon of anhedonia—a decrease in the ability toexperience enjoyment and pleasure. Cognitive symptomsinclude negative thoughts about the self, world, and future(Beck, 1967). Somatic symptoms include increased ordecreased appetite (changes in feeding), increased ordecreased energy and interest in the environment (changesin foraging), irritability and hostility (changes in fighting),and decreased libido. Notable symptoms include changesin concentration and memory, and increase or decrease inpsychomotor activation.

Assessment of major depression episodes requiresevaluation of the severity and subtype of depression, ofcomorbid disorders, of suicidal ideation and disability, andof psychosocial stressors and supports. The differentialdiagnosis of major depressive disorder from bipolardepression is crucial; in this volume, we focus only on theformer disorder. In addition, general medical disorders and

substance use must be excluded as potential causes ofdepressive symptoms.

Regulation of mood/processing of reward

A broad range of cognitive-affective functions may berelevant to understanding depression, and may

Table 2.1Symptoms of major depression. Modified withpermission from the Diagnostic and Statistical Manual of MentalHealth Disorders, Fourth Edition, TextRevision. © 2000 AmericanPsychiatric Association.

5

correspondingly suggest an initial approach to theneurocircuitry of this disorder. Executive functions,including modulation of emotional behaviour, for example,would seem relevant to understanding cognitive features(e.g. depressive schemas) that characterize depression, andare mediated by frontal cortex. The construct of lowpositive affect, which may be relevant to depression, hasbeen associated with hypoactivation of left frontal cortex(Mineka et al, 1998).

Similarly, emotional memory systems are relevant todepression, suggesting amygdala and hippocampalinvolvement. Depressed patients show a bias towardsrecalling negative information, particularly when it is self-referential. This occurs during both explicit memory tasksand implicit tasks (when memory is tested indirectly)(MacLeod and Byrne, 1993). Psychomotor function may berelevant to understanding motoric symptoms of depression,and is mediated by striatal circuits. Feeding, foraging, andso forth, are related to the somatic symptoms ofdepression, and suggest involvement of the hypothalamusand hypothalamic-pituitaryadrenal (HPA) axis. Thesevarious circuits are themselves interlinked.

Much of our understanding of these cognitive-affectivephenomena and their neurocircuitry is based on animalstudies. There are of course limitations in attempting tostudy human mood regulation using animal models.Nevertheless, there is a broad preclinical literatureshowing that pathways linking limbic structures(amygdala, hippocampus, hypothalamus) to a network offrontal, paralimbic (ventral frontal, cingulate, insula,anterior temporal poles), striatal, and brainstem regions iscrucial in mediating affective and motivational behaviour(Damasio, 1996; MacLean, 1949; Rolls, 1990). Byextrapolation similar regions may be involved in humanmood regulation.

Functional imaging studies using humans may shedadditional light on the neurobiology of mood regulation.Several studies of transiently induced sadness in normalsubjects have confirmed the role of a broad network ofcortical-limbic circuits in the modulation of mood (Lane etal, 1997; Mayberg et al, 1999). Ultimately, further work is

6

needed to delineate the contribution of particular neuronalcircuits to normal regulation versus psychopathology, andalso to state versus trait characteristics (Davidson, 1994).

Neurocircuitry and depression

In approaching the neurocircuitry responsible formediating any psychiatric disorder, an important initialapproach, highlighted throughout this volume, is thecareful study of patients with symptoms secondary togeneral medical disorders. Seminal work by Robinson andcolleagues, for example, demonstrated that in patients withstroke, left-sided lesions were more likely to be associatedwith depression, while right-sided lesions were more likelyto be associated with mania. Notably, in depressedpatients with left-sided lesions, lesions closer to the frontalpole were associated with more severe depression(Robinson et al, 1984). Conversely, patients with secondarydepression have evidence of frontal dysfunction (Mayberg,1994), as do patients with late onset depression (in whichneurological insults may be relatively important) (MacFallet al, 2001).

A range of other findings from the literature ondepression secondary to general medical disorders isrelevant to developing a neuroanatomical model of thisdisorder (Byrum et al, 1999). In support of the role ofstriatal neurocircuitry in depression, classical observationsinclude the association between lesions in these circuits(e.g. Parkinson’s disease, vascular depression) anddepressed mood, and between depression andpsychomotor disturbance (Sobin and Sackheim, 1997).Furthermore, it is notable that patients with variousabnormalities of the hypothalamus and HPA axis maysuffer profound depression.

Another useful approach to the psychobiology ofdepression has been to focus on the sequelae of adverseevents, particularly of early adversity. Such work has itsroots in the seminal observations that primate separationultimately results in a ‘depressive’ picture (Bowlby, 1980).Recent work has documented that the sequelae of earlyadversity, in both animals (Sanchez et al, 2001) and

7

humans (Heim and Nemeroff, 2001), include alteredneuroendocrine function and (perhaps as a consequence ofthis) decreased hippocampal volume.

A range of other studies has further contributed todelineating the neurocircuitry of depression. Postmortemmorphometric studies, for example, have found evidence ofatrophy in specific cell regions (Duman et al, 2000).Demonstration of specific cognitive deficits in depressionhas provided clues about the neurocircuitry of depression(Austin et al, 2001). Neurosurgery is occasionally used inthe treatment of refractory depression, again sheddinglight on underlying neurocircuitry. Brain imaging has,however, proved crucial for advancing this area.

Structural imaging studies, for example, have proved keyin delineating the neurocircuitry of depression, identifyingabnormalities in prefrontal, limbic/paralimbic (cingulate,hippocampus), and striatal regions in depressed patients.A recent comprehensive review of functional imagingstudies suggested that depression is characterized bydecreased activity across this range of interconnectingneurocircuits (Videbach, 2000) (Figure 2.1). However, theseconclusions remain somewhat tentative; a number of

8

studies of depres sion have, for example, found not onlyhypoactivity (e.g. in neocortical areas) but also areas ofhyperactivity (e.g. in paralimbic regions) (Mayberg et al,1999).

There are relatively fewer imaging studies of psychoticmood dis-orders. A recent review suggested that whenoverlapping regions are involved, the dysfunctions inpsychotic mood disorders are more severe than those innonpsychotic depression (Wang and Ketter, 2000).Interestingly, a number of studies suggest that in maniathere is heightened activity in a cortical-subcortical neuralcircuit that includes the anterior cingulate and caudate(Blumberg et al, 2000).

Clearly, further work is needed to consolidate aneuroanatomical model of depression. Differences infindings across functional imaging studies may ultimatelybe explicable in terms of methodological variation or interms of subject heterogeneity. Additional research isneeded to delineate the neuroanatomy of different subtypesof depression, and to correlate functional neuroanatomywith clinical dimensions (Bench et al, 1993). More work isalso needed to replicate early findings that functional

Figure 2.1Functional neuroanatomy of depression: decreasedactivity in prefrontal,paralimbic (cingulate), and striatal regions.

9

findings predict pharmacotherapy response (Mayberg et al,1997).

A number of studies have reported that SSRIs and otheragents act to normalize functional imaging in depressedpatients, although exact findings are inconsistent(Kennedy et al, 2001). Exciting research also suggests thatthere is an overlap in the effects of pharmacotherapy andpsychotherapeutic intervention (Brody et al, 2001; Martinet al, 2001). Cognitive therapy may be mediated by a ‘top-down’ cortical influence on limbic pathways, whereasneurosurgical lesions (limbic leucotomy or subcaudatetractotomy) (Malhi and Bartlett, 2000) comprise a ‘bottom-up’ attack on the limbic system. In this view,pharmacotherapy can be understood as ‘mixed’, withprimary brainstem-limbic effects, but also secondarycortical effects (Mayberg et al, 1999).

Role of serotonin

A broad range of studies has demonstrated abnormalitiesof peripheral serotonergic markers in depression, with themost widely reported abnormality involving decreasedserotonin transporter (5-HTTP) binding (Owens andNemeroff, 1994). In addition, a consistent finding in bothpreclinical and clinical studies has been an associationbetween decreased serotonergic function and impulsivity(including aggression and suicide) (Stein et al, 1993).Importantly, functional imaging and postmortem studieshave confirmed decreased 5-HTTP binding in depression/suicide (Malison et al, 1998; Mann et al, 2000). 5-HTTPbinding, for example, was decreased in prefrontal cortex ofsubjects with a history of depression, with binding lower inventral prefrontal cortex in suicides (Mann et al, 2000).

Dynamic studies of the serotonin (5-HT) system inresponse to serotonergic agonists, or to serotonergicdepletion, offer methodological advantages over studies ofstatic measures; a range of data again point towardsaberrant serotonergic neurotransmission in depression(Charney, 1998). Depletion studies, for example, haveshown that decreased 5-HT synthesis precipitatessymptoms of depression in healthly and in remitted

10

depressed patients. In relapsed patients (but not inunaffected subjects), position emission tomography (PET)scanning shows a decrease in activity in dorsolateralprefrontal cortex, orbitofrontal cortex, and thalamus(Bremner et al, 1997). The link between hyposerotoninfunction, decreased frontal activity (hypofrontality), andmood dysfunction is again apparent.

A currently influential hypothesis is that SSRIs exerttheir effects by desensitization of 5-HT1A somato-dendriticautoreceptors (which initially serve as a ‘brake’ onserotonin neurotransmission). This accounts for therelatively slow time course of pharmacotherapy response indepression, with only gradual desensitization andincreased neurotransmission (as the ‘brake’ is effectivelyreleased) (Stahl, 1998). Ultimately there may be increasedserotonergic activity, and reversal of dysfunction(Figure 2.2). There is a growing literature on the molecularimaging of 5-HT receptor subtypes at baseline, and afterSSRI treatment, with some evidence, albeit inconsistent, infavour of such a hypothesis (Becker et al, 2001; Sargent etal, 2000; Staley et al, 1998).

Figure 2.2Effects of SSRIs on the functional neuroanatomy ofdepression: normalizationof activity in prefrontal, paralimbic(cingulate), and striatal regions.

11

There is also growing interest in the relationshipsbetween serotonin genotyping, functional imaging, andpharmacotherapy response. It might be hypothesized, forexample, that the s/s allele of the serotonin transporter (5-HTTP) gene is associated with greater impulsivity,hypofrontality, and worse response to treatment withSSRIs. Nevertheless, to date findings are inconsistent(Mann et al, 2000). Any single gene may only account for arelatively small amount of the variance when consideringthis kind of complex assocation. However, the mapping ofthe human genome, together with the increasingavailability of molecular imaging (Staley et al, 1998), laysthe ways for future advances in this area.

Evolutionary considerations

Evolutionary medicine attempts to supplement standardaccounts of the proximal mechanisms involved in diseasewith hypotheses about the evolutionary origins ofpathology (Nesse and Williams, 1994). Although aspects ofthis framework have long been in existence (Darwin,1965), in recent years there has been renewed theoreticalinterest as well as growing empirical work. The work ofNesse (2000) is perhaps the most sophisticatedevolutionarily based account of depression published todate. In his view, depression is evolutionarily advantageousin the face of a situation where continued effort to pursuea major goal will be likely to result in danger or loss. Suchsituations include, for example, a fight with a dominantfigure, or the failure of a major life enterprise. In thesecases, pessimism and lack of motivation are useful insofaras they inhibit actions which may be dangerous (e.g.attempting to do battle with a much stronger figure) orwasteful (e.g. attempting to start a new enterprise withoutadequate resources).

Such an approach may be useful in providing a distalaccount on which to base other kinds of psychobiologicalknowledge (Duchaine et al, 2001). In this context it isinteresting to note that preclinical and human studiesindicate that changes in serotonergic transmission mediatethresholds for adopting passive or waiting attitudes, or

12

accepting situations that necessitate or create stronginhibitory tendencies (Soubrie, 1986). More generally, ithas been argued that serotonin facilitates gross motoroutput and inhibits sensory information processing(Jacobs and Fornal, 1995).

In the future we can expect that ‘Darwinian psychiatry’(McGuire and Troisi, 1998; Stevens and Price, 1996)becomes supported by data, and is in turn increasinglyused as a rationale for particular kinds of intervention.Nevertheless, in the interim, it is important not only toacknowledge the speculative nature of such work, but alsoto emphasize that many subtypes of depression may not beexplicable in terms of a meaningful response toenvironmental stressors.

Certainly, as indicated earlier, depressive symptoms mayemerge as the direct consequence of a general medicalcondition; depression in the context of stroke, for example,may be more powerfully explained in terms of a brainlesion than in terms of loss. More commonly, individualpsychobiological variations may contribute an importantrisk factor for depression, which in turn may often need tobe conceptualized as a ‘dysfunction’ rather than as afunc tional response. Similarly, evolutionary models ofacute depressive episodes may not apply to chronicsymptoms; Kraepelin was the first to argue thatpsychosocial stressors play a greater role in the initial thanin subsequent episodes of depressive disorders, and a‘kindling’ hypothesis of recurrent depression has receivedsupport from epidemiological (Kendler et al, 2000),biological (Post, 1992), and cognitive (Segal, 1996)perspectives. Different subtypes of depression (bipolardepression, seasonal affective disorder, etc.) may reflectdifferent kinds of underlying dysfunction.

Conclusion

A cognitive-affective neuroscience perspective emphasizesthat mood is ordinarily mediated by specific neuronalcircuits, that similar circuits are dysfunctional in majordepression, that the serotonin system innervates thesecircuits, and that SSRIs act to normalize this

13

neurocircuitry. A model of serotonergic neurocircuitryprovides a way of integrating a broad spectrum of findings,ranging from observations of how early stressors negativelyimpact mood to work on functional brain imaging. Futurework may well provide a more precise delineation of howstressors negatively impact on neurocircuitry, and also onthe neurotrophic effects of antidepressant intervention(Lesch, 2001).

14

3.Generalized anxiety disorder


Generalized anxiety disorder (GAD) is a moderatelycommon anxiety disorder in epidemiological studies(Kessler, 2001) and the most common anxiety disorder inprimary care settings (Maier et al, 2000). A broaderconstruct, akin to the older ‘anxiety neurosis’, is likely tobe even more prevalent in the community and in the clinic.GAD is a chronic and disabling disorder, more common inwomen, with risk for age of onset beginning in the teensand cumulative lifetime prevalence increasing in roughlylinear fashion until the mid forties.

GAD has been a somewhat controversial diagnosis. DSM-III initially characterized GAD as a residual disorder, andsubsequent authors have similarly argued that in view ofits apparently high comorbidity with other disorders GADshould be conceptualized as a prodrome or severity markerof another condition. Nevertheless, there is evidence thatcomorbidity is no higher in GAD than in other disorderssuch as depression, and a growing consensus views GADas an independent condition characterized by specificpsychobiological features (Kessler, 2001).

GAD is characterized by worries that are difficult tocontrol, and by a range of somatic symptoms (Table 3.1).The increased focus in DSM on ‘worry’ at the expense of‘tension’ has been criticized by a number of authors(Rickels and Rynn, 2001). In some ways, the term ‘tensiondisorder’ seems more appropriate: GAD is char acterizedby both psychic tension (worry, irritability, insomnia), and

somatic tension (muscle tension, pains and aches). Suchtension may be followed by worry, which can beunderstood as an avoidance behaviour.

Assessment of GAD requires evaluating the course ofsymptoms, comorbid mood and anxiety disorders, andimpairment in function. Psychosocial stressors andsupports must be determined. Although it is clear that

Table 3.1Symptoms of Generalized Anxiety Disorder. Modifiedwith permission from the Diagnostic and Statistical Manual ofMental Health Disorders, FourthEdition, Text Revision. © 2000American Psychiatric Association.

16

GAD is associated with increased medical utilization, itshould also be remembered that general medicalconditions and substance use may contribute to anxietysymptoms, and these should therefore be excluded.

Planning the future

Whereas depression symptoms characteristically focus onthe past, GAD symptoms revolve around anticipation offuture harm. Indeed, there is an empirical literaturedocumenting that anxious patients show greateranticipation of future negative events (MacLeod and Byrne,1993). A number of regions may conceivably be involved insuch anticipation, including prefrontal cortex, given theimportance of this region in mediating executive functionssuch as planning and predicting.

In addition, GAD symptoms presumably involve generalactivation of what might be termed the ‘basic fear circuit’.Anxious patients attend selectively, and not necessarilywith awareness, to threatening cues (Macleod and Byrne,1993). While fear conditioning is discussed in detail inChapter 5, for the moment we can note that the amygdalaplays a crucial role in this fundamental process. It has alsobeen hypothesized that the bed nucleus of the striaterminalis (part of the extended amygdala) is particularlyimportant in free-floating anxiety (Davis and Whalen,2001). The hippocampus may be particularly relevant inmore complex situations involving conflict (Gray andMcNaughton, 1996) or avoidance.

There is a growing brain imaging literature focusing onthe induction of negative emotions in normal controls.Findings may depend on a range of factors includingprovocation strategy, but activation of inferior frontalcortical and anterior temporal areas has been documentedwith different emotions including anxiety (Kimbrell et al,1999).

Orbitofrontal regions are particularly likely to beactivated by internally (rather than externally) generatedaffects (Zald and Kim, 1996). Anterior temporal lobeactivation during emotional process ing may reflect a rolein assessment of context, whereas amygdala activation

17

may reflect the experience of the affect itself (Dolan et al,2000).

Despite the increasing recognition of GAD as anindependent disorder, it remains important to emphasizethat GAD is often followed by depression. Several kinds ofexplanation have been given to explain this temporalrelationship (Table 3.2). In patients with GAD-depression,symptoms are not restricted to anticipation of futureharm.

Neurocircuitry and GAD

The literature on the pathogenesis of GAD remains at anearly stage. Nevertheless, a number of themes haveemerged. A first question is whether the pathogenesis ofGAD differs in any way from that of depression. Aninfluential twin study indicated that GAD and majordepression (MD) shared common genetic factors, but hadsubstantially different nonfamilial environment risk factorswith different kinds of life events predisposing to anxietyand mood disorders (Kendler et al, 1992). Nevertheless,methodological aspects of this work have been criticized(Kessler et al, 1999), and there is also an empiricalliterature suggesting that GAD and MD are mediated byseparate genetic factors, consistent with the increased

Table 3.2Explanations of the temporal relationship beween GADand depression.

18

acceptance of GAD as an independent disorder. Thus, forexample, in another twin study, there was differentialtransmission of GAD and MD (Torgersen, 1990).

Indeed, preliminary brain imaging studies suggest thatGAD is characterized by a number of specific abnormalities(Figure 3.1). Thus, there may be increased amygdalavolume (De Bellis et al, 2000b), and abnormalbenzodiazepine receptor binding in the temporal pole(Tiihonen et al, 1997) of GAD patients. An earlytopographic electroencephalography study indicateddifferences between GAD and normals in temporal regions(Buchsbaum et al, 1985), and subsequent PET studieshave also shown temporal abnormalities in this disorder(Wu et al, 1991).

GAD may also be characterized by increased activity infrontal regions (Wu et al, 1991). Furthermore, an analysisof pooled PET symptom provocation data in obsessive-compulsive disorder, post-traumatic stress disorder, andsimple phobia found activation of right inferior frontalcortex, paralimbic structures (right posterior medialorbitofrontal cortex, bilateral insular cortex), bilaterallenticulate nuclei, and bilateral brainstem foci (Rauch et

Figure 3.1Functional neuroanatomy of GAD: increased activity inamygdala and perhapsprefrontally.

19

al, 1997b). This finding arguably provides indirect evidencefor a role for inferior frontal cortex and the paralimbicsystem (which serves as a conduit from sensory, motor,and association cortex to the limbic system itself) in arange of anxiety symptoms.

Role of serotonin

Animal studies demonstrate that serotonin hypofunction isassociated with hypersensitivity to environmental cues andincreasing responsiveness to threat (Handley, 1995;Lightowler et al, 1994). Knock-out models reinforce theimportance of a link between particular 5-HT subreceptorsand anxiety (Parks et al, 1998; Ramboz et al, 1998).Furthermore, in the social interaction test, which maydespite its name be a model of GAD, SSRIs have anxiolyticeffects (Lightowler et al, 1994).

In community studies, there may be an associationbetween the s/s allele of the serotonin transporter protein(5HTTP) and trait anxiety, but not all studies areconsistent (Lesch, 2001). In GAD, however, reducedcerebrospinal fluid (CSF) levels of serotonin and reducedplatelet paroxetine binding have been observed (Iny et al,1994). Furthermore, administration of the non-specificserotonin agonist m-chlorophenylpiperazine (mCPP)resulted in increased anxiety and hostility in GAD patients(Germine et al, 1992).

In addition, treatment data support the possibility thatthe serotonin system plays a role in mediating GAD.Buspirone, a 5-HT1A partial agonist, has long been usedfor the treatment of this disorder, despite uncertainty aboutthe robustness of its effects. Moreover, the SSRIs areincreasingly viewed as first-line agents in the managementof GAD (Ballenger et al, 2001).

There is growing interest in imaging the role of theserotonin system in anxiety (Tauscher et al, 2001). Todate, however, there is little available data on the effect ofSSRI treatment on the functional neuroanatomy of GAD.Nevertheless, it may be suggested that, as in other mood/anxiety disorders, these agents are able to normalizedysfunctional circuits (Figure 3.2).

20

Evolutionary considerations: futurealarm

Anxiety is likely to have had a long evolutionary history(Darwin, 1965; Stein and Bouwer, 1997a), occurring inresponse to signals of danger and representing a set ofresponse tendencies that have resulted in avoidance ofsimilar dangers in the history of both the individualorganism and its species. Different kinds of fear responsestend to emerge at the developmental stage where theybecome adaptive.

Even in simple organisms, like Aplysia, simpleconditioning and sensitization occur. In more complexorganisms, the anxiety response includes affective,cognitive, and motoric components. Interestingly, thereappears to be some continuity in neurobiologicalmechanisms; thus in Aplysia conditioning results inincreased release of serotonin with presynaptic facilitation(Kandel, 1983).

It is important to emphasize that genetic andenvironmental factors are inextricably intertwined in fearconditioning. Clearly fear conditioning requires learning.On the other hand, there has already been progress in

Figure 3.2Effects of SSRIs on the functional neuroanatomy of GA:normalization ofactivity in amygdala and prefrontal region.

21

delineating the specific genes responsible for determiningsuch factors as increased susceptibility to fearconditioning and/or poor habituation (Flint, 1997).

GAD can be argued to represent a non-specific anxiety/tension alarm (or future alarm). Indeed, it has been arguedthat general anxiety evolved to deal with threats whosenature could not be defined very clearly, while subtypes ofanxiety evolved to deal with particular dangers (Marks andNesse, 1994). Nevertheless, anxiety defends against manykinds of danger, and subtypes of fear alarms (see laterchapters) are not completely distinct.

Argument for an evolutionarily based false alarm can bestrengthened if evidence of individuals with a maladaptivelow-threshold alarm can be found. It turns out thatprimates who appear to have low harm avoidance and 5-HT hypofunction do have increased morbidity (Higley et al,1996). Similarly it may be speculated that humans withlow thresholds for the anxiety/tension alarm suffer fromantisocial personality traits; such people seem not to showadaptive responses to potentially dangerous situations.

Interestingly, early electrophysiological studies ofantisocial personality disorder (ASPD) suggested thatsubjects failed to show ‘arousal’. More recently, there hasbeen growing evidence for an association betweendecreased frontal lobe activity and increased impulsiveaggression (Davidson et al, 2001; Stein, 2000). This standsin contrast to increased frontal activity in GAD patients.

Conclusion

GAD patients experience a world dominated by worry andtension. In clinical practice, it is also important torecognize the triad of GAD-depression-somatization (Stein,2001a). While this symptom triad is commonly foundthroughout the world, its experience and expression mayvary from place to place, and time to time. ‘Neurasthenia’for example was a common diagnosis in the United Statesin the nineteenth century, and remains in wide use in theEast.

Worry and tension are in turn mediated by the serotoninsystem. Fortunately, both pharmacotherapy and

22

psychotherapy are effective in the treatment of GAD. Itseems entirely reasonable to speculate that early diagnosisand intervention may prevent the subsequent developmentof mood and other comorbid conditions. Primary carepractitioners, who are likely to see the majority of GADpatients, must be trained in the recognition and treatmentof this condition.

23

4.Obsessive-compulsive disorder


Obsessive-compulsive disorder (OCD) was the fourth mostcommon psychiatric disorder in the United StatesEpidemiological Catchment Area (ECA) study (Robins et al,1984), and has a life-time prevalence of 2–3% in manyparts of the world (Weissman et al, 1994). Furthermore, itwas the tenth most disabling of all medical disorders in alandmark mortality and morbidity study (Murray andLopez, 1996). Indeed, OCD has been suggested to cost theeconomy of a country such as the United States severalbillion dollars each year (Hollander et al, 1997).

OCD is characterized by obsessions and compulsions(Table 4.1). Obsessions are intrusive thoughts, ideas, orimages that increase anxiety, whereas compulsions arerepetitive rituals or mental actions performed in responseto obsessions in order to decrease anxiety. While patientsreport a wide range of different kinds of obsessions andcompulsions, there is an impressive consistency of themesbetween patients (Table 4.2) and across cultures (Stein andRapoport, 1996).

Despite this homogeneity, there have been increasingattempts to further the understanding and treatment ofOCD by specifying different subgroups. There is someevidence, for example, that the various symptoms of OCDcan be mapped onto a four-factor solution, with eachsymptom factor being mediated by somewhat differentneurobiological factors (Leckman et al, 1997). Thus, the

Table 4.1Symptoms of OCD. Modified with permission from theDiagnostic and Statistical Manual of Mental Health Disorders,Fourth Edition, Text Revision.© 2000 American PsychiatricAssodation.

25

treatment implications.Another nosological debate involves the putative OCD

spectrum disorders (Table 4.3). Although controversyremains about which disorders belong to this spectrum, anumber of conditions share considerablephenomenological and neurobiological overlap with OCD.In particular, several disorders characterized by intrusiverepetitive symptoms show a selective response to serotoninreup take inhibitors (SRIs) in comparison withnoradrenaline reuptake inhibitors (NRIs); these includeOCD itself, body dysmorphic disorder, possibly olfactoryreference syndrome, hypochondriasis, trichotillomania,and obsessive-compulsive symptoms in Tourette’s

Table 4.2Typical obsessions and consequent compulsions in OCD.

Table 4.3Putative obsessive-compulsive spectrum disorders.

26

presence of comorbid tics in OCD appears to have specific

syndrome (TS) and autism. There may also be overlap inthe neuroanatomy, neuroimmunology, and neurogeneticsof some of these disorders, particularly between OCD andTS (Stein, 2001b).

Procedural strategies

A crucial component of normal cognitive-affectivefunctioning is the selection, maintenance, and initiation ofcognitive and motoric programmes. These programmeshave been given terms such as ‘habit system’ (Mishkin andPetri, 1984), ‘response set’ (Robins and Brown, 1990), or‘procedural mobilization’ (Saint-Cyr et al, 1995). Given thatOCD symptoms involve stereotypical behaviour, animmediate possibility is that OCD involves dysfunction ofprocedural strategies.

It is likely that cortical-striatal-thalamic-cortical systems(CSTC) play a crucial role in the implicit learning ofprocedural strategies, and their subsequent automaticexecution. There are several parallel CSTC circuits, each ofwhich governs a somewhat different spectrum of cognitiveand affective function (Alexander et al, 1986). VentralCSTC circuits appear to play a particularly important rolein recognizing behaviourally significant stimuli (and inerror detection) and in regulating autonomic and goal-directed responses (including response inhibition andsuppression of negative emotion), and are therefore a goodcandidate for involvement in OCD (Davidson et al, 2001;Rauch and Baxter, 1998; Zald and Kim, 1996). InTourette’s syndrome, more motoric symptoms are mediatedby related CSTC circuits (Stern et al, 2000).

Neurocircuitry and OCD

There is a range of evidence that CSTC circuits aredisrupted in OCD. Perhaps the first evidence that OCDmight have a neurologi cal basis was provided during thepandemic of viral encephalitis lethargica early in the lastcentury. Patients with parkinsonian features were observedto have obsessive-compulsive symptoms, tics, and focal

27

brain lesions including involvement of the basal ganglia(Cheyette and Cummings, 1995).

Other neurological disorders with basal gangliainvolvement can also present with obsessive-compulsivesymptoms (Stein et al, 1994) (Table 4.4). Of particularinterest is the association between streptococcal infectionand subsequent OCD and tics (so-called PANDAS, orpaediatric autoimmune neuropsychiatric disordersassociated with Streptococcus), as this association suggeststhat autoimmune factors may be involved in OCD (Swedoet al, 1998).

A range of additional evidence points to cortico-striatalinvolvement in OCD (Stein et al, 1994). OCD patients haveincreased neurological soft signs including tics, consistentwith basal ganglia damage. Neuropsychological testing isarguably also consistent with CSTC dysfunction.Neurosurgery of refractory OCD involves making lesions inCSTC circuitry.

Perhaps most persuasive, however, is evidence fromstructural and functional brain imaging (Rauch andBaxter, 1998). A number of structural imaging studieshave pointed to basal ganglia abnormalities, althoughthere is some inconsistency, with data pointing toincreased as well as decreased basal ganglia volume.Interestingly, patients with PANDAS have been shown tohave larger basal ganglia, with shrinkage perhapsoccurring over time; such possible changes in volume over

Table 4.4Lesions of the basal gangia associated with OCD.

28

time may contribute to some of the inconsistenciesbetween studies.

A range of functional imaging studies have pointed toCSTC involvement in OCD. Studies show that at baselineOCD patients have hyperactivity in orbitofrontal cortex,anterior cingulate, and caudate nucleus (Figure 4.1). Thishyperactivity is exacerbated by exposure to feared stimuli,but is normalized by successful treatment with eithermedication or behavioural therapy (Baxter et al, 1992)(Figure 4.2).

The question of how CSTC dysfunction arises in OCDremains unresolved. A range of mechanisms may beinvolved, including genetic factors, the marked sensitivityof striatal circuits to anoxic damage, the development ofdisordered striatal architecture after emotionaldeprivation, as well as a broad spectrum of potentialneurochemical, neuroimmunological, andneuroendocrinological factors.

Role of serotonin

As noted earlier, OCD is well known to have a selectiveresponse to serotonergic agents. Several other findings

Figure 4.1Functional neuroanatomy of OCD: increased activity inorbitofrontal cortexand ventral striatum.

29

support the role of serotonin in mediating OCD. Theliterature on peripheral markers in OCD is somewhatinconsistent, but an early study of OCD showed that CSF5-HIAA decreased during treatment with clomipramine(Thoren et al, 1980). Studies of serotonergic ‘challenges’ inOCD also show some inconsistency, but there appears tobe a group of patients who respond to serotonin agonistswith exacerbation of OCD symptoms, a phenomenon thatis no longer apparent after SRI treatment (Zohar et al,1988).

The serotonin system is a complex one, and an aim ofrecent research has been to focus on the serotoninsubreceptors most relevant to OCD. Animal work, forexample, has focused on the 5-HT1D receptor; duringadministration of SSRIs this terminal autoreceptor (i.e.brake) is gradually desensitized, so resulting in increasedserotonergic acitivity (the brake is lifted). In animals, 5-HT1D desensitization requires at least 8 weeks of high-doseSRI treatment, so paralleling clinical findings whichindicate that OCD requires higher doses and longerduration of treatment than does depression (El Mansari et

Figure 4.2Effects of SSRIs on the functional neuroanatomy ofOCD: normalization ofactivity in orbitofrontal and ventral striatalregions.

30

al, 1995) (Table 4.5). 5-HT2 receptors may also beimportant (Delgado and Moreno, 1998).Do these findings fit with our knowledge of theneurocircuits involved in OCD? Importantly, CSTC circuitshave significant serotonergic innervation. In animal work,5-HT1D receptors desensitized by relatively high doses anddurations of SSRIs lie in orbitofrontal cortex. In humans,particular 5-HT1D alleles have been associated withincreased risk for OCD (Mundo et al, 2000), andadministration of sumatriptan, a 5-HT1D agonist, isassociated with changes in the activation of cortico-striatalcircuits in OCD (Stein, 1999). Most persuasively, aftertreatment of OCD with SSRIs, there is normalization ofcortico-striatal activity (Baxter et al, 1992) (Figure 4.2).

Evolutionary considerations: groomingalarms

OCD can be viewed in terms of a failure to inhibit cortico-striatally governed procedural strategies from intrudinginto consciousness. Such a view is consistent with (1) the

Table 4.5Treatment principles differ in depression and OCD.

31

limited number of symptom themes in OCD (e.g. washing,hoarding), and their apparent evolutionary importance; (2)dysfunction of CSTC circuits in a range of studies, withactivation of temporal rather than striatal areas duringimplicit cognition (Rauch et al, 1997a); and (3)abnormalities in serotonergic systems which innervatecortico-striatal circuits and which may be associated withdisinhibition.

It is interesting that a range of procedural strategies maybe relevant to OCD. Much of the literature has focused oncontamination and increased grooming. Some animalmodels appear highly relevant to such a formulation,demonstrating a remarkably similar pharmacotherapyresponse profile with OCD (Rapoport et al, 1992; Stein etal, 1992). Indeed, some OCD spectrum disorders canperhaps be conceptualized as grooming disorders (Stein etal, 1999a). However, other procedural strategies may alsobe relevant; these include hoarding (Stein et al, 1999b) andsymmetry assessment (which again appears to be mediatedby specific evolutionary mechanisms, and which may beparticularly pertinent to body dysmorphic disorder).

An important theoretical question in the study of OCD iswhether patients are compulsive or impulsive? Early onFreud argued that OCD was ultimately a disorder ofincreased aggression, with compensatory defencemechanisms (Stein and Stone, 1997). Put in more modernterms, OCD can be understood to involve loss of inhibitorymechanisms (perhaps primarily serotonergic), withapparent compensatory factors (for example, increasedactivity in orbitofrontal cortex, with some evidence ofhyperserotonergic activity). Whereas impulsive patientshave decreased hypofrontal/serotonin function,compulsive patients are also able to show a compensatoryresponse with hyperfrontal/hyperserotonergic function(Stein and Hollander, 1993).

Another significant issue is whether OCD really is ananxiety disorder (Montgomery, 1993). Certainly, theprimary emotion in OCD does not appear to be fear.Indeed, some authors have begun to suggest rather that theemotion that is particularly relevant to OCD is that ofdisgust (Stein et al, 2001). Interestingly, the neurobiology

32

of fear and disgust can be dissociated on neurobiologicalstudies; whereas fear involves the amygdala (seeChapter 5), disgust is mediated by the CSTC circuits whichare so central to OCD.

Conclusion

There have been very great changes in the OCD field; onceconceptualized as a rather rare and treatment-refractorycondition, OCD is now recognized to be one of the mostcommon of the psychiatric disorders, and it often respondsto modern treatments. Whereas OCD once provided a keyexemplar for a model of psychopathology based onunconscious conflict, it can now be seen as a seminalexemplar for a psychiatry based on moderncognitiveaffective neuroscience. Thus, for example, OCDwas the first disorder for which it was shown that bothspecific medications (SRIs) and particularpsychotherapeutic techniques (exposure) normalizedprecise neuroanatomical circuitry. In the future, we canexpect further delineation of the pathogenesis ofdysfunction in these circuits, which will in turn hopefullylead to novel treatments.

33

5.Panic disorder


Panic disorder is present in approximately 2% of thepopulation, with a somewhat higher incidence in females(Kessler et al, 1994). Patients frequently present to primarycare practitioners and non-pychiatric medical specialists,and underdiagnosis/undertreatment and overutilization ofmedical resources remain important issues.

Panic attacks may be present in all of the anxietydisorders. However, in panic disorder they arecharacteristically spontaneous. They are accompanied by arange of symptoms, including respiratory, cardiovascular,gastrointestinal, and occulovestibular symptoms (Tables5.1 and 5.2). Panic attacks vary, however, in their cueing,in their extent, and in the time of onset (Table 5.3).

Patients may go on to develop agoraphobia, or avoidanceof situations which may precipitate panic attacks. Thissequence of anxiety avoidance is a common themethroughout this volume. From a theoretical perspective itraises questions about the different neurocircuits involvedin mediating these phenomena, and from a clinicalperspective it emphasizes the importance of early exposurein preventing later avoidance.

In addition, panic disorder may be accompanied by anumber of mood and anxiety disorders, with depression aparticularly important complication. Panic-depression isthe most common form of mood-anxiety comorbidity (Roy-Byrne et al, 2000), and a number of authors have

emphasized the importance of the link between panic andsuicide (although not all data are consistent).

Conditioned fear

An important recent advance in affective neuroscience hasbeen the delineation of brain circuits involved in fearconditioning. Fear conditioning has been a fundamentalparadigm for clinicians since the eminent behaviouristJohn Watson demonstrated how a fear of fluffy white toyscould be conditioned in an infant by pairing presentationof such toys with an aversive stimulus. Modernneurobiologists have been able to demonstrate theproximate neurobiological factors involved (Davis andWhalen, 2001; Le Doux, 1998).

The amygdala appears to play a particularly importantrole in mediating conditioned fear. Afferents to thebasolateral amygdala include the thalamus which relays

Table 5.1Symptoms of panic attack. Modified with permissionfrom the Diagnostic and Statistical Manual of Mental HealthDisorders, Fourth Edition, Text Revision.© 2000 AmericanPsychiatric Association.

35

sensory information, while efferents from the centralamygdala nucleus and lateral bed nucleus of the striaterminalis (BNST) (or extended amygdala) include a rangeof structures which mediate the ‘fight-or-flight’ response.Such structures include the lateral nucleus (autonomicarousal and sympathetic discharge) and paraventricular

Table 5.2Symptoms of panic disorder. Modified with permissionfrom the Diagnostic and Statistical Manual of Mental HealthDisorders, Fourth Edition, TextRevision. © 2000 AmericanPsychiatric Assodation.

Table 5.3Panic attack subtypes in panic disorder.

36

nucleus (increased adrenocorticoid release) of thehypothalamus, as well as the locus ceruleus (increasednoradrenaline release), parabrachial nucleus (increasedrespiratory rate), periaqueductal grey (defensive behavioursand postural freezing), and nucleus pontine reticularis(startle response) in the brainstem. Given this information,it is easy to speculate that the amgydala is important inmediating panic attack symptoms.

The hippocampus, on the other hand, plays animportant role in processing the context of the fearconditioning. The hippocampus is situated at theconfluence of a dorsal pathway for spatial/perceptualmemory (of ‘where’) and a ventral pathway for object/conceptual memory (of ‘what'), and so plays an importantrole in representing spatial location within a frameworkfixed to the environment (Burgess et al, 1999). Thehippocampus may play a particularly important role inmediating avoidance behaviours in people who haveexperienced panic attacks.

Consistent with evidence that implicit and explicitprocessing are partially localized to somewhat differentneuroanatomical circuits (Salloway et al, 1997), theexplicit memory of where and how fear conditioning tookplace, and the implicit processing involved in the fearconditioning itself, can be dissociated. Thus, an amygdalalesion does not impact on explicit recall, while ahippocampal lesion does not prevent implicit fearconditioning from occurring (Bechara et al, 1995). There isevidence that the implicit pathway is evolutionarily older(Reber, 1993); certainly the development of explicit memoryis likely to have appeared relatively late in the developmentof primates. Conversely, when there are widespread brainlesions (e.g. in Alzheimer’s disorder), explicit processing isaffected first, while implicit memory is more robust and isretained for longer.

Neurocircuitry and panic disorder

Is there any evidence that panic disorder involvesdysregulation of amygdala-hippocampal fear systems?Certainly, there is a small literature demonstrating that

37

panic disorder can be associated with a range of (especiallyright-sided) temporal abnormalities including seizuredisorder (Young et al, 1995). Similarly, stimulation of theamygdala in preclinical and clinical studies (of seizuredisorder patients) is associated with fear responses(Cendes et al, 1994; Davis and Whalen, 2001). Conversely,patients with amygdala lesions demonstrate selectiveimpairment in the recognition of fearful facial expressions,and show an inability to be conditioned to fear—this is theclassical Klüver-Bucy syndrome (Klüver and Bucy, 1939).

Studies in normals show activation of amygdala andperiamygdaloid cortical areas during fear acquisition andextinction (Gorman et al, 2000), although this may berelated to affective processing rather than affect itself(Davis and Whalen, 2001). A structural imaging studysuggested abnormal temporal lobe volume in panic(Vythilingam et al, 2000). In addition, PET scanning duringanxious anticipation in normals (Reiman et al, 1989a), andduring lactate-induced panic attacks in panic disorderpatients (Reiman et al, 1989b) both demonstratedincreased activity in paralimbic regions (temporal poles).Activation of the amygdala leads in turn to activation ofefferent circuits (hypothalamus, brainstem) (Figure 5.1).An early study suggested that only panic disorder patientssusceptible to lactate-induced panic had abnormalasymmetry of a parahippocampal region at rest (Reiman etal, 1986). Subsequent functional imaging studies haveconfirmed dysfunctions of hippocampus orparahippocampal regions in panic disorder, although theprecise abnormalities documented have not always beenconsistent (Gorman et al, 2000).

Hypocapnia-induced vasoconstriction compounds thedifficulty of interpreting some panic disorder imagingstudies in which there is decreased activity in variousregions. Another possibility, however, is that whereasanxiety may be associated with activation of partic ularbrain regions in an attempt to suppress negative emotion,during the height of a panic attack there are in fact regionsof deactivation (perhaps associated with impairment inspecific cognitive-affective phenomena, such asarticulation/verbalization of feelings).

38

Role of serotonin

Several studies of serotonergic markers have shownabnormalities in panic disorder. Administration of mCPP, aserotonin agonist, results in exacerbation of panicsymptoms in patients with panic disorder. Similarly, panicattacks can be precipitated or exacerbated by a range ofsubstances, such as marijuana, that have serotoninagonist effects (Coplan et al, 1992). One possibility is thatpost-synaptic serotonin receptors are upregulated in thiscondition, so that they are supersensitive to suchserotonergic agents.

Conversely, there is now good evidence of the efficacy ofSSRIs in panic disorder, so that these medications aregenerally considered to be a first-line pharmacotherapy forthe management of this condition (provided that relativelylow doses are used at first, so as to avoid unnecessaryagitation). Indeed, a meta-analysis comparing thesemedications with imipramine and benzodiazepinessuggested the superiority of the SSRIs over other agents(Boyer, 1995).

Figure 5.1Neuroanatomy of panic disorder: once the amygdala isactivated, its efferentsto the hypothalamns and brainstem mediatea fear response.

39

Is it possible to integrate these data with the work on theneuroanatomy of fear conditioning? It turns out that theserotonergic system interacts at several points with fearconditioning pathways (Coplan and Lydiard, 1998), soallowing an interweaving of neuroanatomic andneurochemical models. Serotonergic projections from thedorsal raphe nucleus (DRN) generally inhibit the locusceruleus (LC), while projections from the LC stimulate DRNserotonergic neurons and inhibit median raphe nucleus(MRN) neurons. In addition, the DRN sends projections toprefrontal cortex, amygdala, hypothalamus, and theperiaqeductal grey (PAG) amongst other structures.

Modulation of the serotonin system therefore has thepotential to influence the major regions of the panicdisorder circuit, so resulting in decreased noradrenergicactivity, diminished release of corticotropin-release factor,and modification of defence/escape behaviours. While thiskind of model requires additional empirical validation, aninteresting imaging study found that after administration ofthe serotonin releaser and reuptake inhibitor fenfluramine,panic disorder patients had increased parietal-temporalcortex activation (Meyer et al, 2000). Presumably SSRIs areable to normalize functional abnormalities in panicdisorder.

Evolutionary considerations:suffocationalarms

Klein has argued persuasively that panic disorder ischaracterized by a false suffocation alarm (Klein, 1993). Hebegins with the suggestion that the suffocation alarm is anevolved, adaptive response to a lack of oxygen (signaled byincreasing PCO2 and brain lactate). He then goes on tohypothesize, buttressing his argument with a review of therelevant empirical literature, that the threshold for thisalarm is lowered in panic disorder.

First, the most prominent symptom of many panicattacks is that of dyspnea (indicating a specific emergencyreaction to suffocation), and a range of studies documentrespiratory abnormalities in panic (such as increasedsighing, perhaps indicating an attempt to avoid dyspnea by

40

lowering PCO2). This focus also helps bolster Klein’s initialdifferentiation between episodic spontaneous panics (inwhich chronic hyperventilation acts as a predictor oflactateinduced panic) from chronic fear-like anxiety (wherethere is rather HPA activation) (Klein, 1964).

Second, panic attacks increase during a range ofconditions that are characterized by an increase in PCO2(relaxation and sleep, premenstrual period, and respiratoryinsufficiency). Conversely, panic attacks decrease duringconditions characterized by a decrease in PCO2(pregnancy). Interestingly, there appears to be a conditionin which the false suffocation alarm is absent; in congenitalhypoventilation syndrome or Ondine’s curse, patients notonly require treatment with agents (e.g. amphetamines)that act as panicogens in normals, but it turns out thatthey rarely develop panic attacks (Pine et al, 1994).

Another possibility, however, is that panic represents anacute danger alarm, that may be triggered by a range ofunconditioned stimuli including increased PCO2. Indeed,Klein’s original conceptualization of panic attacks was interms of an evolved response to separation anxiety (Klein,1981). Although such a broader view of the panic disorder‘false alarm’ may arguably be weakened by some loss offocus on respiratory systems, it may be more consistentwith the range of other data about environmentalprecipitants of panic attacks (Shear, 1996) and theneurobiology of unconditioned fear responses (Panskepp,1998).

Conclusion

Panic disorder has for many years been a lost disorder,hidden in the broad swath of ‘anxiety neuroses’. Itsdiscovery as a unique entity, characterized by specificneurobiological dysfunction, and responding to selectivetreatment with modern medications and psychotherapies,represents a tremendously important advancefor psychiatry. Delineation of the neurobiology of fearconditioning has been a particularly influentialdevelopment for the clinical conceptualization of panicdisorder. The role of the serotonergic system in innervating

41

the relevant circuits, and of the SSRIs in normalizingunderlying dysfunction in panic disorder, provides thebasis for developing a model of panic that incorporatesclinical experience with this disorder as well as a range ofresearch findings.

42

6.Post-traumatic stress disorder


Post-traumatic stress disorder (PTSD) has long beenconsidered a ‘normal’ response to an ‘abnormal’ event.However, it turns out that rates of exposure to trauma areextremely high, and that only a small percentage of peoplego on to develop chronic PTSD. Thus PTSD is increasinglyseen as an abnormal response to a traumatic event (Yehudaand McFarlane, 1995).

PTSD begins, by definition, in the aftermath of atraumatic experience (Table 6.1). DSM-IV attempts todefine the traumatic event in both objective terms (there isphysical danger) and subjective ones (there is horror, fear).Traumatic events classically associated with PTSD includeinterpersonal traumas such as combat (more common inmen) and rape (more common in women), as well asnatural disasters.

Three characteristic clusters of symptoms then emerge:re-experiencing, avoidance/numbing, and hyperarousal(Table 6.1). Reexperiencing and hyperarousal symptomsare similar in some ways to the ‘positive’ or panickysymptoms seen in various anxiety disorders, although theyare distinguished by their focus on a traumatic event.Avoidance and numbing symptoms are redolent of thevarious ‘negative’ or avoidance symptoms that also cutacross the anxiety disorders, although loss of memory (ofthe traumatic event) is perhaps particularlydistinguishing. (cont.)

44

Danger alarm

The amygdala plays a crucial role in the organism’sresponse to danger. Rapid thalamo-amygdala circuitsimmediately pass sensory information to the centralnucleus of the amygdala (CeN), which then co-ordinates a‘fight-or-flight’ response (see Chapter 5). Efferent fibresfrom the CeN and internal bed nucleus of the striaterminalis (BNST) innervate a number of structures whichmediate this complex response. These include the lateralnucleus (autonomic arousal and sympathetic discharge)and paraventricular nucleus (increased adrenocorticoidrelease) of the hypothalamus, as well as the locus ceruleus(increased noradrenaline release), parabrachial nucleus(increased respiratory rate), and PAG (defensive behavioursand postural freezing) in the brainstem.

The hippocampus, on the other hand, plays animportant role in processing the context of the fearconditioning. As discussed in Chapter 5, the explicitmemory of where and how fear conditioning took place,and the implicit processing involved in the fearconditioning itself, are processes that can be dissociated.Indeed, what is remarkable about PTSD is the extent to

Table 6.1Symptoms of PTSD. Modified with permission from theDiagnostic and Statistical Manual of Mental Health Disorders,Fourth Edition, Text Revision.© 2000 American PsychiatricAssodation.

45

which explicit cognition is taken ‘off-line’. (Notably, PTSDcan develop even when head trauma results in loss ofexplicit memories (Macmillian, 1991).) While suchdissociation may be adaptive at the time of the trauma, itmay interfere with processing of the traumatic event andsubsequent adaptive responses (Brewin, 2001).

The animal literature suggests that a ‘danger alarm’ canbe down-regulated by medial prefrontal cortex (anteriorcingulate) (Le Doux, 1998), and there is some supportivehuman imaging data (Davidson et al, 2001; Hugdahl,1998). From a different perspective, this ‘top-down’ controlcan be understood in terms of the ‘processing’ of thetraumatic event. Implicit processes are integrated togetherwith explicit ones, the traumatic event is articulated andintegrated with the rest of the person’s schemas, and theperson readjusts and adapts. It is possible, however, that arepeated traumatic event will trigger the rapid amygdalo-thalamic fibers, over-riding this frontal processing, andresulting in a return of symptoms (the return of therepressed!). Certainly there is a growing literaturedocumenting the long-lasting psychobiological impact ofearly developmental trauma and of repeated exposure tostressors (Maier, 2001; Sanchez et al, 2001).

Neurocircuitry and PTSD

Brain imaging findings have provided some empiricalevidence that such a ‘danger alarm’ model of PTSD is infact at least partially correct (Figure 6.1). Structuralfindings have focused on decreased hippocampus volume(Rauch et al, 1998). Although not all studies have beenconsistent (Bonne et al, 2001), in some work decreasedhippocampus volume has correlated with trauma exposureor with cognitive impairment. Of course, it is theoreticallypossible that decreased hippocampus volume may be arisk factor for subsequent PTSD. However, there may beother explanations, including the neurotoxic effects ofdysfunctional HPA axis activity (Sapolsky, 2000).

In healthy controls, imaging studies have demonstratedsubcortical processing of masked emotional stimuli by theamygdala. Indeed, in an early study, PTSD patients

46

exposed to audiotaped traumatic and neutral scriptsduring PET were found to have increases in nor malizedblood flow in right-sided limbic, paralimbic, and visualareas, with decreases in left inferior frontal and middletemporal cortex (Rauch et al, 1996). Subsequent work hasbeen more are less consistent (Rauch et al, 1998). Theauthors of this research concluded that emotionsassociated with the PTSD symptomatic state are mediatedby the limbic and paralimbic systems within the righthemisphere, with activation of visual cortex perhapscorresponding to visual re-experiencing. Decreased activityin Broca’s area during exposure to trauma in PTSD, on theother hand, is consistent with patients’ inability to processverbally traumatic memories (Rauch et al, 1996).

A recent study was suggestive of decreasedbenzodiazepine receptor binding in prefrontal cortex ofPTSD patients (Bremner et al, 2000). There is also agrowing empirical literature on the anterior cingulate inparticular in PTSD, with several imaging studiesdemonstrating decreased activity in this region (Hamner et

Figure 6.1Functional neuroanatomy of PTSD: in addition toactivation of the amygdalaand efferent circuitory, there isdecreased activation of Broca’s and perhapsother frontal areas, aswell as impairment in hippocampus function (withapparentreduction in volume).

47

al, 1999). Furthermore, a recent magnetic resonancespectroscopy study demonstrated that the ratio of N-acetylaspartate to creatinine, a putative marker ofneuronal integrity, was significantly lower in the anteriorcingulate of children and adolescents with PTSD than inhealthy controls (De Bellis et al, 2000a). (Interestingly,although conversion disorder, like PTSD, involves a shiftfrom verbal to non-verbal processes, this condition appearsto involve a somewhat different functional neuroanatomy(Halligan et al, 2000).)

Although involvement of the basal ganglia has nottypically been found in functional imaging studies ofPTSD, a study of singlephoton emission computedtomography (SPECT) scans in patients with PTSD and OCDreported that these groups had similarities in comparisonto scans of patients with panic disorder and healthycontrols (Lucey et al, 1997). The authors suggested thatthis might reflect the existence of repetitive intrusivesymptoms in both PTSD and OCD. The possibility ofcertain phenomenological and psychopharmacologicalsimilarities between PTSD and OCD certainly bears furtherthought. Also, interactions between the amygdala andcorticostriatal systems may be be important in mediatingthe transition from emotional reaction to emotional action(Le Doux, 1998).

The role of serotonin

Animal studies have demonstrated that serotonin isinvolved in regulation of the amygdala and connectingstructures at a number of points (Coplan and Lydiard,1998); this may well be relevant to the mediation of PTSDsymptoms.

Clinical studies of abnormal paroxetine binding in PTSD,and exacerbation of PTSD symptoms in response toadministration of the serotonin agonist mCPP are alsoconsistent with a role for serotonin in this disorder(Connor and Davidson, 1998).

Furthermore, there is increasing evidence for the efficacyof SSRIs in the treatment of PTSD (Stein et al, 2000), withsome hints that these agents may even be more effective

48

than other classes of medication such as less serotonergictricyclic antidepressants and benzodiazepines (Penava etal, 1996) (although there are no head-to-head trials todecide this question). There are few studies of the effects ofSSRIs on the functional neuroanatomy of PTSD.Nevertheless, preliminary evidence indicates that theyexert an effect by normalizing temperolimbic activation(Seedat et al, 2000) (Figure 6.2).

Evolutionary considerations: going ‘off-line’

The ability of the brain-mind to go ‘off-line’ in response totrauma presumably has ancient phylogenetic roots, havingevolved as an adaptive response. Nevertheless, it seemsthat in PTSD this process continues to be maintained evenonce the danger has passed.

At a neurobiological level this may reflect sensitization ofneurochemical systems, perhaps with consequent damageto the hippocampus. Essentially, excessive trauma hasresulted in dysfunction. In this perspective, a crucial risk

Figure 6.2Effects of SSRIs on the functional neuroanatomy ofPTSD: normalization ofactivity in Broca’s area and amygdala-mediated nerocircuits.

49

factor for PTSD would be previous exposure to traumaticevents.

A psychobiological conceptualization may emphasizethat in humans language and higher cognitive processesordinarily play an important role. When these higherfunctions are taken ‘off-line’, there is an inability toprocess traumatic events. In PTSD, this problem persists.

Risk factors for PTSD (Table 6.2) can readily beconceptualized in terms of such a view. Thus patients withpre-trauma poor processing skills may be more prone todevelop PTSD. Similarly, patients with peri-traumaticdissociation are more likely to have difficulty in verbalizingtheir responses. Finally, patients who experience guilt,shame, or lack of social support in the aftermath oftraumatic events may have more difficulty in processingsuch experiences (Yehuda, 1999).

Conclusion

Some authors have argued that ‘trauma’ ultimatelyconstitutes the final common pathway that lies at thebottom of all psychopathology. Freud, of course, reversedhis early similar stance, and modern data on individualpsychobiological susceptibilities suggest that these alsoplay a crucial role in determining whether responses totrauma are characterized by resilience and growth, or bypsychopathology and dysfunction. Advances in ourunderstanding of the serotonin system provide a way of atleast partly accounting for the effects of individualsusceptibility, of the effects of trauma on neuronal and

Table 6.2Risk factors for PTSD.

50

cognitive functioning, and of the way in which SSRIs areable to normalize dysfunction and reduce symptoms.

51

7.Social phobia


Social phobia (SP) is, apart from specific phobia, the mostcommon of the anxiety disorders, with prevalence rangingfrom 3 to 16% in various studies (Davidson et al, 1993b;Kessler et al, 1994). It is more common in women incommunity studies, but in clinical studies the ratio of menwith the disorder increases considerably. Cross-nationalcommunity studies show similarities in demographic andclinical features in different parts of the world (Weissman etal, 1996).

SP is characterized by fear of embarrassment orhumiliation in social situations (Table 7.1). Socialsituations comprise social interaction (e.g. talking in smallgroups, dating) and performance (e.g. speaking or eating infront of others). These situations are associated withsymptoms of panic, but the panic attacks of SP are morelikely to be characterized by blushing, tremor, and avertedgaze (Amies et al, 1983).

As in other anxiety disorders, anxiety often leads toavoidance and disability. In SP, the avoidance is of socialsituations. Patients with SP are more likely to remainunmarried, to drop out of school or college, and to earnless money than people without this condition. The term‘social anxiety disorder’ is increasingly recommended aspossibly less stigmatizing, and helps emphasize the overlapbetween the different anxiety disorders.

Important subtypes of SP are generalized and discreteSP. Generalized SP is characterized by fear of most social

situations, whereas discrete SP is limited to one or twoperformance situations. Generalized SP is associated withmore severe and disabling symptoms, and may be morefamilial. It is possible, however, that there exists acontinuum of symptom severity and associated disability,ranging from generalized SP through to discrete SP and onto fear of public speaking (Kessler et al, 1998).

It is important to note the comorbidity of SP withdepression and with substance use disorders (Kessler etal, 1999). Indeed, given that SP often begins early in life,

Table 7.1Symptoms of SP. Modified with permission from theDiagnostic and Statistical Manual of Mental Health Disorders,Fourth Edition, Text Revision. © 2000American PsychiatricAssociation.

53

has a chronic course, and usually precedes otherdisorders, it is fair to argue that SP predisposes to suchconditions. Interestingly, the symptoms of depression in SPare those characteristic of atypical depression (i.e.hyperphagia, hypersomnia, leaden paralysis, and rejectionsensitivity).

Social status

Although SP is characterized by pathological distress anddysfunction, anxiety in social situations is often a normaland adaptive emotion. Indeed, this is arguably the reasonit is has taken so long to recognize SP as a psychiatricdisorder, and why it continues to remain underdiagnosedand undertreated (Schneier et al, 1992). Clinicians,familiar of course with their own feelings of social anxiety,may mistakenly ‘normalize’ their patients’ symptoms of SP.

Indeed, although textbooks of psychopathology oftenquote the example of the lion and the antelope to illustratehuman fright-fight-flight responses, humans are very socialprimates and threatening conspecifics (i.e. other humans)are perhaps more constantly relevant than is theoccasional predator. Humans have presumably evolvedefficient mechanisms for recognizing thoughts andemotions in conspecifics and for responding accordingly.

There is a growing body of knowledge about theneurocircuitry involved in recognizing and processing thefaces, emotions, and gaze of others (Allison et al, 2000). Arange of structures have been suggested to mediate socialcognition, including the amygdala and temporal regions,the striatum, and prefrontal and cingulate cortex (Adolphs,2001).

There is also a growing body of knowledge about theneuroanatomy of anxiety in general, as reviewed earlier.Nevertheless, relatively little is known about theneuroanatomy of social anxiety per se. Humans, as Darwin(1965) pointed out, are the only animals that blush, sothat while the general neuroanatomy of anxiety may berelevant, other more specific circuits also requiredelineation.

54

Amygdala-dorsal striatum ganglia circuitry, for example,may mediate inhibitory avoidance and motor learning(Davis and Whalen, 2001); interestingly, socially anxiouschildren show reduced general facial activity and have amore restricted facial repertoire (Melfsen et al, 2000). Also,amygdala-ventral striatal circuitry may be importantinsofar as it bridges processing of affects and reward-seeking behaviour (Davis and Whalen, 2001)

Shyness and behavioural inhibition (manifesting in socialsituations) may represent predisposing traits for SP, andare thought to have a heritable component. Although theneurobiology of behavioural inhibition is incompletelyunderstood, it might be speculated that prefrontalhyperactivity may contribute (Johnson et al, 1999; Kaganet al, 1988).

Neurocircuitry and SP

Given that social anxiety is a form of anxiety, it would notbe surprising if the amygdala played a role in mediating SP(Figure 7.1). Certainly, a range of panicogenics are able totrigger panic attacks in SP patients, albeit not to the sameextent as in panic disorder (Stein et al, 2002).Furthermore, there is interesting recent data showing thatSP patients demonstrate selective activation of the the

Figure 7.1Functional neuroanatomy of SP: increased amygdalaand cingulate activity,with decreased basal ganglia activities.

55

amygdala when exposed to fear-relevant stimuli(Birbaumer et al, 1998) or tasks (Tillfors et al, 2001), orshow abnormal patterns of amygdala activation duringaversive conditioning (Schneider et al, 1999). Conversely,after lesions of the amygdala (Klüver-Bucy syndrome) theremay be inappropriate loss of fear.

There is also growing evidence that striatal neurocircuitsplay a role in SP. Thus, patients with SP have a greaterreduction in putamen volume with ageing (Potts et al,1994), reduced choline and creatinine signal:noise ratios insubcortical, thalamic, and caudate areas (Davidson et al,1993a), and decreased N-acetylaspartate (NAA) levels and alower ratio of NAA to other metabolites in cortical andsubcortical regions (Davidson et al, 1993a; Tupler et al,1997). Furthermore, striatal dopamine systems may beabnormal in SP.

Finally, frontal areas may play a role in social anxiety.Although not all work is consistent, there is a report ofincreased dorso-lateral prefrontal cortex activity duringsymptom provocation in a PET study of social anxietydisorder (SAD) (Nutt et al, 1998), and of cortical grey matterabnormalities in SAD in particular (Tupler et al, 1997).Anterior cingulate, which is involved in performancemonitoring (McDonald et al, 2000), may play a crucial rolein a number of anxiety disorders, including SP. Inaddition, imaging studies that have pooled or comparedfindings across different anxiety disorders suggest theimportance of increased activation of inferior cortex inmediating anxiety symptoms (Rauch et al, 1997b).

Role of serotonin

Given the potential involvement of these variousneurocircuits (amygdala, basal ganglia, frontal) in SP, itcan be postulated that the serotonin system plays animportant role in the mediation of social anxiety. Asreviewed in Chapters 4 and 5, the serotonin systembranches widely and extends to both amygdala andcorticostriatal neurocircuits (Figure 1.1).

Furthermore, serotonin plays a central role in mediatingsocial behaviour in animal models. Thus, reduction of

56

serotonin function led to avoidance of affiliative socialbehaviours in primates, while enhancement of serotonergicfunction resulted in increased prosocial behaviours(Raleigh et al, 1983). Similarly, free-ranging primates withlow CSF 5-HIAA had less social competence and emigratedfrom their social groups earlier (Mehlman et al, 1995;Raleigh et al, 1985). Interestingly, there are some datasuggesting that increasing serotonergic activity in humansis also associated with an increase in social affiliation,although the data are not altogether consistent. Therelationships between social behaviour and serotoninstatus are, however, complex. For example, removal ofdominant primates from the group results in significantdecreases in their serotonin levels (Raleigh et al, 1984).

There is little evidence for abnormalities in staticperipheral measures of serotonin function in SP (Stein etal, 1995; Tancer et al, 1994). Nevertheless, earlypharmacological ‘challenge’ studies with serotonergicagents, which assess the dynamic responsiveness of theserotonergic system, provided some support for serotonindysfunction in SP (Tancer et al, 1994). A low-activitypolymorphism of the serotonin transporter protein (5-HTTP) gene may be associated with anxiety-related traits,although this particular polymorphism may not berelevant to SP (Stein et al, 1998).

SSRIs are increasingly seen as the medication treatmentof choice for SP (Ballenger et al, 1998; van der Linden etal, 2000). Relatively little is known about the effects ofSSRIs on the functional neuroanatomy of SP. Nevertheless,treatment with SSRIs may be hypothesized to normalizedysfunctional circuitry in SP. Certainly, there is someevidence that SSRI treatment leads to reduced activity inamygdala-hippocampal, frontal, and cingulate regions inSP patients (van der Linden et al, 1999; Furmark et al,2002) (Figure 7.2).

Evolutionary considerations:appeasementalarms

Although it was Darwin who noted that humans are theonly animals to blush, it was Twain who made the astute

57

observation that only humans needed to (Twain, 1897).What indeed is the function of blushing? Given the closerelationship between blushing and SP, an answer to thisquestion might shed light on the distal (evolutionary)mechanisms that underlie this condition.

In the animal world, dominant and submissive statusare signalled by a range of mechanisms. Appeasementdisplays, for example, play an important role in indicatingacceptance of the status quo to a dominant conspecific (DeWaal, 1989). Is it possible that blushing serves as anappeasement display? Certainly, an embarrassed blushaccompanied by lowering of the gaze and a silly grin doesseem reminiscent of certain appeasement displays.Furthermore, empirical studies show that displays ofembarrassment do mitigate the negative reactions of others(Leary et al, 1992).

It turns out that blushing and social phobia havesomewhat similar demographics (more common in femalesand in younger people) and that both are elicited by similartriggers (social attention) (Stein and Bouwer, 1997b). Inaddition, a range of data would indicate that patents withSP misperceive information about the need for social

Figure 7.2Effects of SSRIs on the functional neuroanatomy of SP:normalized activity inamygdala, cingulate and basal ganglia.

58

appeasement (e.g. exaggerated view of the low status of theself, overestimation of social threat). Although theneurobiology of blushing remains incompletelyunderstood, it is possible to speculate that there arecertain overlaps with SP.

Evolutionary hypotheses about false alarms arestrengthened when cases of the pathological absence of aparticular alarm are noted. Are there some conditions inwhich there is insufficient social anxiety? Apart fromKlüver-Bucy syndrome (in which there is a loss of fear), itturns out that people with a hereditary condition know asWilliam’s disorder may be characterized by hypersociability(Bellugi et al, 1999). Such hypersociability can of coursepotentially land patients in all sorts of trouble! Theneurobiology of this condition remains incompletelyunderstood.

Conclusion

The ubiquity of social anxiety may go some way toexplaining the underdiagnosis and undertreatment ofsocial anxiety disorder. In this chapter the neurocircuitry ofsocial anxiety and of SP has been tentatively delineated.Serotonin is once again likely to play an important role inmediating both the symptoms of this disorder, and itsrobust response to treatment with SSRIs.

59

8.Conclusion

This volume has presented models of depression and eachof the major anxiety disorders. Insofar as these modelsintegrate findings in neurochemistry, neuroanatomy, andbehaviour, they can be characterized as representing acognitive-affective neuroscience approach. Whereasclinical work was once dominated by psychodynamicmodels, and academic psychology by a behaviouralapproach, a more integrative approach that incorporatesdata from a range of disciplines appears to be emerging.This approach is not only useful in the research setting,but provides the framework for integrative clinical practice.

Although there is some overlap between the differentmodels presented here, there are also importantdifferences between the neurocircuitry of the mood andanxiety disorders. It is interesting, for example, thatdepression is characterized by many areas of functionalhypoactivity, whereas anxiety disorders are oftencharacterized by areas of functional hyperactivity. This isconsistent with clinical experience, which attests both tothe frequent comorbidity of these conditions, and also tothe unique characteristics of each of them. There iscertainly some truth in the criticism that psychiatricnosology has not sufficiently incorporated psychobiologicalknowledge (van Praag, 1998). In the future we cancertainly expect that the neurocircuitry ofpsychopathological variations and spectrums will be morefully delineated.

In the first chapter of this volume we noted thatserotonergic neurons branched widely throughout thebrain. In subsequent chapters we showed how the SSRIs

are able to normalize a wide range of differentdysfunctions. The serotonin neurotransmitter system has along evolutionary history, and it has come to play a role ina broad range of different behaviours and emotions. Thisarguably goes some way to explaining the power ofserotonergic antidepressants in clinical settings. Certainly,the introduction of these agents into the modernpsychopharmacological armamentarium has constituted adramatic advance, and has provided researchers with theimportant challenge of developing integrative models of thebrain-mind.

61

References

Adolphs R.The neurobiology of social cognition. CurrOpinNeurobiol2001; 11:231–9.

Alexander CE, DeLong MR, Strick PL. Parallel organization offunctionally segregated circuits linking basal ganglia andcortex. Annu Rev Neurosci1986; 9: 357–81.

Allison T, Puce A, McCarthy G.Social perception from visual cues:role of the STS region. Trends Cognitive Sci2000; 4: 267–77.

Amies PL, Gelder MG, Shaw PM. Social phobia: a comparativeclinical study. Br J Psychiatry1983; 142:174–9.

Austin M-P, Mitchell P, Goodwin GM. Cognitive deficits indepression: possible implications for functionalneuropathology. BrJ Psychiatry2001; 178: 200-6.

Ballenger JC, Davidson JA, Lecrubier Y et al. Consensusstatement on social anxiety disorder from the InternationalConsensus Group on Depression and Anxiety. J ClinPsychiatry1998; 59:54–60.

Ballenger JC, Davidson JR, Lecrubier Y et al. Consensusstatement on generalized anxiety disorder from theInternational Consensus Group on Depression and Anxiety. JClin Psychiatry2001; 62S: 53–8.

Baxter LR, Schwartz JM, Bergman KS et al. Caudate glucosemetabolic rate changes with both drug and behavior therapyfor OCD. Arch Gen Psychiatry1992; 49:681-9.

Bechara A, Tranel D, Damasio H et al. Double dissociation ofconditioning and declarative knowledge relative to theamygdala and hippocampus in humans. Science1995; 269:1115–8.

Beck AT.Depression: Clinical, Experimental, andTheoreticalAspects.New York: Harper & Row, 1967.

Becker G, Berg D, Lesch KP et al. Basal limbic system alterationin major depression: a hypothesis supported by transcranialsonography and MRI findings. Int JNeuropsychopharmacol2001; 4: 21–31.

Bellugi U, Adolphs R, Cassady C et al. Towards the neural basisfor hypersociability in a genetic syndrome. Neuroreport1999;10:1653–7.

Bench CJ, Friston KJ, Brown RG et al. Regional cerebral bloodflow in depression measured by positron emissiontomography: the relationship with clinical dimensions.Psychol Med1993; 23: 579–90.

Birbaumer N, Grodd W, Diedrich O et al. fMRI reveals amygdalaactivation to human faces in social phobics.Neuroreport1998; 9: 1223–6.

Blumberg HP, Stern E, Martinez D et al. Increased anteriorcingulate and caudate activity in bipolar mania. BiolPsychiatry2000; 48:1045–52.

Bonne O, Brandes D, Gilboa A et al. Longitudinal MRI sudy ofhippocampal volume in trauma survivors with PTSD. AmJPsychiatry2001; 158:1248–51.

Bowlby J.Attachment and Loss,Vol 3. New York: Basic Books,1980.

Boyer W. Serotonin uptake inhibitors are superior to imipramineand alprazolam in alleviating panic attacks: a meta-analysis.IntClin Psychopharmacol1995; 10:45–9.

Bremner JD, Innis RB, Salomon RM et al. Positron emissiontomography measurement of cerebral metabolic correlates oftrypophan depletion-induced depressive relapse. Arch GenPsychiatry1997; 54:364–74.

Bremner JD, Innis RB, Southwick SM et al. Decreasedbenzodiazepine receptor binding in prefrontal cortex incombat-related posttraumatic stress disorder. Am JPsychiatry2000; 157:1120–6.

Brewin CR. A cognitive neuroscience account of posttraumaticstress disorder and its treatment. Behav Res ther2001; 39:373–93.

Brody AL, Saxena S, Stoessel P et al. Regional brain metabolicchanges in patients with major depression treated with eitherparoxetine or interpersonal therapy. Arch GenPsychiatry2001; 58: 631–40.

Buchsbaum MS, Hazlett E, Sicotte N et al. Topographic EEGchanges with benzodiazepine administration in generalizedanxiety disorder. Biol Psychiatry1985; 20:832–42.

Burgess N, Jeffrey KJ, O’Keefe J. The Hippocampal andParietalFoundations of Spatial Cognition.New York: OxfordUniversity Press, 1999.

Byrum CE, Ahearn EP, Krishnan KR. A neuroanatomic model fordepression. Prog Neuropsychopharmacol Biol Psychiatry1999;23:175–93.

63

Cendes FAFGP, Gambardella A, Lopes-Cendes I et al.Relationship between atrophy of the amygdala and ictal fearin temporal lobe epilepsy. Brain1994; 117:739–46.

Charney DS. Monoamine dysfunction and the pathophysiologyand treatment of depression. J Clin Psychiatry1998; 59S:11–4.

Cheyette SR, Cummings JL. Encephalitis lethargica: lessons forcontemporary neuropsychiatry. J Neuropsychiatry ClinNeurosci1995; 7:125–35.

Connor KM, Davidson JRT. The role of serotonin in posttraumaticstress disorder: neurobiology and pharmacotherapy.CNSSpectrums1998; 3:43–51.

Coplan JD, Gorman JM, Klein DF. Serotonin related functions inpanic-anxiety: a critical overview.Neuropsychopharmacology1992; 6:189–200.

Coplan JD, Lydiard RB. Brain circuits in panic disorder.BiolPsychiatry1998; 44:1264–76.

Damasio AR. The somatic marker hypothesis and the possiblefunctions of the prefrontal cortex. Philos Trans R Soc1996;351: 1413–20.

Darwin C. The Expression of Emotion in Man andAnimals.Chicago, IL: Chicago University Press, 1965.

Davidson JRT, Boyko O, Charles HC et al. Magnetic resonancespectroscopy in social phobia. J Clin Psychiatry1993a; 54S:19–25.

Davidson JRT, Hughes DL, George LK et al. The epidemiology ofsocial phobia: findings from the Duke EpidemiologicCatchment Area study. Psychol Med1993b; 23:709–18.

Davidson RJ. Asymmetric brain function, affective style andpsychopathology: the role of early experience and plasticity.DevPsychopathol1994; 6:741–58.

Davidson RJ, Putnam KM, Larson CL. Dysfunction in the neuralcircuitry of emotion regulation—a possible prelude toviolence. Science2001; 289:591–4.

Davis M, Whalen PJ. The amygdala: vigilance and emotion.MolPsychiatry2001; 6:13–34.

De Bellis M, Keshavan MS, Spencer S et al. N-acetylaspartateconcentration in the anterior cingulate of maltreated childrenand adolescents with PTSD. Am J Psychiatry2000a; 157:1175–7.

De Bellis MD, Casey BJ, Dahl RE et al. A pilot study of amygdalavolumes in pediatric generalized anxiety disorder. BiolPsychiatry2000b; 48:51–7.

64

De Waal F. Peacemaking among Primates.Cambridge, MA:Harvard University Press, 1989.

Delgado PL, Moreno FA. Hallucinogens, serotonin andobsessivecompulsive disorder. J Psychoactive Drugs1998; 30:359–66

Dolan RJ, Lane R, Chua P et al. Dissociable temporal lobeactivations during emotional episodic memory retrieval.Neuroimage2000; 11:203–9.

Duchaine B, Cosmides L, Tooby J. Evolutionary psychology andthe brain. Curr Opin Neurobiol2001; 11:225–30.

Duman RS, Malberg J, Nakagawa S et al. Neuronal plasticity andsurvival in mood disorders. Biol Psychiatry2000; 48:732–9.

El Mansari M, Bouchard C, Blier P. Alteration of serotonin releasein the guinea pig orbito-frontal cortex by selective serotoninreuptake inhibitors. Neuropsychopharmacology1995; 13:117–27.

Flint J. Freeze!Nat Genet1997; 17:250–1.Furmark T, Tillfors M, Marteinsdottir I et al. Common changes in

cerebral blood flow in patients with social phobia treated withcitalopram or cognitive-behavioral therapy. Arch GenPsychiatry2002; 59:425–33.

Germine M, Goddard AW, Woods SW et al. Anger and anxietyresponses to m-chlorophenylpiperazine in generalized anxietydisorder. Biol Psychiatry1992; 32:457–67.

Gorman JM, Kent JM, Sullivan GM et al. Neuroanatomicalhypothesis of panic disorder, revised. Am J Psychiatry2000;157: 493–505.

Gray JA, McNaughton N. The neuropsychology of anxiety. InHope, DA, ed. Perspectives on Anxiety, Panic, and Fear,Vol.43, Nebraska Symposium on Motivation. Omaha: Universityof Nebraska Press, 1996:61–134.

Halligan PW, Bass C, Wade DT. New approaches to conversionhysteria. BMJ2000; 320:1488–9.

Hamner MB, Lorberbaum JP, George MS. Potential role of theanterior cingulate cortex in PTSD: review and hypothesis. JNeuropsychiatry Clin Neurosci1999; 9:1–14.

Handley SL. 5-Hydroxytryptamine pathways in anxiety and itstreatment. Pharmacol ther1995; 66:103–48.

Heim C, Nemeroff CB. The role of childhood trauma in theneurobiology of mood and anxiety disorders: preclinical andclinical studies. Biol Psychiatry2001; 49:1023–9.

Higley JD, Mehlman PT, Taub DM et al. Excessive mortality inyoung free-ranging male non-human primates with low

65

cerebrospinal fluid 5-hydroxyindolacetic acid concentrations.Arch GenPsychiatry1996; 55:537–13.

Hollander E, Stein DJ, Broatch J et al. A pharmacoeconomic andquality of life study of obsessive-compulsive disorder.CNSSpectrums1997; 2:16–25.

Hugdahl K. Cortical control of human classical conditioning:autonomic and PET data. Pyschophysiology1998; 35:170–8.

Iny LJ, Pecknold J, Suranyi-Cadotte BE et al. Studies of aneurochemical link between depression, anxiety, and stressfrom [3H] imipramine and [3H] paroxetine binding on humanplatelets. BiolPsychiatry1994; 36:281–91.

Jacobs BJ, Fornal CA. Serotonin and behavior: a generalhypothesis. In: Bloom FE, Kupfer DJ, eds.Psychopharmacology: FourthGeneration of Progress.New York:Raven Press, 1995.

Johnson DL, Wiebe JS, Gold SM et al. Cerebral blood flow andpersonality: a positron emission tomography study. AmJPsychiatry1999; 156:252–7.

Kagan J, Reznick JS, Gibbons J. Biological basis of childhoodshyness. Science1988; 240:167–71.

Kandel ER. From metapsychology to molecular biology:explorations into the nature of anxiety. Am J Psychiatry1983;140:1277–93.

Kendler KS, Neale MC, Kessler RC et al. Major depression andgeneralized anxiety disorder: same genes, (partly) differentenvironments?Arch Gen Psychiatry1992; 49:716–22.

Kendler KS, Thornton LM, Gardner CO. Stressful life events andprevious episodes in the etiology of major depression inwomen: an evaluation of the ‘kindling’ hypothesis. Am JPsychiatry2000; 157:1243–51.

Kennedy SH, Evans KR, Krüger S et al. Changes in regional brainglucose metabolism measured with positron emissiontomography after paroxetine treatment of major depression.Am J Psychiatry2001; 158:899–905.

Kessler RC. The epidemiology of pure and comorbid generalizedanxiety disorder. A review and evaluation of recent research.ActaPsychiatr Scand2001;102:7–13.

Kessler RC, McGonagle KC, Zhao S et al. Lifetime and 12-monthprevalence of DSM-III-R psychiatric disorders in the UnitedStates: results from the National Comorbidity Survey. ArchGenPsychiatry1994; 51:8–19.

Kessler RC, Stang P, Wittchen H-U et al. Lifetime co-morbiditiesbetween social phobia and mood disorders in the US NationalComorbidity Survey. Psychol Med1999; 29:555–67.

66

Kessler RC, Stein MB, Berglund P. Social phobia subtypes in theNational Comorbidity Study. Am J Psychiatry1998; 155:613–9.

Kimbrell TA, George MS, Parekh PI et al. Regional brain activityduring transient self-induced anxiety and anger in healthyadults. Biol Psychiatry1999; 46:454–65.

Klein DF. Delineation of two drug-responsive anxiety syndromes.Psychopharmacologia1964; 5:397–408.

Klein DF. Anxiety reconceptualized. In: Klein DF, Rabkin J, eds.Anxiety: New Research and Changing Concepts.New York:Raven Press, 1981.

Klein DF. False suffocation alarms, spontaneous panics, andrelated conditions: an integrative hypothesis. Arch GenPsychiatry1993; 50:306–17.

Klüver H, Bucy PC. Preliminary analysis of functions of thetemporal lobes in monkeys. Arch Neurol Psychiatry1939; 42:979–1000.

Lane RD, Reiman EM, Ahern GL et al. Neuroanatomicalcorrelates of happiness, sadness, and disgust. Am JPsychiatry1997; 154: 926–33.

Le Doux J. Fear and the brain: where have we been, and whereare we going?Biol Psychiatry1998; 44:1229–38.

Leary MR, Cutlip WDI, Brit TW et al. Social blushing.Psychological Bull1992; 3:446–60.

Leckman JF, Grice DE, Boardman J et al. Symptoms ofobsessivecompulsive disorder. Am J Psychiatry1997; 154:911–7.

Lesch KP. Serotonergic gene expression and depression:implications for developing novel antidepressants. J AffectDisord2001; 62:57–76.

Lightowler S, Kennet GA, Williamson IJ et al. Anxiolytic-like effectof paroxetine in a rat social interaction test.PharmacolBiochem Behav1994; 49:281–5.

Lucey JV, Costa DC, Adshead G et al. Brain blood flow in anxietydisorders. OCD, panic disorder with agoraphobia, andposttraumatic stress disorder on 99mTcHMPAO singlephoton emission tomography (SPET). Br J Psychiatry1997;171:346–50.

MacFall JR, Payne ME, Provenzale JE et al. Medial orbital frontallesions in late-onset depression. Biol Psychiatry2001; 49:803–6.

MacLean PD. Psychosomatic disease and the visceral brain:recent developments bearing on the Papez theory of emotion.PsychosomMed1949; 11:338–53.

67

MacLeod AK, Byrne A. Anxiety, depression, and the anticipationof future positive and negative experience. J AbnormPsychol1993; 102:238–47.

Macmillian TM. Posttraumatic stress disorder and severe headinjury. Br J Psychiatry1991; 159:431–3.

Maier SF.Exposure to the stressor environment prevents thetemporal dissipation of behavioral depression/learnedhelplessness. Biol Psychiatry2001; 49:763–73.

Maier W, Gansicke A, Freyberger HJ et al. Generalized anxietydisorder (ICD-10) in primary care from a cross-culturalperspective: a valid diagnostic entity?Acta PsychiatrScand2000; 101:29–36.

Malhi GS, Bartlett JR. Depression: a role for neurosurgery?BrJNeurosurg2000; 14:415–22.

Malison RT, Price LH, Berman R et al. Reduced brain serotonintransporter availability in major depression as measured by[123I]-2ß-carbomethoxy-3ß-(4-iodophenyl) tropane and singlephoton emission computed tomography. Biol Psychiatry1998;44:1090–8.

Mann JJ, Huang YY, Underwood MD et al. A serotonintransporter gene promoter polymorphism (5-HTTLPR) andprefrontal cortical binding in major depression and suicide.Arch Gen Psychiatry2000; 57:729–38.

Marks IM, Nesse RM. Fears and fitness: an evolutionary analysisof anxiety disorders. Ethology and Sociobiology1994; 15:247–61.

Martin SD, Martin E, Rai SS et al. Brain blood flow changes indepressed patients treated with interpersonal psychotherapyof venlafaxine hydrochloride. Arch Gen Psychiatry2001; 58:641–8.

Mayberg HS. Frontal lobe dysfunction in secondary depression. JNeuropsychiatry Clin Neurosci1994; 6:428–42.

Mayberg HS, Brannan SK, Mahurin RK et al. Cingulate functionin depression: a potential predictor of treatment response.Neuroreport1997; 8:1057–61.

Mayberg HS, Liotti M, Branna SK et al. Reciprocal limbic-corticalfunction and negative mood: converging PET findings indepression and normal sadness. Am J Psychiatry1999; 156:675–82.

McDonald AW, III, Cohen JD, Stenger VA et al. Dissociating therole of the dorsolateral prefrontal and anterior cingulatecortex in cognitive control. Science2000; 288:1835–8.

McGuire M, Troisi A. Darwinian Psychiatry.New York: OxfordUniversity Press, 1998.

68

Mehlman PT, Higley JD, Faucher I et al. Correlation of CSF 5-HIAA concentration with sociality and the timing ofemigration in free-ranging primates. Am J Psychiatry1995;152:907–13.

Melfsen S, Osterlow J, Florin I. Deliberate emotional expressionsof socially anxious children and their mothers. J AnxDisord2000; 14:249–61.

Meyer JH, Swinson R, Kennedy SH et al. Increased left posteriorparietal-temporal cortex activation after D-fenfluramine inwomen with panic disorder. Psychiatry Res2000; 98:133–43.

Mineka S, Watson D, Clark LA. Comorbidity of anxiety andunipolar mood disorders. Annu Rev Psychol1998; 49:377–412.

Mishkin M, Petri H. Memories and habits: some implications forthe analysis of learning and retention. In: Squire LR, ButtersN. Neuropsychology of Memory.New York: Guilford Press,1984.

Montgomery SA. Obsessive compulsive disorder is not an anxietydisorder. Int Clin Psychopharmacol1993; S8:57–62.

Mundo E, Richter MA, Sam F et al. Is the 5-HT(1Dbeta) receptorgene implicated in the pathogenesis of obsessive-compulsivedisorder?Am J Psychiatry2000; 157:1160–1.

Murray CJL, Lopez AD. Global Burden of Disease:AComprehensive Assessment of Mortality and MorbidityfromDiseases, Injuries and Risk Factors in 1990 and Projectedto 2020,Vol. I. Harvard: World Health Organization, 1996.

Nesse RM. Is depression an adaptation?Arch Gen Psychiatry2000;57:14–20.

Nesse RM, Williams GC. Why We Get Sick: the New ScienceofDarwinian Medicine.New York: Vintage Books, 1994.

Nutt DJ, Bell CJ, Malizia AL. Brain mechanisms of social anxietydisorder. J Clin Psychiatry1998; 59:4–11.

Owens MJ, Nemeroff CB. Role of serotonin in the pathophysiologyof depression: focus on the serotonin transporter. ClinChem1994; 40:288–95.

Panskepp J. Affective Neuroscience: the Foundations of HumanandAnimal Emotions.New York: Oxford University Press, 1998.

Parks CL, Robinson PS, Sibille E et al. Increased anxiety of micelacking the serotonin l A receptor. Proc Natl Acad SciUSA1998; 95:10734–9.

Penava SJ, Otto MW, Pollack MH et al. Current status ofpharmacotherapy for PTSD: an effect size analysis ofcontrolled studies. Depress Anx1996; 4:240–2.

69

Pine DS, Weese-Mayer DE, Silvestri JM et al. Anxiety andcongenital hypoventilation syndrome. Am J Psychiatry1994;151:864–70.

Post RM. Transduction of psychosocial stress into theneurobiology of recurrent affective disorder. Am JPsychiatry1992; 149:999–1010

Potts NL, Davidson JR, Krishnan KR et al. Magnetic resonanceimaging in social phobia. Psychiatry Res1994; 52:35–42.

Raleigh MJ, Brammer GL, McGuire MT. Male dominance,serotonergic systems, and the behavioral and physiologicaleffects of drugs in vervet monkeys (Cercopithecus aethiopssabaeus). In: Miczek KA, ed. Ethopharmacology: PrimateModels ofNeurospychiatric Disorders.New York: Alan R Liss,1983.

Raleigh MJ, Brammer GL, McGuire MT et al. Dominant socialstatus facilitates the behavioral effects of serotonergicagonist. Brain Res1985; 348:274–82.

Raleigh MJ, McGuire MT, Brammer GL et al. Social andenvironmental influences on blood serotonin concentrationsin monkeys. Arch Gen Psychiatry1984; 41:405–10.

Ramboz S, Oosting R, Amara DA et al. Serotonin receptor 1Aknockout: an animal model of anxiety-related disorder. ProcNatlAcad Sci USA1998; 95:14476–81.

Rapoport JL, Ryland DH, Kriete M. Drug treatment of canineacral lick. Arch Gen Psychiatry1992; 48:517–21.

Rauch SL, Baxter LRJ. Neuroimaging in obsessive-compulsivedisorder and related disorders. In: Jenicke MA, Baer L,Minichiello WE eds. Obsessive-Compulsive Disorders:Practical Management,3rd edn. St Louis, MI: Mosby, 1998.

Rauch SL, Savage CR, Alpert NM et al. Probing striatal functionin obsessive compulsive disorder: a PET study of implicitsequence learning. J Neurosci1997a; 9:568–73.

Rauch SL, Savage CR, Alpert NM et al. The functionalneuroanatomy of anxiety: a study of three disorders usingpositron emission tomography and symptom provocation. BiolPsychiatry1997b; 42:446–52.

Rauch SL, Shin LM, Whalen PJ et al. Neuroimaging and theneuroanatomy of posttraumatic stress disorder. CNSSpectrums1998; 3:31–41.

Rauch SL, van der Kolk BA, Fisler RE et al. A symptomprovocation study of posttraumatic stress disorder usingpositron emission tomography and script-driven imagery.Arch Gen Psychiatry1996; 53:380–7.

70

Reber AS. Implicit Learning and Tacit Knowledge: an Essay ontheCognitive Unconscious.New York: Oxford University Press,1993.

Reiman EM, Fusselman MJ, Fox PT et al. Neuroanatomicalcorrelates of anticipatory anxiety. Science1989a; 243:1071–4.

Reiman EM, Raichle ME, Robins E et al. The application ofpositron emission tomography to the study of panic disorder.Am JPsychiatry1986; 143:469–77.

Reiman EM, Raichle ME, Robins E et al. Neuroanatomicalcorrelates of a lactate-induced anxiety attack. Arch GenPsychiatry1989b; 46:493–500.

Rickels K, Rynn MA. What is generalized anxiety disorder?JClinPsychiatry2001; 62S11:4–12.

Robins LN, Helzer JE, Weissman MM et al. Lifetime prevalence ofspecific psychiatric disorders in three sites. Arch GenPsychiatry1984; 41:949–58.

Robins TW, Brown VJ. The role of the striatum in the mentalchronometry of action: a theoretical review. RevNeurosci1990; 2: 181–213.

Robinson RG, Kubos KL, Starr LB et al. Mood disorders in strokepatients: importance of location of lesion. Brain1984; 107:81–93.

Rolls ET. A theory of emotion, and its application tounderstanding the neural basis of emotion. CognitionEmotion1990; 4:161–90.

Rob-Bryne PP, Katon W. Generalized anxiety disorder in primarycare. The precursor/modifier pathway to increased healthcare utilization. JClin Psychiatry1997; 58(Suppl.3): 34–8.

Roy-Byrne PP, Stang P, Wittchen H-U et al. Lifetime panic-depression comorbidity in the National Comorbidity Survey.Br JPsychiatry2000; 176:229–35.

Saint-Cyr JA, Taylor AE, Nicholson K. Behavior and the basalganglia. In: Weiner WJ, Lang AE, eds. Behavioral NeurologyofMovement Disorders.New York: Raven Press, 1995.

Salloway S, Malloy P, Cummings JL. The NeuropsychiatryofLimbic and Subcortical Disorders.Washington, DC: AmericanPsychiatric Press, 1997.

Sanchez MM, Ladd CO, Plotsky PM. Early adverse experience as adevelopmental risk factor for later psychopathology: evidencefrom rodent and primate models. Dev Psychopathology2001;13: 419–49.

71

Sapolsky RM. Glucocorticoids and hippocampal atrophy inneuropsychiatric disorders. Arch Gen Psychiatry2000; 57:925–35.

Sargent PA, Kjaer KH, Bench CJ et al. Brain serotonin 1A receptorbinding measured by positron emission tomography with[11C] WAY-100635: effects of depression and antidepressanttreatment. Arch Gen Psychiatry2000; 57:174–180.

Schneider F, Weiss U, Kessler C et al. Subcortical correlates ofdifferential classical conditioning of aversive emotionalreactions in social phobia. Biol Psychiatry1999; 45:863–71.

Schneier FR, Johnson J, Hornig CD et al. Social phobia:comorbidity and morbidity in an epidemiological sample. ArchGenPsychiatry1992; 49:282–8.

Seedat S, Stein DJ, Warwick J et al. Single photon emissioncomputed tomography before and after treatment with theselective serotonin reuptake inhibitor citalopram. IntJNeuropsychopharmacol2000; 2S: 378.

Segal ZV, Williams JM, Teasdale JD, Gemar M. A cognitivescience perspective on kindling and episode sensitization inrecurrent affective disorder. Psychol Med1996; 26:371–80.

Shear MK. Factors in the etiology and pathogenesis of panicdisorder: revisiting the attachment separation paradigm. AmJPsychiatry1996; 153:125–35.

Sobin C, Sackheim HA. Psychomotor symptoms of depression.AmJ Psychiatry1997; 154:4–17.

Soubrie P. Reconciling the role of central serotonin neurones inhuman and animal behavior. Behav Brain Sci1986; 9:319–64.

Stahl SM.Mechanism of action of serotonin selective reuptakeinhibitors: serotonin receptors and pathways mediatetherapeutic effects and side effects. J Affective Disord1998;51:215–35.

Staley JK, Malison RT, Innis RB. Imaging of the serotonergicsystem: interactions of neuroanatomical and functionalabnormalities of depression. Biol Psychiatry1998; 44:534–49.

Stein DJ. Single photon emission computed tomography of thebrain with tc-99m HMPAO during sumatriptan challenge inobsessive-compulsive disorder: investigating the functionalrole of the serotonin auto-receptor. ProgNeuropsychopharmacol BiolPsychiatry1999; 23:1079–99.

Stein DJ. The neurobiology of evil: psychiatric perspectives onperpetrators. Eth Health2000; 5:305–15.

Stein DJ. Comorbidity in generalized anxiety disorder: impactand implications. J Clin Psychiatry2001a; 62S: 29–36.

72

Stein DJ. Neurobiology of the obsessive-compulsive spectrum ofdisorders. Biol Psychiatry2001b; 47:296–304.

Stein DJ, Bouwer C. A neuro-evolutionary approach to theanxiety disorders. J Anx Disord1997a; 11:409–29.

Stein DJ, Bouwer C. Blushing and social phobia: aneuroethological speculation. Med Hypotheses1997b; 49:101-8.

Stein DJ, Hollander E. Impulsive aggression and obsessive-compulsive disorder. Psych Annals1993; 23: 389–95.

Stein DJ, Hollander E, Cohen L. Neuropsychiatry ofobsessivecompulsive disorder. In: Hollander E, Zohar J,Marazziti D, Olivier B, eds. Current Insights in Obsessive-CompulsiveDisorder.Chichester: Wiley, 1994.

Stein DJ, Hollander E, Liebowitz MR. Neurobiology of impulsivityand impulse control disorders. J Neuropsychiatry ClinNeurosci1993; 5: 9–17.

Stein DJ, Liu Y, Shapira NA et al. The psychobiology ofobsessivecompulsive disorder: how important is the role ofdisgust?CurrPsychiatry Reports2001; 3: 281–7.

Stein DJ, O’Sullivan R, Hollander E. Neurobiology oftrichotillomania. In: Stein DJ, Christenson GA, Hollander E,eds. Trichotillomania.Washington, DC: American PsychiatricPress, 1999a.

Stein DJ, Rapoport JL. Cross-cultural studies and obsessive-compulsive disorder. CNS Spectrums1996; 1: 42-6.

Stein DJ, Seedat S, Potocnik F. Hoarding: a review. IsraelJPsychiatry1999b; 36: 35–46.

Stein DJ, Shoulberg N, Helton K et al. The neuroethological modelof obsessive-compulsive disorder. Compr Psychiatry1992; 33:274–81.

Stein DJ, Stone MH. Essential Papers on Obsessive-CompulsiveDisorders.New York: New York University Press,1997.

Stein DJ, Westenberg H, Liebowitz MR. Social anxiety disorderand generalized anxiety disorder: serotonergic anddopaminergic neurocircuitry. J Clin Psychiatry2002; 63(Suppl.)12-9.

Stein DJ, Zungu-Dirwayi N, van der Linden GJ et al.Pharmacotherapy for posttraumatic stress disorder.CochraneDatabase of Systematic Reviews2000; 4:CD002795:

Stein MB, Chartier MJ, Kozak. Genetic linkage to the serotonintransporter protein and 5HT2A receptor genes excluded ingeneralized social phobia. Psychiatry Res1998; 81:283–91.

73

Stein MB, Delaney SM, Chartier MJ. Platelet [3H]-paroxetinebinding in social phobia: comparison to patients with panicdisorder and healthy volunteers. Biol Psychiatry1995; 37:224–8.

Stern E, Silbersweig DA, Chee K-Y et al. A functionalneuroanatomy of tics in Tourette’s syndrome. Arch GenPsychiatry2000; 57:741–8.

Stevens A, Price J. Evolutionary Psychiatry: a NewBeginning.London: Routledge, 1996.

Swedo SE, Leonard HL, Garvey M et al. Pediatric autoimmuneneuropsychiatric disorders associated with streptococcalinfections: clinical description of the first 50 cases. Am JPsychiatry1998; 155:264–71.

Tancer ME, Mailman RB, Stein MB et al. Neuroendocrinesensitivity to monoaminergic system probes in generalizedsocial phobia. Anxiety1994; 1:216–23.

Tauscher J, Bagby RM, Javanmard M et al. Inverse relationshipbetween serotonin 5-HT1A receptor binding and anxiety: a[11C] WAY-100653 PET investigation in healthy volunteers.Am JPsychiatry2001; 158:1326–8.

Thoren P, Asberg M, Bertilsson L. Clomipramine treatment ofobsessive-compulsive disorder. II. Biochemical aspects. ArchGenPsychiatry1980; 37:1289–94.

Tiihonen JF, Kuikka J, Rasanen P et al. Cerebral benzodiazepinereceptor binding and distribution in generalized anxietydisorder: a fractal analysis. Mol Psychiatry1997; 2:463–71.

Tillfors M, Furmack T, Marteinsdottir I et al. Cerebral blood flowin subjects with social phobia during stressful speakingtasks: a PET study. Am J Psychiatry2001; 158:1220–6.

Torgersen S. Comorbidity of major depression and anxietydisorders in twin pairs. Am J Psychiatry1990; 147:1199–202.

Tupler LA, Davidson JRT, Smith RD et al. A repeat protonmagnetic resonance spectroscopy study in social phobia. BiolPsychiatry1997; 42:419–24.

Twain M. Following the Equator,Vol. I. New York: Harper andBrothers, 1897.

van der Linden G, van Heerden B, Warwick J et al. Functionalbrain imaging and pharmacotherapy in social phobia: singlephoton emission tomography before and after treatment withthe selective serotonin reuptake inhibitor citalopram. ProgNeuropsychopharmacolBiol Psychiatry1999; 24:419–38.

van der Linden GJH, Stein DJ, van Balkom AJLM. The efficacy ofthe selective serotonin reuptake inhibitors for social anxiety

74

disorder (social phobia): a meta-analysis of randomizedcontrolled trials. Int Clin Psychopharmacol2000; 15S2:15–24.

van Praag HM. Inflationary tendencies in judging the yield ofdepression research. Neuropsychobiol1998; 37:130–41.

Videbach P. PET measurements of brain glucose metabolism andblood flow in major depressive disorder: a critical review.ActaPsychiatr Scand2000; 101:11–20.

Vythilingam M, Anderson ER, Goddard A et al. Temporal lobevolume in panic disorder—a quantitative magnetic resonanceimaging study. Psychiatry Res2000; 99:75–82.

Wang PW, Ketter TA. Biology and recent brain imaging studies inaffective psychoses. Curr Psychiatry Reports2000; 2:298–304.

Weissman MM, Bland RC, Canino GJ et al. The cross nationalepidemiology of obsessive compulsive disorder. J ClinPsychiatry1994; 55S: 5–10.

Weissman MM, Bland RC, Canino GJ et al. The cross-nationalepidemiology of social phobia: a preliminary report. IntClinPsychopharmacol1996; 11S:9–14.

Wu JC, Buchsbaum MS, Hershey TG et al. PET in generalizedanxiety disorder. Biol Psychiatry1991; 29:1181–99.

Yehuda R. Risk Factors for Posttraumatic StressDisorder.Washington, DC: American Psychiatric Press, 1999.

Yehuda R, McFarlane AC. Conflict between current knowledgeabout posttraumatic stress disorder and its originalconceptual basis. Am J Psychiatry1995; 152:1705–13.

Young GB, Chandarana PC, Blume WT et al. Mesial temporal lobeseizures presenting as anxiety disorders. J NeuropsychiatryClinNeurosci1995; 7:352–7.

Zald DH, Kim SW. Anatomy and function of the orbital frontalcortex, I: anatomy, neurocircuitry, and obsessive-compulsivedisorder. J Neuropsychiatry Clin Neurosci1996; 8:125–38.

Zohar J, Insel TR, Zohar-Kadouch RC. Serotonergic responsivityin obsessive-compulsive disorder: effects of chronicclomipramine treatment. Arch Gen Psychiatry1988; 45:167–72.

75

Index

adverse events, sequelae of6agoraphobia33amygdala2

anxiety disorders16, 18fear mediation32, 34, 37associated pathways38

reaction-action transition48social phobia53, 54threat perception16, 43

amygdala-dorsal striatumganglia54

amygdala-ventral striatum54amygdalo-thalamic pathways

see thalamo-amygdalapathways

animal studiesamygdala function48fear conditioning46GAD19OCD29, 31serotonergic system48

anterior cingulatePTSD47social phobia55

anterior temporal lobe16antisocial personality

disorder22anxiety disordersv

see also generalized anxietydisorder

appeasement alarms57appetite changes3, 4

avoidance behaviorsOCD33PTSD42, 43social phobia51, 52, 56

basal gangliaimaging studies48lesions27

‘basic fear circuit’16bed nucleus of stria terminalis

in fear mediation35, 43in free-floating anxiety16

behavioral inhibition54benzodiazepine

receptorbinding47biological explanations forGAD-

depression17blushing51, 54, 57brain imaging see imaging

studiesBroca’s area47buspirone19

classification see DSM criteria/classification

clomipramine29cognitive affective

neuroscience2cognitive approach/

explanations17cognitive deficits in

depression7

76

cognitive therapy9community studies

5-HTTP and trait anxiety19social phobia51

comorbiditydepression17, 22, 33, 53GAD17panic disorder33PTSD43social phobia53substance abuse53

compulsions23, 24, 24conditioned fear19, 34–5congenital hypoventilation

syndrome40cortical-striatal-thalamic-

cortical (CSTC) systemsdisgust mediation32in OCD26, 30ventral26

danger alarm43depression

GAD followed by17kindling hypothesis of13major3–12assessment3–4cognitive-affectiveconsiderations5complications3evolutionaryconsiderations11–12neuroanatomical model6neurochemistry10–10

medical precursors12neurocircuitry of6–8subtypes13symptoms3, 4in social phobia53

treatmentv, 30disgust32dorsal raphe neurons39DSM criteria/classification

generalized anxietydisorder15OCD symptoms24panic attacks34panic disorder35post-traumatic stressdisorder42, 43–3

dyspnea40

emotional memory systems5energy changes3, 4environment risk factors17ethological explanations17evolutionary advantage of

depression12evolutionary false alarms40,

58, 22evolutionary medicine12

facial signalling53, 54fear32fear conditioning19, 34, 40

implicit vs explicitprocessing35

functional imaging studies5–6,9

functional neuroanatomy seeneuroanatomy, functional

future alarm19–19future harm, anticipation of16,

17

GAD see generalized anxietydisorder

GAD-depression-somatization22

generalized anxiety disorder(GAD)14–15assessment15–15diagnosis14evolutionaryconsiderations19–20followed by depression17medical aspects16

77

neurocircuitry17–18prevalence14symptoms14, 15–16

genetic factorsin anxiety19GAD/MD17, 18

grooming31alarms31–1

hippocampus2in conflict situations16in fear conditioning/mediation35, 37, 43

hoarding31hostility35-HT see serotonin5-HTTP10, 11, 56hyperarousal42, 43hypersociability58hypoactivity vs hyperactivity59hypothalamus2

abnormalities6hypothalamus-pituitary-

adrenal (HPA) axis6

imaging studies5–6amygdala function46–5depression7–8fear conditioning46functional9GAD16–16, 18OCD28panic disorder37, 39psychotic mood disorders9PTSD46–5, 48in social phobia55

implicit vs explicitprocessing35

impulsivity10infections and OCD27inferior cortex55interest changes3, 4irritability3

kindling hypothesisofdepression13

Klüver-Bucy syndrome37, 55,58

lactate-induced panic37libido3limbic structures, neurological

pathways to5limbic systems47locus ceruleus39low positive affect construct5

major depression seedepression, major

MD see depression, majormedian raphe neurons39medications see

pharmacotherapymemory systems

emotional5in PTSD35, 43

mood regulation5–6morphometric studies,

postmortem7

neuroanatomical model ofdepression6

neuroanatomy, functionalanxiety53depression7GAD16, 18OCD28, 29PTSD47social phobia54, 56, 57

neurobiology of behavioralinhibition54

neurochemistrymajor depression10–10OCD29panic disorder38PTSD48social phobia55

neurocircuitry

78

depression6differences betweendisorders59GAD19generalized anxietydisorder17–19normalization by SSRIs13OCD26, 30panic disorder37, 39social cognition53social phobia54stroke6

numbing of responsiveness42,43

obsessions23, 24, 24obsessive compulsive disorder

(OCD)24as anxiety disorder31–1classification23, 24compulsivity vsimpulsivity31evolutionaryconsiderations31–1neurocircuitry/neurochemistry29PET symptom provocationdata18prevalence23symptoms23, 24treatment30

obsessive compulsivespectrum disorders24–5, 31

OCD see obsessive compulsivedisorder

Ondine’s curse40orbitofrontal regions in

emotional processing16

PANDAS26, 27panic attacks33, 34

lactate-induced37social phobia51subtypes35

panic disorder33–40evolutionaryconsiderations39–9neurocircuitry37and neurochemistry38–8

symptoms33, 35panic-depression33panicogenics54paralimbic circuits47paroxetine binding19, 48pCO239–9PET see positron emission

tomographypharmacotherapy9phobia18positron emission tomography

(PET) scanning10PTSD46–5social phobia55GAD18

positron emission tomography(PET) symptom provocationdata18

post-traumatic stress disorder(PTSD)42–8evolutionaryconsiderations49–8neurochemistry48neurocircuitry46–6risk factors49, 50serotonergic system in48–7symptoms42PET symptom provocationdata18

procedural strategies26, 31psychobiological effects of

trauma/stressors46psychobiological susceptibility

to PTSD48psychomotor function5psychosocial stressors13psychotherapeutic

intervention9

79

PTSD see post-traumaticstress disorder

re-experiencing trauma42, 43repetitive intrusive

symptoms24reward processing5–6

schemas46secondary depression6selective serotonin reuptake

inhibitors (SSRIs)v–2, 60for depression10–10imaging studies9, 10–10

for panic disorder38–8for PTSD48–7for social phobia56for GAD19–19for OCD29, 30

serotonergic markers10serotonergic neurons2serotonergic systemv–2, 60

decreased function10in GAD19–19neurocircuitry13in OCD29, 31in panic disorder38in PTSD48and response to stressors12role in depression10in social phobia55–4

serotoninevolutionary aspects22genotype11

serotonin reuptake inhibitors(SRIs)24

serotonin transporter (5-HTTP)10, 11, 56

shyness54social cognition53social interaction test19social phobia51

evolutionaryconsiderations57–6

neurocircuitry54–3and neurochemistry55–4prevalence51social status53–2subtypes52symptoms51ubiquity58

SSRIs see selective serotoninreuptake inhibitors

stressorsdepression12, 13GAD15

stria terminalis, bed nucleus ofin fear mediation35, 43in free-floating anxiety16

striatal neurocircuitry6, 55striatum2stroke, neurocircuitry of6suffocation alarms39–9suicide3sumatriptan30symmetry assessment31

tension14–14thalamo-amygdala

pathways38, 46thalamus2threat perception16

from conspecifics53Tourette’s syndrome26treatmentvtwin studies17, 18

weight changes4Williams’ disorder58worry14, 15

80

Serotonergic Neurocircuitry in Mood and Anxiety Disorders

Documents