Affective Processing in Major Depressive Disorder: Neuroanatomical Correlates of State and Trait Abnormalities

AFFECTIVE PROCESSING IN MAJOR DEPRESSIVE DISORDER: NEUROANATOMICAL CORRELATES OF STATE AND TRAIT ABNORMALITIES

Jakub Zbigniew Konarski

A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy Graduate Department of Institute of Medical Science University of Toronto

ABSTRACT

Doctor of Philosophy, 2010, Jakub Konarski, Institute of Medical Science, University of Toronto

Patients with MDD demonstrate impairments in various components of affective processing, which are believed to persist in the remitted phase of the illness and are believed to underlie the vulnerability for future relapse. Despite advances in neuropsychiatry, the neuroanatomical site of action of various treatment modalities remains unclear, leaving clinicians without an algorithm to guide optimal treatment selection for individual patients.

This thesis sought to characterize differences in brain activation during affective processing between MDD treatment responders (RS) and non-responders (NR) by combining clinical and neuroimaging variables in a repeat-measure functional magnetic resonance imaging (fMRI) investigation. We induced increases in positive and negative affect using visual stimuli under fMRI conditions in 21 MDD subjects and 18 healthy controls (HC).

Based on previous neuroimaging investigations and preclinical animal data, we hypothesized that increased activation of the amygdala and the pregenual cingulate during negative affect induction (NAI), and decreased activity of the ventral striatum during positive affect induction (PAI), would differentiate ultimate NR from RS. Following the first scan, treatment with fluoxetine and olanzapine was initiated in the MDD group, with follow-up scans at one- and six-weeks thereafter. We hypothesized that decreases in depressive symptoms would be associated with decreased activation of the ventromedial prefrontal cortex (PFC) and amygdala during NAI and increased activation of the hippocampus during PAI.

Eleven MDD subjects met criteria for clinical remission at study endpoint. Based on trait differences between MDD and HC, we hypothesized that differences observed during NAI would be limited to brain regions involved in regulation of the affective state, including the dorsolateral PFC and the anterior midcingulate cortex.

ii The results of the analyses confirmed the a-prior hypotheses and additionally demonstrated differential activation of the insular, medial temporal, and premotor cortex during repeat PAI and NAI between HC, RS, and NR. These findings provide: i) a neuroanatomical target of successful antidepressant therapy during PAI/NAI; ii) a differential effect of depressive symptoms and dispositional affect on brain activation during PAI/NAI; and iii) an a-prior method to differentiate RS from NR, and iv) demonstrate the need for additional treatment to prevent relapse in the remitted state.

iii ACKNOWLEDGMENTS

The collection of data that ultimately culminated in the production of this thesis would not be possible without the considerable aid from a number of colleagues, friends, and family. On a professional level, I would like to thank Drs. Sidney H. Kennedy and Roger S. McIntyre not only for their valuable academic contributions, but on a more personal note, their interest and contributions to the realization of my own career and personal ambitions. I am equally indebted to my extended supervisory team that includes my committee members; Drs. Karen Davis, Larry Grupp, and Jean St.-Cyr, and defense examiners; Drs George Awad, Jeff Daskalakis, Paul Sandor, Gwenn Smith, and Claudio Soares who have served as an extended supervisory family. I would also like to acknowledge the assistance of numerous other research staff in the acquisition and processing of MRI data, including Dr. Adrian Crawley, Mr. Shahryar Rafi-Tari, the Toronto Western Hospital Magnetic Resonance medical imaging staff, and the administrative and research staff of the Mood Disorders Psychopharmacology Unit. The analysis of MRI data has also been greatly facilitated by advice from Dr. Helen S. Mayberg, Emory University, Dr. Terence A. Ketter, Stanford University, Dr. Ralf Veit, University of Tübingen, and Dr. Wouter Depuydt, Katholieke Universiteit Leuven.

iv TABLE OF CONTENTS

ABSTRACT II

ACKNOWLEDGMENTS IV

TABLE OF CONTENTS V

LIST OF TABLES VIII

LIST OF FIGURES X

LIST OF ABBREVIATIONS XII

INTRODUCTION, OBJECTIVES AND AIMS 1

LITERATURE REVIEW 3

MAJOR DEPRESSIVE DISORDER 4 MAJOR DEPRESSIVE EPISODE 4 GLOBAL IMPACT 5 TREATMENT 7 MEASURING SEVERITY OF DEPRESSIVE ILLNESS 10 AFFECTIVE PROCESSING 14 THEORIES OF EMOTION 14 AFFECTIVE PROCESSING IN MDD 16 MEASURING AFFECT 18 INTERACTION BETWEEN DEPRESSIVE SYMPTOMS AND DISPOSITIONAL AFFECT 19 NEUROIMAGING 21 RADIONUCLIDE BASED FUNCTIONAL NEUROIMAGING 22 MAGNETIC RESONANCE IMAGING 23 COUPLING OF METABOLISM AND CIRCULATION 27 ORIGIN OF THE BOLD SIGNAL 28 ANALYSIS OF NEUROIMAGING DATA 30 NEUROANATOMY OF AFFECTIVE PROCESSING 32 NEUROANATOMY OVERVIEW 33 IDENTIFICATION OF EMOTIONAL SIGNIFICANCE 42 PRODUCTION OF AFFECTIVE STATE 44 REGULATION OF AFFECTIVE STATE 50 PHARMACOLOGY OF AFFECTIVE PROCESSING 55 NEUROTRANSMITTERS INVOLVED IN DISPOSITIONAL AFFECT 55 EFFECT OF PSYCHOTROPIC AGENTS OF AFFECTIVE PROCESSING 57 ABNORMALITIES IN BRAIN STRUCTURES MEDIATING AFFECTIVE PROCESSING IN MDD 63 STRUCTURAL FINDINGS 63 METABOLIC AND CEREBRAL PERFUSION 69 AFFECTIVE PROCESSING IN MDD 72 PHARMACOLOGICAL INDUCTION 73 RECALL INDUCTION 75 INDUCTION OF AFFECT WITH VISUAL STIMULI 81

AIMS AND HYPOTHESIS 96

METHODS 100

SUBJECTS 100 SAMPLE SIZE ANALYSIS 100 SUBJECT RECRUITMENT AND COMPENSATION 100 INCLUSION CRITERIA AND EXCLUSION CRITERIA 100 STUDY OUTLINE 101 RECRUITMENT 101 SCREENING VISIT 102 STUDY VISITS 102 TREATMENT 105 AFFECTIVE PROCESSING PARADIGM 105 NEUROIMAGING PARAMETERS 109 DATA ANALYSIS 110 CLINICAL DATA 110 NEUROIMAGING DATA 110

RESULTS 117

CLINICAL RESULTS 117 PRIMARY AIM A: 124 POSITIVE AFFECT NEUROIMAGING RESULTS AT BASELINE VISIT 124 NEGATIVE AFFECT NEUROIMAGING RESULTS AT BASELINE VISIT 133 PRIMARY AIM B: 142 REPEAT INDUCTION OF POSITIVE AFFECT – INTER SCAN CHANGES IN BOLD SIGNAL 142 REPEAT INDUCTION OF NEGATIVE AFFECT – INTER SCAN CHANGES IN BOLD SIGNAL 150 SECONDARY AIM A: 160 INFLUENCE OF DEPRESSIVE SYMPTOMS SEVERITY ON POSITIVE AFFECTIVE PROCESSING 160 INFLUENCE OF DEPRESSIVE SYMPTOMS SEVERITY ON NEGATIVE AFFECTIVE PROCESSING 166 SECONDARY AIM B: 170 INFLUENCE OF DISPOSITIONAL AFFECT ON POSITIVE AFFECTIVE PROCESSING 170 INFLUENCE OF DISPOSITIONAL AFFECT ON NEGATIVE AFFECTIVE PROCESSING 171 CORRELATION BETWEEN INDUCED AFFECT AND POSITIVE AFFECTIVE PROCESSING 175 CORRELATION BETWEEN INDUCED AFFECT AND NEGATIVE AFFECTIVE PROCESSING 177

DISCUSSION 181

SUMMARY OF FINDINGS 181 CONVERGENCE WITH PREVIOUS INVESTIGATIONS 187 DIFFERENCES IN POSITIVE AFFECTIVE PROCESSING: DEPRESSED VS. CONTROL SUBJECTS 187 DIFFERENCES IN NEGATIVE AFFECTIVE PROCESSING: DEPRESSED VS. CONTROL SUBJECTS 189 EFFECT OF EXPOSURE TO ANTIDEPRESSANT 193 EFFECT OF RESPONSE TO ANTIDEPRESSANT 194 AFFECTIVE PROCESSING IN EUTHYMIA 197 RESPONSE PREDICTION 199 EFFECT OF LATERALITY 200 PHARMACODYNAMICS OF OLANZAPINE FLUOXETINE COMBINATION 201 AFFECTIVE PROCESSING MODELS 204 OCCIPITOTEMPORAL CORTEX 205 PREMOTOR CORTEX 208 PRECUNEUS 210 CEREBELLUM 211 LIMITATIONS 216 SAMPLE SIZE 216 STATISTICAL PARAMETRIC MAPPING 217 UNIVARIATE STATISTICS 218 MULTIVARIATE ANALYSIS 219 PARAMETRIC DESIGN 221 BOLD SIGNAL INTERPRETATION 222 CLINICAL LIMITATIONS 223

CONCLUSIONS 224

FUTURE DIRECTIONS 228

REFERENCES 232

APPENDICES 307

APPENDIX 1 - AFFECTIVE PICTURE RATING SCALE 307 APPENDIX 2 - IAPS PHOTOGRAPH DESCRIPTIONS AND RATINGS – MALE – RUN1 308 APPENDIX 3 - IAPS PHOTOGRAPH DESCRIPTIONS AND RATINGS – MALE – RUN2 309 APPENDIX 4 - IAPS PHOTOGRAPH DESCRIPTIONS AND RATINGS – FEMALE – RUN1 310 APPENDIX 5 - IAPS PHOTOGRAPH DESCRIPTIONS AND RATINGS – FEMALE – RUN2 311

vii LIST OF TABLES

Table 1 – Study Participants Clinical and Demographic Characteristics: Comparison of HC and MDD Groups ...... 118 Table 2 – Study Participants Clinical and Demographic Characteristics: Comparison of RS and NR Groups ...... 118 Table 3 – Between Group Comparison of Study Participants’ Measures of Depression Severity, Dispositional Affect, and Induced Affect at the Baseline, Second, and Final Visit ...... 120 Table 4 – Between- and Within-Group Comparison of Study Participants’ Button Press Accuracy and Response Times at the Baseline, Second, and Final Visit for Positive-Neutral Affect and Negative-Neutral Affect Runs ...... 123 Table 5 – Activation and Deactivation in BOLD Signal Evoked by Positive Affective Visual Stimuli in All Subjects ...... 125 Table 6 – Changes in BOLD Signal between the Early and Late Components of the Positive Affective Block in All Subjects ...... 127 Table 7 – Between Group Differences in BOLD Signal Evoked by Positive Affective Stimuli during the Baseline Visit ...... 129 Table 8 – Group x Intra-Scan Habituation Interaction in BOLD Signal Evoked by Positive Affective Stimuli during the Baseline Visit ...... 131 Table 9 – Activation and Deactivation in BOLD Signal Evoked by Negative Affective Visual Stimuli in All Subjects ...... 133 Table 10 – Changes in BOLD Signal between the Early and Late Components of the Negative Affective Block in All Subjects ...... 135 Table 11 – Changes in BOLD Signal between the First and Subsequent Negative Affective Block in All Subjects ...... 135 Table 12 – Between Group Differences in BOLD Signal Evoked by Negative Affective Stimuli during the Baseline Visit ...... 138 Table 13 – Group x Intra-Scan Habituation Interaction in BOLD Signal Evoked by Negative Affective Stimuli during the Baseline Visit ...... 140 Table 14 – Inter-Scan Changes in BOLD Signal Evoked by Repeat Presentation of Positive Affective Visual Stimuli in HC ...... 142 Table 15 – Inter-Scan Changes in BOLD Signal Evoked by Repeat Presentation of Positive Affective Visual Stimuli in RS ...... 144 Table 16 – Inter-Scan Changes in Intra-Scan BOLD Signal Habituation Evoked by Positive Affective Visual Stimuli in RS (Time x Intra-Scan Habituation) ...... 144 Table 17 – Inter-Scan Changes in BOLD Signal Evoked by Repeat Presentation of Positive Affective Visual Stimuli in NR ...... 146 Table 18 – Inter-Scan Changes in Intra-Scan BOLD Signal Habituation Evoked by Positive Affective Visual Stimuli in NR (Time x Intra-Scan Habituation) ...... 146 Table 19 – Group by Time Interaction in BOLD Signal Evoked by Positive Affective Visual Stimuli in (Group x Time) ...... 148 Table 20 – Inter-Scan Changes in BOLD Signal Evoked by Repeat Presentation of Negative Affective Visual Stimuli in HC ...... 150 Table 21 – Inter-Scan Changes in Intra-Scan BOLD Signal Habituation Evoked by Negative Affective Visual Stimuli in HC ...... 152 Table 22 – Inter-Scan Changes in BOLD Signal Evoked by Repeat Presentation of Negative Affective Visual Stimuli in RS ...... 154 Table 23 – Inter-Scan Changes in Intra-Scan BOLD Signal Habituation Evoked by Negative Affective Visual Stimuli in RS ...... 154 Table 24 – Inter-Scan Changes in BOLD Signal Evoked by Repeat Presentation of Negative Affective Visual Stimuli in NR ...... 156 Table 25 – Inter-Scan Changes in Intra-Scan BOLD Signal Habituation Evoked by Negative Affective Visual Stimuli in NR ...... 156 Table 26 – Group by Time Interaction in BOLD Signal Evoked by Negative Affective Visual Stimuli in (Group x Time) ...... 158 Table 27 – Correlations between BOLD Signal Evoked by Positive Affective Visual Stimuli and Measures of Depressive Symptom Severity in MDD ...... 160 Table 28 – Between Group Comparison of Study Participants’ Measures of Depression Severity, Dispositional Affect, and Induced Affect at the Final Visit ...... 163 Table 29 – Between Group Differences in BOLD Signal Evoked by Positive Affective Stimuli during the Final Visit ...... 163 Table 30 – Correlations between BOLD Signal Evoked by Negative Affective Visual Stimuli and Measures of Depressive Symptom Severity in MDD ...... 166 Table 31 – Between Group Differences in BOLD Signal Evoked by Negative Affective Stimuli during the Final Visit ...... 168 Table 32 – Correlations between BOLD Signal Evoked by Positive Affective Visual Stimuli and Measures of Dispositional Affect in MDD ...... 170 Table 33 – Correlations between BOLD Signal Evoked by Negative Affective Visual Stimuli and Measures of Dispositional Affect in MDD ...... 173 Table 34 – Correlations between BOLD Signal Evoked by Positive Affective Visual Stimuli and Measures of Induced Affect in MDD While Viewing Positive Affective Visual Stimuli ...... 175 Table 35 – Correlations between BOLD Signal Evoked by Positive Affective Visual Stimuli and Measures of Induced Affect in HC While Viewing Positive Affective Visual Stimuli ...... 177 Table 36 – Correlations between BOLD Signal Evoked by Negative Affective Visual Stimuli and Measures of Induced Affect in MDD While Viewing Negative Affective Visual Stimuli . 177 Table 37 – Correlations between BOLD Signal Evoked by Negative Affective Visual Stimuli and Measures of Induced Affect in HC While Viewing Negative Affective Visual Stimuli ..... 179 Table 38 – Summary of Findings for Positive Affect – All Participants ...... 183 Table 39 – Summary of Findings for Positive Affect – Between Group Comparisons ...... 183 Table 40 – Summary of Findings for Negative Affect – All Participants ...... 184 Table 41 – Summary of Findings for Negative Affect – Between Group Comparisons ...... 184 Table 42 – Summary of Clinical-Neuroimaging Correlative Analyses in MDD ...... 185 Table 43 – Summary of Induced Affect Correlations in HC and MDD ...... 186

ix LIST OF FIGURES

Figure 1 – Brodmann Area (BA) Classification of the Cerebral Cortex (Gray, 1918) ...... 34 Figure 2 – Cingulate Gyrus Nomenclature – Vogt Classification (Vogt, 2005) ...... 36 Figure 3 – Subdivision of the Visual Cortex...... 38 Figure 4 – Subdivision of the Motor Cortex ...... 39 Figure 5 – Experimental Timeline – Timing of Research and Clinical Data Collection ...... 104 Figure 6 – Affective Induction Neuroimaging Paradigm ...... 108 Figure 7 – Modeling of Inter- and Intra- Block Habituation ...... 113 Figure 8 – Longitudinal Course of Depressive Symptom Severity, Dispositional Affect, and Induced Affect in the Study Participants...... 122 Figure 9 – Activation and Deactivations in BOLD Signal Evoked by Positive Affective Visual Stimuli in All Subjects ...... 126 Figure 10 – Main Effect of Intra-Block Changes in BOLD Signal between the Early and Late Components of the Positive Affective Block in All Subjects ...... 128 Figure 11 – Main Effect of Group: Between Group Differences in BOLD Signal Evoked by Positive Affective Stimuli during the Baseline Visit ...... 130 Figure 12 – Group x Intra-Scan Habituation Interaction in BOLD Signal Evoked by Positive Affective Stimuli during the Baseline Visit ...... 132 Figure 13 – Activation and Deactivations in BOLD Signal Evoked by Negative Affective Visual Stimuli in All Subjects ...... 134 Figure 14 – Main Effect of Intra-Block Changes in BOLD Signal between the Early and Late Components of the Negative Affective Block in All Subjects ...... 136 Figure 15 – Main Effect of Inter-Block Changes in BOLD Signal between the First and Subsequent Negative Affective Block in All Subjects ...... 137 Figure 16 – Main Effect of Group: Between Group Differences in BOLD Signal Evoked by Negative Affective Stimuli during the Baseline Visit ...... 139 Figure 17 – Group x Intra-Scan Habituation Interaction in BOLD Signal Evoked by Negative Affective Stimuli during the Baseline Visit ...... 141 Figure 18 – Inter-Scan Changes in BOLD Signal Evoked by Repeat Presentation of Positive Affective Visual Stimuli in HC ...... 143 Figure 19 – Inter-Scan Changes in BOLD Signal Evoked by Repeat Presentation of Positive Affective Visual Stimuli in RS and Inter-Scan Changes in Intra-Scan BOLD Signal Habituation (Time x Intra-Scan Habituation) ...... 145 Figure 20 – Inter-Scan Changes in BOLD Signal Evoked by Repeat Presentation of Positive Affective Visual Stimuli in NR and Inter-Scan Changes in Intra-Scan BOLD Signal Habituation (Time x Intra-Scan Habituation) ...... 147 Figure 21 – Inter-Scan Changes in BOLD Signal Evoked by Repeat Presentation of Positive Affective Visual Stimuli – Between Group Differences ...... 149 Figure 22 – Inter-Scan Changes in BOLD Signal Evoked by Repeat Presentation of Negative Affective Visual Stimuli in HC ...... 151 Figure 23 – Inter-Scan Changes in BOLD Signal Evoked by Repeat Presentation of Positive Affective Visual Stimuli in NR and Inter-Scan Changes in Intra-Scan BOLD Signal Habituation (Time x Intra-Scan Habituation) ...... 153 Figure 24 – Inter-Scan Changes in BOLD Signal Evoked by Repeat Presentation of Negative Affective Visual Stimuli in RS ...... 155 Figure 25 – Inter-Scan Changes in BOLD Signal Evoked by Repeat Presentation of Negative Affective Visual Stimuli in NR ...... 157 Figure 26 – Inter-Scan Changes in BOLD Signal Evoked by Repeat Presentation of Negative Affective Visual Stimuli – Between Group Differences ...... 159 Figure 27 – Correlations between BOLD Signal Evoked by Positive Affective Visual Stimuli and Measures of Depressive Symptom Severity in MDD ...... 162 Figure 28 – Main Effect of Group: Between Group Differences in BOLD Signal Evoked by Positive Affective Stimuli during the Final Visit ...... 165 Figure 29 – Correlations between BOLD Signal Evoked by Negative Affective Visual Stimuli and Measures of Depressive Symptom Severity in MDD ...... 167 Figure 30 – Main Effect of Group: Between Group Differences in BOLD Signal Evoked by Negative Affective Stimuli during the Final Visit ...... 169 Figure 31 – Correlations between BOLD Signal Evoked by Positive Affective Visual Stimuli and Measures of Dispositional Affect in MDD ...... 172 Figure 32 – Correlations between BOLD Signal Evoked by Negative Affective Visual Stimuli and Measures of Dispositional Affect in MDD ...... 174 Figure 33 – Correlations between BOLD Signal Evoked by Positive Affective Visual Stimuli and Measures of Induced Affect in MDD While Viewing Positive Affective Visual Stimuli ... 176 Figure 34 – Correlations between BOLD Signal Evoked by Negative Affective Visual Stimuli and Measures of Induced Affect in MDD While Viewing Negative Affective Visual Stimuli . 178

xi LIST OF ABBREVIATIONS

131Xe: Radionuclide xenon 133 15 O-H2O: Radionuclide oxygen 15 labeled water 18FDG: Radionuclide fluorine 18 labeled deoxyglucose 5-HT: 5-hydroxytryptamine 99mTc: Metastable nuclear isomer of technetium ACC: Anterior cingulate cortex aMCC: Anterior midcingulate cortex Amg: Amygdala AMDP: Association for methodology and documentation in psychiatry AMPT: α-methyl-para-tyrosine ANOVA: Analysis of Variance APRS: Affective picture rating scale ATP: Adenosine triphosphate BA: Brodmann area BD: Bipolar disorder BDI: Beck depression inventory BOLD: Blood oxygenation level dependent CAD: Coronary artery disease Cb: Cerebellum CBF: Cerebral blood flow CBT: Cognitive behavioural therapy CCHS: Canadian Community Healthcare Survey Cd: Caudate CGI-I: Clinical Global Impression - Improvement CGI-S: Clinical Global Impression - Severity CNS: Central nervous system CRF: Clinical research folder CT: Computed tomography Cu: Cuneus Cx: Cortex DA: Dopamine dACC: Dorsal anterior cingulate cortex dPCC: Dorsal posterior cingulate cortex DALY: Disability adjusted life years DCM: Dynamic causal modeling DCT: Discrete cosine transform DICOM: Digital Imaging and Communications in Medicine dlPFC: Dorsolateral prefrontal cortex dmPFC: Dorsomedial prefrontal cortex DSM-IV: Diagnostic and Statistical Manual for Diagnosis of Mental Disorders DTI: Diffusion Tensor Imaging ECT: Electroconvulsive Therapy EPI: Echoplanar imaging FDR: False discovery rate fMRI: Functional magnetic resonance imaging FWE: Family-wise error FWHM: Full-width at half maximum GABA: γ-amino-butyric acid HC: Healthy control subjects HDL: High density lipoproteins HDRS: Hamilton Depression Rating Scale Hi: Hippocampus Hy: Hypothalamus IAPS: International Affective Picture System ICA: Independent component analysis ICBM: International Consortium for Brain Mapping ITT: Intent to treat LDL: Low density lipoproteins LFBF: Low-frequency blood oxygen level-dependent fluctuations LFP: Local field potential LOCF: Last observation carried forward MAOI: Monoamine Oxidase Inhibitor MDD: Major depressive disorder MDE: Major depressive episode

xiii mEFP: Mean excitatory field potential MINI: Mini International Neuropsychiatric Interview MMRS: Mixed effects model repeated measure MNI: Montreal Neurological Institute MPL: Membrane phospholipid MRI: Magnetic resonance imaging MRS: Magnetic resonance spectroscopy MUA: Multi unit activity NAA: N-acetyl-aspartate NAI: Negative affect induction NAcs: Nucleus accumbens NCS: National Comorbidity Survey NE: Norepinephrine NIMH: National Institute of Mental Health NRI: Norepinephrine Reuptake Inhibitor NR: Major depressive disorder, non-responder OCD: Obsessive compulsive disorder OFC: Orbitofrontal cortex OPT: Occipitotemporal cortex pACC: Pregenual anterior cingulate cortex PAI: Positive affect induction PANAS: Positive and negative affect schedule PCA: Principal component analysis PCr: Phosphocreatinine PDE: Phosphodiester PET: Positron emission tomography PFC: Prefrontal cortex Pi: Inorganic phosphate pMCC: Posterior midcingulate cortex PME: Phosphomonoester rem Röngten equivalents in men REM: Rapid eye movement RM: Major depressive disorder, remitter

xiv ROI: Region of interest RS: Major depressive disorder, responder RSC: Retrosplenial cortex SD: Standard deviation SEM: Structural equation modelling sACC: Subgenual anterior cingulate cortex SGPFC: Subgenual prefrontal cortex SN: Substantia nigra SNRI: Serotonin Norepinephrine Reuptake Inhibitor SPECT: Single photon emission computed tomography SPM: Statistical parametric mapping SPSS: Statistical Package for the Social Sciences SSRI: Selective Serotonin Reuptake Inhibitor STAR*D: Sequenced Treatment Alternatives to Relieve Depression TCA: Tricyclic antidepressant TE: Echo time Th: Thalamus TR: Repetition time TSH: Thyroid stimulating hormone U: Uncus V1: Visual cortex area 1 V2: Visual cortex area 2 V3: Visual cortex area 3 V4: Visual cortex area 4 V5: Visual cortex area 5 VBR: Ventricular brain ratio vlPFC: Ventrolateral prefrontal cortex vmPFC: Ventromedial prefrontal cortex vPCC: Ventral posterior cingulate cortex WHM: White matter hyperintensities WM: Working memory YLD: Years lived with disability

xv INTRODUCTION, OBJECTIVES AND AIMS

Major depressive disorder (MDD) is a leading cause of disability globally and an important risk factor for the development of major medical disorders such as coronary artery disease (CAD) (Moussavi et al., 2007). The 1-year MDD prevalence in Canada has been estimated at 3.2% to 4.6% (Parikh and Lam, 2001) with a lifetime prevalence of 10.7% (n=4110) (McIntyre et al., 2006). Although the cardinal feature of MDD is persistently depressed mood(American Psychiatric Association, 1994), patients with major depressive disorder (MDD) also show impairments in various components of affective processing including the i) identification of emotional value of stimuli, ii) production and iii) regulation of affective state. These dysfunctional attitudes are believed to persist in the remitted phase of the illness and are believed to underlie the vulnerability of remitted MDD patients to relapse.

This thesis sought to characterize differences in brain activation during affective processing between treatment responder (RS) and non-responders (NR) by combining clinical and neuroimaging variables in MDD subjects undergoing pharmacotherapy by using a repeat- measure functional magnetic resonance imaging (MRI) study design. Specific aims were to i): evaluate differences in brain activation during affective processing between HC and MDD, and between RS and NR, ii) evaluate differences in inter-scan habituation on brain activation during repeat affective processing between HC, RS, and NR, iii) evaluate the association between depressive symptom reduction and brain activation during repeat affective processing in MDD, and iv) evaluate the association between self-rated dispositional and induced affect on brain activation during affective processing.

While it is generally accepted that regional differences in monoamine concentration and function underlie the pathophysiology of MDD, and that antidepressant medications pharmaco- dynamically modulate their levels, the neuroanatomical site of action of various treatment modalities remains unclear (Meyer, 2008;Nutt, 2008). The clinical translation of this mechanistic uncertainty resonates with results from multi-center trials which indicate that roughly 50% of patients do not adequately respond to first-line antidepressant treatment and only a third achieve asymptomatic remission (Trivedi et al., 2006b;Rush et al., 2006a). Moreover, there is no current algorithm to guide optimal treatment selection for individual patients (Konarski et al., 2009). This thesis sought to use fMRI data obtained during affective processing to identify treatment

1 non-responders before treatment initiation. Selecting patients who will ultimately not respond a- prior before initiating treatment has tremendous clinical translation value.

Previous neuroimaging investigations of affective processing have been limited by several methodological shortcomings. The majority of neuroimaging investigations in MDD have focused primarily on negative affect processing, in spite of clinical and pharmacological evidence suggesting that abnormalities in positive reward processing may underlie a biological vulnerability to relapse in MDD. Furthermore, in spite of recent efforts to examine pre-post differences in depressed patients, there has been an absence of neuroimaging studies that have assessed brain activation during affective processing during a period of symptom remission. A glaring oversight as the persistence of affective biases may predispose the patients to future relapse. To our knowledge, this thesis is the first investigation to explore affective processing in MDD under fMRI conditions during a period of asymptomatic remission.

Previous neuroimaging investigations of affective processing in MDD have not explored the stability of the BOLD signal during affective processing obfuscating the possibility of recording differences in signal habituation between groups. This thesis is the first to explore differences in BOLD signal habituation during affective processing within and between scans. This thesis addresses the limitations of earlier studies with a repeat measure design that permits delineation of the effect of trait, state, medication, and habituation on brain activation and deactivation during the induction of positive and negative affect. Characterization of differences in the brain’s response to affective visual stimuli between MDD and HC provides a neuroanatomical substrate for the clinical symptoms of MDD. The results of neuroimaging investigations have been a critical step in the development of alternative treatment strategies, including deep brain stimulation (Mayberg et al., 2005;Giacobbe et al., 2006;Lozano et al., 2008). Further integration of these brain regions with regional differences in receptor subtype expression in the brain can lead to novel selective pharmacological interventions.

Traditional approaches to the treatment of MDD rely on treating depressive symptoms with minimal attention devoted to restoration of positive affect and reduction of negative affect. The results presented herein suggest that the treatment of MDD begins with the reduction of depressive symptoms, but should not end until the restoration of the positive-negative affect equilibrium is achieved, to prevent future relapse.

2 LITERATURE REVIEW

This thesis evaluated the neural correlates of affective processing in patients diagnosed with MDD. This literature review is pertinent to the generation of the hypotheses being tested. Accordingly, the review begins with diagnostic definitions of MDD, its global impact, and a summary of treatment recommendations. Neuropsychological evidence supporting abnormalities in affective processing in MDD are presented next.

The theoretical and neuroanatomical models of emotion and affective processing are explored, specifically in the context of parsing affective processing into three simpler processes; identifying the emotional salience of a stimulus, production of an affective state, and regulation of the affective state. As contemporary investigations of affective processes rely heavily on neuroimaging techniques, a concise review of available modalities is also provided.

The neuroimaging evidence implicating affective processing abnormalities in MDD is reviewed chronologically, beginning with volumetric investigations, proceeding into measures of metabolism and blood flow. Finally, a review of previous neuroimaging investigations evaluating affective processing in MDD is provided, with a special emphasis on investigations employing fMRI scanning technology. The three most frequently utilized methods of inducing changes in affect are reviewed; pharmacological approaches, induction based on self-recall, and the use of visual presentations.

3 Major Depressive Disorder

Psychiatric disorders where a disturbance in mood is the predominant feature are diagnostically classified as mood disorders according to the Diagnostic and Statistical Manual of Mental Disorder – Fourth Edition (DSM-IV)(American Psychiatric Association, 1994). Mood disorders are categorized as Depressive Disorders (including Major Depressive Disorder, Dysthymic Disorder, Depressive Disorder Not Otherwise Specified), Bipolar Disorders (Bipolar I Disorder, Bipolar II Disorder, Cyclothymic Disorder, Bipolar Disorder Not Otherwise Specified), and two other disorders based on etiology; Mood Disorder Due to a General Medical Condition and Substance-Induced Mood Disorders. The Depressive Disorders are distinguished from Bipolar Disorders by the absence of a manic, mixed, or hypomanic episode (American Psychiatric Association, 1994).

Major Depressive Disorder (MDD) is characterized by the presence of one or more Major Depressive Episodes (MDE). Dysthymic disorder is characterized by at least two years of depressed mood for the majority of the time, accompanied by depressive symptoms that do not meet criteria for a MDE. Depressive Disorder Not Otherwise Specified is characterized by depressive symptoms that do not meet criteria for MDD, Dysthymic Disorder, or Adjustment Disorder with Depressed Mood (American Psychiatric Association, 1994).

Major Depressive Episode

The essential feature of a MDE is a period of at least two-weeks with depressed mood or the loss of interest or pleasure in nearly all activities. Additionally, the affected individual must experience at least three additional symptoms including change in appetite or weight, sleep, and psychomotor activity; decreased energy; feelings of worthlessness or guilt; difficulty in thinking, concentrating, or making decisions; recurrent thoughts of death or suicidal ideation, plans, or attempts. These symptoms must also meet the following criteria; be newly present or clearly worsened compared to the individual’s pre-episode status; persist for most of the day; and persist for nearly every day of at least two consecutive weeks. Finally, an MDE must be accompanied by a clinically significant impairment in social or occupational domain of function (American Psychiatric Association, 1994).

4 Global Impact

Major depressive disorder (MDD) is a prevalent medical disorder largely diagnosed and treated in primary-care settings (Von Korff et al., 2001). Currently, MDD is a leading cause of disability globally, and there is increasing evidence that it is an important risk factor for the development of major medical disorders such as coronary artery disease (CAD) (Glassman and Shapiro, 1998;Goodwin, 2006). Compared to other chronic medical disorders, MDD is associated with a significantly poorer long-term prognosis (Wells et al., 1989;Keller, 2004). More effective treatment of depression has been identified as a national health priority in the United States and elsewhere (Kessler et al., 2003;Demyttenaere et al., 2004).

Globally, depression has been listed as the fourth most important contributor to the burden of disease (Murray and Lopez, 1997). In the year 2000, MDD accounted for 4.5% of total worldwide disability adjusted life years (DALYs), and accounted for 12% of the total number of years lived with disability (YLD) worldwide (Ustun et al., 2004). Epidemiological investigations indicate that women have a two-fold higher risk for MDD compared to men (WHO, 2001). Indeed, the global incidence of MDD appeared to be more prevalent in women (4930 vs. 3199 per 100 000 year), with depressive disorders being the fourth most common cause of disease burden in women, compared to the seventh position among men (Meyer, 2004).

The 1-year MDD prevalence in Canada has been estimated at 3.2% to 4.6%; similar to rates reported internationally (Parikh and Lam, 2001). Recently, the lifetime MDD prevalence was estimated using results from Statistics Canada’s Canadian Community Health Survey: Mental Health and Well-Being (CCHS 1.2); a survey of 36,984 respondents (age 15-85) drawn from the Canadian household-dwelling population. Using diagnostic criteria outlined in the DSM-IV to screen respondents, the lifetime prevalence of MDD was estimated at 10.7% (n=4110) (McIntyre et al., 2006).

American estimates using The Epidemiological Catchment Area and National Comorbidity Survey (NCS) estimated that almost half (48 %) of the American population (age 15–54) will suffer from a mental disorder on a lifetime basis (Regier et al., 1988). It has been suggested that the high estimate may be over-represented by mild excursions of mood in response to mild distress or transient homeostatic responses to internal or external stimuli, rather than genuine

5 psychiatric disorders (Regier et al., 1998). Prevalence estimates from pharmaceutical prescription surveillance indicate that 11% of women and 5% of men in the United States are receiving antidepressants (Moncrieff, 2006).

In a recent NCS replication study, a criterion of ‘clinical significance’ was added to enhance the face validity (Kessing, 2007). Using DSM-IV criteria for psychiatric diagnoses, face-to-face interviews of 43,093 responders produced a response rate of 81 %. In an effort to differentiate between primary and secondary depressive syndromes, the temporal relationship between depressive symptoms, substance abuse or medical disorders was dissected (Hasin et al., 2005). The 12-month MDD prevalence was 5.3% (3.6% for men and 6.9% for women), with a lifetime prevalence of 13.3% (9.0% for men and 17.1% for women) (Hasin et al., 2005).

Subanalyses of the survey results indicated that the prevalence was highest for younger patients, aged 18–29 years and lowest for patients aged 65 years or more. Widowed, separated, divorced, and never married status, along with presence in the lower-income bracket were all associated with increased prevalence rates. On average, the age of MDD onset was in the early 30’s, with MDD responders reporting 4.7 lifetime MDEs. A variety of psychiatric comorbidities including personality disorders (38%), anxiety disorders (36%), and alcohol use disorders (14%) were frequently associated with MDD. Only 60% of MDD patients reported seeking treatment with lower rates among male respondents.

Depressive syndromes are clinically heterogeneous with variations in the directions and types of symptoms across several domains (i.e. emotional, cognitive, behavioural and somatic symptoms (McIntyre et al., 2004;Fava, 2002)). The frequency and response characteristics of the disparate depressive symptoms are dissimilar. For example, a large multinational survey noted that 69% of primary-care depressed patients reported only somatic symptoms as the primary reason for contacting their health-care provider (Simon et al., 1999;Corruble and Guelfi, 2000). Moreover, analyses within patients who are from non-industrialized nations suggest that somatic symptoms are more prevalent than psychological symptoms (Tseng, 2001).

6 Treatment

Although results from randomized controlled trials indicate that all of the commercially available antidepressants are reliably efficacious, the long-term outcome of this illness remains rather disappointing (Rudolph, 1999;Alexopoulos et al., 2005). Several modifiable deficiencies in the management of the depressed patient have been proposed to explain this chasm. These deficiencies include, but are not limited to: under-recognition of depression, failure to employ evidence-based treatments, poor adherence to therapy, insufficient management of co-morbidity, ineffectual therapeutic paradigms (e.g. brief treatment trials, insufficient dose), and failure to treat to full symptomatic remission (Rudolph, 1999). A number of clinical guidelines have been prepared to systematically guide treatment selection in the treatment of MDD.

In 2009 the Canadian Psychiatric Association and the Canadian Network for Mood and Anxiety Treatments (CANMAT) critically evaluated the quality of scientific evidence, and in conjunction with expert consensus clinical opinion, prepared standardized guidelines for the management of MDD (Kennedy et al., 2009a); focusing on classification, burden and principles of management (Patten et al., 2009), pharmacotherapy (Lam et al., 2009), psychotherapy in combination with medication (Parikh et al., 2009), neurostimulation therapies (Kennedy et al., 2009b), and complimentary/alternative medicine treatments (Ravindran et al., 2009a).

The majority of MDD patients remain treated with monoaminergic-based pharmacotherapy. Despite encouraging efficacy and tolerability data from registration trials of newer antidepressants, variability in patient response does not permit the selection of a specific first choice medications for all patients (Lam et al., 2009). The second generation antidepressants (selective serotonin reuptake inhibitors (SSRIs), newer dual-action antidepressants (venlafaxine, duloxetine, mirtazapine, and bupropion) all present Level 1 evidence for efficacy and tolerability and are all considered first-line MDD treatments. Meanwhile, the first-generation antidepressants ( tricyclic and monoamine oxidase inhibitors) may be considered as second- or third-line treatments. The CANMAT guidelines also recommend switching strategies and add-on strategies including lithium and atypical antipsychotics for inadequate or incomplete response (Lam et al., 2009).

The most frequently reported symptomatic outcome-measure in antidepressant clinical-trials has

7 been response to treatment, arbitrarily defined as >50% reduction from pretreatment in total symptom severity (Frank et al., 1991). A limitation of this categorical outcome is that many ‘responders’ fail to achieve a fully asymptomatic state that includes patients with ongoing clinically-significant disease activity. Subjects who show improvement in depressive symptom severity, but are not asymptomatic, are at risk for developing chronic depression, as well as, continued vulnerability for poor outcome and comorbid medical disorders (Judd et al., 1998a;Brown et al., 2004;Paykel, 1998). Achieving full remission is the treatment goal in major depressive disorder (Keller, 2003;Judd et al., 1998b;Judd et al., 1998a;Judd et al., 2000a;Judd et al., 2000b;Keller et al., 1992b;Keller et al., 1992a).

The Sequenced Treatment Alternatives to Relieve Depression (STAR*D) effectiveness trial was primarily designed to ascertain the optimal next medication step in patients non-remitting with SSRI monotherapy; the options were one of seven augmentation or switching strategies (Fava et al., 2003). The study employed a clinical equipoise, stratified, randomized design which enhanced the ecological validity of the results.

In the second level, patients not remitting with citalopram monotherapy (n=1997) were assigned to either a switch of antidepressant therapy (sertraline n=287, bupropion n=287, venlafaxine n =287, or cognitive psychotherapy n=204) or an augmentation of the citalopram (bupropion n=354, buspirone n=354, or cognitive psychotherapy n=224). Within the switch arm, remission rates between the three treatments did not differ significantly between treatments (bupropion 21.3 percent, sertraline 17.6 percent, and 24.8%) (Rush et al., 2006b). Similarly, within the augmentation arm comparable remission rates were also observed (bupropion 29.7 percent, buspirone 30.1 percent) (Trivedi et al., 2006a). Participants assigned to receive cognitive therapy (either alone or in combination with citalopram) experienced comparable remission rates to pharmacotherapy groups albeit at delayed time frames. Within the augmentation arm, the use of pharmacotherapy as an augmentation agent resulted in significantly more rapid remission than augmentation with cognitive therapy. In the switch arm, there were no significant differences in outcome, although patients who switched to another form of pharmacotherapy (versus cognitive therapy) reported significantly more side effects (Thase et al., 2007b).

The results of STAR*D trial demonstrate that after unsuccessful treatment with an SSRI, only one in four patients will experience remission of symptoms after switching to another

8 antidepressant medication. Although augmentation results in similar remission rates, an SSRI with another psychotropic resulted in numerically greater remission rates, the design of the trial did not a comparison of switching versus augmentation (i.e. most subject chose not to be randomized to all options). Future analyses will evaluate newer psychotropic agents with antidepressant properties, including the atypical antipsychotics (Rush, 2007).

A burgeoning literature base supports the use of atypical antipsychotics as augmenters of SSRI pharmacotherapy in the management of MDD. Early reports suggested that the combination of an atypical antipsychotic and an SSRI may have superior antidepressant effects to SSRI monotherapy in treatment resistant MDD populations. Positive trial results have been obtained with nearly all the commercially available agents in treatment resistant MDD including olanzapine (Shelton et al., 2001), risperidone (Ostroff and Nelson, 1999), aripiprazole, (Barbee et al., 2004), and ziprasidone (Papakostas et al., 2004).

The combination of fluoxetine and olanzapine has been one of the most studied atypical- antipsychotic combinations in treatment resistant MDD with superior efficacy compared to fluoxetine or olanzapine monotherapy (Corya et al., 2003;Thase et al., 2007a), the tricyclic antidepressant nortriptyline (Shelton et al., 2005), and the SNRI venlafaxine (Corya et al., 2006). Preliminary reports suggest that the combination of olanzapine and fluoxetine may offer higher efficacy in non-treatment resistant MDD populations. In a subanalysis of the combination of olanzapine and fluoxetine in MDD patients with and without prior treatment resistance, higher response (65% vs. 53%) and remission (60% vs. 44%) rates were noted in patients with MDD (Corya et al., 2003;Bobo and Shelton, 2009).

Recent randomized, placebo-controlled, double-blind, registration trials have demonstrated that the use of the atypical antipsychotic, quetiapine, provides rapid and sustained symptomatic improvement in the acute and maintenance treatment of MDD (McIntyre et al., 2009). Moreover, quetiapine XR may also offer advantages relative to duloxetine in time to onset of antidepressant action, and a tolerability advantages on measures of sexual dysfunction (McIntyre et al., 2009).

9 Measuring Severity of Depressive Illness

Standardized methods of clinical examination are necessary to objectively quantify psychopathological phenomena. The Hamilton Depression Rating Scale (HDRS) was originally designed as a 21-question multiple choice clinician-rated scale to assess the severity of depressive symptoms (Hamilton, 1960). The scale rates the severity of symptoms commonly observed in depressed patients, specifically depressed mood (1), guilt (2), suicidal thoughts (3), insomnia (4-6), work and interests (7), retardation (8), agitation (9), psychic (10) and somatic (11) anxiety, gastrointestinal (12) and general (13) somatic symptoms, genital symptoms (14), hypochondriasis (15), loss of insight (16) and weight (17). Questions 18-21 record the presence of diurnal variation (18), depersonalization (19), paranoid (20), or obsessional symptoms (21). Each of the 21 items has 3-5 possible responses (0, 1, 2, 3, and 4) with an increased score corresponding to increased severity.

The most frequently used version of the HDRS employing the first 17 questions (HDRS-17) has been the gold standard in the assessment of depressive symptom severity for over four decades. It has become the standard scale for use in clinical trials and has been estimated as the most commonly used measure of depression (Bagby et al., 2004;Demyttenaere and De Fruyt, 2003). In a recent critical review of its psychometric properties – reliability, item-response characteristics, and validity – it was concluded that internal, inter-rater, and retest reliability estimates for the overall scale are satisfactory, as are internal reliability estimates for each item. Criteria for convergent, discriminant, and predictive validity were also satisfied (Bagby et al., 2004;Beck et al., 1961).

The most frequently reported symptomatic outcome measure in clinical trials has been response to treatment, arbitrarily defined as a reduction of 50% or more in total symptom severity from a pretreatment assessment of the patient's depression . Over the past decade, a fully asymptomatic state of remission has been emphasized as a critical end point in the management of depressed patients. A universally agreed-upon criterion for remission, however, does not currently exist, however, the proposed definition and operational criteria for remission (HRSD-17 ≤ 7) put forth by the McArthur Foundation group, which is the definition of remission most cited, has served as a useful heuristic (Frank et al., 1991).

10 Self-rated instruments can also be used by patients to systematically describe past or current behavior. They are also economical for the assessor, and provide a method of eliminating observer bias .Their use, however, also has the potential of introducing both conscious and unconscious tendencies to falsify response. For example, patients may have a tendency to exaggerate certain characteristics of their illness, while concealing other symptoms. Positive response bias and social desirability factors may represent a further limitation of these instruments (Moller, 2009).

The Beck Depression Inventory (BDI) is a self-rated twenty-one item questionnaire used in the assessment of depressive symptom severity (Beck et al., 1961). (Steer et al., 1999). The original scale consisted of twenty-one questions evaluating the subject in the past week. Each question had a set of at least four possible answer choices (0, 1, 2, and 3), with increased severity represented by a higher score. A total score below 9 is indicative of a non-depressed individual; 10-18 indicates mild-moderate depression, 19-29 indicates moderate-severe depression and 30- 63 signifies severe depression.

The original scale (BDI) has been subsequently updated, first in 1971 (BDI-IA) to remove some items for which two different answers had the same score (i.e. 2a & 2b), and later in 1996 (BDI- II) in response to the revised criteria for MDD by the Diagnostic and Statistical Manual of Mental Disorders Fourth Edition (DSM-IV). The current iteration evaluates both cognitive and somatic components of depression. The cognitive subscale includes the items pessimism, past failures, guilty feelings, punishment feelings, self-dislike, suicidal thoughts, and worthlessness. The somatic subscale evaluates sadness, loss of pleasure, crying, agitation, loss of interest, indecisiveness, loss of energy, change in sleep patterns, irritability, change in appetite, concentration difficulties, tiredness, and loss of interest in sex (Edwards et al., 1984).

Although some clinicians have criticized the BDI for being a liberal measure of treatment outcome, others contend that interviewer-rating scales are too conservative measures of treatment gain. In a meta-analysis of 19 studies with 1,150 subjects that compared the predictive validity of the HDRS and the BDI, Edwards and colleagues found that the HDRS was more sensitive to change (Lambert et al., 1988). Lambert and colleagues, however, found that the BDI was more likely to show treatment effects later than the HDRS (Guy, 1976). The BDI-II was administered in the trial, and will be referred to as the BDI from this point forward.

More recent depression scales have tried to improve upon some of the limitations of the HDRS- 17 and BDI. The Montgomery Asberg Depression Rating Scale (MADRS) was developed to be particularly sensitive to treatment effect (Montgomery and Asberg, 1979), with an increased capacity to differentiate responders from non-responders. A retrospective chart analysis using receiver operating curves indicated that MADRS score of 31 was better than the HDRS-17 in separating between moderate and severe depression according to CGI criteria with 93.5 sensitivity and 83.3% specificity (Muller et al., 2003). Demyttenaere and De Fruyt relate differences between the HDRS-17 and MADRS according to their historical origin; specifically the HDRS and the MADRS may reflect antidepressant activity while the BDI reflects psychotherapy (Demyttenaere and De Fruyt, 2003). The two authors also caution about the limitations of each scale, with the risk of the HDRS-17 combining 'all depressions in one basket', while the MADRS may put 'all antidepressants in one basket' (Demyttenaere and De Fruyt, 2003).

The briefer 7 item HDRS, otherwise known as the HAMD-7, has also been validated as a psychiatric tool for symptom measurement that can indicate when remission, with psychometric properties comparable to the HRSD-17 and the MADRS (McIntyre et al., 2005;McIntyre et al., 2007) Several other brief rating scales for depression have also been validated and reviewed (Faries et al., 2000;Maier et al., 1988;Spitzer et al., 1999;Zimmerman et al., 2004).

Whereas the HDRS-17 and the BDI measure the intensity of depressive symptoms, they do not provide a global impression of the illness. The clinical global impression (CGI) is a three-item scale traditionally used to assess treatment response in psychiatric patients. The three items include the severity of illness (CGI-S), global improvement (CGI-I), and the efficacy index (CGI-EI). The CGI-S is rated on a seven-point scale from (1=normal to 7=extremely ill); the CGI-I on a seven-point scale (1=very much improved to 7=very much worse); and the CGI-EI on a four-point scale (from 'none' to 'outweighs therapeutic effect') (Guy, 1976).

The CGI-S is an evaluation of the severity of the patient's illness at the time of assessment, based on the clinician's past experience with patients having the same diagnosis. The options, and applicable scores, are: not at all ill (1); borderline mentally ill (2); mildly ill (3); moderately ill (4); markedly ill (5); severely ill (6); or extremely ill (7). The CGI-I is an evaluation of how the

12 patient's illness has improved or worsened relative to a baseline state. The options, and applicable scores, are: very much improved (1); much improved (2); minimally improved (3); no change (4); minimally worse (5); much worse (6); or very much worse (7). In a retrospective chart review on 139 MDD subjects who participated in six randomized, placebo-controlled, clinical trials, Khan and colleagues found that that the effect sizes of the HDRS-17 and CGI-I/S were significantly similar to each other in their ability to detect antidepressant-placebo differences (Khan et al., 2002;Khan et al., 2004).

Despite poor clinical outcomes associated with current antidepressant treatment, limited emphasis has been placed on psychosocial or functional outcomes. Kennedy and colleagues reviewed published data on long-term social functioning after depression and its temporal relationship with clinical recovery (Kennedy et al., 2007a). Although there is a dearth of investigations to evaluate long-term functional impairment, the extant literature suggests that psychosocial impairment persists even after clinical remission. Moreover, residual symptomatology and neurocognitive deficits may lead to enduring psychosocial impairment. While Axis I and II comorbidities were found to predict a poor psychosocial outcome, the number of depressive episodes did not appear to be associated with personality 'scarring' (Kennedy et al., 2007a).

13 Affective Processing

Theories of Emotion

The ability to efficiently identify, integrate, and form appropriate responses to emotionally salient information in the environment, including danger and reward, is critical to survival (Darwin C, 1872;Plutchik, 1984). In the latter part of the 19th century, the prevailing thought was that emotions were the result of the autonomic nervous system coordinating physiological events such as muscular tension, a rise in heart rate, perspiration, and dryness of the mouth in response to external stimuli. Emotions were perceived as the subjective experience resulting from these physiological changes, rather than being their cause (James, 1884). A deficiency of the model was that it could not account for the difference in the onset of the visceral response and the reported emotion, and for the multiplicity of emotional states generated by comparable autonomic responses.

In the first half of the 20th century, Walter Canon and Philip Bard proposed that individuals react to emotions first and produce an appropriate autonomic response thereafter. The theory departed from James-Lange’s theory in abdicating a mechanism for emotion, and advocating that one reacts to a specific stimulus and experiences the corresponding emotion simultaneously, and that one can only react to a stimulus after experiencing the related emotion (Cannon, 1927;Cannon, 1929). In the early 1960s, Stanley Schachter and Jerome Singer brought forth the Two Factor Theory of Emotion proposing that emotion is the integration of the cognitive interpretation and the physiological arousal (Schachter and Singer, 1962).

In a pivotal experiment in social psychology, 184 college students were randomized to receive either adrenaline or saline injections, with instructions that they had received vitamins to test their vision. Sympathetic autonomic effects including tachycardia, increased breathing, and increased blood flow to the muscles and brain were observed exclusively in the active (adrenaline arm). A third of the subjects were informed about the physiological side-effects of the adrenaline, a third were purposefully deceived and informed about a dull headache and numbness, while another third did not receive any preparatory information. Following the injections, all subjects interacted with members of the investigation team who displayed either a playful or angry affect. Subjects who had been misinformed about the injection's side-effects

14 were found to behave similarly to the investigator’s aide, taking cues from the induced situation to interpret their arousal level and determine their emotional state. Conversely, subjects who were anticipating the autonomic effects of adrenaline, did not manifest emotional mirroring (Schachter and Singer, 1962).

Richard Lazarus believed that appraisal is a necessary as well as sufficient cause of emotion and that knowledge is necessary but not sufficient. He contrasted automatic processing without awareness with deliberate and conscious processing, and advocated the concept of resonance between an individual’s needs and environmental stimuli (Lazarus, 1991a), and invoked cognitive-motivational-relational theory to explain and predict anger, anxiety, sadness, and pride (Lazarus, 1991b). Contemporary proponents of appraisalist theories of emotion have emphasized that appraisal or identification of stimulus salience, whether conscious or subconscious, precedes the generation of emotional response (Clore and Ortony, 2000;Davidson and Irwin, 1999).

Antonio Damasio described emotion as a “collection of changes in both body state and brain state responding to content of one’s thoughts relative to a particular event” (Damasio, 1995). Moreover, Damasio made the distinction between responses toward the body-that resulted in a specific body state, and those toward the brain which resulted in a specific mode of operation that leads to a change in cognitive style. Both responses may also produce physiological modifications, which are perceptible to an external observer, including changes in skin color, body posture, and facial expression, this externally displayed behaviour is referred to as affect (Davidson, 1998).

Positive and negative affect consistently emerge as the first two factors in factor analyses of self- rated mood, facial expressions, and mood terms (Diener et al., 1985;Russell, 1980;Russell, 1991;Stone, 1981;Zevon and Tellegen, 1982;Watson et al., 1984). In light of their strong negative correlation, positive and negative affect are highly distinctive dimensions, represented as orthogonal dimensions in factor analytic studies of affect (Watson and Tellegen, 1985). According to Auke Tellegen, positive affect reflects the extent to which a person feels enthusiastic, active, and alert, with high positive affect being characterized by high energy, concentration, and pleasurable engagement. Negative affect, on the other hand is a general dimension of subjective distress and unpleasurable engagement that encompasses a variety of aversive mood states, including anger, contempt, disgust, fear, guilt, and nervousness (Tellegen,

15 1985).

Two basic systems mediating different positive and negative affect have been proposed; the approach and withdrawal systems (Cacioppo and Gardner, 1999;Davidson and Fox, 1982;Lang et al., 1990). The approach system is believed to generate certain positive affect that are approach-related and generated in the context of moving toward a desired goal (e.g. enthusiasm, pride, etc.) (Stein and Trabasso, 1992). The corollary to the approach system, the withdrawal system coordinates the withdrawal of an individual from sources of aversive stimuli, and the generation of certain forms of negative affect that are withdrawal-related; fear and disgust increase the distance between the organism and the source of the aversive stimuli (Davidson and Irwin, 1999). Affective processing is a broad term that encompasses multiple processes representing affect in the absence of immediate rewards and punishments, as well as, different aspects of emotional regulation (Davidson, 2002).

Philips and colleagues have modeled emotions as a series of mental process occurring after the presentation of a stimulus with emotional salience, which allows the generation of complex affective states, and behaviours: 1) the appraisal and identification of the emotional significance of the stimulus; 2) the production of a specific affective state in response to the stimulus, and 3) the regulation of the affective state and emotional behaviour (Phillips et al., 2003a). The brain regions implicated in each of these processes are summarized following a neuroanatomical overview of the limbic system.

Affective Processing in MDD

Although, disturbances in both cognitive and affective processing domains have been described in MDD patients (Elliott et al., 1998;DeBattista, 2005;Murphy et al., 2001), the evidentiary database supporting cognitive deficits (assessed with neuropsychological test batteries evaluating attention and psychomotor speed, verbal and visual memory, and problem solving) in MDD is equivocal, with some studies reporting no statistically significant difference compared to psychiatrically unaffected controls, (Goldberg et al., 1993;Sweeney et al., 2000) and significantly less dysfunction as compared to depressed bipolar patients (Borkowska and Rybakowski, 2001). It is likely that cognitive impairments are partially depressive state-dependent as positive correlations between the severity of depressive symptoms (Smith et al., 1994) and degree of

16 cognitive impairment have been reported. Moreover, there may be a cumulative effect of the depressive illness, as the duration and chronicity of the depressive illness correlates with the severity of reported cognitive deficits (Borkowska and Rybakowski, 2001;MacQueen et al., 2003).

A more robust, well-established literature base supports impaired affective processing in MDD. Specifically, deficits in the identification of the emotive value of a stimulus, the production of affective states, and the regulation of affective states has been observed. Rubinow and Post found that depressed subjects made significantly fewer correct matches for sad, happy, and interested face items during a facial affect recognition task (Rubinow and Post, 1992). Moreover, neuropsychological tests support the idea of a bias towards negative affective stimuli, or decreased sensitivity to positive stimuli (David and Cutting, 1990;Teasdale and Fogarty, 1979;Teasdale and Russell, 1983). In facial discrimination tasks, Gur and colleagues reported that MDD subjects balanced for age and sociodemographic characteristics with psychiatrically unaffected HC, were impaired on measures of sensitivity for happy discrimination and specificity for sad discrimination (Gur et al., 1992). Additionally, severity of negative affect was correlated with poorer performance for subjects.

When bipolar manic and MDD subjects were directly compared with controls on an affective shifting task requiring inhibitory control over different components of cognitive and emotional processing, Murphy and colleagues found that depressed subjects were impaired in their ability to shift the focus of attention. Additionally, depressed subjects also exhibited an affective bias for negative stimuli (Murphy et al., 1999). Modification of a Stroop color naming task to include words of self-descriptive emotional information, allows one to study the processing of affective self-referential stimuli (Williams et al., 1996). In a comparison of individuals with MDD (n=58) and HC participants (n=44), Segal and colleagues demonstrated that MDD participants exhibited increased reaction times for self-descriptive negative targets than for any other prime-target condition (Segal et al., 1995). In comparison, HC did not show any effect of prime-target relation for material in either positive or negative valence.

By examining the impact of distracting emotional information during performance on a working memory task, Ladouceur and colleagues investigated mechanisms mediating affective regulation in i) psychiatrically unaffected children, and children meeting criteria for ii) MDD, iii) an anxiety

17 disorder, or iv) MDD comorbid with anxiety (Ladouceur et al., 2005). Both the MDD and comorbid MDD+anxiety groups demonstrated significantly longer reaction times to negative affective versus neutral backgrounds. As the E-n-back task that was used engages higher-order cognitive processes, the authors concluded that alterations in affective processing, specifically regulation, interferes with cognitive processes that govern how attentional resources are allocated (Ladouceur et al., 2005).

In summary, MDD is a prevalent, chronic, and costly psychiatric disorder. Extant clinical trial data suggest that response to a conventional unimodal antidepressant is unsatisfactory, with asymptomatic remission being achieved by less than a third of medicated patients. The practice of augmenting traditional antidepressants with other psychotropics, particularly atypical antipsychotics, has demonstrated increased efficacy over monoaminergic antidepressant monotherapy. Neuropsychological analyses provide compelling evidence to suggest altered affective processing in MDD patients, however, the depressive state dependent, and depressive state independent (trait) components of these abnormalities have not been well elucidated. A theoretical discussion of emotional responses, particularly the production and regulation of affective states, follows.

Measuring Affect

The Positive Affect and Negative Affect Schedule (PANAS) scale was developed in an effort to produce a reliable quantification of positive and negative dispositional affect. The PANAS is a self-report questionnaire composed of two 10-item scales that separately assess positive and negative affect in the past week, or the amount of dispositional positive and negative affect.. The positive subscale includes the following items: interested, excited, strong, determined, enthusiastic, proud, alert, inspired, attentive, and active. Meanwhile, the negative subscale includes the following items: distressed, upset, guilty, hostile, irritable, ashamed, nervous, scared, jittery, and afraid. Each item is rated on a five point scale from very slight, or not at all (1), a little (2), moderately (3), quite a bit (4), and extremely (5) (Watson et al., 1988).

In the original validation, Watson and colleagues reported that the scales were brief and easy to administer. The authors also noted that over a 2 month period the scales were stable, exhibit high internal consistency, and largely uncorrelated. Satisfactory levels of factorial, convergent, and

18 discriminant validity were also observed (Crawford and Henry, 2004). More recently, Crawford and Henry administered the PANAS to a large non-clinical sample, broadly representative of the general adult UK population (N = 1,003) (Mayberg et al., 2005;Anderson and Phelps, 2002;Davidson et al., 2003). The authors noted that PANAS continues to be a reliable and valid measure of positive and negative affect, although there is evidence that the positive and negative subscales may not be completely independent of each other. The PANAS has previously also been successfully employed in the measurement of affect in neuroimaging investigations of MDD subjects (Malhi et al., 2004b;Malhi et al., 2004a).

In this study, the method described by Malhi and colleagues has been used to construct an instrument to measure the valence and amount of affect generated by the affective induction. Malhi and colleagues advised their subjects that they would be asked to complete a questionnaire after the scanning session evaluating the stimuli that were presented. They asked subjects to rate each stimulus on a nine point scale (−4 to +4), according to the amount of negative or positive affect that was generated at the time of the scan, with a score of 0 indicating none at all, 1 slight, 2 moderate levels, 3 very much, and 4 extreme (Kumari et al., 2003).

Interaction between Depressive Symptoms and Dispositional Affect

Few studies have systematically documented changes in dispositional affect that accompany treatment of a major depressive episode, and most of these studies evaluated changes in dispositional affect as part of a larger antidepressant clinical trial. For example Davidson and colleagues found that eight weeks of venlafaxine pharmacotherapy (a serotonin-norepinephrine reuptake inhibitor) was associated with statistically significant increase in positive affect and decrease in negative affect in 5 MDD subjects (Davidson et al., 2003). More importantly, in comparison to the HC group, the researchers noted that there were no corresponding changes in either positive or negative affect in the control group. Moreover, after eight weeks of treatment, negative affect levels in both groups were comparable, whereas levels of positive affect continued to be significantly lower in the depressed group.

The affect regulating properties of venlafaxine have also been compared against traditional specific serotonin reuptake inhibitors (Dichter et al., 2005). In a double blind study, 20 MDD subjects were randomly assigned to 12 weeks of treatment with either venlafaxine (225mg/day)

19 or paroxetine (30 mg/day). The authors hypothesized that while the serotonergic actions of both antidepressants would yield comparable therapeutic effects on negative affect, the catecholaminergic effects of venlafaxine would be associated with greater improvements in positive affect. Contrary to their hypothesis, both agents were equally effective in reducing negative affect and increasing positive affect. The authors concluded that increases in positive affect may proceed through two different pharmacological pathways, or paroxetine may have a clinically relevant noradrenergic effect at 30 mg/day. An alternative explanation is that changes in positive affect may be secondary to decreases in negative affect.

The therapeutic effects of bupropion, a norepinephrine-dopamine reuptake inhibitor were also assessed on mood, anxiety symptoms, and positive dispositional affect (Tomarken et al., 2004). In this placebo-controlled design, 19 MDD subjects were initially treated with bupropion (300 mg/day) or placebo for 6 weeks. During the second open-label phase, previously bupropion- medicated subjects previously had their dose increased to 400 mg/day, while the placebo began treatment with 300 mg/day of bupropion. The investigators noted that during the blind phase of the trial bupropion treatment was associated with greater improvement in mood, anxiety levels, and ratings of positive affect. Interestingly, the weakest placebo effects were noted in the reduction of anhedonia, suggesting the specificity of the treatment on increasing levels of positive affect. Further analysis revealed that the first phase of treatment was associated with a reduction of anhedonia, whereas the second phase was associated with an increase in positive affect.

Evaluation of change in dispositional affect with various forms of psychotherapy in MDD, allows the omission of the pharmacological compound of antidepressant medication. Kring and colleagues evaluated the interaction between changes in affect and changes in depressive and anxiety symptoms in 41 MDD subjects undergoing cognitive behaviour therapy (Kring et al., 2007). The authors noted that change in negative affect was associated with both anxiety and depressive symptom, while changes in positive affect were more strongly coupled to depressive versus anxiety, symptoms. Moreover, while symptoms of depression and anxiety and negative affect all decreased during treatment, ratings of positive affect only increased in a subset of patients who demonstrated a significant decline in depression and only over an extended period of treatment. These findings further support the notion that increases in positive affect may be secondary to the elimination of excessive negative affect.

There is evidence to suggest that pharmacotherapy and psychotherapy may have different effects on dispositional affect (Schmid et al., 2002;DeRubeis et al., 2005). Positive and negative dispositional affect were measured in 240 MDD randomized to either 16 weeks of paroxetine, 16 weeks of cognitive therapy, or 8 weeks of placebo. Irrespective of treatment modality, all subjects reported significant decrease in negative affect and increases in positive affect over the course of treatment. Although there were between-group differences at study endpoint, changes in negative and positive affect were more rapid for subjects receiving medication.

This growing body of literature suggests that normalization of dispositional affect (increases in positive affect and decreases in negative affect) accompanies successful treatment of a MDE, irrespective of treatment modality. The reviewed studies, however, have not sufficiently characterized the temporal relationship between changes in positive and negative affect. Moreover, the state versus trait characterization of dispositional affect has also not been formally evaluated.

Neuroimaging

Neuroimaging encompasses a variety of techniques to measure the structure, physiology, and pharmacology of the brain. Beginning with the introduction of computed tomography (CT) (Hounsfield, 1973), the discipline has seen an advancement in the quality of the imaging data collected and an improvement in the safety of the subjects undergoing evaluation (Raichle, 2000). Neuroimaging techniques can be broadly categorized as either structural, denoting analyses of gross or cellular morphology, and functional, denoting continuous imaging of physiological processes at multiple time points during a single scan. Functional neuroimaging techniques can be further subdivided based on the source of the imaging data. The two most commonly employed techniques rely either on the localization of an injected radionuclide or the magnetic resonance of the endogenous host tissues.

21 Radionuclide Based Functional Neuroimaging

The localization of an injected radioactive neurotransmitter-derivative serves as the mechanism by which PET and SPECT produce three-dimensional images of the brain. With PET, these synthetic ligands are labeled with a rapidly decaying radioactive atom, usually Carbon-11, Fluorine-18, Oxygen-15, or Nitrogen-13. Single-photon emission computerized tomography (SPECT) is a technique similar to PET, with radioactive nuclei that have a longer half-life than those used in PET, and emit single, instead of double, gamma rays (Xenon-133, Technetium-99, Iodine-123) (Levin, 2005).

A subject, in the supine position, is injected with a radioactive tracer that incrementally progresses through the PET or SPECT camera. A gamma ray detector array captures the gamma rays that are produced at the collision site between a positron emitted from the radioactive substance and an electron in the tissue (in PET), or directly from the radionuclide (in the case of SPECT). As in CT scanning, the process is repeated, producing a series of two-dimensional thin slices of the brain that are later converted to a three-dimensional representation.

Although, SPECT is relatively less expensive than PET, its sensitivity and spatial resolution are inferior. A pragmatic advantage of SPECT scanners is that they do not require juxtaposition to a particle accelerator center. Analysis of regional blood flow with SPECT has generally been 15 replaced by O-H2O-PET or functional MRI (see below). Positron emission tomography can provide data on blood flow (i.e. hypo/hyperperfusions) or other biochemical functions, depending on the identity of the radioactively tagged molecule. Group differences in neuronal glucose metabolism can be evaluated via injection of a fluorine-tagged, non-hydrolyzable form of glucose, 18F-2-fluoro-2-deoxyglucose (FDG).

15 Radioactively labeled water ( O-H2O) provides an elegant technique for evaluating regional brain differences in cerebral blood flow (CBF). The relatively short half-life of 15O (~2 min) provides an opportunity to administer a new bolus every 12-15 minutes and to acquire a new snapshot of blood flow within the same scanning session. Soon after the tracer enters the smaller vessels in the brain, data acquisition can begin and usually lasts 60-90 seconds. The acquisition of multiple data points during a single scanning session allows the possibility of provocation paradigms, and the exposition of differences in brain that may not be apparent under resting

22 baseline conditions.

A ligand is a molecule with an affinity for a unique biological target, most often a protein receptor or transporter. Developments in PET and SPECT methodologies incorporating ligands provide an opportunity to carefully scrutinize the cellular pharmacodynamics of psychotropic medications (Eckelman et al., 1984;Wagner, Jr. et al., 1983;Smith et al., 2003). Ligands provide a surrogate of drug activity by measuring the ratio of ligand-receptor occupancy, versus drug occupancy. Several radiotracers have been developed for human imaging studies, targeting disparate neurotransmitters (e.g. acetylcholine (Kuhl et al., 1994;Iyo et al., 1997;Podruchny et al., 2003), glutamate (Erlandsson et al., 2003), dopamine (Farde et al., 1986;Costa et al., 1990) and serotonin (Wong et al., 1984;Passchier et al., 2000;Szabo et al., 1995;Houle et al., 2000;Meyer et al., 2001;Tauscher et al., 2001).

Both pre-synaptic and post-synaptic neuronal sites can be labeled with a radiotracer. Pre-synaptic sites can be involved in the regulation of neurotransmitter release from nerve terminals, while post-synaptic sites are at the beginning of the cascade of molecular events that will lead to the biological response (Maziere, 1995). Therefore, the binding of different radiotracers pre- or post- synaptically may reveal different stages in diseases involving these systems. For example the short-term effect of SSRI treatment is attenuated 5-HT neurotransmission (Blier and de Montigny, 1994). Long-term treatment is hypothesized to desensitize the inhibitory presynaptic 5-HT1 autoreceptors resulting in enhanced 5-HT neurotransmission (Blier and Ward, 2003). Recent advances in the development of high-resolution research tomography (HRRT), have resulted in dedicated human brain scanners with improved spatial resolution and reduced partial volume effects in receptor-ligand and glucose metabolism investigations (Varrone et al., 2009;van Velden et al., 2009;Chow et al., 2009;Eggers et al., 2009).

Magnetic Resonance Imaging

Magnetic resonance imaging (MRI) is based on the principle that differential magnetic properties exist amongst hydrogen atoms across different biological tissues (Paushter et al., 1984). Although a thorough comprehension of magnetic resonance necessitates an understanding of quantum mechanics, and is beyond the scope of this thesis, the basic concepts and application of magnetic resonance in imaging can be conveyed through classical physics.

The basis of the medical resonance signal is an intrinsic property of all protons, known as ‘spin.’ Spin refers to the angular momentum of an elementary particle (Manji, 2009). When protons are placed in a magnetic field (Bo), their spins become aligned with the direction of the magnetic field, so as to enter their lowest energy state. At typical clinical magnet strengths of 1.5 Tesla, only 0.001% of all hydrogen nuclei spins become aligned with the induced magnetic field, nevertheless, this small fraction is sufficient to create a weak equilibrium magnetization (Mo). In the presence of an addition magnetic field (B1), that is perpendicular to the main field, and oscillating at a specific resonant frequency, the equilibrium magnetization of the sample can be tipped over. The resonant frequency, is equal to the rate at which the proton precesses (spin axis of the proton rotates around the axis of the magnetic field) and is known as the Larmor frequency (Birkhoff, 1943).

The magnitude of the flip angle is modulated by the amplitude and/or duration of the RF pulse. A 90˚ flip angle will result in the magnetization being completely tipped in the transverse plane, completely eliminating the longitudinal component. A 180˚ flip angle, also known as an inversion pulse, will invert the longitudinal component, without altering the transverse plane. As the transverse component is not altered, this technique will not yield a traditional MR signal. However, this technique may be used to mark magnetization at a particular time, a particularly useful manipulation in arterial spin labeling (Williams et al., 1992).

Once the RF pulse is terminated, the hydrogen nuclei will slowly return to the lowest energy state, aligning the Mo back with Bo. The longitudinal component of magnetization will exponentially return back to its equilibrium value according to a longitudinal relaxation time constant (T1), while the transverse component decays exponentially with the transverse relaxation constant (T2). For example, using a 1.5 Tesla magnet, the T1 value for arterial blood is approximately, 1300 ms, while the T2 value is about 240 ms, meaning that longitudinal magnetization is fully relaxed after 6.5 seconds (5 time constants), while the transverse magnetization will take 1.2 seconds to completely decay.

After the RF pulse is turned off, the decay in transverse magnetization is measured during a time frame referred to as echo time (TE). The location of the signal is coded by applying field gradients to spread the signals into different frequencies, with different frequencies

24 corresponding to different locations on a grid. The interval between successive RF pulse is called the repetition time (TR). The process is repeated at different planes, or slices, through the head until the complete brain has been sampled. The images from successive slices are later reconstructed into a single three-dimensional volume. Different tissues types exhibit different T1 and T2 times and this is the basis for the contrast provided by structural MRI (Brown et al., 2007).

Current structural MRI techniques offer a spatial resolution exceeding 1 mm3 affording the possibility of visualizing and quantifying smaller brain structures compared to previous computed tomography (CT) investigations. Additional advantages over CT include the absence of ionizing radiation, and with three-dimensional MRI acquisition technology, an opportunity for higher resolution images of brain regions of interest (Maravilla and Sory, 1986).

A limitation however, of MRI is the elucidation of anatomical boundaries of deep brain gray matter structures, including the basal ganglia and deep brain nuclei (Helms et al., 2009).Other limitations of the MRI technique relate to the subjects being evaluated. Because of the small bore of the magnet, some patients may experience claustrophobia and have difficulty in cooperating during the study. Moreover, some obese patients cannot be examined, some patients, particularly acutely ill patients, cannot cooperate and movement artifacts may result. Patients with pacemakers and certain ferromagnetic appliances also cannot be studied. Finally, MRI units require careful shielding and greater technological expertise is required for utilization of MRI than for most other imaging methods (Manji, 2009).

Through a modification of conventional MRI scanning characteristics, it is possible to study the dynamics of brain function. Although it is assumed that the Bo remains stable throughout the imaging space, the molecular structure of surrounding biological substance can disrupt the field giving rise to regional differences, known as a magnetic susceptibility effect (Elster and Burdette, 2001). Although this effect is usually an unwanted artifact resulting in decreased MR signal at air/tissue and water/tissue boundaries, it is also the source of the signal in Blood- Oxygen-Level-Dependent (BOLD) neuroimaging.

When the hemoglobin molecule is bound to oxygen (oxyhemoglobin), as it presents in arterial blood, it has no unpaired electrons and is said to be diamagnetic. However, after releasing

25 oxygen at the capillary level, the deoxyhemoglobin molecule becomes paramagnetic with unpaired electrons in the central iron atom. This diamagnetic property of deoxyhemoglobin ultimately alters the magnetic field lines inside the blood vessels and surrounding tissue, and is responsible for dephasing the spins of hydrogen nuclei in the near vicinity (Buxton, 2002). The magnitude of the intravascular susceptibility effect was found to be directly related to the ratio of oxyhemoglobin to deoxyhemoglobin (Thulborn et al., 1982). Moreover, it has been found that at a magnet strength of 1.5 Tesla, approximately 50% of the BOLD signal is intravascular in origin (Boxerman et al., 1995), with the largest contribution to the BOLD signal near veins draining the activated brain region (Ogawa et al., 1993).

As the BOLD signal is sensitive to the ratio of oxyhemoglobin:deoxyhemoglobin, one would expect that increased extraction of oxygen from blood vessels by tissue demand would result in a negative BOLD signal. Contrary to this, increases in brain activation, were found to correspond with a positive BOLD signal (Ogawa et al., 1990;Bandettini et al., 1992). The current model posits that there is a compensatory increase in oxygenated blood flow beyond the additional need of the neural cells (Logothetis, 2002) (Buxton, 2002). In light of this model, the BOLD technique has also been referred to as the deoxyhemoglobin washout technique (Brown et al., 2007). The time course of the BOLD signal begins with an initial drop in the first two seconds, followed by a rapid increase to a maximum, a decline, and an undershoot before returning to baseline, corresponding to an initial decrease in oxyhemoglobin, followed by compensatory increases in CBF (Uludag et al., 2006).

Functional neuroimaging using the fMRI BOLD has superior spatial and temporal resolution to visualize activation during cognitive events in cortical and subcortical brain structures compared to radionuclide based neuroimaging modalities. Moreover, unlike PET and SPECT, most fMRI techniques are noninvasive and radiation-free, enabling repeat scans in individual subjects in different disease states (e.g. imaging a bipolar patient in manic, depressive, and euthymic states) (Lyoo et al., 2006). Repeat BOLD fMRI evaluations may also be used to compare regional brain activity between symptomatic and asymptomatic states (Fu et al., 2004;Siegle et al., 2006;Davidson et al., 2003).

In addition to the magnetic susceptibility effect which may complicate the examination of regions at the base of the brain such as the orbitofrontal and medial temporal cortices (Elster and

26 Burdette, 2001), BOLD fMRI also presents with other limitations. Because the BOLD contrast always represents a comparison of a signal across multiple behavioural states or patient groups, the functional contrast is relative. Second, the BOLD signal does not measure a single physiological process, but rather the composite effect of changes in blood, blood volume, and tissue oxygen utilization. Finally, the comparatively delayed response of the cerebrovascular smoothes the underlying neural signal (Brown et al., 2007). The influence of clinical and pathological conditions on the BOLD signal, including pharmacotherapy, is an active, yet largely nascent, area of research (Lowen et al., 2009;Biswal et al., 2007).

Coupling of Metabolism and Circulation

Although the human brain contributes less than 2% of the overall body mass, it is responsible for almost 20% of its oxygen consumption (Shulman et al., 2004). Not surprisingly, changes in neural activity depend on a timely coupling between oxygen consumption, glucose metabolism, and cerebral blood flow (CBF). Already by the end of the 19th century, Roy and Sherrington proposed that ‘intrinsic mechanisms’ are responsible for the coupling of neural activity to blood flow (Roy and Sherrington, 1890). Later investigations have confirmed these initial observations with the use of nitrous oxide to evaluate cerebral blood flow (Kety and Schmidt, 1948). With the use of carbon-14 radioactively labeled deoxyglucose, a number of investigators have also been able to demonstrate a correlation between regional brain activity and glucose metabolism (Sokoloff, 1981).

Structural support for this coupling has been noted with a close correlation between the five layers of intracortical vasculature and the six cytoarchitectonically defined layers of Brodmann’s laminae (Duvernoy et al., 1981;Brodmann, 1905). Based on the differential density of capillary beds and neuronal synapses in different laminae, it has been hypothesized that vascular density may be coupled with perisynaptic elements, rather than neuronal or glial soma (Zheng et al., 1991;Logothetis and Pfeuffer, 2004). Harrison and colleagues found that the signals evoked by acoustic stimulation (Grinvald et al., 1986), correlated directly with discrete capillary beds that were later revealed by electron microscopy of endovascular corrosion casts (Harrison et al., 2002). Moreover, the authors noted that within the cortex both arterial supply and the capillary network contained various flow control structures.

27 Changes in glutamatergic signaling have been associated with corresponding changes in glucose metabolism (Magistretti and Pellerin, 1999). Astrocytes, situated proximal to both cerebral capillary beds and neuronal synapses, play a pivotal role in the coupling of glucose uptake and glutamatergic neurotransmission (Petroff, 2002). Following the release of glutamate by the presynaptic neuron, its activity on the postsynaptic surface is quickly terminated by glutamate reuptake into the surrounding glia (Danbolt, 1994). Subsequent glutamate reuptake by glia stimulates glucose uptake and metabolism by astrocytes, adenosine triphosphate-generating processes which ultimately lead to the recycling of glutamate back to glutamine (Takahashi et al., 1995). The cerebral metabolic rate can also be expressed in terms of oxygen consumption as the majority of glucose metabolism is aerobic (Ames, III, 2000).

Ongoing research initiatives aimed at elucidating the primary subcellular source of the energy expenditure have identified both presynaptic (Jueptner and Weiller, 1995) and postsynaptic targets (Attwell and Laughlin, 2001). Alternative models contend that neurotransmitter-induced signaling is correlated with changes in energy expenditure, but is not triggered by the utilization of energy. In this model, hemodynamic responses are driven by post-synaptic cellular events induced following neurotransmitter release, rather than the direct local energy requirements (Attwell and Iadecola, 2002). Taken together, extant data suggest that perisynaptic activity, including neurotransmitter metabolism, changes in membrane potentials, and restoration of gradients all make significant contributions to energy consumption in the CNS (Logothetis and Pfeuffer, 2004).

Origin of the BOLD Signal

Multiple-modality imaging combining electrophysiology and magnetic resonance imaging has helped to define the neural basis of the BOLD signal (Logothetis and Wandell, 2004). The mean extracellular field potential (mEFP) represents the sum of inputs and outputs of multiple cells recorded by an extracellular electrode (Nicolelis and Ribeiro, 2002). If the recording electrode is placed at a sufficient distance from individual cells, then one can record both integrative dendritic processes and combined spikes generated by multiple neurons (Bishop and O'Leary, 1942).

If this signal is high-pass filtered at 400 Hz, then one can measure multiple unit spiking activity

28 (MUA) (Legatt et al., 1980). It is believed that the MUA most likely represents the weighted (based on neuron size) sum of extracellular action potentials within a sphere of approximately 300 μm (Legatt et al., 1980). If one examines the 300 Hz lowpass filtered component of the signal, one can measure local field potentials (LFPs); slow events believed to underlie cooperative activity in neural populations (Fromm and Bond, 1964;Mitzdorf, 1987). Electrophysiological investigations have further characterized LFPs as representing the combined input and local processing of a particular cortical area, including the activity of both excitatory and inhibitory interneurons (Kamondi et al., 1998;Gustafsson, 1984;Kobayashi et al., 1997;Logothetis and Wandell, 2004). Electrophysiological investigations have further characterized LFPs as representing the combined input and local processing of a particular cortical area, including the activity of both excitatory and inhibitory interneurons (Logothetis et al., 2001). Therefore, increases in BOLD signal may represent either excitation or inhibition depending on the predominant neurotransmitter used by the neurons with increased BOLD signal.

Simultaneous recording of LFPs and MUA with the hemodynamic response found strong correlation between both LFPs, MUA, and the BOLD signal (Logothetis et al., 2001). However, in cases where dissociation between LFPs and spiking activity was observed, the BOLD signal was only predicted by LFPs. The investigators also found a difference in the temporal dynamics, with MUA returning to baseline shortly after stimulus onset, while BOLD and LFPs signals remained elevated for the complete duration of the visual stimulus presentation (Logothetis et al., 2001). Moreover, Mathiesen and colleagues using laser Dopler flow technique, reported that activity-dependent CBF increases are dependent on synaptic excitation, whereas the net activity of the principal neurons is unimportant for the vascular response (Mathiesen et al., 1998).

It is worth noting that as the BOLD signal represents inputs from both excitatory and inhibitory neurons and local processing of these signals. When the predominant input is excitatory, a correlation between MUA, LFP, and BOLD will be noted. Conversely, if the excitatory input is matched by equal inhibitory input, then MUA will be a poor predictor of BOLD and LFP response (Logothetis and Pfeuffer, 2004). In other words, BOLD activation represents a measure of regionally specific cortical input and processing, but not necessarily cortical output.

Recent evidence suggests that the BOLD signal may be modulated by a host of other autonomic

29 changes independent of neuronal activity (Birn et al., 2009). However, normalizing the fMRI data to variation in basal physiological parameters, such as venous oxygenation, allows the detection of more significant voxels (Lu et al., 2009). The authors found that normalization of fMRI data to venous oxygenation was superior than adjusting for other parameters including blood pressure, heart rate, arterial oxygenation, and end-tidal CO2. These normalizations, however, should be performed with caution as different age groups may present with different physiological BOLD scaling parameters (Ances et al., 2009).

Increase in the strength of magnets used in fMRI experiments may also have a modulatory role on the BOLD signal. In a comparison of a simple motor task under three different magnet strengths (1.5T, 3T, 7T), it was report that the number of pixels that met statistical criteria, and the t-values that were calculated, and the percentage signal change all increase significantly with field strength (van der Zwaag et al., 2009). More importantly, the authors noted an earlier onset of the hemodynamic response at higher field strengths, leading them to propose a reduced venous contribution to the BOLD signal at higher magnet strengths. The reduction of effect of the venous contribution to the BOLD signal may have important consequences in helping to understand the reasons for the nonlinear characteristics of the vascular response that does not correlate with neural activity. Zhang and colleagues suppressed large-vessel BOLD contributions to report that the BOLD signal from the micro-vascular activity is replicable in response to replicated neuronal activities even very short inter stimulus interval (Zhang et al., 2009).

Analysis of Neuroimaging Data

Neuroimaging data analysis involves modeling the data to segment the observed neurophysiological signal into components of interest, confounds, and error, and subsequently making inferences about effects of interest using the variances of the partitions (Friston, 2005). The smallest unit of neuroimaging resolution is called a voxel, or volume element, and represents a distinct location on a three-dimensional (-x -y -z) coordinate system. A number of different analyses techniques have been proposed for the analysis of functional neuroimaging data. The choice of data analysis technique employed is influenced by the specific questions and preferences; sensitivity to detect small differences in a specific locus, or the ability to survey the entire brain volume for statistically significant differences.

30 The region of interest (ROI) analysis approach offer good sensitivity for detecting abnormalities within a specific brain region, and allows for targeted hypothesis testing. Most ROI approaches involve overlaying the PET/SPECT/fMRI functional data on an anatomic MRI image, and manually demarcating the region (Nieto-Castanon et al., 2003). The inherent variability in ROI criteria between studies, however, and the absence of ROI validation, provide the impetus for an approach that avoids the problems of not validated ROIs through an unbiased survey of the entire brain at the voxel level (Friston KJ, 1994). It is worth noting however, that a number of techniques for defining ROIs have been developed including the use of a-priori functional activations, electronic and traditional atlases (Etzel et al., 2009;Tsang et al., 2008).

Statistical parametric mapping is a spatially extended statistical process evaluating hypotheses about regionally specific effects (Friston et al., 1991). Statistical parametric maps (SPMs) are three-dimensional representations of voxel values that are distributed according to a known probability density function (student's T or F distributions) under the null hypothesis. Statistical parametric mapping analyzes every voxel using standard univariate statistical tests, evaluating the activation or regression of some explanatory variable. The SPMs are interpreted as spatially extended statistical processes with reference to the probabilistic behaviour of random fields theory (Adler, 1981;Friston et al., 1991;Worsley et al., 1992).

Random field may be conceptualized as random numbers whose values are mapped onto an n- dimensional space. Spatial correlation in a random field implies that adjacent values do not differ as much as values that are further apart (Besag, 1974). Within SPM, random fields model the univariate probabilistic characteristics of any voxel and the spatial covariance structure under the null hypothesis. Probabilistically unlikely deviations of SPM are interpreted as regionally specific effects, attributable to the factors of interest being manipulated experimentally (Friston et al., 2002a).

The use of statistical parametric mapping most frequently involves the combination of the general linear model (GLM) and Gaussian random field (GRF) theory to make classical inferences about spatially extended data (Friston K et al., 1995). The GLM follows the simple model to estimate parameters that could explain the spatially continuous data comparable to a conventional analysis of discrete data. To resolve the problem of multiple-comparisons when making inferences over a volume of the brain, the GRF theory provides a method for adjusting p

31 values to controlling for false positive rates. Whereas the Bonferonni provides a correction for a family of discontinuous or discrete statistical tests, the GRF theory serves as an analogue for continuous data like functional neuroimaging time-series (Kiebel et al., 1999).

Within SPM, the general linear model (y = Xβ + ε) forms the backbone of a number of related neuroimaging analyses including i) student’s T-tests comparing scans assigned to one condition or another, ii) correlation coefficients between observed between responses and parametric stimuli, iii) inferences using multiple linear regression, iv) evoked responses estimated using linear time invariant models, and v) selective averaging to estimate event-related responses (Friston K et al., 1995). The observed response (y) is modeled as a linear combination of an explanatory variable matrix (X) plus a well-characterized error term (ε).

The design matrix (X) is constructed by specifying the level of each experimental factor for each functional image series acquired. Each column of the design matrix is reserved for some factorial effect built into the experiment or that may confound the results; explanatory variables, covariates, or regressors. The relative contribution of each of these columns to the response is estimated by the β parameter, using standard least squares, with inferences being evaluated by T or F statistics, if a particular linear combination or interaction is being evaluated respectively. According to the GRF, the error term (ε) is an independently and identically distributed Gaussian random variable.

Inferences about the parameters rely on the estimated variance. Test of null hypotheses that a particular linear combination (e.g., a subtraction) of an estimate is zero is evaluated using an SPM T-test. The T statistic is calculated by dividing the contrast or compound (specified by contrast weights) of the parameter estimates by the standard error of that compound. When multiple contrasts of parameter estimates are jointly being evaluated, SPM F-tests are used and are specified with a matrix of contrast weights that can be thought of as a collection of "T contrasts" that one wants to test together (Friston, 2005). Limitations of the conjoint use of GLM and GRF are explored further in the discussion.

Neuroanatomy of Affective Processing

32 Neuroanatomy Overview

Differences in the arrangement of cellular elements into layers, the cell size, type, or packing density across layers, and the relative thickness of the layers, have all been used as the basis for cortical categorization. Early parcellation efforts, however, focused on histological observations with reports of differential cortex colour in the visual cortex of primates (Vicq d'Azyr, 1786) (Baillarger, 1840).

Microscopic examination of fixed, sectioned and cell-stained tissue revealed that the cerebral cortex is not a homogeneous sheet of grey matter. Early pioneering work by Meynert first demonstrated cellular differences between the rhinencephalic region and the neocortex (Meynert, 1867;Meynert, 1885). Later investigations characterized different regions of cerebral cortex (Betz, 1874;Lewis and Clarke, 1878;Lewis and Clarke, 1878), culminating with the first complete cytoarchitectonic map of the human cerebral cortex at the turn of the century (Campbell, 1905). In the same year, Korbinian Brodmann published a complete architectonic map of the monkey cerebral cortex (Brodmann, 1905), followed shortly by an architectonic map of the human cerebral cortex (Brodmann, 1909), effectively dividing the cerebral cortex into 52 different regions based on similarities in cytoarchitecture (Figure 1).

The increasing use of functional neuroimaging by the neuroscience community has popularized the use of Brodmann’s map. Further contributing to the popularity of the Brodmann map has been its adoption in the Talairach & Tournoux (1988) proportional stereotaxic atlas of the human brain (Talairach and Tournoux, 1988). It is important to note that that the location of cortical areas in the Talairach and Tournoux atlas was based on a simple projection of the Brodmann map onto the brain sections of the atlas and not on architectonic analysis of those particular sections. More recent initiatives have attempted to rectify these shortcomings through evaluation of the architecture of the cerebral cortex in several brains and description of the variability in the location of cortical areas in the proportional Talairach stereotaxic space (Amunts et al., 1999;Morosan et al., 2001).

33 Figure 1 – Brodmann Area (BA) Classification of the Cerebral Cortex (Gray, 1918)

LEGEND: Medial and lateral views of the human brain, with approximate location of Brodmann Areas (BA) indicated in square text boxes. Subcortical structures have been removed in the medial view.

34 Crucial to the validity of neuroimaging studies is effective crosstalk between functional neuroimaging investigations of the human brain and experimental work in the monkey. Unfortunately, the numerical designations employed by Brodmann in his maps of the human and the monkey brain are not always consistent. Moreover, while maps of the human cerebral cortex have remained largely unchanged since their publication in the first half of the 20th century, ((Brodmann, 1909;Economo and Koskinas, 1925;Sarkissov et al., 1955), maps of the monkey cortex continue to evolve throughout the second half of the twentieth century (Petrides, 2005).

On the medial side of the cortex, the cingulate gyrus has been sectioned into four regions, with associated subregions, each making a qualitatively unique contribution to brain functions; i) the subgenual and pregenual anterior cingulate cortex (sACC and pACC), ii) the anterior and posterior midcingulate cortex (aMCC and pMCC), iii) the dorsal and ventral posterior cingulate cortex (dPCC and vPCC), and iv) the retrosplenial cortex (RSC) (Vogt, 2005) (Figure 2). It is worth noting, however, that this nomenclature is relatively recent. Prior to Vogt’s re- categorization, the modern aMCC and pMCC were simply referred to as the ACC, while the dPCC, vPCC were lumped together as the PCC (Bush et al., 2000). As a result, a number of previous investigations that report activation in the dorsal or caudal ACC were actually describing the pMCC.

In addition to the sACC/pACC/aMCC nomenclature, the ACC has also been divided into dorsal cognitive and ventral affective components (Bush et al., 2000). This ventral region of the ACC comprises the sACC (also referred to as the subcallosal gyrus, Brodmann area 25), Brodmann area 33, and pACC or rostral ACC (rostral regions of Brodmann areas 24) (Devinsky et al., 1995). The dorsal portion of the cingulate gyrus may also have a role in motor control. Three cingulate motor areas have also been described with the rostral (CMAr) located on the dorsal surface of the aMCC (BA 24/32), the dorsal (CMAd) located inferior to the SMA on the medial frontal gyrus (BA 6), and the ventral (CMAv) located on the dorsal side of the dPCC (BA 31) (He et al., 1995) (Figure 4).

35 Figure 2 – Cingulate Gyrus Nomenclature – Vogt Classification (Vogt, 2005)

LEGEND: Medial view of the human brain with approximate location of cingulate areas indicated. sACC – subgenual anterior cingulate cortex; pACC – pregenual anterior cingulate cortex; aMCC – anterior midcingulate cortex; pMCC – posterior midcingulate cortex; dPCC – dorsal posterior cingulate cortex; vPCC – ventral posterior cingulate cortex; RSC – retrosplenial cortex (RSC).

36 The visual cortex has been further subdivided into the primary or striate visual cortex (also referred to as visual area 1, V1) and extrastriate visual cortex (areas V2, V3, V4, and V5). The primary visual cortex is equivalent to BA 17, with V2 being represented by BA18, and V3, V4, and V5 contained in BA 19 (Payne, 1993;Kaas, 1996) (Figure 3). The primary visual cortex, V1, receives information directly from the lateral geniculate nucleus of the thalamus before relaying information into two concurrent streams; dorsal and ventral. The dorsal pathway starts in V1, proceeding dorsally through V2, and the dorsomedial area, into the inferior parietal lobule. The dorsal stream is believed to mediate the required sensorimotor transformations for visually guided actions directed towards objects (Goodale and Milner, 1992). The ventral stream proceeds laterally from V1, through V2 and V4 before reaching the inferior temporal lobe, and may be more involved in the perceptual identification of objects.

A burgeoning database of neuroimaging literature has documented the interaction between the processing of affective states and activity in the visual cortex. Activity in the visual cortex may be subjected to both i)attentional modulation - when the context of a particular task alters visual processing of the same stimuli and ii) emotional modulation - when affective properties of the stimuli influence processing (Vuilleumier and Pourtois, 2007).

37 Figure 3 – Subdivision of the Visual Cortex

LEGEND: Medial view of the human brain with approximate location of visual areas indicated (Kaas, 1996). V1 (pink), V2 (yellow), V3 (green), V4 (blue), and V5 (red) correspond to visual areas 1-5. Striate visual cortex is composed of V1, while the extrastriate cortex comprises V2- V5.

38 Figure 4 – Subdivision of the Motor Cortex

LEGEND: Medial view of the human brain with approximate location of the supplementary and pres-supplementary and cingulate motor areas indicated (He et al., 1995). SMA – supplementary motor area; preSMA – pre supplementary motor area; rCMA – rostral cingulate motor area, dCMA – dorsal cingulate motor area, vCMA – ventral cingulate motor area.

39 Converging lines of evidence from physiological and anatomical investigations demonstrated the existence of multiple motor areas in the primate cortex (Graziano and Aflalo, 2007). The motor cortex can be broadly divided into the primary motor cortex, located in the dorsal part of the precentral gyrus and the anterior bank of the central sulcus (BA 4), and a number of second motor cortices including the premotor cortex, the supplementary motor area, the presupplementary area, the posterior parietal cortex, and the rostral, dorsal, and ventral cingulate motor areas (Tsuji et al., 2006) (Figure 4). Just anterior to the precentral sulcus, on the lateral cortical edge is the premotor cortex (BA 6) that has been further subdivided into a dorsal (PMd) and ventral (PMv) component (Chouinard and Paus, 2006).

The medial wall of the premotor cortex is home to the supplemental motor area (SMA) (Woolsey et al., 1952), and the pre-supplementary motor area (pre-SMA) immediately anterior to the SMA. Neuronal activity in these premotor regions has been associated with various motor tasks, including learning new motor skills, the performance of a sequential motor task based on memory, and reward-based selection (Kollias et al., 2001). Although partial somatotopical representation may exist in these premotor regions, a clear body representation has only been documented for the SMA (Macpherson et al., 1982).

The posterior parietal cortex (PPC) (BA 5/7) also appears to be a component of the motor system (Fogassi and Luppino, 2005). Emerging data indicate its fundamental role in visuomotor transformations (Bremmer et al., 2001), spatially guided behaviour (Selemon and Goldman- Rakic, 1988), and movement planning (Snyder et al., 1997;Kleiser et al., 2009;Hu et al., 2009). Activation of the PPC has also been noted during task preparation and episodic retrieval (Knops et al., 2009), and short-term memory maintenance (Wendelken et al., 2008;Magen et al., 2009). A number of affective processes, including the imaginary experience of pain (Ogino et al., 2007;Benuzzi et al., 2008), and the processing of facial expressions (van der et al., 2007), and other affective imagery (Brazdil et al., 2009) have also been reported to engage the PPC.

Following its introduction in 1937 (Papez, 1937), and advocacy in the early 1950s (MacLean, 1952), the limbic system has been frequently linked to emotional processes (LeDoux, 1996). The foundation of the limbic system is the limbic lobe an anatomical concept restricted to the cortical mantle (Broca, 1878). The ‘greater limbic lobe’ also includes the olfactory cortex, hippocampus and adjacent cortical areas, including not only the caudal orbital and medial prefrontal cortex,

40 part of the temporal polar cortex and a large anteroventral, agranular and dysgranular part of the insula (Morgane et al., 1982;Yakovlev, 1972;Isaacson, 1972;MacLean, 1990;Mesulam, 2000) (Figure 5). Although it has certainly advanced research in many fields of neuroscience including psychiatry and psychology, the concept of the limbic system is not without its criticisms, partially due to initial faulty labeling logic Whereas visual, motor and various memory systems have been defined based on identifying structures responsible for a particular function, an attempt was made to fit several emotional functions into a number of closely related anatomical structures in the limbic system (Kaada, 1960;Brodal, 1969). Heimer and Van Hoesen have proposed that the limbic system label comprise the greater limbic lobe, the basal nucleus of Meynert, the ventral striatopallidal system, and the extended amygdala (Heimer and van Hoesen, 2006).

The basal nucleus of Meynert is actually a collection of aggregated and non-aggregated cholinergic and GABA-ergic neurons stretching from the rostral forebrain to the caudal globus pallidus that project to the cerebral cortex and other components of the limbic lobe (De Lacalle and Saper, 1997;Halliday et al., 1993;Mesulam and Geula, 1988;Mesulam and van Hoesen, 1976). The main cortical input into the nucleus of Myenert arrives from the limbic lobe BA 24/25/31/32), and as such, it may be viewed as an interface between the limbic lobe and the entire cerebral cortex (Mesulam and Mufson, 1984). Additional ascending inputs from the catecholamine systems have also been documented (Zaborszky et al., 1993).

The ventral striatopallidal system refers to the interdigitation of the ventral striatum and ventral pallidal components extending towards the olfactory tubercle (Heimer and Wilson, 1975). Originally defined based on its input from the olfactory cortex , hippocampus, and amygdala, recent evidence indicates additional input from the entorhinal cortex (BA 27/28/34), pACC/sACC (BA 24/25/32), OFC (BA 11/12), and the insula (Chikama et al., 1997;Ferry et al., 2000;Haber et al., 1995). Using tracer methods, four cortical-basal ganglia-thalamocortical reentrant circuits have been identified that link the basal ganglia with PFC; an executive circuit originating in the dlPFC, an anterior cingulate circuit originating in the ACC, and a medial and a lateral orbitofrontal circuit originating in the medial and lateral OFC (Alexander et al., 1986;Haber et al., 1985). Whereas the executive circuit is implicated in planning and working memory, the remaining three PFC circuits form the ventromedial emotional–motivational domain of the striatum and are responsible for choosing and mobilizing appropriate adaptive

41 behavior (e.g. reward-guided choice behavior) (Schultz et al., 2000;Ferry et al., 2000).

The extended amygdala consists of the central and medial amygdaloid nuclei along with their extensions in the subpallidal or sublenticular region, and along the arc of the stria terminalis which link the centromedial amygdala to the bed nucleus of stria terminalis (de Olmos, 2004;Heimer et al., 1999;Martin et al., 1991;de Olmos and Ingram, 1972;Johnston, 1923). The primary inputs into the extended amygdala arrive from the basal lateral amygdala and other parts of the limbic lobe (McDonald et al., 1999). Meanwhile, efferent projections to autonomic, somatomotor, endocrine-related centers in hypothalamus and the brainstem enable the extended amygdala to coordinate activities in multiple limbic lobe regions for the development of behavioral responses (Alheid and Heimer, 1988;Heimer et al., 1999).

Identification of Emotional Significance

In spite of the difficulties in parsing out brain regions involved in the identification of emotional stimuli, rather than the processing of emotional stimuli, investigators have repeatedly reported on the importance of a prefrontal – subcortical circuit. Consisting of the bilateral anterior insula, the amygdala, the vmPFC, hypothalamus, and the periaqueductal gray, this network is thought to be involved in perceiving and organizing autonomic responses to aversive or threatening stimuli (Alexander et al., 1990;Phillips et al., 2003a;Ongur and Price, 2000;Augustine, 1996).

Amygdalar cellular responses to faces and eye gaze directions have been reported in studies of nonhuman primates (Brothers and Ring, 1993). Similar findings (recognition of emotionally salient stimuli) in humans have also implicated the amygdala. Both stimulation studies (Heit et al., 1988) and investigations of subjects with amygdalar lesions (Young et al., 1995) report activation of the amygdala in response to visual (Adolphs et al., 1994;Phelps and Anderson, 1997) and auditory (Scott et al., 1997) fearful stimuli. Stimulation of the human anterior insula has also been associated with the perception of unpleasant taste (Penfield and Faulk, 1955;Craig, 2004), while direct stimulation of the posterior part of the insula in humans produces both painful and non-painful sensations (Ostrowsky et al., 2002).

Evaluations of subjects who sustained subcortical lesions further support the role of the insula and striatum in recognizing emotional salience. Calder and colleagues reported an impairment in

42 the recognition of facial and vocal expressions of disgust in a patient with a focal insula lesion (Calder et al., 2001), while Anderson and Phelps reported reduced enhanced perception of emotionally salient information (Anderson and Phelps, 2001) in subjects with damaged left amygdala. Impairment in the recognition of facial expressions depicting disgust has also been reported in subjects with Huntington’s disease, where degeneration of the caudate nucleus is a documented abnormality (Sprengelmeyer et al., 1996;Calder et al., 2000).

Results from neuroimaging investigations have confirmed earlier results from lesion and stimulation studies. The amygdala plays a key role in the initial identification of stimuli as emotionally salient. Increased activity has been reported within the amygdala in response to viewing new faces (Dubois et al., 1999;Wright et al., 2006;Schwartz et al., 2003), the detection of eye gaze (Kawashima et al., 1999;Hooker et al., 2003), and faces with fearful (Breiter et al., 1996;Morris et al., 1996;Phillips et al., 1997;Phillips et al., 2001;Wright et al., 2001;Thomas et al., 2001b) and sad (Blair et al., 1999;Britton et al., 2006;Hariri et al., 2002) facial expression. Presentations of threatening words (Isenberg et al., 1999;Siegle et al., 2003), fearful vocalizations (Phillips et al., 1998b), unpleasant olfactory (Zald and Pardo, 1997) and gustatory (O'Doherty et al., 2001b), auditory, and visual (Anand et al., 2005a;Baumgartner et al., 2006;Bermpohl et al., 2006b) stimuli, have also been associated with amygdala activity.

Interestingly amygdala activation was also noted in response to viewing happy facial expressions (Campbell et al., 2006;Breiter et al., 1996;Lawrence et al., 2004;Surguladze et al., 2005), emotive scenes and film excerpts (Reiman et al., 1997;Taylor et al., 2000), and emotional verbal stimuli (Hamann and Mao, 2002). The film clips were all silent, with the first neutral in affective tone, while the next two showed a puppy playing with flowers, and monkeys and gorillas engaging in amusing activities. The last two clips were portions of training videos for nurses and showed a leg amputation and a third-degree burn victim (Tomarken et al., 1990). The film clips were all silent, with the first neutral in affective tone, while the next two showed a puppy playing with flowers, and monkeys and gorillas engaging in amusing activities. The last two clips were portions of training videos for nurses and showed a leg amputation and a third-degree burn victim (Cahill et al., 1996;Canli et al., 2000;Hamann et al., 1999;McGaugh, 2004).

Activation of the amygdala has also been reported during the memory encoding of emotionally salient information (Cahill et al., 1996;Canli et al., 2000;Hamann et al., 1999;McGaugh, 2004). Preliminary evidence appears that there is a time-dependent habituation of the amygdalar response to repeated presentations of visual stimuli with emotional content (Protopopescu et al., 2005;Strauss et al., 2005a;Phillips et al., 2001;Wright et al., 2001;Fu et al., 2004;Sheline et al., 2001) with decreased activation with repeated presentations. Moreover, sex differences in rates of habituation have also been reported, with greater habituation in men, compared to women. (Killgore et al., 2001;Thomas et al., 2001b). Davis and Whalen have suggested that the amygdala may have a specific role in the modulation of vigilance and attention to emotionally relevant stimuli (Davis and Whalen, 2001). Projections from the central nucleus of the amygdala to cortical targets may lower cortical neuronal activation thresholds and facilitate cortical processing (Whalen et al., 1998b;LeDoux, 1993;LeDoux, 2000).

Neuroimaging investigation demonstrates bilateral activation of the anterior insula in delayed fear conditioning (Buchel et al., 1999;Surguladze et al., 2003a;Ohman, 2005) and during anticipation of an aversive stimulus (Phelps et al., 2001;Porro et al., 2002;Nitschke et al., 2006). The insula may thus assist in the representation of aversive sensory information to the amygdala. Evidence indicates that the insula, in concert with the ventral striatum and thalamus, may also be relevant in the identification of facial expressions of disgust (Phillips et al., 1997;Sambataro et al., 2006;Phillips et al., 2004) and in taste perception (Small et al., 1999;Rothemund et al., 2007;Wagner et al., 2007). The insular cortex has established connectivity with primary and secondary somatosensory areas, anterior cingulate cortex, the amygdala, the hippocampus, the prefrontal cortex and the motor cortex (Nagai et al., 2007;Taylor et al., 2008). The anterior insula appears to be involved in affective processing (Jabbi et al., 2008;Jabbi and Keysers, 2008;Rilling et al., 2008) while the mid-posterior insula is implicated in awareness and interoception (Karnath et al., 2005;Kondo and Kashino, 2007;Mochizuki et al., 2007;Seminowicz and Davis, 2007b).

Production of Affective State

The production of affective states refers to a number of processes that assist in the production of a conscious perception of the affective state including change in physiological (autonomic) and psychological parameters. Positive dispositional affect is characterized by the following human

44 properties; interested, excited, strong, determined, enthusiastic, proud, alert, inspired, attentive, and active. Preclinical animal data with reward processing may provide a proxy to some of the positive affect components.

Early preclinical evidence indicated that dopaminergic stimulation of the ventral tegmental area, the nucleus accumbens and the medial prefrontal cortex was found to enhance response to reward in rats (Spanagel and Weiss, 1999). Additionally, cholinergic neurotransmission through the nucleus basalis may improve the neural representations of behaviourally important stimuli while ignoring irrelevant stimuli (Kilgard and Merzenich, 1998). Comparable brain regions, including the ventral striatum have been identified in non-human primate models of reward processing (Heimer and Alheid, 1991;Ongur and Price, 2000).

All of the aforementioned brain regions (ventral tegmental area, nucleus accumbens, and mPFC) extend projections to the amygdala. Significant changes in social behaviour in monkeys, including hyperorality, social disinhibition, an absence of emotional motor and vocal reactions have been observed following lesions of the amygdala (Kluver and Bucy, 1939;Zola-Morgan et al., 1991). Moreover, autonomic responses to emotional stimuli are also disrupted in amygdalae- lesioned mammals during anxiety-provoking situations (Bagshaw MH et al., 1965) and fear conditioning (Gallagher et al., 1990;Blanchard and Blanchard, 1972). Electrical and pharmacological manipulations of the amygdala also elicit affect-related endocrinological and behavioural changes. Electrical stimulation increases plasma levels of corticosterone, and elicits autonomic symptoms of fear and anxiety, (Applegate et al., 1983;Kapp et al., 1994), while infusion of the γ-aminobutyric acid (A) receptor (GABA-AR) antagonist into the amygdala increases blood pressure and heart rate (Sanders and Shekhar, 1991).

Case reports of amygdala lesions in humans describe emotional blunting (Aggleton and Brown, 1999) and impaired fear conditioning (Bechara et al., 1995). Electrical stimulation of the human amygdala is associated with subjective feelings and autonomic responses characteristic of fear (Gloor P, 1992). Similarly, activation of the amygdala in patients with temporal lobe epilepsy also leads to fear-like behavioural responses (Gloor P, 1992). In addition to the amygdala, animal studies point to the insula in the production of affective states. Taste aversion has been reported in rats with insular lesions (Dunn and Everitt, 1988), while lesions including the vlPFC reduce fear reactivity (Morgan and LeDoux, 1999).

Anatomical projections from the sACC and pACC to the amygdala, periaqueductal gray, mediodorsal and anterior thalamic nuclei, nucleus accumbens and ventral striatum, have been documented and are believe to subserve emotional behaviour and related autonomic function (Devinsky et al., 1995;Paus, 2001). With reciprocal connections to the dorsolateral prefrontal and parietal cortex, the premotor and supplementary motor areas, the dorsal section corresponds to aMCC and modulates various attentional and executive functions including motivation, error detection, working memory; and anticipation (Vogt et al., 1995;Carter et al., 1999)(Vogt, 2005).

Correspondingly, lesions in the pACC and sACC are associated with i) an impairment in the response of the autonomic system to conditioned stimuli, ii) abolishing conditioned emotional vocalizations to painful stimuli (Frysztak and Neafsey, 1991), iii) emotional blunting, and iv) changes in maternal–infant interactions (MacLean and Newman, 1988). More recent evidence indicates that lesions of the ACC may have a clinical use in the treatment of intractable and chronic pain (Davis et al., 1998;Wilkinson, 2000;Yen et al., 2005). In animals, stimulation of this region is associated with changes in autonomic and endocrine function (Buchanan SL and Powell DA, 1993).

A variety of affective states (including apathy, depression, disinhibition and anxiety) have been reported following lesions in the sACC in humans (Angelini et al., 1981;Levin and Duchowny, 1991). Both autonomic, visceromotor, and affective changes have been reported in human stimulation studies and in patients with seizure foci in the sACC and pACC (Bancaud and Talairach, 1992). Recently, it was reported that in subjects with treatment resistant depression, stimulation of white matter tracts in Brodmann areas 25 was associated with a reduction in the severity of depressed symptoms (Mayberg et al., 2005;Giacobbe et al., 2006;Lozano et al., 2008).

Orbital and medial (ventromedial) prefrontal cortex has also been implicated in the production of emotional states and behaviour. The orbitofrontal cortex includes Brodmann areas 11, 12, and 13 and the medial portion of area 47, and may also be defined as including subgenual and pregenual portions of the ACC, Brodmann areas 25, 24, and Brodmann area 32 (Price, 1999). It has direct connections from the basolateral nucleus of the amygdala (Davis and Whalen, 2001) and appears to play a critical role in the representation of the reward value of a stimulus and the way in which

46 this representation guides goal-directed behaviour (Francis et al., 1999;Rolls et al., 2003).

The orbitofrontal cortex (OFC) has been implicated in a number of functions related to reward processing. Reduced aggression, changes in food-selection behaviour, and impairments in dissociating stimulus and reward have been reported in monkeys with OFC lesions (Meunier et al., 1997). Schoenbaum and colleagues demonstrated activation of OFC neurons in rats when processing the reward value of stimuli to guide behaviour (Schoenbaum et al., 1998). Moreover, in monkeys, OFC neuronal activation has been recorded during taste (Rolls, 1994) and smell (Critchley and Rolls, 1996) perception, and visual stimuli associated with reward (Thorpe et al., 1983).

Single-neuron recordings in the human OFC have documented increased activity in response to aversive stimuli (Kawasaki et al., 2001), whereas OFC lesions have led to personality changes including impaired emotion misidentification, disinhibition, and impulsiveness (Damasio, 1994;Hornak et al., 1996;Rolls, 2004). Lesions of vmPFC cortex have also been associated with impaired judgment in decision-making tasks evaluating high- versus low-risk (Bechara et al., 1998).

Both cortical and subcortical structures participate in during the induction of affective states. Activation of the ventral striatum has been observed in craving (Breiter et al., 1997;Franklin et al., 2007;Kilts et al., 2004), reward prediction (Pagnoni et al., 2002;Rolls et al., 2007;Tanaka et al., 2006), anticipation of reward (Knutson et al., 2001;Liu et al., 2007;Galvan et al., 2005), and in romantic love (Bartels and Zeki, 2000;Fisher et al., 2005;Aron et al., 2005). Furthermore, a correlation has been reported between the euphoric response to amphetamine and dopamine release in ventral striatal areas (Drevets et al., 2001;Willeit et al., 2007;Riccardi et al., 2006).

A wealth of data support the engagement of the amygdala in response to induction of positive and negative affect (Reiman et al., 1997;Siegle et al., 2003), and fear-conditioning paradigms (Buchel et al., 1999;LaBar et al., 1998;Knight et al., 2005). Amygdalar activation has also been reported in response to presentations of masked affective facial expressions where the observer reports no conscious awareness (Morris et al., 1998c;Whalen et al., 1998b) and when faces are shown at ignored locations (Anderson et al., 2003;Vuilleumier et al., 2001), and suppressed by binocular rivalry (Pasley et al., 2004;Williams et al., 2004).

In addition to activation of the anterior insula during induced sadness and anticipatory anxiety in HC, during lactate- or cholecystokinin-induced panic attacks in HC (Javanmard et al., 1999) and during exposure to trauma-related stimuli in posttraumatic stress disorder subjects (Charney and Drevets, 2002), functional imaging studies also support the involvement of the anterior insula in pain processing (Seminowicz and Davis, 2007a). Specifically, insular activation during the encoding of both the intensity (Coghill et al., 1999;Craig et al., 2000;Frot and Mauguiere, 2003),and laterality(Bingel et al., 2003;Brooks et al., 2002) of painful and non-painful thermal stimuli (Craig et al., 2000;Hua et al., 2005), and in affective pain processing (i.e. the unpleasantness of tonic pain) (Critchley et al., 2005;Singer et al., 2004b;Davis et al., 1998;Davis, 2003;Downar et al., 2003;Downar et al., 2002;Seminowicz and Davis, 2006;Taylor et al., 2008) (Schreckenberger et al., 2005).

The insular cortex may also serve a sensory integrative function for pain, tactile, vestibular, taste, and other visceral sensations (Augustine, 1996). Within the posterior and anterior insula, the somatosensory region is part of a sensory limbic pathway from the secondary somatosensory cortex to the amygdala, with nociceptive input reaching the anterior parts of the insula directly via the posterior part of ventromedial nucleus (Craig et al., 1994). In addition, owing to its projection to the amygdala, the insula has also been proposed to be involved in affective and emotional processes, specifically the induction of mood states (Reiman et al., 1997), and the induction of affect with words (Canli et al., 2004;Siegle et al., 2007), facial expressions, (Keedwell et al., 2005a;Britton et al., 2006;Fu et al., 2004;Gotlib et al., 2005), other affective visual stimuli (Kumari et al., 2003;Anand et al., 2005a;Baumgartner et al., 2006;Beauregard et al., 2006;Bermpohl et al., 2006b;Eugene et al., 2003), and during the experience of guilt (Shin et al., 2000).

The ventral, or affective, subdivision of the ACC, currently identified as the sACC and pACC also may have an important role in both the initial processing of emotional stimuli and subsequent production of affective states. The affective component of the ACC, (corresponding to sACC and pACC) has been activated by affect inducing tasks, including studies of emotional processing in HC and symptom provocation studies in psychiatric populations (anxiety, simple phobia and obsessive–compulsive disorder) (Bush et al., 2000;Phan et al., 2002). Activation of the sACC and pACC has also been reported by induced negative affect in HC (Anand et al.,

48 2005a;Greicius et al., 2007;Bermpohl et al., 2006b;Britton et al., 2006;Eugene et al., 2003;Hariri et al., 2003;Lee et al., 2004;Mayberg et al., 1999;Shin et al., 2000;Downar et al., 2003). Conversely, when HC are instructed to switch from affective processing to cognitive processing tasks, deactivation in sACC and pACC has also been reported (Bush et al., 2000;Drevets and Raichle, 1998;Mohanty et al., 2007;Whalen et al., 1998a).

Various portions of the prefrontal cortex have also been implicated in the production of affective states. The vmPFC and the OFC may mediate autonomic changes accompanying affective state production and provide the basis for an automatic regulation of affect induced behaviours. For example, increased activation (BOLD and CBF) of the OFC is reported to accompany the perception of pleasant and unpleasant odors (de, I et al., 2003;Small et al., 2004), flavors (Frank et al., 2003;Marciani et al., 2006), and tactile stimuli (Francis et al., 1999;O'Doherty et al., 2001b;Zald and Pardo, 1997). The perception of foods not eaten to satiety (O'Doherty et al., 2000;Santel et al., 2006;Smeets et al., 2006), the performance of gambling tasks (O'Doherty et al., 2001a;Tobler et al., 2007;van et al., 2006), guessing and decision-making on the basis of reward (Elliott et al., 2000;Jensen et al., 2007;Hsu et al., 2005;Kringelbach, 2005), the consideration of moral dilemmas with affective components (Singer et al., 2004a;Heekeren et al., 2003;Greene et al., 2001) all report activation of the OFC and surrounding vmPFC. Increased CBF in the OFC has also been demonstrated during events precipitating anger (Dougherty et al., 1999;Kimbrell et al., 1999;Coccaro et al., 2007;Sander et al., 2005) and imagined restraint of physical aggression (Pietrini et al., 2000).

The vlPFC corresponding to Broca's area (BA 44,45), located dorsolaterally to the OFC (Ongur and Price, 2000) has also been involved in the production of affect, but predominantly negative affect. Increased activation (BOLD and CBF) has been reported in this region during the induction of sad mood (Pardo et al., 1993;Habel et al., 2005;Levesque et al., 2003b) and guilt (Shin et al., 2000;Finger et al., 2006), the recall of personal memories (Fink et al., 1996) and emotional material (Reiman et al., 1997;Kuchinke et al., 2006;Grimm et al., 2006), the presentations of negative facial expressions (Sprengelmeyer et al., 1996;Lange et al., 2003;Monk et al., 2006;Taylor et al., 2006b;Nelson et al., 2003). Activation of the vlPFC in the generation of humour has also been reported (Mobbs et al., 2005), along with vlPFC activation during the evaluation of positive attitudes (Wood et al., 2005). Activation of the temporofrontal junction and proximate vlPFC have additionally been associated with autobiographical memory retrieval

49 (Markowitsch, 1997;Markowitsch et al., 2003;Maguire et al., 2001).

Regulation of Affective State

Occupying a strategic locus between the reactive-based ventrolateral prefrontal cortex (vlPFC), and the planning-based dorsomedial prefrontal cortex (dmPFC), the ventromedial prefrontal cortex (vmPFC) is ideally situated to modulate the production of affect (Sanides, 1970;Mega et al., 1997). Ventral prefrontal regions may modulate affect at an automatic level through reciprocal connections to the hypothalamus, amygdala, thalamus, ventral striatum, and brainstem nuclei (Kaufman et al., 2000). The OFC has a role in the inhibition of autonomic and defensive behaviours induced by amygdalar stimulation (Timms, 1977). Moreover, lesions in the left vmPFC inhibit autonomic, behavioural, and endocrine responses to fear-conditioned stimuli (Sullivan and Gratton, 1999).

The role of the hippocampus in affective regulation may be related to its inhibition of the stress response through inhibitory connections to subcortical structures activated by stress (Lopez et al., 1999). Various theories modeling the neurocircuitry of affective regulation of the hippocampus have been put forth, emphasizing the role of spatial cognition (O'Keefe and Nadel, 1978) and episodic memory (Squire et al., 1992). A unitary explanation may involve the hippocampus as a monitor of conflict between different goal-directed behaviours, with facilitation of exploratory rather than defensive patterns of behaviour, to allow resolution of goal conflict (Gray and McNaughton, 2000;Phillips et al., 2003a).

Dorsal regions of the PFC are activated by cognitive versus affective components of a stimulus, and by planning-based regulation of emotional behaviour (Tucker et al., 1995). Lesion studies in rats demonstrate the modulatory potential of these brain regions with increased cardiovascular response to fear conditioning (Frysztak and Neafsey, 1994). Conversely, electrical stimulation reduced the behavioural effects of amygdalar stimulation (Frysztak and Neafsey, 1994). The dorsal cognitive division of the ACC, otherwise known as the aMCC and pMCC shares connections with the lateral prefrontal cortex, primary motor cortex and supplementary motor area (Vogt et al., 1992). Preclinical evidence has implicated the MCC in early learning, error monitoring and detection, and avoidance learning (Gemba et al., 1986;Gabriel, 1990)

50 Case studies of subjects with MCC lesions report attention deficits and impaired performance on tasks including the generation or selection of novel sequences (Ochsner et al., 2001). Other studies have (Simpson, Jr. et al., 2001) demonstrated a focus within the MCC for the error- related negativity observed with event-related potentials and associated with errors during task performance (Dehaene et al., 1994).

Intriguingly, the motor cortex of primates also has an association with the limbic lobe with aMCC in the lower bank and fundus of the cingulate sulcus; the cingulate motor areas, M3 and M4 (Biber et al., 1978;Braak, 1976;Morecraft et al., 2001). These regions are highly interconnected in terms of body part representation with the supplementary motor cortex and the primary motor cortex in the precentral gyrus (Morecraft and van Hoesen, 1992), with M3 receives input from the majority of the limbic lobe, particularly the basolateral complex of the amygdala (Morecraft and van Hoesen, 1998).

Recent evidence from human brain lesion patients highlights the importance of the regulation of the affective state. Evaluating the clinical presentation of a variety of patients with brain lesion, the authors were able to calculate the risk of depression based on lesion location and found that indicates that vmPFC damage confers resistance to depression, whereas dorsal PFC damage confers vulnerability (Koenigs et al., 2008).

Preclinical lesion and stimulation investigations, along with clinical case reports, implicate the amygdala, ventral striatum, and insula in the detection of emotional salience of a stimulus. The amygdala, along with neighbouring cortical sites including the orbital/medial PFC, and the subgenual ACC appear to be involved in the actual production of an affective state. Meanwhile, cortical regulation of the affective state proceeds through engagement of the vmPFC, dorsal ACC, and dlPFC. Additional subcortical modulation may also proceed through the hippocampus. Functional neuroimaging investigations have provided corroborative evidence in replicating the neuroanatomical localization established with exploratory lesion and stimulation studies.

Neuroimaging investigations specifically probing the effortful regulation of affect have emphasized dlPFC, but portions of the vlPFC may also be salient in the regulation of autonomic responses. The dorsal and rostral divisions of the ACC may additionally integrate general monitoring with self-referential representation. There is growing evidence to support the

51 functional dissociation between the dorsal and affective subdivisions of the ACC (Bush et al., 2000;Drevets and Raichle, 1998). The established role of the dorsal ACC in effortful control (Raichle et al., 1994), error-performance monitoring (Carter et al., 1998), conflict processing (Botvinick et al., 1999;Mohanty et al., 2007) and difficult task performance (Paus et al., 1998) highlight the potential role of the dorsal ACC in affective state regulation.

Investigations noting activation of the MCC in the perception of the unpleasantness of pain (Casey et al., 1994;Rainville et al., 1997;Vogt, 2005;Downar et al., 2003;Downar et al., 2002;Hutchison et al., 1999;Downar et al., 2000), anticipatory arousal and uncertainty (Critchley et al., 2001a), intentional regulation of autonomic arousal (Critchley et al., 2001b), support the role of this brain region in the monitoring of affective states. More specifically, increased MCC CBF has been reported during attention to subjective emotional states and experiences (Gusnard et al., 2001a;Lane et al., 1998) and inhibition of sexual arousal generated by erotic stimuli (Beauregard et al., 2001). Increased BOLD signal in the aMCC has been observed in HC and MDD subjects during the effortful down regulation of negative affect induced by visual stimuli in children (Levesque et al., 2004) and adults (Beauregard et al., 2006).

Neuroimaging investigations of the neighbouring dmPFC cortex implicate this cortical region in the regulation of autonomic responses and arousal associated with affective processing. Simpson and colleagues demonstrated an inverse relationship between CBF and measures of autonomic function and anxiety during a novel cognitive task (Simpson, Jr. et al., 2001). Increased activation has also been noted during self-referential judgments made when viewing stimuli with emotional content (Gusnard et al., 2001a;Fossati et al., 2003) and during the anticipation of pain (Ploghaus et al., 1999). Activation of the dmPFC has also been reported during the voluntary suppression of emotional reaction in response to comparable stimuli in children (Levesque et al., 2004;Urry et al., 2006) and adults

Activation of the dlPFC (BA 9, 44, 46) in relation to regulation of affective processes may be related to the representation and manipulation of non-emotive visuospatial and verbal components of emotional stimuli. Hariri and colleagues found that emotion-labeling tasks compared with more implicit tasks during the presentation of positive and negative facial expression were associated with increased activation in the dlPFC regions, traditionally reserved for working memory (Goldman-Rakic, 1988;Hariri et al., 2000). Activation of the dlPFC has

52 also been reported during the voluntary suppression of sadness (Levesque et al., 2003a;Eippert et al., 2007) and sexual arousal (Beauregard et al., 2001) induced by visual stimuli in adults.

Compelling evidence suggests that affective information, particularly responses to fearful faces in the amygdala may persist under some conditions of inattention or unawareness. For example, persistent amygdalar activation has been noted even when faces are i) shown at ignored locations (Anderson et al., 2003;Vuilleumier et al., 2001), ii) masked (Morris et al., 1998c;Whalen, 1998), iii) suppressed by binocular rivalry (Pasley et al., 2004;Williams et al., 2004), iv) presented on the neglected side in parietal patients (Vuilleumier et al., 2002) or in the blind field of patients with occipital lesions (Anders et al., 2004;Hamm et al., 2003;Morris et al., 2001). These findings raise the intriguing question of how can the presence of visual information reach the amygdala despite reduced, or absent, activation of the visual cortex or destruction of the visual cortical pathways?

Two possible neuroanatomical solutions have been proposed to solve this paradox. On the basis of accounts of sustained vision following occipital lesions (Cowey and Stoerig, 1991;Weiskrantz, 1956) and the existence of direct connection linking the thalamus and amygdala in rodents (Cowey and Stoerig, 1991;Linke et al., 1999) one possibility may involve a subcortical route that may provide a preliminary output to the amygdala through extra-geniculate projections bypassing the striate visual cortex (de et al., 2000;Hamm et al., 2003). Alternatively, other alternative routes to extrastriate areas, may mediate an initial feed forward sweep of inputs within the cortex, before the information is ‘carefully’ analyzed within V1 (Bar, 2003;Bullier, 2001).

Neuroimaging data suggest that amygdalar responses to fearful faces may be primarily driven by coarse visual features in facial expressions of affect (Vuilleumier et al., 2003;Winston et al., 2003;Whalen et al., 2004). These outlines of facial expression are carried in the low spatial

53 frequency spectrum of visual images that can be transmitted by magnocellular visual channels in subcortical pathways (Leventhal et al., 1985;Schiller and Tehovnik, 2001).

54 Pharmacology of Affective Processing

Neurotransmitters Involved in Dispositional Affect

Research aimed at uncovering the mechanism of action of antidepressant treatments has begun to shed light on the neurotransmitters involved in affective processing. Approximately half of the core symptoms of depression including loss of interest, loss of energy and loss of motivation are associated with decreased positive affect. Depressed mood, increased guilt, agitation, and anxiety, meanwhile are associated with increased negative affect (Clark and Watson, 1991;Watson et al., 1995b;Watson et al., 1995a).

Selective serotonin reuptake inhibitors (SSRI) are the most common pharmacological compounds currently used in the treatment of MDD (Kennedy et al., 2009a). Although their pharmacodynamic profile is primarily characterized by the rapid block of serotonin reuptake rapidly, their therapeutic action is delayed. It is hypothesized that the increase in synaptic 5-HT activates feedback mechanisms mediated by both 5-HT1A (cell body) and 5-HT1B (terminal) autoreceptors, which reduce the firing in 5-HT neurons and decrease the amount of 5-HT released, respectively. Thus, the short-term effect is attenuated 5-HT neurotransmission (Blier and de Montigny, 1994). Long-term treatment is hypothesized to desensitize the inhibitory 5- HT1 autoreceptors resulting in enhanced 5-HT neurotransmission (Blier and Ward, 2003). Excitatory inputs from noradrenergic neurons further modulate the firing of 5-HT neurons in the raphe nuclei. Specifically, antagonism of inhibitory alpha2-adrenoceptors may represent an additional antidepressant pharmacodynamic mechanism (Tremblay and Blier, 2006).

Evidence from antidepressant pharmacodynamic profiles suggests that a major depressive episode characterized by increased negative affect may be more responsive to agents with a serotonergic mechanism of action. Increases in multiple dimensions of positive affect, particularly increases in motivation, pleasure, and reward, are associated with medications that increase synaptic levels of dopamine (Belmaker, 2008). Antidepressant medications that increase noradrenergic synaptic availability appear to increase both positive affect by increasing attention, alertness, and decreasing negative affect by improving mood (Delgado, 2006).

55 Although these observations may assist in the pharmacotherapy selection strategy in MDD, the monoaminergic hypothesis of affect dysregulation is not without its shortcomings. First, no monoamine-related abnormality has been identified as a diagnostic screening tool for depression, or other mental disorders characterized by disturbances in mood or affect (Belmaker, 2008). Second, the depletion of serotonin or norepinephrine in HC has, for the most part, not been associated with significant changes in mood (Delgado, 2006).

Whereas, the pharmacological manipulation of serotonin and norepinephrine has had minimal effect on mood or affect in HC, administration of psychostimulants (amphetamines), which lead to increased synaptic release of dopamine, has been associated with significant reinforcing effects (Hart et al., 2001). Indeed, the mesocorticolimbic dopaminergic pathway has been extensively evaluated in human studies of nicotine, cocaine, and dextroamphetamine reward (Stein et al., 1998;Volkow et al., 1999). Future research should evaluate the role of other neurotransmitters that show promise as antidepressant medication.

Electroconvulsive therapy (ECT) is a non-pharmacological treatment modality usually reserved for subjects with poor response to antidepressant pharmacotherapy (Lisanby, 2007). Treatment with ECT is associated with similar profile of depressive symptom resolution, albeit the therapeutic result is usually more pronounced with a faster time of onset. Nutt and Glue reviewed clinical and preclinical data on ECT to correlate changes in symptomatology with changes in monoamine levels accompanying a full course of ECT treatment (Nutt and Glue, 1992).

Immediately after the first two seizures there is an increase in dopaminergic receptor function, with subjects experience a transient period of increased appetite and drive. With additional seizures these transient increases become longer, with slow increases in appetite and drive. Sometime between the third and fifth seizures, an increase in synaptic norepinephrine can be identified along with increases in the activity level of subjects. It is hypothesized that this effect may be mediated by changes in the auto-inhibitory regulatory receptors that inhibit norepinephrine release, which may become desensitized after a few seizures, leading to increase in synaptically available norepinephrine (Nutt, 2008). Following 6 seizures, the investigators noted increases in serotonergic function and clinical improvements in cognition, loss of negativity in cognition, and resolution of anxiety(Nutt and Glue, 1992).

56 Corroborating evidence from antidepressant pharmacodynamic profiles suggests that a major depressive episode characterized by increased negative affect may be more responsive to agents with a serotonergic mechanism of action. Increases in multiple dimensions of positive affect, particularly increases in motivation, pleasure, and reward, are associated with medications that increase synaptic levels of dopamine (Nutt et al., 2007). Antidepressant medications that increase noradrenergic synaptic availability appear to increase both positive affect by increasing attention, alertness, and decreasing negative affect by improving mood (Foote and Ashton-Jones, 1995;Stahl, 2003).

Effect of Psychotropic Agents of Affective Processing

The marked effects of dietary tryptophan depletion on affective processing in HC provide a rationale to examine the effects of antidepressants in this psychiatrically unaffected sample. Murphy and colleagues examined the effects of tryptophan depletion on both cognitive and affective processing in healthy volunteers in a double-blind, placebo-controlled, crossover design (Murphy et al., 2002). The investigators found that a-tryptophan-free amino acid mixture significantly lowered plasma total and free tryptophan and produced deficits similar to those observed in depressive disorders. Moreover, tryptophan depletion increased response times to happy, but not sad, targets in an affective go/no-go task and decreased response time to a visual discrimination and reversal learning task.

Although a number of neuroimaging investigations have evaluated the effects of antidepressant treatment on affective processing in depressed populations, fewer investigations have evaluated their effects without the confound of a depressive illness. Changes in affective processing following antidepressant treatment in depressed patients may either be due to the pharmacological actions of the medications, the effect of decreased depressive symptom severity, or more likely a combination of the two. Recent studies have shown that a single dose of an antidepressant can have modulatory effects on both positive and negative affective stimuli in non-depressed volunteers. In keeping with this view, antidepressant medication can be conceptualized as engaging similar neurobiological substrates as various psychosocial treatments that aim to correct affective biases in information processing (Serra et al., 2006).

It has been proposed that selective serotonin reuptake inhibitors (SSRIs) directly modulate 57 amygdala-dependent processing, with a normalization of excessive activity in depression and anxiety (Drevets, 2000). Recently, the effects of acute SSRI administration (citalopram) were evaluated on the processing of happy facial expressions and fear recognition (Harmer et al., 2003a). Compared to placebo, acute administration of citalopram enhanced both the response speed and accuracy of recognition of happy facial expressions. Furthermore, recognition of stimuli with low emotional intensity was more reliably performed in the citalopram group. Interestingly, the investigators found no effect of citalopram on any other emotions (sadness, anger or disgust); supporting the hypothesis that positive and negative affective processing involves dissociable neural circuits. Similar findings have been obtained with the use of tryptophan (Attenburrow et al., 2003).

Examination of the effects of citalopram on cortical electrophysiology during positive affective processing indicates increased occipitoparietal activity compared to placebo (Kemp et al., 2004). Whereas, negative affective processing was associated with significant occipitoparietal and frontal activation, citalopram's effects included attenuation of these regions, particularly the right anterior frontal and left anterior temporal regions. These findings suggest that the effect of citalopram may be to potentiate positive affect and suppress negative affect. Interestingly, a gender x affective valence was reported with significantly greater effects of citalopram on positive affect in men and negative affect in women. Moreover, significant changes in heart rate were reported for the placebo treatment condition in response to affective induction, but not for the citalopram condition, suggesting that citalopram may modulate the autonomic response to affective processing (Kemp et al., 2004;Thayer and Lane, 2000). Prolonged exposure to citalopram has also been associated with a reduction in the processing of negative affective stimuli. Following 7 days of citalopram administration at clinical doses (20 mg daily) HC were noted to display reduced recognition of negative facial expressions and improved memory for positive emotional information (Harmer et al., 2002).

The neural correlates of chronic citalopram administration have also been investigated in healthy controls performing a working memory task (Rose et al., 2006). Seven days of escitalopram (20 mg daily) had no significant effect on working memory accuracy or reaction time. Similarly, analysis of the neuroimaging data revealed no significant differences in memory-load-dependent activation between conditions.

58 More recently, the effects of 21 days of citalopram administration on amygdalar activation during a face emotion recognition have also been evaluated in a placebo-controlled design in unaffected HC (Arce et al., 2008). The investigators monitored medication adherence with weekly urinalysis and determined a reduction in amygdala activation in the escitalopram condition compared to placebo. Similar decreases in amygdalar reactivity were also noted after acute citalopram administration (Anderson et al., 2007).

Even more recent reports of a four-week antidepressant trial in HC have also been reported (Cardoso, Jr. et al., 2009).In this investigation, the fMRI response to positive and negative affect induction with the IAPS was compared before and after 4 weeks of low dose clomipramine treatment. In addition to confirming earlier reports of decreased BOLD signal in the amygdala during negative affect processing, the authors also noted decreased BOLD signal to both negative and positive visual stimuli in the aMCC and the anterior insula. A one-week trial of citalopram, was also successful in increasing the BOLD signal in the amygdala of HC during the presentation of happy faces (Norbury et al., 2009).

This is the first report of modulatory effects of repeated antidepressant use on the central representation of somatic states in response to emotions of both negative and positive valences in healthy individuals. Also, our results corroborate findings of antidepressant-induced temporolimbic activity changes to emotion-provoking stimuli obtained in studies of subjects treated acutely with such agents

A growing literature supports the efficacy of noradrenergic medications in the treatment of depression (Andreoli et al., 2002;Hajos et al., 2004). Noradrenergic neurotransmission may also be essential in memory performance of emotional material. For example, in the presence of amygdala activation, norepinephrine increases the likelihood that non-affective social or sensory stimuli are interpreted as aversive or emotionally arousing (Davidson and Irwin, 1999).

In a similar analysis to experiments conducted with SSRIs, Harmer and colleagues evaluated the effects of acute administration of the norepinephrine reuptake inhibitor (NRI), reboxetine, on affective processing. Tests of facial expression recognition, emotional categorization, and emotional memory in 24 psychiatrically unaffected HC revealed that reboxetine biased perception toward positive, rather than negative, without concurrent changes in non affective

59 processes or mood (Harmer et al., 2003b).

As with citalopram, greater recognition of happy facial expressions was reported with reboxetine compared with placebo, without changes in the recognition of other basic emotions. Moreover, in a surprise recall test, subjects receiving reboxetine recalled an equal proportion of positive and negative valenced words, while the group receiving placebo remembered significantly fewer positive words. The effects of reboxetine were specific to affective processes as non-emotional information processing was not affected by reboxetine.

Functional neuroimaging studies of psychiatrically unaffected HC receiving the beta adrenoreceptor antagonist, propranolol, further support the role of norepinephrine in manipulating affective processing. Van Stegeren and colleagues reported that with placebo administration, fMRI BOLD activation in the amygdala correlated with the affective intensity of the presented visual stimuli. Alternatively, administration of propranolol selectively decreased amygdala activation to affective, but not neutral, stimuli (van Stegeren et al., 2005).

Earlier work indicates that affective memory performance may be modulated by changes in central norepinephrine. For example, propranolol (pharmacologically active at both central and peripheral targets) impaired memory for affective stimuli while nadolol, an antagonist with exclusively peripheral actions had no effect on memory (Cahill et al., 1994;van Stegeren et al., 1998). Moreover yohimbine, the noradrenergic agonist, was found to enhance recall and recognition of affective material, while propranolol had the opposite effect (O'Carroll et al., 1999).

Evaluation of two antidepressants with different pharmacological profiles allows an inference on the neuropsychological basis of improved affective processing in subjects receiving pharmacotherapy. One study has compared the effect of a serotonergic (citalopram) versus noradrenergic (reboxetine) antidepressant on affective processing, and this study comparing 7 days of citalopram (20 mg daily), reboxetine (8 mg daily), or placebo treatment in 42 psychiatrically unaffected controls (Harmer et al., 2004). Compared to placebo, both antidepressants reduced the recognition of fearful and angry facial expressions, with citalopram additionally reducing recognition of disgust and surprise, although misclassification of fear as surprise was more common in the citalopram and reboxetine groups compared to placebo.

60 Correcting a possible negative bias (Phillips et al., 2003b), the groups receiving antidepressants were more likely than the placebo group to misclassify all 3 of these negative emotions as happy, and incidental memory for positive self-referent social information was also improved without any between-group differences, in terms of speed of correct responses (Harmer et al., 2004). There were also differences between antidepressants; citalopram produced a wider effect in the facial expression recognition task, while the reboxetine-treated group was relatively quicker to classify positive versus negative personality.

There is a paucity of studies evaluating the behavioural effect of acute or chronic antipsychotic administration. The currently published investigations involving healthy controls are limited to evaluations of atypical antipsychotic pharmacokinetics (Thyrum et al., 2000;Jann et al., 2006;Macias et al., 1998) and effects on sleep architecture (Gimenez et al., 2007;Lindberg et al., 2002;Sharpley et al., 2000;Sharpley et al., 2003).

The effect of the atypical antipsychotics amisulpride and risperidone on cognitive processing in a psychiatrically unaffected group was recently compared to the typical antipsychotic, chlorpromazine. Using a parallel group, placebo-controlled study, Barrett and colleagues investigated auditory and visual latent inhibition, prepulse inhibition of the acoustic startle response, executive function and eye movements in a large group of HC (n=136) receiving an acute dose of an antipsychotic (Barrett et al., 2004). Although there was no effect of the medication on auditory latent inhibition, prepulse inhibition or executive functioning, there was a trend towards disruption of visual latent inhibition by the atypical antipsychotics. Specifically, both risperidone and chlorpromazine decreased eye scan movement and increased visual processing error rates. In the case of the atypical antipsychotic, risperidone, visual impairments were correlated with observed severity of akathisia (Barrett et al., 2004).

Compared to the first-generation conventional antipsychotics, the newer generation ‘atypical’ antipsychotics have been associated with increased efficacy in the treatment of negative symptoms including apathy and anhedonia (Javitt, 2001). In an effort to explore the neural correlates of this preferential effect on negative symptoms, Juckel and colleagues measures changes in brain activation during a reward anticipation paradigm in schizophrenic subjects mediated with either conventional or atypical antipsychotics. The investigators noted activation of the ventral striatum in atypical antipsychotic treated schizophrenic subjects and an HC

61 comparator group, with, decreases in activation noted in the schizophrenic group treated with conventional antipsychotics (Juckel et al., 2006).

In summary, monoaminergic based antidepressants may enhance positive affect processing, and suppress components of negative processing. These effects appear to be observed with acute administration, and may be independent of changes in mood. The atypical antipsychotics appear to have a therapeutic effect on negative symptoms, and may not impair positive affect processing like their first-generation predecessors. Preliminary evidence suggests that atypical antipsychotics may have an effect on visual processing.

62 Abnormalities in Brain Structures Mediating Affective Processing in MDD

Neuroimaging investigations have identified structural and functional abnormalities in brain regions identified in the earlier part of the thesis as mediating the identification, processing, and regulation of affective states. The majority of structural MRI studies comparing MDD to HC have demonstrated volume reductions, within many of the regions involved in affective and cognitive processing including the amygdala and ventral striatum, subgenual cingulate gyrus, prefrontal cortical regions and hippocampus. Differences in resting state blood flow and glucose metabolism have also been identified in MDD subjects during an MDE compared to HC using PET and SPECT. Selective normalization of brain regions differentiating depressed MDD and HC has been observed following the administration of various modalities of antidepressants treatment and repeat neuroimaging data acquisition. Relatively fewer studies have explored differential brain metabolism and perfusion in remitted MDD subjects compared to HC.

Structural Findings

Computed tomography studies in subjects with mood disorders have consistently identified several volumetric abnormalities in brain neurocircuits that putatively subserve affect regulation and emotional expression (Strakowski et al., 2002a). Findings from postmortem studies have indicated reduced density and number of glial cells in the amygdala (Bowley et al., 2002), the sACC and orbitofrontal cortex (Cotter et al., 2001;Ongur et al., 1998), and the dlPFC cortex (Rajkowska et al., 1999), and reduced neuronal cell density in the vlPFC and dlPFC (Rajkowska et al., 1999), in subjects with MDD. Neuropathological abnormalities have also been reported in these subjects in the entorhinal cortex (Bernstein et al., 1998).

For the most part, MRI investigations have reported similar global cerebral volumes in MDD (Husain et al., 1991;Krishnan et al., 1992;Axelson et al., 1993;McIntosh et al., 2001;Coffey et al., 1993;Dupont et al., 1995;Sheline et al., 1996;Kumar et al., 1998;Pillay et al., 1997;Ashtari et al., 1999;Bremner et al., 2000;MacQueen et al., 2003) subjects compared to healthy controls. Studies which evaluated white and gray matter separately indicate that compared to the general population, regional abnormalities in white and gray matter may be a consistent finding in MDD compared to investigations where white and gray matter are not segmented (Salokangas et al., 2002;Shah et al., 2002;Kasai et al., 2003;Ballmaier et al., 2004a;Konarski et al., 2008).

Investigations scrutinizing ventricle to brain ratios (VBR) have consistently reported relative increases in ventricular size in heterogeneous samples of mood disorder populations (Swayze et al., 1990;Strakowski et al., 1993;Zipursky et al., 1997;Lim et al., 1999;Hauser et al., 2000;Strakowski et al., 2000;Strakowski et al., 2002b;Salokangas et al., 2002;Morys et al., 2003;Davis et al., 2004;Strakowski et al., 1993;Pearlson et al., 1997). Conspicuously, there has been no consistent association between ventricular abnormalities and any historical illness variables (e.g. age of onset, number of episodes duration of illness) possibly due to the lack of regional specificity of ventricular volume data (Strakowski et al., 2000).

A consistently reported finding in MDD populations has been volumetric reduction of the prefrontal cortex (Coffman et al., 1990;Krishnan et al., 1992;Coffey et al., 1993;Kumar et al., 1998;Sax et al., 1999;Lavretsky et al., 2004). The prefrontal lobe has been segmented into smaller regions of analysis including dlPFC, OFC, subgenual prefrontal cortex (SGPFC), and the ventral and dorsal regions of the ACC.

The proposed roles of the DLPFC include temporally organizing behaviour (Fuster, 1997), working memory (Goldman-Rakic, 1995), attention to different kinds of information (Petrides, 1994), integration of diverse cognitive demands (Duncan and Owen, 2000) and response selection (Rowe et al., 2000). The observed hypoactivity of the DLPFC in functional neuroimaging studies of depressed MDD and BD subjects (Kennedy et al., 2001;Kruger et al., 2003), and its reduced size in structural analyses, may be relevant to the cognitive impairments in mood disorders (Quraishi and Frangou, 2002;Ottowitz et al., 2002;Rajkowska, 2000;Soares, 2003).

Current evidence implicates the OFC in decoding and representing primary reinforcers such as taste and touch; in learning and reversing associations of visual and other stimuli to primary reinforcers; and in controlling and correcting reward-related and punishment-related behaviour (Rolls, 2004;Rolls et al., 2003;De Araujo et al., 2003;O'Doherty et al., 2001a). Preclinical animal data indicate that the OFC plays an important role in regulation of emotional responses through inhibitory inputs to the amygdala and other brain regions believed to modulate reward and punishment representation (Morgan and LeDoux, 1999;Morgan et al., 1993). The blunting of affect following frontal lobotomy - where lesions include the OFC – also suggests a role for this

64 area in the regulation of emotion and mood (Bremner et al., 2002), although the clinical procedure of lobotomy may affect other adjacent brain regions as well. Indeed, reduced OFC volumes in MDD (Lai et al., 2000;Bremner et al., 2002;Lee et al., 2003;Steffens et al., 2003;Taylor et al., 2003;Ballmaier et al., 2004b;Lavretsky et al., 2004) have been frequently reported. Some investigators, however, have failed to find differences or even noted increases in total gray matter volume in the OFC (Steingard et al., 2002;Salokangas et al., 2002;Ballmaier et al., 2004a;Ballmaier et al., 2004b;Lacerda et al., 2004;Hastings et al., 2004;Janssen et al., 2004). This equivocal evidence suggests that reductions in volume may be associated with a specific clinical presentation of MDD that is not universally expressed by all MDD patients.

Like the neighbouring OFC, the SGPFC also shares extensive connections with structures implicated in affect regulation and autonomic responses to stressors (Sesack et al., 1989;Carmichael and Price, 1995b;Carmichael and Price, 1995a). Human lesion studies indicate that damage to these regions is associated with abnormal autonomic responses to emotional challenges, impaired reactivity to emotional cues, and poor judgment (Damasio et al., 1990;Bechara et al., 1996). It has been hypothesized that disordered interaction between the SGPFC and interconnected structures may contribute to the pathological guilt or anxiety characterizing MDD, and to the rapid shifting between anger and euphoria characteristic of mania in BD (Drevets et al., 1997).

Following the initial report of a smaller, yet hyperactive, SGPFC in BD and MDD (Drevets et al., 1997), these volumetric results have been corroborated in MDD (Botteron et al., 2002). Structural abnormalities in the SGPFC may be the result of multiple affective episodes, as BD adolescents and children were not found to have volumes different from age-matched controls (Sanches et al., 2005). Alternatively, it is possible that individuals with BD have a different developmental change in a direction opposite to what is seen in psychiatrically unaffected control subjects.

The ventral region ACC comprises the sACC and the pACC (Devinsky et al., 1995). There are extensive connections between the sACC and the amygdala, periaqueductal gray, mediodorsal and anterior thalamic nuclei, nucleus accumbens and ventral striatum – connections that have been associated with autonomic function and emotional behaviour (Devinsky et al., 1995;Paus,

65 2001). Lesions of the sACC impair the ability of the autonomic system to respond to conditioned stimuli and result in variety of different changes in emotional behaviour, ranging from apathy and depression to disinhibition and anxiety (Angelini et al., 1981;Levin and Duchowny, 1991;Frysztak and Neafsey, 1991;MacLean and Newman, 1988). Stimulation of this region is also associated with changes in autonomic and endocrine function (Phillips et al., 2003a). Functional neuroimaging data demonstrate increased cerebral blood flow and activation within the subgenual and pregenual ACC during mood induction paradigms (Mayberg et al., 1999;Shin et al., 2000), suggesting an important role for the region in the production of affective states and in the subsequent processing of emotional information during arousal (Phillips et al., 2003a).

Lesions of the MCC – putatively implicated in cognitive functioning – are associated with attentional deficits and impaired performance on tasks requiring controlled processing (Ochsner et al., 2001;Paulus et al., 2004). Functional neuroimaging studies have demonstrated increased blood flow within the MCC, but decreased blood flow within sACC during the performance of tasks requiring selective attention and novel response selection, and vice versa for performance of emotional tasks (Bush et al., 2000). Thus, atrophy of the ACC may impair the recognition, processing, and the effortful regulation of mood and affect in subjects with MDD.

Aberrant function of the ACC has been among the most consistent findings in mood disorders, particularly MDD (Mayberg, 2003a). A few reports of decreased volume (Lochhead et al., 2004;Hastings et al., 2004) and grey matter (Ballmaier et al., 2004b;Lyoo et al., 2004;Wilke et al., 2004) in BD and MDD suggest that both functional and structural abnormalities may be operative.

There is no consensus regarding the polarity of volumetric changes in the temporal cortex in MDD. Measures of the temporal lobe in MDD are restricted to one study in the elderly that reported a unilateral increase (Ballmaier et al., 2004a). It has been hypothesized that the hippocampus serves as a general-purpose comparator, with a central role in determining the extent of conflict between different goal-directed behaviours, in facilitating exploratory rather than defensive patterns of behaviour, and allowing resolution of goal conflict (McNaughton and Wickens, 2003). The hippocampus has also been implicated in affect regulation through facilitation and inhibition of defensive behaviour and anxiety in response to threatening or potentially threatening environmental contexts. The contribution of the hippocampus to spatial

66 cognition (O'Keefe and Nadel, 1978) and episodic memory (Squire et al., 1992;Squire, 1992) must also be noted in reference to the impairments seen in subjects with mood disorders (Danion et al., 1991;Burt et al., 1995). In concert with impaired signaling to DLPFC and ACC, deficits in hippocampal structure and function may underlie the cognitive dysfunction associated with depression (Bremner, 2002).

The majority of studies examining hippocampal volume in MDD populations report decreased volume (Sheline et al., 1996;Shah et al., 1998;Sheline et al., 1999;Bremner et al., 2000;Bell- McGinty et al., 2002;MacQueen et al., 2003;Frodl et al., 2004a;Frodl et al., 2004b;MacMaster and Kusumakar, 2004;Caetano et al., 2004;Janssen et al., 2004), and while some studies have not been able to detect a statistically significant difference (Mervaala et al., 2000;Vakili et al., 2000;Posener et al., 2003;Hastings et al., 2004), there are no reports of a larger hippocampus in MDD. Indeed, a recently conducted meta-analysis concluded that the pooled effect size of depression was significant in both hemispheres for subjects with MDD, however, the number of depressive episodes significantly correlated with decreases only in the right hippocampus (Videbech and Ravnkilde, 2004).

The amygdala is an important component of the system involved in fear recognition (Adolphs et al., 1994) and the acquisition, storage, and expression of fear memory (LeDoux, 2000). Amygdala lesions in humans result in emotional blunting (Aggleton and Brown, 1999) and reduced fear recognition (Adolphs et al., 1994) and conditioning (Bechara et al., 1995). Functional neuroimaging studies describe amygdala activation in response to induction of positive and negative emotional states (Reiman et al., 1997), fear (LaBar et al., 1998;Buchel et al., 1999), and presentations of fearful and angry facial expressions (Morris et al., 1998c;Whalen et al., 1998b).

It has been proposed that a smaller, but hyperactive, amygdala, as often visualized in MDD, may result in a restricted emotional range, but with a bias toward the identification of negative rather than positive emotions (Phillips et al., 2003b). Smaller amygdala have been reported more frequently in MDD (Sheline et al., 1998;Caetano et al., 2004;Siegle et al., 2003;Hastings et al., 2004) than enlarged amygdala (Bremner et al., 2000;Frodl et al., 2002;MacMillan et al., 2003). One study conducted in adolescents and young adults found that with increasing age left amygdala volumes increased in subjects, but decreased in healthy controls (Chen et al., 2004a).

67 This finding is consistent with studies including younger populations detecting amygdala volume decreases in BD (Blumberg et al., 2003a;Chen et al., 2004a;DelBello et al., 2004).

The basal ganglia are composed of the caudate nucleus, putamen, globus pallidus, subthalamic nucleus, and substantia nigra. The extant data do not permit an evaluation of each region in BD and MDD; however some trends are beginning to emerge.

In concert with the insula, and thalamus, the ventral striatum has been implicated in the identification of displays of disgust (Phillips et al., 1997;Jensen et al., 2003); taste perception (Small et al., 1999); craving (Breiter et al., 1997); time-locked processing of reward prediction (Pagnoni et al., 2002); anticipation of reward (Knutson et al., 2001), and in romantic love (Bartels and Zeki, 2000). Loss of control over these processes has the potential to mimic the clinical characteristic of mania, and perhaps related to this, subjects with BD show abnormal activation of these regions during timed trials (Caligiuri et al., 2003) and emotional processing (Lawrence et al., 2004;Malhi et al., 2004b). Hyperactivity of the caudate has also been one of the most consistent findings in anxiety disorders, particularly obsessive-compulsive disorder (OCD) (Saxena et al., 2002;Saxena et al., 2003;Aouizerate et al., 2004), a common comorbidity in BD (McElroy et al., 2001). There are indications that these brain regions may be smaller in MDD (Husain et al., 1991;Krishnan et al., 1992;Greenwald et al., 1997).

Traditionally perceived as a modulator of motor control, the role of the cerebellum in neuropsychiatry is being re-evaluated amidst reports of cerebellar abnormalities in subjects with mood disorders as revealed by functional imaging and histopathology (Harrison, 2000;Ketter et al., 2001;Kimbrell et al., 2002;Konarski et al., 2004). Limited data from structural neuroimaging studies are limited to a single investigation in MDD failing to find differences in cerebellar gray matter content (Pillay et al., 1997).

The production and release of hormones by the anterior and posterior pituitary is modulated through feedback loops, involving the adrenal, thyroid, gonads, and discrete brain regions like the hypothalamus (Wilson et al., 1998). Abnormalities in some of the neuroendocrine loops involving the pituitary have also been reported in mood disorders (Herman and Cullinan, 1997;Anisman and Merali, 2003). It remains to be elucidated whether the structural abnormalities observed in the pituitary precede, accompany, or follow the onset of mood

68 disorders.

Three investigations have reported pituitary volume changes in mood disorders, one indicating a larger pituitary in MDD (MacMaster and Kusumakar, 2004) and the other, a reduction in BD (Sassi et al., 2001). As with the SGPFC, decreases in pituitary volumes may depend on the course of the illness, as adolescent bipolar subjects were found to have pituitary volumes comparable to healthy controls (Chen et al., 2004b).

Metabolic and Cerebral Perfusion

The first reports of cerebral perfusion in MDD made use of inhaled 133Xe inhalation to reveal decreased perfusion bilaterally in the prefrontal lobe (Mathew et al., 1980) in MDD subjects. Moreover, reductions in CBF correlated significantly with severity of depressive symptoms. Later investigators employed 99TC SPECT to confirm reductions of CBF in the anterolateral prefrontal cortex (Ebert et al., 1991;Mayberg et al., 1994). In a large comparison of 40 MDD subjects and 20 HC, Austin and colleagues reported in the MDD group in the majority of cortical and sub-cortical regions examined, most significantly in temporal, inferior frontal and parietal areas (Austin et al., 1992).

Early investigations of glucose metabolism in MDD, HC, and schizophrenic subjects reported decreased ratio of anterior: posterior brain metabolic activity, particularly at superior slice locations (Buchsbaum et al., 1984a;Buchsbaum et al., 1984b). Improved resolution of PET glucose metabolic activity led to a series of reports of decreased metabolic activity in the dlPFC (Dolan et al., 1992;Buchsbaum et al., 1997;Baxter, Jr. et al., 1989;Ho et al., 1996;Biver et al., 15 1994;Pizzagalli et al., 2004). Similarly, decreases in CBF detected with O-H2O PET were also noted in the dlPFC (Bench et al., 1993;Bench et al., 1995). Contrary to findings in the superior portions of the prefrontal lobe, increased metabolism and blood flow has been reported within the vlPFC cortex (Drevets et al., 1992;Nofzinger et al., 1999;Osuch et al., 2000;Pizzagalli et al., 2004;Biver et al., 1994;Kennedy et al., 2007b).

Compared to HC, changes in resting state metabolism have been reported in both directions in the dorsal ACC in MDD (Dolan et al., 1992;Pizzagalli et al., 2004;Mayberg et al., 1997;Biver et al., 1994). Changes in blood flow, on the other hand, have documented exclusively decreased

69 perfusion (Bench et al., 1993;Bench et al., 1995). Although hypermetabolism (Wu et al., 1999;Nofzinger et al., 2005;Nofzinger et al., 2005;Dougherty et al., 2003) and increased CBF (Mayberg et al., 2005) in the ventral, or subgenual portion of the ACC have been found in treatment resistant MDD compared to HC.

Increased blood flow within the amygdala (Drevets et al., 1992), thalamus (Baxter, Jr. et al., 1989) (Drevets et al., 1992), and ventral limbic regions, including the anterior insula and ventral striatum (Mayberg et al., 1999), and a positive correlation between amygdalar metabolism and the severity of depressed mood (Abercrombie et al., 1998) have been reported in subjects with major depressive disorder. Decreased metabolism in other portions of the temporal lobe have also been reported (Dunn et al., 2002;Post et al., 1987;Pizzagalli et al., 2004;Kumar et al., 1993), along with decreased metabolism in the caudate (Kumar et al., 1993;Saxena et al., 2001;Baxter, Jr. et al., 1985;Goldman et al., 1995).

Several longitudinal investigations have sought to characterize the effect of psychotropic administration on volumetric changes. For example, paroxetine treatment of OCD has been associated with a normalization (reduction) of amygdala (Szeszko et al., 2004) and thalamic volumes (Gilbert et al., 2000). Cross sectional investigation evaluating unmedicated BD subjects and their chronically- medicated counterparts revealed a larger SGPFC volume (Ketter, 2005) and an increase in prefrontal grey matter volume (Moore et al., 2000a) associated with mood stabilizer use.

Recovery from a major depressive episode after successful pharmacological treatment has been associated predominantly with increased metabolism and blood flow within dmPFC and dlPFC (Baxter, Jr. et al., 1989;Brody et al., 1999;Buchsbaum et al., 1997;Kennedy et al., 2001;Mayberg et al., 1999;Mayberg et al., 2000) and the dorsal ACC (Kennedy et al., 2001). Studies examining the effect of pharmacologic and interpersonal therapies on prefrontal activity in subjects with major depressive disorder have either failed to demonstrate increased prefrontal blood flow (Martin et al., 2001) or have reported decreased prefrontal metabolism (Brody et al., 2001a;Goldapple et al., 2004) after these treatments.

Other changes reported after treatment include reduced metabolism in the sACC (Drevets et al., 2002;Mayberg et al., 1999;Mayberg et al., 2000;Kennedy et al., 2007b) and in other regions

70 important for the generation of emotional states, including the thalamus, ventral striatum, and insula (Mayberg et al., 1999;Nobler et al., 1994;Smith et al., 1999;Mayberg et al., 2000;Kennedy et al., 2007b). Within the hippocampus, increased metabolism has been associated with 1-week treatment with fluoxetine, whereas decreased metabolism has been associated with a response to 6 weeks of treatment with this medication (Mayberg et al., 2000).

There are discrepant findings regarding the role of the p ACC in major depression. Although there are reports of increased activity in this region during depressive episodes (Drevets, 1999), metabolism has been demonstrated to be abnormally decreased in this region in depressed subjects who had poor responses to treatment (Mayberg et al., 1997). Furthermore, although a response to medication at 6 weeks of treatment with selective serotonin reuptake inhibitor medication is associated with decreased regional cerebral blood flow in the subgenual ACC (Mayberg et al., 1999;Mayberg et al., 2000), this is also associated with increased regional cerebral blood flow in the pregenual ACC (rostral region of Brodmann Area 24; (Mayberg et al., 1999;Mayberg et al., 2000). Decreased regional cerebral blood flow within this region has also been demonstrated in healthy volunteers during the induction of sad mood (Mayberg et al., 1999).

Decreases in glucose metabolism in ventral regions of the prefrontal cortex (Nobler et al., 1994;Brody et al., 1999;Brody et al., 2001a;Kennedy et al., 2001;Brody et al., 1999) and increases in the temporal cortex (Brody et al., 2001b;Buchsbaum et al., 1997) have been previously associated with response to SSRIs. Additional pre-post changes in sACC (BA25), vlPFC, RSC (BA29) and putamen have also been reported with non-SSRI antidepressant pharmacotherapy (Davies et al., 2003;Martin et al., 2001).

It has been suggested that pharmacologic interventions may have a direct effect in reducing activity within regions important for the generation of emotional states, including the limbic structures described above, resulting in a “relaxation” of activity within the vlPFC cortex. Non- pharmacological interventions, on the other hand, may serve to increase activity within the vlPFC cortex to enhance the role of this structure in the regulation of emotional behaviour (Drevets, 2000). The co-localization of common regional brain metabolic changes associated with response to either psychotherapy or pharmacotherapy may represent treatment-independent effects of clinical response. In a randomized controlled trial of venlafaxine versus CBT, response

71 to either treatment modality was associated with decreased glucose metabolism bilaterally in the orbitofrontal cortex and left medial prefrontal cortex, along with increased metabolism in the right occipital-temporal cortex (Kennedy et al., 2007b).

The currently published evaluations of pharmacotherapy treatment in mood disorders, particularly in major depressive disorder, have been largely limited to traditional (monoamine level elevating) antidepressants (i.e. TCAs, MAOIs, SSRIs). In spite of studies in psychotic disorders, particularly schizophrenia, that have evaluated cognitive function before and after the administration of an atypical antipsychotic (Fusar-Poli et al., 2007;Wolf et al., 2007;Meisenzahl et al., 2006;Kumari et al., 2006;Juckel et al., 2006;Snitz et al., 2005;Davis et al., 2005), there are no cross-sectional, or longitudinal investigations evaluating the effects of atypical antipsychotics in mood disorders, let alone, MDD. Thus, there is a practical need to evaluate the antidepressant neuroimaging profile of atypical antipsychotics, especially in light of recent , placebo-controlled evidence of antidepressant efficacy of atypical antipsychotics even when administered monotherapy (Vieta et al., 2007;Brugue and Vieta, 2007).

In summary, differences in brain regions mediating i) the recognition of affective stimuli, ii) the production of affective states, and iii) the regulation of affective states have been reported between MDD and HC. Moreover, these differences have been reported using structural, and a variety of functional neuroimaging modalities. The functional consequence of these abnormalities is explored further in the next section, which reviews functional neuroimaging investigations comparing various components of affective processing between HC and MDD groups.

Affective Processing in MDD

The methodology underlying induction of a mood or affect in MDD subjects has for the most part followed investigations conducted in HC. Although various methodologies have been employed, they can be broadly categorized into recall induction, visual induction, and pharmacological induction. Investigators examining affective processing in MDD have typically adapted methodologies used earlier in HC samples. The presentation of aberrant affective processing in MDD will first be prefaced by findings in HC.

72 Pharmacological Induction

The monoamine hypothesis of depression posits that depletion in synaptic levels of serotonin (5- hydroxytryptamine, 5-HT), norepinephrine (NE), and/or dopamine (DA) may be salient to the pathophysiology and treatment of MDD (Gottfries et al., 1974;Delgado, 2000). The mechanistic basis of monoamine depletion involves either synthesis inhibition (NE or DA) (Sjoerdsma et al., 1965), or synthesis reduction (5-HT) (De Marte and Enesco, 1985). Although the depletion of serotonin or norepinephrine in HC has, for the most part, not been associated with significant changes in mood or cognition, the technique has been useful in characterizing the monoaminergic specificity of the therapeutic effect offered by SSRI and NRI antidepressant.

Serotonin When male HC ingested amino acid mixtures that were either i) tryptophan-free, ii) balanced, or iii) contained excess tryptophan, corresponding changes in change in plasma tryptophan were observed (Young et al., 1985). Moreover behavioural effects were also observed as subjects in the tryptophan-free group reported significantly elevated depression severity scores, and decreased performance on a proofreading task with dysphoric distractors (Delgado et al., 1989).

Tryptophan depletion has also been studied in medication free euthymic MDD (Smith et al., 1997;Leyton et al., 1997), symptomatic MDD (Delgado et al., 1991;Delgado et al., 1994), and MDD subjects treated with serotonergic and noradrenergic (Miller et al., 1996a;Delgado et al., 1999;Delgado et al., 1990;Delgado et al., 1993) antidepressants. The most common behavioural effect of tryptophan depletion in MDD has been negligible, with the notable exception of MDD subjects treated with serotonergic antidepressant (i.e. SSRIs). In MDD subjects successfully treated with an SSRI or SNRI, there is a 50-80% chance of a transient return of depressive symptoms (Delgado, 2006). The relapse of depressive symptoms appears was delimited to MDD subjects successfully treated with serotonergic antidepressants, as subjects treated with a noradrenergic antidepressants appear immune to the depressogenic effects of tryptophan depletion (Moore et al., 2000b).

The neural correlates of the of tryptophan depletion-induced depressive relapse have been evaluated with 18FDG in a group of MDD subjects stabilized on SSRIs who were subsequently randomized to a tryptophan-depleting diet or placebo (Bremner et al., 1997). A comparison of

73 brain metabolism in subjects with and without depressive relapse revealed decreased brain metabolism in the dlPFC, thalamus, and OFC. Moreover, there was an inverse correlation between brain metabolism in these regions and the severity of depressive symptoms.

Norepinephrine The administration of α-methyl-para-tyrosine (AMPT) has been shown to reversibly inhibit the activity of tyrosine hydroxylase, the rate limiting enzyme in the conversion of the amino acid tyrosine to L-3,4-dihydroxyphenylalanine – a precursor of NE and DA (Engelman et al., 1968;Bunney, Jr. et al., 1971). Analogous to results obtained with tryptophan depletion in MDD subjects treated with serotonergic antidepressants, administration of AMPT also induces a relapse of depressive symptoms; 80% of MDD subjects treated with noradrenergic antidepressants, compared to 20% relapse rates in SSRI treated subjects (Delgado et al., 1990;Miller et al., 1996b).

The neural correlates of AMPT induced depressive relapse were investigated in a trial comparable to the tryptophan depletion studies (Bremner et al., 1997). In this investigation, asymptomatic MDD subjects responding to noradrenergic antidepressant pharmacotherapy were randomized to receive either AMPT or placebo. Comparison of regional brain metabolism rates in subjects with and without AMPT-induced return of depressive symptoms, revealed widespread decreased cortical metabolism with maximal decreased noted in the OFC, dlPFC (P =.03) and subcortically in the thalamus. A retrospective analysis of baseline scans identified increased metabolism in prefrontal and limbic areas as markers of vulnerability to the return of depressive symptoms (Bremner et al., 2003).

Dopamine The mesocorticolimbic dopamine system has been implicated in human studies of nicotine, cocaine, and dextroamphetamine reward (Stein et al., 1998;Volkow et al., 1999). The psychostimulant amphetamines have been reported to increase neuronal dopamine release (Baird and Lewis, 1964) and are associated with significant reinforcing effects (Hart et al., 2001). Tremblay and colleagues conducted a double-blind, placebo-controlled, randomized, parallel study, to investigate the behavioural and physiological effects of acute dextroamphetamine in 40 medicated MDD subjects and 36 HC (Tremblay et al., 2002). A significant correlation on the reported measures of positive affect and severity of depression was observed suggesting the

74 presence of a hypersensitive, but hypofunctional brain reward system in MDD.

In a follow-up investigation, fMRI imaging of affective visual stimuli was performed in concert with a dextroamphetamine dopaminergic probe in both MDD and HC groups (Tremblay et al., 2005). In addition to the previously reported hypersensitive response to the rewarding effects of dextroamphetamine (2-fold higher in the MDD group), altered brain activation to affective stimuli was also reported. At peak drug concentrations, there was decreased activation to affective (positive and negative) stimuli in the right vlPFC (BA 10), left OFC (BA 11), right mPFC (BA 25), and bilaterally in the caudate and putamen. Bilateral increases in activation in the temporal pole in the MDD group were also reported.

Recall Induction

Early neuroimaging investigations of affective manipulations relied on the mood recall methodology. During these sessions, subjects are generally asked to recall or re-experience personal life episodes marked by sadness, happiness, anger or fear under neuroimaging conditions.

15 One of the first investigations by Pardo and colleagues used O-H2O PET to measure CBF in seven HC under two different conditions (Pardo et al., 1993). During the control condition, subjects were resting with their eyes closed. During the active condition, they were then asked to imagine or recall a situation producing profound sadness, while maintaining their eyes closed. Specific instructions were provided for subjects to produce feelings of sadness devoid of any anger or anxiety. In this preliminary analysis, women demonstrated increased bilateral inferior and orbitofrontal CBF, whereas men displayed predominantly left-sided CBF increases. In a similar investigation, eleven male HC were also asked to recall sad personal memories in the active condition and either recall affectively neutral personal events, or simply rest without any 15 recall (Gemar et al., 1996). Decreases in H2O PET CBF were noted in the left dlPFC, left mPFC, and left temporal cortex in response to memories with a negative affect component.

One of the earliest neuroimaging investigations evaluating the effect of affective components on cognitive function, instructed 12 HC to imagine and plan behaviour in situations with and 15 without negative affect under O-H2O PET scanning conditions (Partiot et al., 1995). Planning

75 in the absence of an affective context was characterized by activation of the dl PFC and posterior temporal cortex, while the mPFC and anterior temporal cortex were more activated in the presence of imagined sadness. These early studies highlighted the general deactivation of the dlPFC in tasks where negative emotional stimuli are introduced.

Subsequent studies involving recall induction investigated the correlates of both positive and 15 negative affect. George and colleagues measured O-H2O PET CBF in 11 female HC recalling affect-appropriate (happy, sad, and neutral) life events (George et al., 1995). Compared with the neutral condition, negative affect laden memories were associated with increased CBF in bilateral limbic and paralimbic structures (pACC/aMCC, medial prefrontal, and mesial temporal cortex), as well as brainstem, thalamus, and caudate/putamen. Conversely, the recall of positive affect life events was not associated with any significant increases in CBF, but was instead associated with widespread decreases in cortical CBF, with maximum decreases in the right prefrontal and bilateral temporal-parietal cortex.

To investigate sex differences in affective processing, George and colleagues replicated their previous investigations with 10 male HC and 20 female HC (George et al., 1996). Decreases in 15 H2O PET CBF were noted in the left dlPFC, left mPFC, and left temporal cortex in response to memories with a negative affect component.Although there were no between-group differences in difficulty, effort required, or degree of positive or negative affect induced, women activated a significantly wider portion of the limbic system during the negative affect induction. Moreover at rest, women exhibited decreased temporal and prefrontal cortex CBF, and increased brainstem CBF. Results from neuroimaging investigations evaluating both positive and negative affect induced by personal recall highlighted that positive and negative affect processing engages different brain regions, as opposed to, opposite activity in identical brain regions.

In one of the largest investigations evaluating affective processing in psychiatrically unaffected HC, the neural basis of emotion and feeling was investigated in 41 HC who were asked to recall 15 and relive personal life events characterized by sadness, happiness, anger or fear under O- H2O PET CBF scanning conditions (Damasio et al., 2000). Analysis of common increases in CBF during all affective processes indicated increased blood flow in the somatosensory cortex, and upper brainstem nuclei. These results were used to support the author’s hypothesis that affective processing is partly grounded in neurocircuitry representing several aspects of an organism's

76 continuously changing internal state of homeostasis (Damasio et al., 2000).

In an elegant study designed to demonstrate that depressed mood in MDD is comparable to self- 15 induced negative affect in HC, Mayberg and colleagues compared changes in O-H2O PET CBF accompanying provocation of negative affect in HC to resolution of chronic dysphoric symptoms in MDD with 18FDG (Mayberg et al., 1999). A reciprocal series of changes in CBF consisting of increased perfusion in limbic-paralimbic blood regions and decreases in neocortical regions (right dlPFC, inferior parietal) were reported in the HC group. Meanwhile, recovery from depression was associated with the reverse pattern, in the same regions; limbic metabolic decreases and neocortical increases. Moreover, a significant inverse correlation between sACC and right dlPFC activity was reported in HC and MDD groups in CBF and FDG, respectively.

Further investigation into the role of limbic-cortical activation under transient emotional stress as 15 a function of personality style was investigated in a group of HC with O-H2O PET CBF. An sACC centered limbic-cortical network was identified using the multivariate partial least squares technique (Keightley et al., 2003). When the HC group was divided into temperamentally negative and temperamentally positive groups, a divergent SGPFC (BA 25)-mediated network was found to differentiate the two. Specifically, using SGPFC as a seed in a network analysis, a prefrontal network occupying the vlPFC, dlPFC, and dmPFC was found to correlate positively with the SGPFC activity during a mood challenge in the temperamentally negative group, while a negative correlation was reported in the other group.

A separate investigation using self-recall to induce negative affect, hitherto a rarely used self- recall methodology, evaluated three groups; individual with remitted MDD, acutely symptomatic 15 MDD, and HC with O-H2O PET CBF during the reflection on a personal events evoking negative affect (Liotti et al., 2002). In contrast to the HC group, changes in affect through the mood provocation in both depressed groups were associated with decreases in regional CBF in medial OFC (BA 10, 11). Moreover, in the HC group CBF increases in the sACC (BA 25) and a decrease in right dlPFC (BA 9) were not observed in the depressed groups. An analysis of the difference between acutely depressed and remitted depressed subjects identified a decrease in CBF decrease in pregenual ACC (BA 24), and in increase in the globus pallidus (GP) in the remitted MDD group. Conversely, in response to mood provocation, the acutely depressed MDD group exhibited decreases in CBF in the ventral insula, the parahippocampal gyrus (BA 35), and

77 the superior and middle temporal cortex (BA 21, 22) (Liotti et al., 2002).

Studies of affective processing using the recall induction method have for the most part made use 15 of the O-H2O PET CBF neuroimaging modality. In addition to unwanted effects of exposure to a radionuclide, PET measurements of CBF dictate the temporal dynamics of the experimental design. As the acquisition of a single functional volume is on the order of minutes, activation epochs or conditions must also be of similar duration. Uncertainty over the magnitude and stability of the induced change in affect are an obvious limitation of this relatively lengthy acquisition time.

Effects of Induction Mode and Cognitive Demand In a meta-analysis of neuroimaging studies examining affective processing, Phan and colleagues found that affective induction through recollection or recall activated the ACC (specifically the aMCC); more frequently than visual and auditory induction studies (Reiman, 1997;Teasdale et al., 1999;Phan et al., 2002). These observations confirm the hypothesis that recruitment of the MCC may be specific to cognitively demanding emotional (Phan et al., 2002) as subjects must recall and imagine emotionally laden memories and internally generate intense target affect.

Moreover, there was a three-fold higher likelihood of increased CBF in the anterior insula in recall-induced changes in positive and negative affect compared to the induction of affect through visual stimuli (Reiman et al., 1997;Lane et al., 1997c). When the two modes of affective induction were compared, emotional recall, but not the viewing of emotional film activated the insula (Damasio et al., 2000). In an investigation of a variety of individual emotions all induced by recall, activations in the pACC, aMCC, insula, and brainstem were found in response to happiness, sadness, fear, and disgust (Damasio et al., 2000). These neuroimaging findings support the notion of the insula as monitoring the internal affective state of the organism (Reiman, 1997), and acting as an alarm center for appropriate homeostatic changes (Lane et al., 1997a).

Conversely, activation of the occipital cortex (BA 18, 19) was almost exclusively reported in neuroimaging investigations where changes in affect were elicited by visual stimuli (Amaral and Price, 1984;Corbetta et al., 1993). As activations in the visual association areas have been reported for both positive and negative affective stimuli, it has been proposed that

78 occipitotemporal (OPT) cortical regions may be involved in the evaluation of visual stimuli with emotional relevance (Lane et al., 1997a; Beauregard et al., 1998). Alternatively, recruitment of the occipital cortex may due to the highly arousing nature of the stimuli with activation of other perceptual areas (Lang et al., 1998; Taylor et al., 2000). Differential eye movements between affective and neutral stimuli may account for some activation in these areas, although when eye- movement was measured, Lang and colleagues et al. (1998) did not find any statistical significant differences in the duration or magnitude of scanning eye movements between the stimuli.

A neuromodulatory function of the amygdala on the extrastriate cortex has been proposed on projections from the amygdala into the primary and associative visual cortex (Iwai et al., 1987;Aggleton and Mishkin, 1990) and input into the amygdala from the temporal visual- association areas (Sprengelmeyer et al., 1998;Morris et al., 1998a). In support of this model, are reports of an inverse correlation between amygdala and fusiform activity in response to fearful faces (Phan et al., 2002;Damasio et al., 2000;Reiman, 1997) . In addition to activation of the V1- V5, the induction of affect with visual stimuli has also been reported to preferentially activate the amygdalae versus recall and auditory-generated affective states (Reiman et al., 1997;Teasdale et al., 1999).

For example, during the presentation of neutral and fearful faces in either low-pass filtered, high- pass filtered, or intact images, amygdala activation to fearful expression was greater for intact or low-pass than high-pass faces, even though activation in the fusiform cortex was actually greater for intact or high-pass faces irrespective of displayed affect (Vuilleumier et al., 2003). Intriguingly, the modulation of activation in the fusiform cortex by affective expression was only observed with intact or low-pass faces, not with high-pass faces, despite the heightened cortical sensitivity to high-pass visual cues. Based on these results it would appear that the modulation of fusiform responses to faces arises through feedback influences from the amygdala, where affective information had been extracted as a low-pass signal from the intact image. In support of affective processing bypassing the extrastriate visual cortex are reports of amygdalar responses by crude visual cues including masked stimuli containing only fearful eyes (Whalen et al., 2004), or by chimeric faces with fearful eyes and happy mouth (Morris et al., 2002).

Lack of activation of the amygdala in response to recall-driven affective states has led some to

79 speculate that the amygdala is preferentially responsible for the processing of externally-cued perceptual emotional stimuli (George et al., 1995). Due to their uncontrolled nature, recalled memories are prone to contain multiple emotions, thus reducing the ability to detect a specific affective state (Mayberg, 1997). Moreover, the recall of personal events in a PET or MRI scanner is phenomenologically different from spontaneous changes in affect encountered in everyday settings (Phan et al., 2002) (Mayberg et al., 1997).

One technique that neuroimaging investigations have employed to ensure subject attention to presented stimuli is the concurrent execution of cognitive tasks. Phan and colleagues compared neuroimaging investigation with and without cognitive demand. They found that activation of the ACC was three-fold higher in tasks with cognitive components (Phan et al., 2002). Whereas the medial PFC may respond to cognitive aspects that are potentially common to the induction and processing of an affective state (attention, appraisal, interpretation) explicit cognitive tasks requiring subject response (sex identification, recognition, encoding or rating of stimuli) additionally activate the ACC (Desmond and Glover, 2002).

Other investigators have evaluated the opposite interaction, the effect of affective distraction on cognitive processing. Wang and colleagues investigated the neural correlates of emotional oddball task in 19 MDD and 20 HC. The emotional oddball task requires the detection of infrequently presented circle targets while sad and neutral pictures are irrelevant novel distractors (Wang et al., 2008). In addition to slower responses in the ‘Target-after-Sad’ versus ‘Target- after-Neutral’ stimuli, MDD revealed greater activation compared to HC in the left pACC, left anterior insula, and bilaterally in the vlPFC (BA 10/47), and the right dlPFC (BA 45). Moreover, the investigators noted a correlation in the patients group between activation in the right dlPFC and reaction time, strongly suggesting a role of this region in coping with emotional distraction.

A similar approach was employed by Fales and colleagues who investigated whether the negativity bias in depression is reflected by dysfunction in emotional processing or impaired cognitive control over emotion (Fales et al., 2008). Functional MRI data were acquired from 27 MDD and 24 HC, during a matching task (same/different house) while subjects were exposed to emotional interference (neutral/fearful faces). Compared to the HC group, MDD displayed greater amygdalar activation to unattended fear stimuli, while simultaneously displaying decreased activation in the dlPFC leading the investigators to conclude suggest that patients did

80 not suppress emotional responses to fear-related distracters. Surprisingly, in the in the fear attend condition, the opposite pattern was observed with decreased activation by HC in the dlPFC and increased activation in the amygdala. By contrast, depressed subjects in this condition displayed increased dlPFC and deactivated amygdala indicative of successful affect regulation. The investigators concluded that MDD may be impaired in cognitive control over emotion, when the attention is not focused on the emotion.

A confluence of neuroimaging and lesion investigations suggests that the interaction of visual and affective processing may proceed at the subconscious level. The extant data support this first-pass evaluation in the processing of facial expression. Future research vistas should include the evaluation of this first-pass visual information scan with other emotionally salient stimuli (e.g. affective imagery). The neuroimaging modalities used to evaluate structural and functional neuroanatomy are described in the next section.

Induction of Affect with Visual Stimuli

Numerous neuroimaging investigations have used visual stimuli to investigate affective processing in HC, including words with an affective valence, affective facial expressions, affective imagery, affective videos, and gambling simulations. Several of these methodological approaches have been adapted for use in MDD populations. Below is a summary of comparative findings between HC and MDD, categorized according to the affective stimulus employed.

One of the first fMRI investigations to compare affective processing between HC and MDD involved the passive viewing of an emotionally laden film clip contrasted with the viewing of an emotionally neutral film (Beauregard et al., 1998). Common to both groups, transient sadness was associated with activation in the medial (BA 9) and inferior prefrontal cortex (BA 47), the middle temporal cortex (BA 37), the cerebellum and the caudate. Greater activation in the left medial prefrontal cortex (BA 9) and in the right aMCC (BA 24) differentiated the MDD group from the HC group.

In a follow up investigation, the investigators scanned MDD and HC while they attempted to voluntarily down-regulate sad feelings generated by sad film excerpts (Beauregard et al., 2006). Although the level of induced sadness was comparable between groups, the degree of difficulty

81 experienced while attempting to down-regulate sadness was significantly higher in the MDD group. Moreover, a positive correlation between depression severity and the degree of difficulty reported in the down-regulation was reported.

During the down-regulation phase, significantly greater activation by the MDD group was noted in the aMCC (BA 24), the right anterior temporal pole (BA 21), right amygdala, and right insula (BA 13). To address the possibility that persistent hyperactivity in these regions may be responsible for the increases observed in the self-regulation phase, the authors noted that the BOLD activity in the amygdala and insula was not significantly greater in depressed participants than in controls for the initial induction phase. Within the MDD group, the degree of difficulty reported during the down regulation condition, was positively correlated with activation in the right mPFC (BA 10), right anterior temporal pole (BA 21) and the left amygdala.

As major depressive disorder is frequently associated with sexual dysfunction independently of medication (Kennedy et al., 1999), the neural correlates of sexual dysfunction were investigated in a sample of seven depressed and nine non-depressed women with erotic video stimuli under fMRI scanning conditions (Yang et al., 2008). Increased activity was noted in the control group in the middle occipital gyrus, the middle temporal gyrus, the inferior frontal gyrus, thalamus, and amygdala. Activity in the same brain regions was noted in the depressed group, albeit with significantly reduced activity in the hypothalamus, the sACC, and the parahippocampal gyrus. These preliminary reports indicate that affective video stimuli represent a reliable method of invoking changes in positive and negative affect.

Gambling Paradigms Compared to negative affect induction, the reliable induction of positive affect with visual stimuli has proven to be more difficult. An alternative to studying reward processing in the brain is to examine the behavioural and neural responses to a reward decision-making task. Forbes and colleagues scanned adolescent HC and MDD subjects during the execution of such a task where choices about possible rewards involving varying magnitude and probability of reward were presented. Although their experimental design permitted the dissection of decision/anticipation and outcome phases of reward processing, they found that MDD subjects exhibited reduced activation than the HC group in reward-related brain areas (amygdala, OFC, vmPFC, caudate) during both phases of the task. Through comparison to an age-matched group

82 with anxiety disorders, they concluded that between-group differences were not mediated by the presence of anxiety (Forbes et al., 2006).

According to Edmund Rolls, the omission of predicted positive reinforcers leads to changes in affect including frustration, through anger to rage, or sadness through grief to depression. Conversely, the omission of predicted negative reinforcers is associated with affective changes ranging from relief through pleasure to elation (Rolls, 2000). The presence of predominant negative bias in MDD led Steele and colleagues to hypothesize that MDD subjects would be characterized by a different pattern of error signals than HC in a gambling task (Steele et al., 2004). Consistent with the hypothesis, the investigators reported that MDD subjects exhibited an increased error signal in the rostral ACC, the parahippocampal gyrus, and the cerebellum.

Preliminary results from gambling paradigms appear to indicate impaired reward processing pathways in MDD. Specifically decreased activation in response to the processing of a prospective reward and its anticipation has been found, along with increased signals of negativity when an incorrect choice is made.

To exclude the potentially confounding nature of antidepressant medication on positive affect processing, the neural correlates of monetary incentives were evaluated in 14 unmedicated MDD and 12 HC (Knutson et al., 2008). While there were no differences in activation of the nucleus accumbens (NAcc) during gain anticipation, MDD exhibited increasing activation in the aMCC with increasing gains, whereas HC displayed increasing aMCC during anticipation of increasing loss. Results from this study suggest that MDD have the potential to experience comparable transient increases in positive affect during a scanning session to levels in HC. Moreover, the authors concluded the differential conditions that elicited affective conflict - during anticipation of avoidable losses in HC versus during anticipation of attainable gains in MDD – may represent a neural marker of MDD (Knutson et al., 2008). It should be noted, that, dysregulation of reward processing has also been noted in Parkinson’s disease, where pathological gambling has been reported in patients treated with dopamine agonists (Crockford et al., 2008).

Affective Words The use of words with an affective component is one of the more popular neuroimaging techniques to compare changes in affect between HC and MDD populations. Investigators have

83 presented specific words to aid recall, provoke, or distract subjects. In one of the first fMRI investigation using this modality, Siegle and colleagues examined brain activation in response to alternating emotional processing (valence identification) and non-emotional processing (Sternberg memory) trials (Sternberg and Jarvik, 1976). A differential time course of amygdalar activity was noted between the two groups. Whereas in the HC group amygdalar responses to affective stimuli decayed within 10 seconds, the MDD group exhibited sustained amygdala responses that carried over to the subsequent non-emotional processing trial 25 seconds later (Siegle et al., 2002). Confirming neuropsychological evidence, MDD subjects confirmed rumination during the working memory trials.

Using an equivalent paradigm, the investigators used a larger sample size to demonstrate that MDD subjects whose affective reactivity profile included low activation of sACC (BA 25) and high in the amygdala would go on to manifest the most robust improvement with CBT (Siegle et al., 2006). When the task was modified to require HC and MDD subjects to state the personal relevance rating of words, or perform a different cognitive task (digit sorting), sustained amygdala reactivity was still observed in the MDD, versus the HC, group. Interestingly, decreased dlPFC (BA 9) activity on the digit-sorting task was also reported in the MDD group, and an examination of the time-course of amygdala-dlPFC functional connectivity revealed decreased coupling between these two regions compared to HC (Siegle et al., 2007).

Paradigms including the presentation of affective words have also been used to identify the correlates of positive affect. The affective processing of positive, negative, and neutral words was evaluated with fMRI in 10 medication-free MDD and 12 HC. In response to positively valenced stimuli, reduced activation in the ventral striatum was recorded in the MDD group correlating with decreased interest in the performance of general activities. Additional decreases in the dmPFC (BA9) region were associated with processing of self-related stimuli (Epstein et al., 2006). In a comparable experimental design, Canli and colleagues evaluated 15 HC and 15 MDD subjects performing a lexical decision tasks (deciding if letter strings were words or non- words) while viewing neutral, happy, sad, and threat-related words under fMRI conditions. The presentation of happy words was again associated with decreased activation in the MDD group in the left vlPFC (BA 47) and temporal cortex (BA 21), right insula, right amygdala, and the right cerebellum. In the sad words condition, increased activation in the MDD group was localized to the parietal lobule (BA 7), the superior temporal gyrus (BA 22) and the cerebellar

84 nodule (Canli et al., 2004).

Presentation of affective words has also been combined with different mood induction techniques. Ramel and colleagues used the combination of a Self-Referent Encoding Evaluation Task (pressing a button to indicate whether a presented word is self-descriptive, SRET) with a multimedia mood induction (re-experiencing an autobiographical sad personal event while listening to somber music) in HC and medication-free, remitted, MDD subjects. Post-hoc analyses revealed that following sad mood induction, bilateral amygdala response during encoding of valenced words correlated with the subsequent recall of negative self-referent words. The association was absent before the sad mood induction and in the HC group regardless of mood state (Ramel et al., 2007).

The combination of mood induction through emotionally laden words along with the presentation of facial expression of affect has also been successfully employed in comparing affective processing between HC and MDD groups by Keedwell and colleagues. Their analysis of 12 HC and 12 MDD subjects compared activation to happy and sad emotional stimuli presented as autobiographical prompts and mood congruent facial expressions. Within the vmPFC (BA 10/32) a significant group x affect interaction was noted; increased activity in the MDD group to negative stimuli and in the HC group to positive stimuli, with decreased activity in the correspondingly opposite valence for both groups (Keedwell et al., 2005a)

In an identical experimental design, the investigators attempted to identify the neural correlates of affective processing in a group of 12 MDD subjects with varying degrees of anhedonia. A statistically significant correlation was found between activation in the vmPFC (BA 10) to positive affective stimuli and the severity of anhedonia, but not the overall severity of depressive symptoms. Moreover, negative correlations between activation in the amygdala, ventral striatum and measures of anhedonia were also noted (Keedwell et al., 2005b).. As the presence of has anhedonia has been documented to be predictive of antidepressant response, (Klein, 1974) higher activation to positive stimuli in the vmPFC may represent activation comparable to what has been observed in HC (Keedwell et al., 2005a).

Recently, the use of affective words has been combined with the emotional Stroop task to evaluate the interference effects between emotional expression conflict and cognitive colour-

85 naming instructions in 17 HC and 17 MDD (Mitterschiffthaler et al., 2008). The presentation of sad words was associated with greater activation of the left pACC (BA 32), the right precuneus (BA 7) by MDD versus HC. Highlighting the importance of the pACC in affective processing, activation in this brain region was directly correlated with the latencies of negative words in the MDD group.

The use of affective words has also been used in the network analysis of affective processing. Yoshimura and colleagues recruited 13 HC and 13 MDD subjects to participate in self-referential judgments of positive and negative valenced personality trait words under fMRI conditions (Yoshimura et al., 2009). The investigators noted that not only did depressed patients show increased BOLD signal in the pACC during the self-referential processing of negative words, the activity of these regions d was correlated with the depressive symptom severity. Based on these results, the authors suggested that pACC activity mediated the correlation between the ventromedial PFC activity and depressive symptoms.

Affective words have also been used to evaluate the depressive-state independent component of affective processing in a group of remitted MDD subjects and HC(Hooley et al., 2009). Both groups were scanned while they heard praising, critical, and neutral comments from their own mothers. Compared to HC, remitted MDD subjects responded to criticism with greater activation in the amygdala and less activation in the dorsolateral PFC, and aMCC. Interestingly, there were no statistically significant differences between the groups during praise and neutral commentary.

Affective Faces Presentation of affective facial expressions has been one of the most reliable means of inducing affect-related changes in the amygdala in HC and MDD populations (Morris et al., 1998b;Hariri et al., 2002). Thomas and colleagues examined the specificity of the amygdala response to fearful and neutral facial expressions in six adults and twelve children under fMRI scanning conditions. The authors noted predominantly left amygdala activity during the presentation of fearful faces relative to fixation and a decrease in activation with repeated exposure. In adults increased left amygdala activity was reported for fearful faces relative to neutral faces, while in children greater amygdala activity was associated with neutral versus fearful faces (Thomas et al., 2001b). In a follow-up investigation, the investigators used an identical paradigm with five MDD children compared to five HC children, and five with anxiety disorders. Compared to

86 healthy children, children with anxiety disorder displayed an exaggerated amygdala response to fearful faces, while children with MDD displayed a blunted amygdala response to fearful faces (Thomas et al., 2001a).

An early investigation in adult MDD subjects compared activation of the amygdala in response to the presentation of masked emotional faces (a fearful or happy face presented for 40 msec followed by a 160-msec presentation of a neutral face). Compared to age-matched HC, adult MDD produced exaggerated left amygdala activation to all faces with the greatest effect observed for fearful faces, while activation in the right amygdala did not differ from HC (Sheline et al., 2001). Following antidepressant treatment, patients displayed decreased bilateral amygdala activation to all faces, while HC did not display any inter-scan differences. In a related prospective study, it was reported that greater amygdala activation to emotional facial expressions among depressed subjects predicts symptom reduction eight months later, even when controlling for initial depression severity and medication status (Canli et al., 2005).

Functional specialization of sACC in affective processing has also been influenced by the presentation of facial expressions. Under fMRI scanning conditions happy and sad facial expressions of affect were presented to 18 MDD and 18 HC (Gotlib et al., 2005). In response to negative facial expressions increased activation in the MDD group was located in the sACC (BA 25), while increased activation to positive facial expression were reported in a more rostral component of the sACC.

Decreases in the BOLD signal during the presentation of sad and angry facial expressions have also been reported in MDD. When pictures of sad and angry faces were presented to 21 MDD and 15 HC, deactivations were noted in the medial (BA 11) and lateral orbitofrontal cortex (BA 10/11). Presentations of sad faces additionally produced deactivations in the medial OFC and bilaterally in the caudate body and the hippocampus (Lee et al., 2008).

The genetic underpinning of amygdala hyper-reactivity to negative stimuli in MDD was investigated in monozygotic twin pairs discordant for the risk of anxiety and depression (n=10 pairs) and in monozygotic twin pairs concordant for high (n=7 pairs) and low (n=15 pairs) risk for anxiety and depression (Wolfensberger et al., 2008). Increased amygdalar activation was reported during the presentation of negatively valenced stimuli (angry/anxious faces) in high-risk

87 twins compared to their low-risk co-twins. These results argue for an environmental factor in mediating increased amygdalar reactivity towards negative stimuli.

The environmental factor may be an automatic judgment bias in MDD. Dannlowski and colleagues recorded amygdala activity under fMRI scanning conditions in response to masked displays of angry, sad and happy facial expressions (Dannlowski et al., 2007). Following the scanning session, patients where exposed to the identical stimuli outside the scanner while performing an affective priming task characterizing the degree of automatic emotion processing (evaluation of each stimuli on a 7-point scale). The investigators noted a significant association between (right) amygdala reactivity and automatic negative judgmental bias.

A natural evolution of the facial affect expression paradigm includes the presentation of varying intensities of fear, happiness, and sadness (Lawrence et al., 2004). In a comparison of depressive episodes between BD, MDD and HC, investigators reported that compared to both depressed groups, HC displayed increased subcortical activation to facial expressions of intense happiness and mild fear, along with increased activation of dlPFC to intense sad expressions. Conversely, MDD subjects demonstrated diminished activation to all facial expressions except mild sadness, findings concordant with a limited emotional range and a negative bias (Phillips et al., 2003b). In a follow-up investigation, a direct comparison of HC and MDD subjects undergoing an identical paradigm demonstrated a linear correlation between response in bilateral fusiform gyri (BA 19) and right putamen and expressions of increasing happiness in the HC group. In contrast, a positive correlation was found between activation in left putamen, left parahippocampal gyrus/amygdala, and right fusiform (BA 19) to expressions of increasing sadness (Surguladze et al., 2005). A more recent report evaluating 17 MDD subjects subjected to the varying intensities of sadness paradigm under fMRI conditions has identified its clinically useful prognostic value. Increased activation in the ventral ACC was associated with faster improvement in response to fluoxetine antidepressant therapy (Chen et al., 2007).

Use of varying intensities of facial expression representing sadness and happiness, also allows for the measurement of the dynamic range of brain response, in addition to the overall capacity (mean difference between affective and baseline trials). Using a repeat scan trial design, Fu and colleagues investigated the effects of antidepressant treatment and symptomatic response on happy and sad facial affect processing in 19 MDD subjects and 19 HC (Fu et al., 2004;Fu et al.,

88 2007). Exposure to antidepressant treatment was associated with reduced capacity for activation to negative affect in the left amygdala, ventral striatum, and frontoparietal cortex (BA 6, 40) and an increased of dynamic range in the prefrontal cortex (BA 9, 44). Specifically, symptomatic improvement was associated with decreased dynamic range in the pACC (BA 24, 32), ventral striatum, and cerebellum (Fu et al., 2004).

In an analysis of changes in positive facial affect associated with antidepressant treatment, decreased dynamic range in the temporal-occipital (BA 19) and parietal (BA 7) cortex and subcortically in the hippocampus, thalamus, putamen, and in the brainstem and cerebellum was reported in the depressed subjects, relative to the comparison subjects. Moreover, reduced extrastriate cortical activation (BA 18, 19) at baseline was increased following antidepressant treatment, with symptomatic improvement being correlated with greater capacity in the hippocampal and extrastriate regions (Fu et al., 2007).

To ensure participant compliance during a neuroimaging investigation with affective facial expressions, most investigators have instructed subjects to indicate the sex of each presented facial stimuli. Alternatively, a more demanding cognitive task can be employed, such as the facial encoding. In this task, subjects to view a series of actors depicting a specific face-emotion during the imaging, and are instructed to recognize the same actors when they depicting a neutral emotion outside the scanner (Pine et al., 2004). Roberson and colleagues used the self-encoding task to fMRI to compare between-group differences among anxious adolescents, non-anxious MDD, and a group of HC during successful vs. unsuccessful face encoding (Roberson-Nay et al., 2006). Compared to the HC group, the MDD group exhibited poorer recall for faces, moreover, analysis comparing successful vs. unsuccessful face encoding, demonstrated increased activation of the left amygdala compared to HC and anxious adolescents.

Affective Pictures

Use of evocative imagery, particularly from the International Affective Picture System (IAPS)(Lang et al., 1990;Lang et al., 1993;Bradley et al., 1996;Sutton et al., 1997) has become one of the most often employed paradigms in the neuroimaging of affective processes. The IAPS emerged from a rich database of neuropsychological (Britton et al., 2006;Northoff et al., 2000;Northoff et al., 2000) and neuroimaging investigation evidence conducted in a socio-

89 demographically diverse psychiatrically unaffected groups and neuroimaging sensitivity comparable to self-recall and facial affective expression paradigms (Mayberg et al., 1999).

One of the first fMRI analyses of the neural correlates of the IAPS was conducted using a combination of fMRI and magnetoencephalography measures of positive and negative affective changes in a group of 5 male and 5 female HC (Northoff et al., 2000). Both negative and positive affective conditions differed from the neutral condition by exhibiting strong activation in the OFC (BA 11, 12) and lateral PFC (BA 9, 45-47). Negative affective processing was additionally characterized by robust activation of the medial OFC (BA 11), while positive affective processing exhibited a stronger activation of the in lateral PFC (BA 9, 45-47). Based on concomitantly acquired MEG data, the investigators concluded that negative affective processing may be characterized by robust, early, medial OFC activation, in contrast to later and weaker activation of the lateral PFC in response to positive affect. The relatively weaker activation in response to positive changes in affect (vs. negative affect) with IAPS may affect sensitivity to detect statistically significant changes in positive affect activation with this visual induction paradigm (Northoff et al., 2004) .

15 The IAPS paradigm has also been validated against an earlier O- H2O PET self-recall mood induction paradigm that demonstrated reciprocal modulation and attenuation of dlPFC in response to emotional-cognitive interaction (Davidson et al., 1990). Using fMRI and the IAPS, Northoff instructed a group of HC to perform affective and non-affective judgment conditions for each visual stimuli picture presented (Grimm et al., 2006). While the presentation of affective IAPS were associated with increases vmPFC and dmPFC, a comparison between the affective and non-affective judgments resulted in specific increases in vmPFC and dmPFC in response to affective judgments, compared to increases activation in vlPFC and dlPFC in response to non- affective judgments.

In a further dissection of affective and cognitive processes underlying the presentation of the IAPS stimuli, Grimm and colleagues took additional steps to isolate the confounds of judgment and preceding attention (Davidson and Fox, 1982;Davidson, 1995). A group of 29 HC were presented with either affective stimuli from the IAPS or a fixation cross; for half of the pictures, HC were asked whether positive or negative affect was being presented. Rating of the valence of affective of stimuli correlated positively with activation vmPFC (BA 10, 11) and negatively with

90 dlPFC (BA 9), while the intensity of the activation correlated with vlPFC (BA 45-47) and dmPFC (BA 9) activation, and recognition of the valence correlation with the rostral ACC (BA 24, 25, 32).

Davidson and colleagues have proposed that changes in affect can be broadly categorized as positive (approach-related) and negative (withdrawal related) (Lee et al., 2004). Moreover, they have also proposed that differential engagement of the cerebral hemispheres may reflect this dichotomy; with negative affect preferentially activating the right hemisphere, while positive affect being associated with left hemispheric activation. This hypothesis was evaluated in 10 HC under fMRI scanning conditions using negative, positive, and neutral affective stimuli from the IAPS paradigm (Bermpohl et al., 2006a). While all affective stimuli induced activation of vmPFC (BA 10), ACC (BA 24), dlPFC (BA 9) amygdala, anterior temporal cortex, and the cerebellum, negative affective pictures induced greater activation in the right hemisphere, with positive affective pictures inducing preferential activation of the left hemisphere.

The distinction between the neural correlates of emotional expectancy and emotional perception have also been evaluated with the IAPS using an experimental design segmented into i) emotional expectancy, ii) neutral expectancy, iii) emotional picture perception, and iv) neutral picture perception (Bermpohl et al., 2006b). Activation of pACC (BA 24, 32), CMAr (BA 24), and occipitoparietal sulcus (BA 7, 19, 31), and the temporal cortex (BA 22) was associated with expectation of affective stimuli. The actual perception of affective stimuli, on the other hand, recruited a network comprising the amygdala, insula, vmPFC (BA 10), vlPFC (BA 44, 45), OPT cortex (BA 17, 18, 19, 20).

An extension of the above paradigm investigated the effect of expectant versus unexpectant affective IAPS stimuli and the role of affective priming. The investigators used the same four conditions (emotional expectancy, neutral expectancy, emotional neutral picture perception), however, in this investigation some affective and neutral pictures were presented without a preceding expectancy cue (Kalin et al., 1997). Within the mPFC (BA 9, 10), amygdala, and dorsal midbrain, investigators noted that expectancy augmented the neural response to affective but not to neutral stimuli. Moreover, extraction of the time course for activated clusters confirmed that this augmented activation was not preceded by baseline increases during affective expectancy phase.

Neuroimaging investigations employing IAPS in a MDD population have been used to compare affective processing between HC and MDD group in different symptomatic states, following antidepressant treatment, and to delineate functional connectivity mediating affective processing. In a proof of concept investigation, Kalin and colleagues employed fMRI neuroimaging in an evaluation of affective processing in MDD. In this preliminary investigation, positive and negative stimuli from the IAPS were presented to two HC and two MDD subjects at baseline and two week later, with MDD subjects commencing antidepressant treatment with venlafaxine after the first scan. Although underpowered, reduced activation induced by the negative pictures was reported during the second scan in both groups. Differential time-dependent activation to positive stimuli was also reported, while HC groups reported reduced activation to positive affective stimuli, the opposite trend was noted in the MDD group in the visual association areas (BA 18,19) (Kalin et al., 1997).

Using a larger sample size (12 MDD and 5 HC) and a longer inter-scan duration (8 weeks), Davidson and colleagues extended the results of Kalin and colleagues (Davidson et al., 2003). In response to negative affective stimuli, both HC and MDD groups displayed bilateral activation in the occipital cortex (BA 37), dlPFC, and amygdala in response to the negative versus neutral stimuli however, increased activation was noted in MDD subjects in the occipital cortex and decreased activation in the left dlPFC. Significant reductions in depressive symptoms were noted following eight weeks of venlafaxine treatment along with group-by-time interactions (increases in MDD and decreases in HC) in response to the negative stimuli in the left insula cortex and left dorsal ACC (BA 24). Moreover, MDD subjects with greater relative activation in the ACC at baseline were more likely to exhibit a robust treatment response.

To assess positive affective processing, the investigators performed another neutral-positive affect IAPS stimuli presentation 22 weeks following the baseline scan (Schaefer et al., 2006). In this subanalysis of the changes induced by positive affective processing in the MDD subjects, the investigators categorized the positive affective stimuli into scenes representing social interaction, human faces, and erotic sexual images. For all categories of positive stimuli, the observed reduction in activation at baseline in the MDD group, increased towards levels found in HC following antidepressant treatment. Specific group x time interactions to positive social interaction stimuli were reported in the left dmPFC (BA 6), dlPFC (BA 8, 46), bilateral ventral

92 striatum, OPT cortex (BA 22, 37) and left occipitoparietal cortex (BA 19, 23). Changes in response to positive facial affect also included the dmPFC (BA 8), vlPFC (BA 45), premotor cortex (BA 6), the insula, medial dorsal thalamus, and the fusiform gyrus (BA 18). The most widespread group x time changes were found in response to erotic positive stimuli with changes in the vmPFC (BA 10), vlPFC (45,46), aMCC (BA 24), premotor area (BA 6), insula, hypothalamus, and the OPT cortex (BA 18,19,38). Based on the substantial restoration of positive affective processing following antidepressant therapy, it was suggested that impaired positive affective processing represents a state-dependent phenomenon. Results from positive affect induction using the IAPS in HC, report robust activation of the left nucleus accumbens ((Meseguer et al., 2007).

An inherent weakness of the IAPS neuroimaging investigations by Davidson’s group is an absence of measures of the amount of affect that was induced inside the scanner. Lee and colleagues overcame this limitation by recording the amount of affective change retrospectively following the presentation of IAPS to 15 MDD and 15 HC (Lee et al., 2007). The investigators reported decreased activation in the left anterior insula and the right sACC during the induction of positive affect in MDD versus HC. Negative affective processing was also associated with decreased activation by MDD in the right hippocampus and the right insula. It is worth noting that the amount affect that was induced by the positive, negative, and neutral stimuli was comparable between the two groups. Increased activation in the amygdala, aMCC and insula compared to controls, has also been reported during negative affect processing in subjects with dysthymia (Ravindran et al., 2009b).

The effects of a non-serotonergic antidepressant, bupropion, on negative affective processing have also been evaluated. Repeat presentations of the Emotional Oddball Task were performed at baseline and eight weeks later in 10 MDD subjects medicated with bupropion (Robertson et al., 2007). The Emotional Oddball involves the execution of an attentional executive function task in the presence of emotional distractors. Along with treatment associated improvements in depression severity, reduced activation during emotional distracters was noted in right OFC (BA 11), left dmPFC (BA 9), right vmPFC (BA 10), right aMCC (BA 24), right inferior frontal cortex (BA 46), right amygdala, right caudate, right fusiform gyrus, and left dPCC. Although there was no HC group, there was a correlation between symptomatic improvement and change in amygdala activation.

The rewarding value of images of chocolate compared to an aversive condition consisting of the sight of moldy strawberries and a corresponding unpleasant taste were compared under fMRI conditions between 14 HC and 14 MDD subjects who met criteria for clinical remisssion (McCabe et al., 2009). In site of comparable ratings of induced affect between the groups, the MDD group was characterized by decreased signal in the ventral striatum and increases in the caudate nucleus to the aversive stimulus.

Functional connectivity has also been evaluated with the IAPS paradigm. Using negative and neutral affective pictures from the IAPS, Anand and colleagues investigated differences in corticolimbic activity and connectivity between 15 MDD and 15 HC using low frequency BOLD related fluctuations (LFBF) (Anand et al., 2005a). In response to the negative affective stimuli, increased activation of cortical and limbic regions was noted in the MDD group. Additionally, corticolimbic LFBF correlations between ACC, amygdala, pallidostriatum, and medial thalamus were decreased during the presentation of neutral and affective visual stimuli. In contrast,

In a subset of 12 MDD and 11 HC, a repeat scan was acquired six weeks later, a period during which MDD subjects were treated with the antidepressant sertraline (Anand et al., 2005b). Significant time-dependent changes were reported in the MDD group. During exposure to neutral and positive pictures and during rest, LFBF correlation between the ACC and limbic regions was significantly increased in MDD subjects after treatment. On exposure to negative pictures, however, corticolimbic LFBF correlations remained decreased. Use of the IAPS has also been applied under fMRI conditions to demonstrate compelling evidence of functional connectivity between the amygdalae in MDD subjects, although only weak connectivity was apparent in HC (Irwin et al., 2004).

In a further analysis of the default mode network, negative BOLD responses (NBRs) in cortical midline structures were compared between 29 HC and 20 MDD (Grimm et al., 2008b). Statistically significant reductions in NBRs were identified in several regions of the default- mode network including the pACC, the vmPFC, and the dPCC. Moreover, decreased NBRs in MDD subjects were directly correlated with depression severity and feelings of hopelessness.

94 In a subsequent reanalysis of the same data set, the investigators compared the BOLD response between the left and right dlPFC during attended and unattended judgment of emotions (Grimm et al., 2008a). Hypoactivity in the left dlPFC was noted in MDD during both unattended and attended emotional judgment, while hyperactivity in the right DLPFC was recorded during attended emotional judgment. In the HC group left dlPFC activity was parametrically modulated by negative emotional valence during emotional judgment, whereas in the MDD group it was inversely modulated by positive emotional valence. Hyperactivity in the right DLPFC was also correlated with depression severity. The authors concluded that left dlPFC hypoactivity may be associated with negative emotional judgment, while right DLPFC hyperactivity may be linked to attentional modulation.

In summary, the IAPS represents a well validated data set that reliably elicits changes in positive and negative affect. Complementing its use in MDD subjects is an accumulating database of normative fMRI data obtained using HC that has dissected affective processing into a number of simpler constituent processes. Moreover, the documented sensitivity of the IAPS to a variety of antidepressant treatments is particularly pertinent to the resolution of affective processing abnormalities into state and trait components in the MDD population.

95 AIMS AND HYPOTHESIS

PRIMARY AIM A: EVALUATE DIFFERENCES IN BRAIN ACTIVATION EVOKED BY AFFECTIVE PROCESSING BETWEEN HC AND MDD, AND BETWEEN RS AND NR

HYPOTHESIS: It is hypothesized that positive affect processing in study participant will be characterized by higher BOLD signal in the ventral striatum of HC, compared to the MDD group. It is further hypothesized that the production of positive affect will be altered by increased psychomotor slowing in the NR group with decreased self-referential processing during the induction of positive affect leading to decreased activation in ventromedial (BA 10) and dorsomedial PFC (BA 9), the precuneus (BA 7) and dPCC (BA 31). Moreover, psychomotor slowing may be responsible for decreased visual analysis of the visual stimuli with decreased activation in the extrastriate visual cortex (BA 18/19) and the neighbouring OPT cortex (BA 20/ 37) and delayed BOLD signal in the ventral striatum and anterior insula.

It is hypothesized that increased BOLD signal will be evoked during negative affect induction in the MDD groups in the amygdala and sACC (BA 25), and in the dorsomedial and dorsolateral PFC (BA 9). Within the MDD group, non-response will associated with increased production of the negative affective state and increased BOLD signal in the amygdala, the vlPFC (BA 10), and the pACC (BA 24/32), and increased need for regulation of the affective state with increased BOLD signal in the dlPFC (BA 9).

Neuropsychological testing has revealed further differences between SSRI responders and nonresponders compared to responders (Gorlyn et al., 2008;Dunkin et al., 2000;Taylor et al., 2006a). Retrospective neuroimaging analyses of MDD subjects before undergoing pharmacotherapy have also revealed differences in brain metabolism between SSRI treatment responders and non-responders (Brody et al., 2001a;Little et al., 1996;Saxena et al., 2003;Mayberg et al., 1997;Pizzagalli et al., 2004;Evans et al., 2006;Mayberg, 2003a)

96 PRIMARY AIM B: EVALUATE DIFFERENCES IN INTER-SCAN HABITUATION ON BRAIN ACTIVATION EVOKED BY REPEAT AFFECTIVE PROCESSING BETWEEN HC, RS, AND NR

HYPOTHESIS: It is hypothesized that differences in BOLD signal evoked by positive affective processing between HC and MDD are depressive state dependent. Therefore, repeat induction of positive affect in RS is hypothesized to lead to increased activation in brain regions responsible for the identification of affective value and the production of the affective state, meaning increased BOLD signal in the anterior insula, ventromedial PFC (BA 10), and ventral striatum. Conversely, it is hypothesized that there will be no differences in BOLD signal during repeat inductions of negative affect in NR is not hypothesized to be associated with only increased activation in the extrastriate visual cortex (BA 18/19) and the ventral OPT cortex (BA 20/37), reflecting inter-scan habituation common to all three groups. It is also hypothesized that repeat presentations of positive visual stimuli to healthy controls over the course of the study will be associated with increased visual evaluation of the presented stimuli with increased activation in the OPT (BA 19/37).

Based on previous neuroimaging investigation, and preclinical animal data, it is hypothesized that the repeat induction of negative affect in RS will be associated with decreased identification and production of the affective state, leading to decreased BOLD signal in the amygdala, the anterior insula, and ventromedial (BA 10/11) and ventrolateral (BA 10/45) PFC. Meanwhile, it is hypothesized that inter-scan changes in the NR group will be limited to brain regions responsible for the identification of negative affective stimuli with decreased BOLD signal limited to the anterior insula and amygdala. It is also hypothesized that repeat presentations of positive visual stimuli to healthy controls over the course of the study will be associated with the same changes as listed for the RS group with additional decreases in BOLD signal in brain regions responsible for regulation of negative affect with decreased BOLD signal in the aMCC (BA 24) and dorsolateral PFC (BA 9)..

In the RS group, the effects of habituation will be superimposed on the effect of dispositional affect normalization (increases in general positive affect, and decreases in negative affect), depressive symptom severity, and the administration of psychotropic medication. Time- dependent changes observed in the NR group represent the non-therapeutic effects of medication, sub-therapeutic clinical improvements, in combination with inter-scan habituation.

97 SECONDARY AIM A: EVALUATE THE ASSOCIATION BETWEEN DEPRESSIVE SYMPTOM REDUCTION AND BRAIN ACTIVATION EVOKED BY REPEAT AFFECTIVE PROCESSING IN MDD

HYPOTHESIS: It is hypothesized that reduction of depressive symptoms will be associated with elimination of statistically significant differences in BOLD signal between remitted depressed subjects (RM) and HC. Preliminary evidence indicates that impairments in positive affective processing are mostly a state dependent feature of MDD (Schaefer et al., 2006;Fu et al., 2007).Therefore decreases in depressive symptom severity are hypothesized to be positively correlated to increased BOLD signal evoked by positive affective processing in the ventral striatum, the anterior insula, and hippocampus.

Decreases in depressive symptom severity will be associated with decreased activation in the ventromedial PFC (BA10) and in the amygdala during the presentation of negative affective visual stimuli. Moreover, it is hypothesized that differences in BOLD signal evoked by decreases in negative affect processing will persist in a state of clinical remission. Compared to HC, hyperactivation by RM is hypothesized to persist in brain regions involved in the regulation of negative affect; the dorsolateral PFC (BA 9), and in autobiographical memory, medial temporal lobe (BA 34/35).

Differences in rCBF have been previously reported between remitted MDD subjects and healthy control group during negative affective processing (Liotti et al., 2002). An accumulating neuropsychological literature base indicates that remitted depressed subjects continue to manifest deficits in attentional and executive functions compared to healthy controls (Paelecke- Habermann et al., 2005;Kessing, 1998).Although antidepressant treatment is effective in decreasing the activation of brain regions involved in the production of the affective state, it may not be effective in normalizing activation in brain regions involved in the regulation of the affective state (Segal et al., 2006).

98 SECONDARY AIM B: EVALUATE THE ASSOCIATION BETWEEN DISPOSITIONAL AND INDUCED AFFECT AND BRAIN ACTIVATION EVOKED BY AFFECTIVE PROCESSING

HYPOTHESIS: It is hypothesized that increases in positive dispositional affect will be associated with increased alertness and attention. During positive affect processing higher PANAS-P scores will be associated with increased BOLD signal in the OPT (BA 19/37), and the hippocampus. During negative affect processing, higher PANAS-P scores will be associated with increased BOLD signal in the aMCC (BA 24) and dlPFC (BA 24).Clinical studies indicate that treatment with fluoxetine increases positive affect and decreases negative affect concurrently with improvements in mood (Aikens et al., 2008;Morilak and Frazer, 2004) and also independently of mood (Cook et al., 2004).

As neuropsychological tests indicate that increased negative affect in MDD is associated with increased emotional reactivity (Myin-Germeys et al., 2003;Lethbridge and Allen, 2008;Segal et al., 2006), it is hypothesized that decreases in negative affect will be associated with decreased activation of brain regions involved in the production of the affective state, particularly the amygdala and the anterior insula during negative affect processing. Moreover, it is hypothesized that the presence of negative affect will limit activation during positive affective processing in brain regions involved in the visual analysis of the presented stimuli, the OPT (BA 19/37), and brain regions involved in autobiographical memory recall, hippocampus/parahippocampal gyrus.

It is hypothesized that the ratings of induced affect will be positively correlated with BOLD signal in brain regions responsible for the production of affective states and the encoding of memory. Ratings of induced affect during positive affective processing will be positively correlated with BOLD activity in the ventrolateral PFC (BA 10/47) and the medial temporal lobe. Conversely, raring of induced during negative affective processing will be positively correlated with BOLD signal in the amygdala and the sACC (BA 25).

99 METHODS

Subjects

Sample Size Analysis

The proposed investigation was designed with 80% power to detect statistically significant differences at the p=0.05 level. Desmond and colleagues analyzed power curves to conclude that a block-design fMRI study, employing subsequent random effects analyses of signal changes approximately 0.5%, requires a minimum of 12 subjects to ensure 80% power at alpha = 0.05 at the single voxel level (Desmond and Glover, 2002). The initial hypothesis was centred on difference in brain activation between MDD and HC, with planned recruitment of 15 subjects for each group.

Subject Recruitment and Compensation

Subjects with MDD in the depressed phase (n=21) were recruited from the Mood Disorders Psychopharmacology Unit (MDPU) at the University Health Network through Toronto Western Hospital and Toronto General Hospital units. Age- and sex-matched healthy control volunteers (HC, n=18) were also recruited from the community through local media advertisements. All subjects provided written informed consent and were compensated for the time required to complete fMRI scans for a total of $200.00 (and additionally for transportation costs). The investigation was approved by the University Health Network Research Ethics Board and was compliant with the Declaration of Helsinki governing medical research (World Medical Association, 1964).

Inclusion Criteria and Exclusion Criteria

The study participants included both HC and MDD subjects and satisfied the following inclusion criteria: 18 to 55 years old, self-reported right-handedness, judged by the investigator to be in generally good health, capable of effective communication with the investigator to provide informed consent, and the capacity to undergo an MRI scan as judged by the Toronto Western Hospital department of radiology

100

Additional inclusion criteria for the MDD group included; meeting DSM-IV criteria for a diagnosis of Major Depressive Disorder, currently meeting criteria for a major depressive episode, and scoring at least seventeen on the Hamilton Depression Rating Scale 17-Item (HDRS-17).

Exclusion criteria for MDD and HC were; DSM-IV-TR criteria for substance abuse or dependence (except nicotine or caffeine) within the past 3 months; comorbid medical or psychiatric illness including schizophrenia, dementia or other major psychiatric disorders as defined in the DSM-IV; history of neurological trauma resulting in loss of consciousness; uncorrected hypothyroidism or hyperthyroidism, including elevated thyroid stimulating hormone (TSH); pregnant or nursing status.

Additional exclusion criteria for the MDD group included; prior failure to respond to fluoxetine and olanzapine in combination at adequate dose and duration; failure to respond to two adequate antidepressant trials in the current episode; evidence of serious risk of suicide based on clinician assessment and/or HDRS-17 suicide item > 3; course of ECT (electroconvulsive therapy) in the preceding 6 months; and hyperglycemia or diabetes mellitus as defined by a fasting blood glucose value of > 125 mg/dl.

Additional exclusion criteria for the HC group included a family history of mood disorders among first-degree relatives based on diagnosis or treatment; or a previous family history of antidepressant, anxiolytic, antipsychotic, or hypnotic use as ascertained by the Mini International Neuropsychiatric Institute (MINI) structured interview.

Study Outline

Recruitment

Subjects were recruited through local media announcements and physician referral. During a consultation visit, potential MDD subjects were presented with a study brochure outlining the possibility of participating in a pharmacotherapy-neuroimaging investigation as one of several different treatment avenues. If the subjects expressed interest in the aforementioned treatment

101 modality, they were introduced to clinically trained research-staff who presented an informed consent form and clarified any immediate patient queries. Subjects were given the informed consent for further consideration in a non-clinical environment. Similarly, HC were presented with an informed consent form after responding to advertisements in the community and had a subsequent appointment scheduled.

Screening Visit

During the screening visit, the signed informed consent was collected from all subjects. Subjects with a working MDD diagnosis were seen by an MDPU psychiatrist with a specialty in mood disorders who confirmed the diagnosis through administration of the Mini International Neuropsychiatric Institute (MINI) structured interview, quantified the severity of the mood symptoms with the Hamilton Depression Rating Scale 17-item (HDRS-17), and assessed global illness severity with the CGI-S. The clinician also administered a structured general physical exam (including circumference of the waist and hips), reviewed the medical, psychiatric and treatment history, and determined the eligibility of the subject for the study through completion of the inclusion/exclusion criteria form. All subjects receiving treatment at the time of enrollment had the treatment gradually discontinued. The initiation of investigational treatment did not commence until clearance of the index regimen (i.e. after five half-lives)

The research staff acquired of patient vital (diastolic, systolic blood pressure and heart rate) and anthropometric (weight and height) measurements. Subjects were then directed to the hospital laboratories for urinalysis and blood work. Laboratory indices of fasting glucose, HDL, LDL, total cholesterol, triglyceride, creatinine, urea, potassium, sodium, thyroid, complete blood count, recreational drugs, alanine aminotransferase and aspartate aminotransferase levels were monitored by the clinician, who used clinical judgment to determine when it is in the subject’s best interest to be discontinued from the investigation. Before leaving, all subjects were taken to the MR scanning facilities to acquaint them with the scanning location and the methodology involved.

Study Visits

At the baseline visit all subjects completed the self-rated BDI and PANAS scales. Additionally,

102 MDD subjects were seen by a psychiatrist who repeated the HDRS-17 and reassessed the Clinical Global Impression of Severity (CGI-S). The clinician also completed baseline measures of Sexual Dysfunction with the Sex-FX scale and medication side effects with the AMDP-5 scale. Combination pharmacotherapy with olanzapine (5mg) and fluoxetine (20mg) was initiated at this visit. The presence of any adverse events and concomitant medications incurred since the consultation visit was recorded by a graduate student (JK) along with measured baseline abdominal adiposity with waist and hip circumference measurements. Subjects also had their weight and baseline laboratory values of fasting glucose, HDL, LDL, total cholesterol and triglycerides measured. Before leaving, subjects were accompanied by JK to the medical imaging facilities where they underwent a positive/negative affect induction under fMRI scanning conditions (details of affect induction and fMRI data acquisition outlined in subsequent sections). Immediately following the fMRI scan, subjects rated the 22 blocks of images on the Affective Picture Rating Scale (APRS), a modification of scale previously used by Kumari and colleagues (see Appendix 1). The APRS-P subscale measured the induced affect during the presentations of the positive affect blocks, while the APRS-N subscale measured the induced affect during the negative affect blocks. Meanwhile, the APRS-0P/APRS-0N measure the induced affect during the neutral blocks presented between the positive and negative affect blocks, respectively.

Subsequent study visits followed a similar pattern to the baseline visits, and were characterized by an appointment with a psychiatrist, completion of self-rated questionnaires, documentation of vital signs, and completion of the neuroimaging session (Figure 6).

103 Figure 5 – Experimental Timeline – Timing of Research and Clinical Data Collection

Visit Label Screening Baseline Follow- Follow- Follow- Follow- Follow- Endpoint Time (wks) -3 to -1 0 1 2 3 4 5 6

Clinical Procedures Psychiatric Consultation X Informed Consent O Patient Psychiatric History X Vital Signs X X X X X Laboratory Blood Measures X X Psychiatric Appointment X X X X X X X X

Treatment Discontinue Previous Medication X Olanzapine + Fluoxetine X X X Dose Adjustment X Adverse Event Monitoring X X X X X X X

Psychometric Instrumentation HDRS-17, CGI-S X X X CGI-I X X X SEX-FX X X PANAS, BDI O O O APRS O O O

Neuroimaging Session MRI Data Acquisition O O O

LEGEND: X: Completed only by MDD Subjects O: Completed by MDD and HC Subjects

104 Treatment

Subjects with MDD began open-label treatment with fluoxetine (20 mg/day) and olanzapine (5 mg/day) after the completion of the first neuroimaging session. In the absence of significant side effects, or evidence of clinical improvement, the doses were increased according to a predefined pharmacotherapy schedule. Fluoxetine was administered at 20 mg/day for the first two weeks, thereafter increasing to 40 mg/day for the next two weeks and finally to 60 mg/day for weeks 5 and 6. Olanzapine was started at 5 mg/day for the first two weeks, increasing to 10 mg/day for weeks 3 and 4, and 12.5 mg/day for the last two weeks. Clinicians were permitted to vary the dose of either medication independently based on an assessment of therapeutic benefit and tolerability profile. Benzodiazepines and non-benzodiazepine sedatives (i.e. zopiclone 3.75 –7.50 mg) were offered as hypnotics. Patient compliance with the medication was monitored by direct patient interviews, and pill counting at each clinic appointment, when MDD subjects returned to refill their prescriptions.

Olanzapine was selected because of its documented antidepressant effects in both bipolar disorder (BD) and MDD populations (Shelton et al., 2001;Matthews et al., 2002;Corya et al., 2003;Tohen et al., 2003b). Fluoxetine, is effective as monotherapy in MDD and in combination with olanzapine in the BD and MDD population (Bremner, 1984;Amsterdam et al., 2004;Amsterdam et al., 1998;Cohn et al., 1989;Cohn and Wilcox, 1985;Chouinard, 1985;Tohen et al., 2000;Shelton et al., 2001;Matthews et al., 2002;Corya et al., 2003;Tohen et al., 2003a;Tohen et al., 2004). Other atypical antipsychotics have been approved as monotherapy and in combination with antidepressants for the treatment of bipolar mania and depression respectively and treatment-resistant depression (Thase and Denko, 2008;Philip et al., 2008;DeBattista and Hawkins, 2009). The dosing regimen that was employed in the study were the result of 1) doses of Symbyax® (commercial combination of olanzapine and fluoxetine, available in the United States for the management of treatment resistant MDD), 2) registration trials for Symbyax®, and 3) the clinical experience of the investigators to ensure maximal therapeutic efficacy and ideal tolerability.

Affective Processing Paradigm

The experimental study of affect necessitates a set of quantitative stimulus tools to reliably evoke

105 affective states (Bradley et al., 1996;Sutton et al., 1997). The International Affective Picture System (IAPS) is a series of 114 35mm color slides (digital electronic picture files were used in this thesis) specifically selected to elicit a range of positive, neutral and negative affective states (CSEA-NIMH, 1999;Lang et al., 1990).

This collection of visual stimuli is one of most frequently used for the study of affective processes, and there is strong evidence that its utilization reliably elicits affective responses (Lang et al., 1993) in both random (Bradley et al., 1996;Sutton et al., 1997) and blocked (Greenwald et al., 1989) presentations. Moreover, electromyography evidence indicates that use of the IAPS in conjunction with self report of affective state elicits activity in zygomaticus and corrugater muscle regions of the face (Davidson et al., 2003;Ekman et al., 1980). The IAPS has also been validated in neuroimaging investigations of affective processes in psychiatrically unaffected HC (Britton et al., 2006;Grimm et al., 2006;Hariri et al., 2002;Hariri et al., 2003;Lee et al., 2004;Northoff et al., 2000;Baumgartner et al., 2006) and in MDD subjects (Kalin et al., 1997;Anand et al., 2005a;Irwin et al., 2004;Tremblay et al., 2005;Wagner et al., 2004;Davidson et al., 2003)

Negative (e.g., mutilated face), positive (e.g., Olympic victories), and neutral (e.g., wicker basket) stimuli were selected based on z score transformed normative ratings of arousal (low to high) and valence (negative to positive), separately for male and female ratings, in order to obtain a derived set of stimuli that reliably elicits maximal affective responses for both sexes (Lang and Greenwald, 1988a;Lang and Greenwald, 1988b). The positive (N=30) and negative (N=30) stimuli were matched for high, positive arousal ratings and valence ratings that are maximally opposite in sign. The neutral (N=72) stimuli were selected to have low, negative arousal ratings and valence ratings that were approximately zero. Positive, negative, and neutral affective visual stimuli were presented in a blocked design. During the imaging session, stimuli were presented in two runs, or data acquisition sets. The first run was composed of alternating blocks (groups) of neutral and positive valence pictures. The second run was composed of alternating blocks of neutral and negative valence pictures (Figure 7.

Each block consisted of 12 photographs with constant affective valence; either negative, positive, or neutral. Within each block, each photograph was displayed for 4.0 seconds, with no inter- photograph transition delay, resulting in a block length of 48.0 seconds. The positive-neutral run

106 (data-acquisition set) consisted of 11 alternating positive-neutral blocks, beginning and ending with neutral blocks for a total duration of 540.0 seconds (Figure 7. Similarly, the negative- neutral run was also comprised of 11 alternating negative-neutral blocks for 540.0 seconds. In total 132 visual stimuli were presented in each run. Each stimulus appeared twice per trial. The stimuli were presented in a quasi-random order such that 1) a given stimulus was never repeated with fewer than 12 intervening stimuli, and 2) novel stimuli appeared up to three- quarters of the way through the trial. Each trial began and ended with neutral stimulus blocks.

Subjects were asked to react naturally to the stimuli that were presented on the screen, without any voluntary rumination on previously presented stimuli, and without any conscious effort to identify or regulate their affect. To ensure viewing compliance, subjects were additionally asked to make a quick non-emotional judgment for each picture presented whether it depicted indoor or outdoor action through a pressing, as quickly as possible, one of two buttons on an MR- compatible button pad (Current Designs, Philadelphia PA).

To allow time for the MR signal to reach a steady-state, subjects viewed the word “BEGIN” and a 12 second countdown timer before the presentation of the first neutral block. The runs were presented in a fixed order to all subjects; neutral-positive followed by neutral-negative. The positive and negative trials were not counterbalanced because of evidence that a negative affective state may linger and interfere with the elicitation of a positive affective state (Ekman et al., 1980;Davidson et al., 2003).The order of the images presented, as well as the official IAPS description, and ratings of valence and arousal appear in Appendices 2-5. Visual stimuli were displayed with an XGA LCD projector onto a projection screen that was visible from within the bore of the magnet through a periscope mirror attached in the headcoil. The presentation of stimuli, recording of reaction times and appropriate response button (indoor/outdoor) was implemented in MATLAB 7.4.0 (Natick, MA) using a custom script written in COGENT2000.

107 Figure 6 – Affective Induction Neuroimaging Paradigm

108 Neuroimaging Parameters

All neuroimaging was performed with a 1.5 Tesla GE Echospeed magnetic resonance imaging system (GE Medical Systems, Milwaukee, WI) fitted with a standard quadrature head coil. Subjects were placed into the scanner in a supine position with dense foam padding around the head to minimize movement within the scanner. The MRI compatible button box was attached to the subject’s right quadriceps with medical tape, and was operable using the subject’s right index and middle finger. Prior to entry into the scanner, all subjects were acquainted with use of the button box with the presentation of 20 photographs selected from the IAPS, and appropriate button press recording.

Two experimental runs of 9 minutes were performed after a 4.5 minute high-resolution three- dimensional (3D) anatomic scan of the whole head was acquired. Half of the subjects additionally performed a finger-tapping neuroimaging session, between the anatomical and functional ‘affective’ runs. The results of this functional run will not be explored in this thesis.

For the anatomical scan, 120 axial slices, with a 256 x 256 matrix, and a 20 x 20 cm field of view were acquired using a T1-weighted 3D spoiled gradient. Voxel sizes with 0.78 x 0.78 x 1.10 mm dimensions were obtained using an echo sequence with a flip angle of 45°, echo time (TE), 5 ms, repetition time (TR), 25 ms, and a slice thickness of 1.1 mm. Whole-brain functional imaging employed a spiral gradient echo imaging (Glover and Lee, 1995) of 25 axial slices with a 64 x 64 matrix, and a 20 x 20 cm field of view. Voxel sizes with 3.13 x 3.13 x 4.40 mm dimensions were obtained using a T2*-weighted sequence with a flip angle of 85°, TE = 25 ms, and TR = 2000 ms, and a slice thickness of 4.4 mm. A total of 270 functional volumes were acquired for each run. The first three scan volumes were automatically removed to allow for signal equilibration, three additional volumes corresponding to last six seconds of the viewing of the BEGIN countdown timer were additionally removed.

109 Data Analysis

Clinical Data

Collected clinical data was transferred from individual clinical research folders (CRF) and entered into an electronic database implemented in Statistical Package for the Social Sciences (SPSS v. 13). An intent to treat analysis (ITT) was undertaken along with last-observation- carried forward (LOCF) method to account for missing data (Lane, 2007). Alternative strategies to account for missing data were also considered. A complete case (CC) analysis was rejected on the grounds that it would have led to omission of a significant number of subjects, and an overestimate of the treatment effect. Similarly, implementation of a mixed-effects model repeated measures (MMRM) approach, although desirable, was not realistic due to inadequacy of sample size (Mallinckrodt et al., 2004)

For all clinical measures, means are presented along with the computed standard deviations. Statistical significance between continuous clinical measures was assessed using the student’s t- test with the resultant t value and degrees of freedom presented in subscript. Repeated measures analysis of variance was employed to evaluate time dependent changes in continuous measures with the F value, along with the degrees in freedom in subscript are presented. For binary outcomes, statistical significance of difference in proportions was computed using the exact Fischer’s test.

Neuroimaging Data

Preprocessing All neuroimaging data were collected in single-slice Digital Imaging and Communications in Medicine (DICOM) file format. Following network transfer to a UNIX-based workstation, the raw data were reconstructed into multi-slice volumes using an in-house C-shell script. Data were subsequently converted into the Analyze 7.5 file format with statistical parametric mapping software 2 (SPM2) (Wellcome Department of Cognitive Neurology). All neuroimaging data were subsequently processed with SPM2 implemented in MATLAB 7.4.0 x64 on a Linux based workstation.

110 For each subject, the time series were corrected with the image realignment algorithm. This routine realigned the time-series from each subject using a least-squares approach and a six- parameter (translation and rotation in the x, y, and z plane). The first functional image was used as a reference to which all subsequent scans were realigned with 4th-degree bilinear interpolation for resampling (Friston et al., 1995;Ashburner and Friston, 1999;Friston et al., 1995;Ashburner and Friston, 1999;Ashburner et al., 1997). Any data set with any motion vectors exceeding 2 mm was discarded and not used in the analysis.

Second, the realigned time series were spatially normalized into a standard space as defined by the SPM/MNI canonical EPI template and the acquired structural MRI. The template images supplied with SPM conform to the space defined by the International Consortium for Brain Mapping (ICBM) (Mazziotta et al., 1995) and approximate the Talairach and Tournoux atlas space (Talairach and Tournoux, 1988). The algorithm proceeds by initially determining the optimum 12-parameter affine transformation, initially by matching the head, then by matching the brains. The affine transformation is followed by estimating nonlinear deformation defined by a linear combination of three dimensional discrete cosine transform (DCT) basis functions (Worsley et al., 1996). Unless otherwise specified, all neuroimaging coordinates are in MNI space.

Third, to improve the signal-to-noise ratio during inter-subject averaging (Kiebel et al., 1999), to allow the use of the Gaussian random field theory (Friston K et al., 1995;Worsley and Friston, 1995;Friston KJ et al., 1999), and to normalize the distribution of error terms, data smoothing was performed. The realigned and normalized volumes were smoothed with an 8 mm full-width at half maximum (FWHM) Gaussian smoothing kernel.

Design Matrix & Contrasts To identify paradigm-dependent MR signal changes, the image series from each run were entered into a general linear model. The design matrix modeled voxel activity as the dependent variable with affective stimuli presence as the independent regressor. A total of 264 scans were entered for each subject run with an inter-scan interval of 2000 ms which corresponded to TR. The onset of affective (vs. neutral) stimuli was defined as starting on scans 24, 72, 120, 168, and 216. The duration of each affective epoch (or block) was defined at 24 scans. As inferences about compounds of interest were sought, the time series from each voxel was realigned,

111 normalized, and smoothed data set were modeled by a boxcar reference function convolved with SPM’s hemodynamic reference. Additionally, vector data collected from the initial preprocessing realignment were included as covariates. Data were additionally high-pass filtered at 128 seconds, without serial autocorrelations, or global scaling.

To model changes in brain activation at the block level, blocks with positive and negative visual stimuli were divided into two components; early and late. The early and late components were identical in length (24 sec) and consistency (12 valence-constant photographs). The onset of the early component of affective (vs. neutral) stimuli was also defined as starting on scans 24, 72, 120, 168, and 216, however the duration of each affective epoch (or block) was defined at 12 scans. The onset of the late component of affective (vs. neutral) stimuli was then defined as starting on scans 36, 84, 132, 180, and 228, with a block duration of 12 scans (Figure 8). Intra- block habituation was modeled by directly comparing the early components with the late components, whereas inter-block habituation was modeled by modeling progressive increases and decreases between the first and last block.

112 Figure 7 – Modeling of Inter- and Intra- Block Habituation

113

Between Group Analyses

The limitation of fixed effects model, is that the factors being evaluated are assumed fixed, and of interest only in their own right. The only random component is the residual error, or residual variance. In a fixed-effects model with repeated measurements on a number of subjects, the presence of a subject by response interaction implies that any inference from a fixed effects analysis of the group average is limited to the particular subjects studied (Holmes and Friston, 1998). To extend inference to the population level, a random effects analysis is required to acknowledge both inter-scan error variance and inter-subject component of variance (Woods, 1996). A random effect model acknowledges that it is not the average response of study subjects that is of interest, but instead the average response for the population from which they were drawn (Friston et al., 2002a).

To extend inference to the population level, a second-stage random effects design was constructed using the previously computed within-subject contrasts for each component of each block of the neutral-positive session and the neutral-negative session. These contrasts represent the differential BOLD signal in response to the affective- versus neutral- stimuli in the early and late portion of each block. Factorial effects of interest were identified in a number of different factorial designs. To identify differences between the RS, NR, and HC group, a three-way mixed-design ANOVA was setup with the independent factors group (HC, RS, NR), and the dependent factors intra block position (early, late), and the time (baseline, +1 week, + 6 weeks), and covariates age (years), and sex (male/female).

To evaluate baseline differences in the time course of the BOLD signal during affective processing, a one-way ANOVA analysis along with a group x intra block position interaction was evaluated with baseline contrast images, and a group x inter block position interaction was evaluated with baseline contrast images. To evaluate differential changes to repeat presentations of affective stimuli between groups, a group x time interaction was evaluated. These group by- time interaction analyses highlight changes in activation patterns in the patients (presumably modulated by treatment and response) for which the effects of initial and repeated exposure to the stimuli are controlled. Persistent differences in the neural correlates of affective processing in RS group meeting criteria for remission were analyzed with a main effect of group, and group x

114 intra-block position interaction delimited to endpoint contrast images derived from HC and remitted MDD subjects (RM).

Correlative Analysis A regression model was implemented in SPM for each group which used clinical data as regressors for neuroimaging data. The results of this analysis yielded brain regions that demonstrated a statistically significant association with the clinical data. Data were extracted from brain clusters meeting a-priori established height and extent thresholds and were further evaluated to assess the directionality and magnitude of the association, as indicated by the non- parametric Spearman rho rank correlation (rs).

Assessment of Statistical Significance All statistical tests were conducted with a height threshold set at the two-tailed p<0.05 False Discovery Rate (FDR) corrected value. As two-tailed analyses were performed (BOLD activation or deactivation), the usual p<0.05 value was adjusted for both comparisons, and an extent threshold set at k>15 voxels. As each voxel was normalized into a 2 mm x 2 mm x 2mm k-space, this represented a volume of 120 mm3. Additionally, all activated clusters met the minimum “expected voxels per cluster” statistic as reported by SPM.

Data Presentation Guidelines For each evaluated contrast, brain clusters meeting a-priori established height (p<0.05 FDR) and extent (k>15) thresholds were summarized in tabular format with MNI coordinates of the peak voxel, along with the results of transformation into Talairach space (Brett, 1999). Semi- automated localization of the activation clusters was performed with MNI Space Utility (Pakhomov, 2001). For cortical activation, the name of nearest gyrus is given, if another adjacent gyrus contributes to more than 25% of the cluster, it was also included. A similar rule was applied in the designation of Brodmann areas.

Within-group changes are presented as activation graphs on a Cartesian coordinate system with % change in BOLD signal on the ordinate (y-axis) and time (weeks post baseline) on the abscissa (x coordinate), with MNI coordinates of the peak voxel of the cluster used for title. Similarly, between-group differences are also presented as vertical bar graphs, with % change in BOLD signal on the ordinate (y-axis) and group (HC, RS, NR) on the abscissa (x coordinate), with MNI

115 coordinates of the peak voxel of the cluster used for title. Extracted cluster data are presented as the mean cluster value +/- standard error of measurement (SD/√n).

Correlative analyses are presented on a Cartesian coordinate system with % change in BOLD signal on the ordinate (y-axis) and clinical instrument reading on the abscissa (x coordinate), with MNI coordinates of the peak voxel of the cluster used for title, and a linear regression line of best fit.

Brain maps provide a graphical display of SPM contrast results overlaid on 2mm slices axial slices from a high resolution T1 weighted structural MRI image that has been normalized to the MNI template. Increases in BOLD signal are presented with the hot color map (red-orange- yellow), while decreases in BOLD signal are presented with the cool color map (cyan-blue- indigo). Results from Analysis of Variance designs are presented in the spring color map (green- yellow-white). The MNI z coordinate is located in the bottom right of each brain slice.

To assist in the cross-localization of statistically significant brain clusters between the tabular and graphical displays, brain clusters are presented in tabular format sorted by the z MNI coordinate with inferior brain regions appearing first.

116 RESULTS

Clinical Results

A total of 21 subjects with a DSM-IV-TR diagnosis of MDD were enrolled and completed at least one neuroimaging session. Following the final study visit, 12 MDD subjects were categorized as antidepressant treatment responders (RS; subjects who experienced >50% reduction in the HDRS-17), while nine were categorized as antidepressant treatment nonresponders (NR; subjects who experienced <50% reduction in the HDRS-17). Eleven RS completed all scanning sessions. The only discontinuation was the result of a scheduling conflict (n=1). Only four NR completed all of the scanning sessions, and six NR participated in at least two neuroimaging sessions. Reasons for early discontinuation in the NR group included adverse events (peripheral edema, n=2), lack of efficacy (n=1), suicidal ideation (n=1), and scheduling conflicts (n=1).

A total of 18 psychiatrically unaffected HC were also enrolled and completed at least one neuroimaging session. Fourteen of the HC completed at least two neuroimaging sessions and ten HC participated in all three scanning sessions. Within the HC group, reasons for discontinuation and absence from all four scanning sessions were exclusively due to scheduling conflicts. There was no statistically significant difference between the HC and MDD subjects in age (36.2 +/-

10.3 (SD) vs. 38.9 +/- 11.4 (SD), t37=0.78, p=0.44) or sex (14/21 female vs. 12/18 females, 2=0, p=1.00) (Table 1). Similarly, there were no differences between the RS and NR groups in either age (40.2 +/- 10.7 (SD) vs. 37.3 +/- 12.8 (SD), t19=0.55, p=0.59), or sex distribution (7/12 female vs. 7/9 females, 2=0.88, p=0.35) (Table 2). Within the MDD group, the average duration of the current major depressive episode was 15.7 +/- 15.5 (SD) months. The majority of patients were experiencing (18/21) a recurrent MDE and the majority reported a positive history for mood disorders among first-degree relatives (17/21) (Table 1). There were no statistically significant differences in these clinical variables between the RS and NR groups (Table 2).

117 Table 1 – Study Participants Clinical and Demographic Characteristics: Comparison of HC and MDD Groups

Variable MDD, n=21 HC, n=18 p Age 38.9 (11.4) (SD) 36.2 (10.3) (SD) 0.44 Sex (Female) 14/21 12/18 1.00

HC – healthy control subjects, MDD – major depressive disorder subject

For continuous variables, means are presented first, followed by standard deviations in parentheses. For dichotomous variables, the number of positive cases is presented as a fraction of the total cases in that cell. The third column represents the p-value from either: a) Student’s t-test (2-tailed, unequal variance) which evaluated the null hypothesis of no statistically significant difference in age between HC and MDD b) Fischer’s exact test (2-tailed) which evaluated the null hypothesis of no statistically significant difference in sex distribution between HC and MDD

Table 2 – Study Participants Clinical and Demographic Characteristics: Comparison of RS and NR Groups

Variable NR, n=9 RS, n=12 p Age 37.3 (12.8) (SD) 40.2 (10.7) (SD) 0.59 Sex (Female: Male) 7/9 7/12 0.53 Duration Current MDE (months) 15.4 (20.5) (SD) 15.9 (11.4) (SD) 0.96 Recurrent MDE (n) 8/9 10/12 0.61 Mood Disorder Family History 8/9 9/12 0.41

RS – MDD treatment responder, NR – MDD treatment non-responder

For continuous variables, means are presented first, followed by standard deviations in parentheses. For dichotomous variables, the number of positive cases is presented as a fraction of the total cases in that cell. The third column represents the p-value from either: a) Student’s t-test (2-tailed, unequal variance) which evaluated the null hypothesis of no statistically significant difference in a continuous measure between RS and NR b) Fischer’s exact test (2-tailed) which evaluated the null hypothesis of no statistically significant difference in a dichotomous measure between RS and NR

118

A comparison between the RS and NR groups before treatment initiation revealed that while there were no between-group differences in the severity of the depressive symptoms or the amount of positive dispositional affect, NR group reported significantly more negative dispositional affect than RS (p=0.018) (Table 3). By the second scan, differences in general negative affect between NR and RS were no longer significant, however, NR reported significantly higher severity of self-rated depressive symptoms (p=0.021) (Table 3). At study endpoint, NR also reported higher severity of self-rated- (p<0.001), and clinician-rated- depressive symptoms (p<0.001), more negative dispositional affect (p=0.011), and less positive dispositional affect (p=0.015).

As expected, HC reported fewer depressive symptoms (p<0.001) and less negative dispositional affect (p<0.001) than either the RS or NR group throughout the duration of the study. Moreover, HC also reported higher amount of general positive affect than either the RS or NR group, throughout the entire study (p<0.001).

During all post-scan evaluation of the visual stimuli presented in the positive-neutral run, RS and NR reported significantly more positive affect with the positive visual stimuli versus the neutral stimuli (9.9 +/- 3.0 vs. 2.8 +/- 1.8., t11=3.03, p=0.014, RS and 11.8 +/- 2.2 vs. 2.0 +/- 1.2., t8=7.84, p<0.001, NR). Similarly during the negative-neutral run, there was significantly more negative affect reported (-16.5 +/- 0.8 vs. 2.3 +/- 1.2, t11= 12.43, p<0.001, RS, and (-16.9 +/- 1.9 vs. 1.6 +/- 2.3, t8= 5.95, p=0.001, NR). Comparable measures of induced affect were also reported by the HC group for both positive-neutral and negative-neutral runs (Table 3). A repeated-measure ANOVA analysis revealed a comparable magnitude of induced affect during both positive-neutral and negative-neutral runs throughout the course of the study (Table 4).

119 Table 3 – Between Group Comparison of Study Participants’ Measures of Depression Severity, Dispositional Affect, and Induced Affect at the Baseline, Second, and Final Visit

Baseline Measurements Scale HC RS NR p Post Hoc APRS-0P 3.6 +/- 1.4 2.3 +/- 1.7 2.0 +/- 1.2 0.70 - APRS-0N 2.8 +/- 1.1 2.3 +/- 1.2 1.6 +/- 2.3 0.85 - APRS-N -13.9 +/- 1.1 -16.5 +/- 0.8 -16.9 +/- 1.9 0.18 - APRS-P 12.8 +/- 1.1 9.9 +/- 3.0 11.8 +/- 2.2 0.57 - HDRS-17 - 21.1 +/- 1.1 22.3 +/- 1.9 0.57 - BDI 1.94 +/- 0.7 26.0 +/- 3.0 28.7 +/- 2.8 <0.001 H>R, H>N PANAS-P 33.2 +- 1.8 20.2 +/- 2.3 18.7 +/- 1.8 <0.001 H>R, H>N PANAS-N 13.8 +/- 1.0 25.0 +/- 2.3 32.1 +/- 2.7 <0.001 H

Second Visit Measurements Scale HC RS NR P Post Hoc APRS-0P 3.6 +/- 1.7 3.0 +/- 1.2 2.3 +/- 1.4 0.84 - APRS-0N 2.2 +/- 1.0 0.5 +/- 1.2 1.8 +/- 2.4 0.65 - APRS-N -13.3 +/- 1.0 -14.3 +/- 2.0 -18.1 +/- 1.1 0.09 - APRS-P 11.9 +/- 1.1 11.5 +/- 1.9 11.4 +/- 1.1 0.96 - HDRS-17 . +/- . 12.8 +/- 1.8 16.2 +/- 2.0 0.22 - BDI 1.9 +/- 0.7 18.2 +/- 2.8 25.9 +/- 3.0 <0.001 H

Final Visit Measurements Scale HC RS NR P Post Hoc APRS-0P 1.9 +/- 1.6 3.1 +/- 1.2 3.0 +/- 1.9 0.81 - APRS-0N 0.1 +/- 1.1 0.8 +/- 0.9 1.2 +/- 3.1 0.87 - APRS-N -14.0 +/- 1.5 -13.9 +/- 1.7 -19.4 +/- 0.6 0.10 - APRS-P 10.7 +/- 1.5 9.9 +/- 1.4 10.4 +/- 2.5 0.93 - HDRS-17 . +/- . 5.8 +/- 0.9 17.9 +/- 2.3 <0.001 R

Means are presented first, followed by +/- standard error of measurement. The final two columns represent the p-value from a one-way ANOVA which evaluated the null hypothesis of no statistically significant difference the psychometric scale along with the results of a Bonferroni post-hoc test to evaluate the location and direction of the between-group difference.

120

Antidepressant treatment with fluoxetine (20 mg/day) and olanzapine (5 mg/day) began immediately after the first scan. The dosing for both medications was flexible under the clinical supervision of a trained psychiatrist with a recommended titration to 40 mg of fluoxetine and 10 mg of olanzapine after two weeks, and 60 mg of fluoxetine and 12.5 mg of olanzapine after 4 weeks. At study endpoint, the average (+/- standard deviation) medication doses for the RS group were 8.9 +/- 2.5 mg olanzapine and 41.7 +/- 13.4 mg fluoxetine, comparable to dosing in the NR group, 8.8 +/- 2.6 mg and 34.4 +/- 13.3 mg, respectively.

Overall, RS experienced a statistically significant decrease in depressive symptom severity as assessed by both the HDRS-17 (baseline 21.1 +/- 3.9, endpoint 5.8 +/- 3.0, F3,33=33.48, p<0.001), and the BDI (baseline 26.0 +/- 10.3, endpoint 13.0 +/- 5.8, F3,33=49.40, p<0.001). Antidepressant treatment was also associated with improvements in ratings of dispositional affect (Table 3). There was a statistically significant increase in general positive affect ratings from baseline to endpoint (20.2 +/- 2.3 to 27.1 +/- 2.3, F3,33=3.35, p=0.031), along with a numerical decrease in general negative affect (25.0 +/- 2.3 to 19.1 +/- 2.0, F3,33=2.24, p=0.102) (Figure 8). Although not as robust as in the RS group, the NR also experienced a statistically significant change in depressive symptom severity as assessed by the HDRS-17 (baseline 22.3

+/- 1.9, endpoint 17.9 +/- 2.3, F3,24=5.71, p=0.004). Antidepressant treatment was also associated with mild improvements in ratings of general negative affect (Table 3, Figure 8), however, these numerical differences were not statistically significant (Figure 8).

Healthy control subjects (HC) reported a constant and time-independent amount of positive dispositional affect (PANAS-P) and negative affect (PANA-N) over the six week course of the study (Table 2). A repeated-measure ANOVA analysis confirmed that there was no statistically significant change in self-rated positive affect PANAS-P (F3,48=0.48, p=0.695) or self-rated negative affect PANAS-N (F3,48=0.28, p=0.840) from baseline to study endpoint. Consistently low ratings of depressive symptoms were reported by HC throughout the course of the study without any statistically significant change over the course of the study (F3,48=0.83, p=0.48).

Button press accuracy and response time to indicate if the presented photographs were indoor/outdoor were comparable between the three groups, without any statistically significant within-group or between-group differences (Table 4).

121

Figure 8 – Longitudinal Course of Depressive Symptom Severity, Dispositional Affect, and Induced Affect in the Study Participants.

Group means of the psychometric scales are presented at each time point along with standard error of measurement. Statistically significant changes from the baseline measure are indicated with an appropriately coloured asterisk for each group (Bonferroni post-hoc test of within-group repeated measures ANOVA).

122 Table 4 – Between- and Within-Group Comparison of Study Participants’ Button Press Accuracy and Response Times at the Baseline, Second, and Final Visit for Positive-Neutral Affect and Negative-Neutral Affect Runs

Positive-Neutral Stimuli Run

Group Time Subjects Accuracy Within- Response Within- (weeks) (n) (%) Group (p) Time (ms) Group (p) HC 0 18 82.5 +/- 1.2 964 +/- 26 HC 1 14 84.0 +/- 1.9 0.788 916 +/- 40 0.536 HC 2 10 83.4 +/- 1.9 969 +/- 47 RS 0 11 84.3 +/- 1.9 957 +/- 34 RS 1 11 85.5 +/- 3.2 0.780 965 +/- 27 0.964 RS 6 11 86.7 +/- 2.0 969 +/- 35 NR 0 6 82.2 +/-1.6 1038 +/- 68 NR 1 5 83.4 +/- 2.0 0.762 985 +/- 65 0.414 NR 6 4 84.0 +/-3.2 903 +/-66 Between 0 0.635 0.383 Group 1 0.862 0.503 (p) 6 0.857 0.666

Negative-Neutral Stimuli Run

Group Time Subjects Accuracy Within- Response Within- (weeks) (n) (%) Group (p) Time (ms) Group (p) HC 0 18 87.3 +/- 1.2 1048 +/- 34 HC 1 14 87.4 +/- 1.7 0.998 971 +/- 33 0.068 HC 2 10 87.2 +/- 1.9 1092 +/- 29 RS 0 11 86.1 +/- 6.2 968 +/- 34 RS 1 11 89.2 +/- 2.9 0.524 1057 +/- 28 0.118 RS 6 11 87.0 +/- 2.2 1008 +/- 26 NR 0 6 86.0 +/- 2.7 968 +/- 40 NR 1 5 88.0 +/- 2.8 0.842 1060 +/- 62 0.251 NR 6 4 86.0 +/- 2.1 1079 +/- 41 Between 0 0.806 0.195 Group 1 0.840 0.139 (p) 6 0.861 0.102

Means are presented first, followed by +/- standard error of measurement. The within-group columns represent the p-value from a repeated measures within-group ANOVA which evaluated the null hypothesis of no statistically significant difference between the three time points. The between-group row on the bottom represents the p-value from a one-way ANOVA which evaluated the null hypothesis of no statistically significant difference between the three time groups at each time point.

123 PRIMARY AIM A:

EVALUATE DIFFERENCES IN BRAIN ACTIVATION EVOKED BY AFFECTIVE PROCESSING BETWEEN HC AND MDD, AND BETWEEN RS AND NR

Positive Affect Neuroimaging Results at Baseline Visit

The presentation of positive-affect visual stimuli compared to neutral affect stimuli resulted in increased BOLD signal in a number of cortical and subcortical regions in the study participants (HC and MDD subjects) (Table 5, Figure 9). Common activation among all study participants was identified bilaterally in the occipitotemporal cortex (BA 19/37), the precuneus (BA 7), in the medial temporal lobe (BA 35/37), and in the temporal lobe. Activations were also identified in the orbitofrontal cortex/ ventromedial PFC (BA 10/11) and in the dorsomedial PFC (BA 9/10). Subcortical activation was localized to the left hippocampus, and ventral hypothalamus in the vicinity of the mammillary bodies, and in the vermis of the cerebellum. The presentation of positive versus neutral affective visual stimuli also evoked BOLD signal deactivations throughout the ventrolateral (BA 11/46/47) and dorsolateral PFC (BA 8/9) and in the anterior midcingulate (BA 24) extending through the rCMA and in the preSMA. Subcortical deactivations were also noted in the dorsal cerebellar hemispheres and in the lateral dorsal nuclei of the thalamus.

To analyze the time course of the BOLD signal change during the presentation of positive affective visual stimuli (the positive blocks), the first half of the positive blocks (early) were contrasted with the second half (late) in the study participants to evaluate intra-block habituation (Table 6). Statistically significant decreases in activation were identified bilaterally in the OPT (BA 19/37), in the precuneus (BA 7), the superior parietal lobule (BA 7), and the ventromedial PFC (BA 10) (Figure 10). In addition, decreases in BOLD signal deactivation were also noted in ventrolateral PFC (BA 45/46) and bilaterally in the cerebellar hemispheres.

To further analyze the time course of the BOLD signal change during positive block presentation, a contrast was made between the five positive blocks in the study participants to evaluate inter-block BOLD signal habituation. There were no statistically significant inter-block changes in the BOLD signal that were common to all study participants.

124 Table 5 – Activation and Deactivation in BOLD Signal Evoked by Positive Affective Visual Stimuli in All Subjects

Gyrus/Region BA Side T-Score Size MNI TAL (voxels) x y z x y Z INCREASES Mid Occipital 19/37 R 15.18 2442 52 -72 0 51 -70 3 Mid Occipital 19 L 9.26 1814 -50 -78 4 -50 -75 7 Precuneus 7 R 7.65 2697 4 -62 34 4 -59 34 Med Frontal 10/11 - 4.89 258 0 56 -10 0 54 -11 Fusiform 37 L4.37 55 -42 -54 -20 -42 -53 -14 Mid Temporal 21 L 4.29 130 -48 8 -34 -48 6 -29 Fusiform 37 R4.07 19 40 -48 -20 40 -47 -14 Mid Temporal 38 R 3.99 58 44 14 -38 44 12 -33 Sup Frontal 10 L 3.83 227 -10 60 18 -10 59 14 Hypothalamus – Cm - L 3.59 37 -8 -6 -16 -8 -6 -13 Parahippocampal 35 R 3.55 93 24 -10 -28 24 -11 -23 Mid/Sup Temporal 38 L 3.51 24 -32 12 -36 -32 10 -31 Cerebellum – vermis - L 3.41 32 -8 -42 -18 -8 -41 -13 Hippocampus - L 3.18 18 -30 -38 -8 -30 -37 -5 Cerebellum – vermis - L 3.07 19 -14 -36 -10 -14 -35 -7 DECREASES Cerebellum – hemisphere L 7.37 273 -28 -52 -22 -28 -51 -16 Inf Frontal 46 L 6.33 2709 -46 36 14 -46 36 11 Cerebellum – hemisphere R 5.97 333 30 -56 -16 30 -55 -11 Med Frontal 8 R 4.98 1503 4 26 50 4 27 45 Inf Frontal 47 R 4.90 657 44 44 -10 44 42 -11 Mid Temporal 21 L 4.75 654 -58 -28 -10 -57 -28 -7 Sup Temporal 22 R 4.36 148 52 16 -6 51 15 -6 Mid Occipital 18/19 L 4.24 192 -26 -82 6 -26 -79 9 Mid Temporal 21 R 4.20 246 52 -36 -12 51 -35 -8 Cerebellum - hemisphere L 4.06 263 -44 -46 -44 -44 -46 -35 Ant Mid Cingulate 24 L 3.94 320 -2 4 32 -2 5 29 Ant Insula R 3.87 97 34 22 8 34 22 6 Thalamus – lateral dorsal R 3.80 113 8 -16 16 8 -15 15 Sup Occipital 19 L 3.68 184 -24 -74 36 -24 -70 37 Inf Parietal Lobule 40 R 3.51 178 50 -58 40 50 -54 40 Precentral 6 R 3.45 86 56 8 14 55 8 12 Thalamus – lateral dorsal L 3.34 34 -10 -18 16 -10 -17 16 Mid Frontal 9 R 3.33 43 40 36 38 40 37 33 Mid Frontal 6 R 3.33 161 28 8 58 28 10 53 Mid Frontal 11 L 3.23 39 -22 42 -14 -22 40 -14

LEGEND: Changes in BOLD signal meeting height threshold p<0.05 (FDR-corrected), and extent threshold k>15 voxels (120 mm3). Anatomical localization based on peak voxel coordinates in Talairach space. Cluster size corresponds to number of (8 mm3) voxels.

125 Figure 9 – Activation and Deactivations in BOLD Signal Evoked by Positive Affective Visual Stimuli in All Subjects

LEGEND: Changes in BOLD signal meeting height threshold p<0.05 (FDR-corrected), and extent threshold k>15 voxels (15 mm3) overlay on representative structural MRI image, with approximate MNI z-axis coordinate in bottom right corner. Hot colour flare represents T-score of increased BOLD signal; cold colour flare represents T-score of decreased BOLD signal from Table 5.

126 Table 6 – Changes in BOLD Signal between the Early and Late Components of the Positive Affective Block in All Subjects

Gyrus/Region BA Side F-Score Size MNI TAL (voxels) x y z x y z DECREASES Mid Occipital 19/37 R 34.95 506 54 -70 0 53 -68 4 Mid Occipital 19 L 26.40 424 -42 -88 0 -42 -85 4 Mid Frontal 10 L 19.65 57 -26 46 -2 -26 45 -4 Precuneus 7/19 R 16.98 115 4 -62 32 4 -59 32 Sup Parietal Lobule 7 R 16.91 84 26 -60 62 26 -55 60 Precuneus 7 L 15.05 27 -8 -52 44 -8 -48 43 INCREASES Cerebellum - hemisphere R 33.36 272 24 -44 -18 24 -43 -13 Cerebellum - hemisphere L 16.40 68 -28 -56 -16 -28 -55 -11 Inf Frontal 45/46 R 15.31 51 40 32 10 40 31 8

127 Figure 10 – Main Effect of Intra-Block Changes in BOLD Signal between the Early and Late Components of the Positive Affective Block in All Subjects

128 To evaluate difference in brain activation and deactivation during the induction of positive affect, one-way ANOVA was performed to revealed statistically significant differences between HC, RS, and NR (Table 7). Post-hoc analyses on extracted data from the statistically significant clusters identified the directionality of these differences (Figure 11). All of the identified clusters were the result of decreased BOLD signal by the NR group. Decreased BOLD signal compared to the RS and HC groups was noted in the ventromedial PFC (BA 10) and the precuneus (BA 7). On the other hand, the RS group was marked by increased BOLD signal compared to the HC and NR group throughout the OPT cortex (BA 21/37/39) and in the dorsomedial and dorsolateral PFC (BA 9). In fact, the BOLD signal in these regions was actually higher than in the HC group.

Table 7 – Between Group Differences in BOLD Signal Evoked by Positive Affective Stimuli during the Baseline Visit

Gyrus/Region BA Side F-Score Size MNI TAL (voxels) x y z x y z RS, HC > NR Med Frontal 10 - 8.05 16 0 54 14 0 53 10 Precuneus 7 R 10.89 125 12 -56 34 12 -53 34 RS > HC > NR Mid Temporal 39 R 14.00 267 58 -62 12 57 -60 14 Dor Post Cingulate 31 L 8.98 30 -4 -46 32 -4 -43 32 RS > HC, NR Mid Temporal 21/37 L 16.21 188 -42 -60 6 -42 -58 8 Inf Temporal 37 L 10.80 68 -48 -54 -14 -47 -53 -9 Mid Frontal 9/10 R 9.51 27 40 50 26 40 49 21 Sup Frontal 9 R 9.23 47 24 54 32 24 54 27 Sup Frontal 9 L 9.16 39 -4 58 34 -4 58 28

129 Figure 11 – Main Effect of Group: Between Group Differences in BOLD Signal Evoked by Positive Affective Stimuli during the Baseline Visit

LEGEND: Differences in BOLD signal meeting height threshold p<0.05 (FDR-corrected), and extent threshold k>15 voxels (120 mm3) overlay on representative structural MRI image, with approximate MNI z-axis coordinate in bottom right corner. Colour flare represents F-score of a statistical test evaluating the effect of group in BOLD signal in the positive affective block from Table 7.

130 To evaluate differential BOLD signal habituation during the induction of positive affect, two group x intra-scan habituation interactions were evaluated (Table 8). Differences in inter-block BOLD signal habituation were recorded on the medial inferior temporal lobe between HC, RS and NR. Whereas HC and RS groups demonstrated comparable BOLD activation throughout the five positive blocks, there was a significant increase in BOLD activation between the first and last positive block in the NR group in the temporal lobe (BA 28) (Figure 12).

The NR group was also unique in its pattern of intra-block habituation. While HC and RS demonstrated decreases in BOLD signal between the early and late components of the positive block in the OPT (BA 39), NR were characterized by increases in BOLD signal. A similar pattern was noted in the right putamen, where the NR demonstrated maximal activation during the second half of the positive block, while the opposite pattern was recorded in the RS. Interestingly, HC demonstrated comparable levels of activation in the putamen for the complete duration of the positive block.

In summary, positive affect processing at baseline was characterized by decreased BOLD signal by the NR group, in the OPT and OPP cortex. Increased BOLD signal by RS compared to the NR (and even HC) was noted in the dorsolateral and dorsomedial PFC. Between-group differences in intra-scan habituation were identified in the NR group with delayed time to maximum activation at the intra-block and inter-block level in the putamen and medial temporal lobe respectively.

Table 8 – Group x Intra-Scan Habituation Interaction in BOLD Signal Evoked by Positive Affective Stimuli during the Baseline Visit

Gyrus/Region BA Side F-Score Size MNI TAL (voxels) x y z x y z Intra-Block Putamen R 10.50 47 22 10 2 22 10 1 Sup Temporal 39 L 9.62 30 -54 -54 28 -53 -51 28 Inter-Block Uncus 28 L 24.11 46 -20 0 -28 -20 -2 -24

131 Figure 12 – Group x Intra-Scan Habituation Interaction in BOLD Signal Evoked by Positive Affective Stimuli during the Baseline Visit

LEGEND: Changes in BOLD signal meeting height threshold p<0.05 (FDR-corrected), and extent threshold k>15 voxels (120 mm3) overlay on representative structural MRI image, with approximate MNI z-axis coordinate in bottom right corner. Pink-yellow colour flare represents F-score of a statistical test evaluating the effect of group x inter-block habituation in BOLD signal during the baseline visit; green-yellow colour flare represents the F-score of a statistical test evaluating the effect of group x intra-block habituation in BOLD signal during the baseline visit in Table 8.

132 Negative Affect Neuroimaging Results at Baseline Visit

The induction of negative affect was associated with a more robust activation than the induction of positive affect (Table 9, Figure 13). Subcortical activation was recorded bilaterally in the amygdala, and the thalamus. Cortical activation was most pronounced in two clusters corresponding to the OPT cortex (BA19/37). Additional activation was recorded in the ventrolateral (11/47), dorsolateral (BA 9/46), and dorsomedial PFC (BA 9/10), and in the aMCC (BA 24) and the dPCC (BA 31). Statistically significant deactivations in BOLD signal were also identified in the posterior insular cortex.

Table 9 – Activation and Deactivation in BOLD Signal Evoked by Negative Affective Visual Stimuli in All Subjects

Gyrus/Region BA Side T-Score Size MNI TAL (voxels) x y z x y z INCREASES Inf Temporal/Occipital 19 R 16.53 26369 46 -80 -4 46 -78 1 Inf/Mid Frontal 9/44 R 11.05 8499 44 12 28 44 13 25 Fusiform/Amygdala 20 L 8.36 5598 -42 20 -26 -42 18 -23 Inf Parietal Lobule 40 L 4.24 236 -58 -32 42 -57 -29 40 Fusiform/Amygdala 21 R 4.19 72 58 -6 -28 57 -7 -23 Dor Postcingulate 31 R 4.08 309 6 -58 28 6 -55 29 Inf Temporal/Uncus 20 L 3.96 37 -30 -8 -40 -30 -9 -33 Ant Midcingulate 24 R 3.46 39 2 0 30 2 1 28 DECREASES Post Insula - L 4.79 274 -40 -18 20 -40 -17 19 Post Insula - R 3.62 92 40 -18 12 40 -17 12

133 Figure 13 – Activation and Deactivations in BOLD Signal Evoked by Negative Affective Visual Stimuli in All Subjects

134 As was done previously for the positive affective processing data set, the time course of the BOLD signal change during the presentation of negative affective visual stimuli were evaluated for both intra-block and inter-block habituation in all study participants. Statistically significant increases in activation were identified in the pACC (BA 24/32) and in the precuneus (BA 7) (Figure 14, Table 10). Conversely, decreases in BOLD signal deactivation were also noted in the components of dorsal visual pathway (BA 7/18) and in the substantia nigra. A contrast between the five negative blocks in the study participants’ BOLD signal also revealed statistically significant inter-block habituation (Table 11). With each successive negative block, the study participants demonstrated decreased BOLD signal in the pACC (BA 24) and dorsolateral temporoparietal cortex (BA 39/40).

Table 10 – Changes in BOLD Signal between the Early and Late Components of the Negative Affective Block in All Subjects

Gyrus/Region BA Side F-Score Size MNI TAL (voxels) x y z x y z INCREASES Pregenual Cingulate 24/32 R 18.62 226 16 38 16 16 38 13 Precuneus 7 R 16.16 142 4 -76 34 4 -72 35 DECREASES Sup Parietal Lobule 7 R 33.82 656 26 -64 48 26 -60 47 Mid Occipital 18 L 29.41 685 -32 -88 4 -32 -85 8 Substantia Nigra - L 17.81 60 -10 -18 -10 -10 -18 -8

Table 11 – Changes in BOLD Signal between the First and Subsequent Negative Affective Block in All Subjects

Gyrus/Region BA Side F-Score Size MNI TAL (voxels) x y z x y z DECREASES Sup Temporal 39 L 7.58 251 -56 -54 26 -55 -51 27 Pregenual Cingulate 24 - 7.33 160 0 28 18 0 28 15 Inf Parietal Lobule 40 R 6.89 167 48 -46 40 48 -43 39

135 Figure 14 – Main Effect of Intra-Block Changes in BOLD Signal between the Early and Late Components of the Negative Affective Block in All Subjects

136 Figure 15 – Main Effect of Inter-Block Changes in BOLD Signal between the First and Subsequent Negative Affective Block in All Subjects

137 To evaluate difference in brain activation and deactivation during the induction of negative affect, another one-way ANOVA was performed to revealed statistically significant differences between HC, RS, and NR (Table 12). Post-hoc analyses on extracted data from the statistically significant clusters identified the directionality of these differences (Figure 16). All of the identified clusters were the result of increased BOLD signal by the MDD group relative to HC (Figure 16, Table 12).

Increased BOLD signal by the MDD groups compared to HC groups was noted throughout the PFC in the dorsolateral (BA 9), dorsomedial (BA 9), ventromedial (BA 11), and ventrolateral (BA 10/46) compartments, the pACC (BA 24/32), the left amygdala, and in the OPT cortex (BA 39). From within these clusters, the NR group also demonstrate increased BOLD signal relative to the RS in the pACC (BA 24), the dorsomedial PFC (BA 9), and ventrolateral PFC (BA 10/46). Moreover, NR also demonstrated increased BOLD signal in the left amygdala against both RS and HC, however, the difference was only statistically significant versus HC. .

Table 12 – Between Group Differences in BOLD Signal Evoked by Negative Affective Stimuli during the Baseline Visit

Gyrus/Region BA Side F-Score Size MNI TAL (voxels) x y z x y z NR > HC, RS Pregenual cingulate 24/32 L 7.59 59 -6 42 18 -6 42 15 Sup Frontal 9 L 9.01 54 -28 46 34 -28 46 29 Mid Temporal 21 L 9.1 61 -58 -18 -10 -57 -18 -8 NR > RS > HC Inf/Mid Frontal 10/46 R 13.95 1030 44 46 16 44 45 12 NR, RS > HC Mid Frontal 9 L 8.69 57 -42 20 36 -42 21 32 NR > HC Amygdala - L7.92 38 -18 -6 -22 -18 -7 -18 RS > HC Med Frontal 11 - 10.41 136 0 50 -10 0 48 -11 RS > HC, NR Mid Temporal 39 R 12.93 63 58 -64 14 57 -61 16

138 Figure 16 – Main Effect of Group: Between Group Differences in BOLD Signal Evoked by Negative Affective Stimuli during the Baseline Visit

LEGEND: Differences in BOLD signal meeting height threshold p<0.05 (FDR-corrected), and extent threshold k>15 voxels (120 mm3) overlay on representative structural MRI image, with approximate MNI z-axis coordinate in bottom right corner. Colour flare represents F- score of a statistical test evaluating the effect of group in BOLD signal in the negative affective block from Table 7.

139 To evaluate potential differences in BOLD signal habituation during the induction of negative affect, another two group x intra-scan habituation interactions were evaluated (Table 13). Differences in inter-block BOLD signal habituation were identified bilaterally in the dorsolateral PFC (BA 9) (Figure 17). Whereas HC demonstrated decreases BOLD activation throughout the five positive blocks, there was a significant increase in BOLD activation between the first and last positive block in the RS group (Figure 17).

When intra block habituation was evaluated at the baseline visit, both MDD groups were found to have statistically significant delays in reaching maximum activation (Table 13, Figure 18). While HC demonstrated maximum activation in the medial temporal lobe (BA 36) in the first half of the negative blocks, followed by BOLD signal deactivation, NR and RS demonstrated maximum activation during the second half of the block.

In summary, negative affect processing at baseline was characterized by increased BOLD signal by the MDD group, in the PFC. Increased BOLD signal by NR compared to the RS was noted in the pACC and ventrolateral PFC. Delayed times to maximum activation were noted in the MDD groups at the intra-block level in the medial temporal lobe, and in the RS groups at the inter- block level in dorsolateral PFC.

Table 13 – Group x Intra-Scan Habituation Interaction in BOLD Signal Evoked by Negative Affective Stimuli during the Baseline Visit

Gyrus/Region BA Side F-Score Size MNI TAL (voxels) x y z x y z Intra-Block Parahippocampal 36 R 10.33 34 16 -36 -18 16 -36 -18 Inter-Block Mid Frontal 9 R 2.99 32 46 38 36 46 38 31 Mid Frontal 9 L 2.98 39 -48 32 34 -48 33 30

140 Figure 17 – Group x Intra-Scan Habituation Interaction in BOLD Signal Evoked by Negative Affective Stimuli during the Baseline Visit

LEGEND: Changes in BOLD signal meeting height threshold p<0.05 (FDR-corrected), and extent threshold k>15 voxels (120 mm3) overlay on representative structural MRI image, with approximate MNI z-axis coordinate in bottom right corner. Green-yellow colour flare represents F-score of a statistical test evaluating the effect of group x inter-block habituation in BOLD signal during the baseline visit (left); and an F-score of a statistical test evaluating the effect of group x intra-block habituation in BOLD signal during the baseline visit (right).

141 PRIMARY AIM B:

EVALUATE DIFFERENCES IN INTER-SCAN HABITUATION ON BRAIN ACTIVATION EVOKED BY REPEAT AFFECTIVE PROCESSING BETWEEN HC, RS, AND NR

Repeat Induction of Positive Affect – Inter Scan Changes in BOLD Signal

In order to identify the effect of successful treatment on the BOLD signal evoked by affective processing, it was first necessary to identify the effect of repeat scans without any treatment; in the HC. To calculate statistically significant changes in activation and deactivation throughout the course of the study, while avoiding the problem of multiple comparisons (i.e. calculation of all possible pair wise comparisons), an F-test was computed to identify the main effect of time on activation and deactivation during the induction of positive affect (Table 14). In addition inter-scan changes in intra-scan habituation were also computed with a time-x-intra-scan habituation interaction F-test.

Two clusters met statistical significance criteria; the left OPT cortex (BA 37) and a portion of the midbrain corresponding roughly to the substantia nigra (Figure 18). Both of these regions represented time-dependent increases in BOLD signal; between the second and final scan in the OPT, and between the first and final scan in the midbrain. There were no statistically significant changes in intra-scan habituation between the first and final scan.

Table 14 – Inter-Scan Changes in BOLD Signal Evoked by Repeat Presentation of Positive Affective Visual Stimuli in HC

Gyrus/Region BA Side F-Score Size MNI TAL (voxels) X y z x y z INTER-SCAN INCREASES Inf Temporal 19/37 L18.76 741 -40 -72 -4 -40 -70 0 Substantia Nigra L 11.68 63 -12 -24 -10 -12 -24 -7

142 Figure 18 – Inter-Scan Changes in BOLD Signal Evoked by Repeat Presentation of Positive Affective Visual Stimuli in HC

LEGEND: Changes in BOLD signal meeting height threshold p<0.05 (FDR-corrected), and extent threshold k>15 voxels (120 mm3) overlay on representative structural MRI image, with approximate MNI z-axis coordinate in bottom right corner. Green-yellow colour flare represents F-score of a statistical test evaluating the main effect of time in BOLD signal in the HC group from Table 14.The asterisk is placed midway between the two time-points where a Bonferroni post-hoc test identified statistically significant changes in BOLD signal.

143

Inter-scan changes in the BOLD signal in the RS group can be modeled as a combination of i) inter-scan habituation (potentially comparable to HC), ii) therapeutic effects of medication, and iii) non-therapeutic medication effect (e.g. somatic side effects). Indeed, repeat induction of positive affect in RS at the second and final visit was accompanied by statistically significant changes in BOLD signal in the prefrontal and temporal lobe (Figure 19, Table 15). Roughly comparable to the HC findings, increased BOLD signal between the first and final scan was observed in the left OPT cortex (BA 37). Decreased BOLD signal between the first and final scan was also observed in the dorsolateral PFC (BA 9/46), a finding not observed in the HC group. Time-dependent changes in intra-scan habituation were also noted in the RS group after treatment initiation. There was a progressive decrease in the BOLD signal difference between the early and late components of the positive block in the sACC (BA 24) culminating in comparable and low activation at the time of the final scan (Figure 19, Table 16).

Table 15 – Inter-Scan Changes in BOLD Signal Evoked by Repeat Presentation of Positive Affective Visual Stimuli in RS

Gyrus/Region BA Side F-Score Size MNI TAL (voxels) X y z x y z INCREASES Inf Temporal 37 L 9.42 24 -58 -56 -12 -57 -55 -7 DECREASES Mid Frontal 9/46 R 7.60 34 38 48 28 38 48 23

Table 16 – Inter-Scan Changes in Intra-Scan BOLD Signal Habituation Evoked by Positive Affective Visual Stimuli in RS (Time x Intra-Scan Habituation)

Gyrus/Region BA Side F-Score Size MNI TAL (voxels) x y z x y z Inf Frontal 46 L 9.01 262 -34 32 14 -34 32 11 Subgenual Cingulate 24 L 7.35 37 -12 36 -4 -12 35 -5 Inf Frontal 45 R 7.03 41 48 24 20 48 24 17

144 Figure 19 – Inter-Scan Changes in BOLD Signal Evoked by Repeat Presentation of Positive Affective Visual Stimuli in RS and Inter-Scan Changes in Intra-Scan BOLD Signal Habituation (Time x Intra-Scan Habituation)

LEGEND: Changes in BOLD signal meeting height threshold p<0.05 (FDR-corrected), and extent threshold k>15 voxels (120 mm3) overlay on representative structural MRI image, with approximate MNI z-axis coordinate in bottom right corner. Green- yellow colour flare represents F-score of a statistical test evaluating the main effect of time in BOLD signal in the RS group from Table 15.The asterisk is placed midway between the two time-points where a Bonferroni post-hoc test identified statistically significant changes in BOLD signal. Pink-yellow colour flare represents F-score of a statistical test evaluating the interaction between time and intra-scan habituation in BOLD signal in the RS group from Table 16.

145

Evaluation of the differences in BOLD signal in the NR group between the first, second and final scans allows an inference about the effect of non-therapeutic medication exposure on affective processing along with an insight into inter-scan habituation in actively depressed subjects. Contrary to the time-dependent inter-scan increases that were identified in the HC and RS groups in the left OPT (BA 19/37), analysis of the NR group actually identified a pronounced decrease in BOLD signal between the first and final scan (Figure 20, Table 17).

Time dependent changes in intra-scan habituation were also noted in the NR group after treatment initiation in the head of the right caudate (Table 18). At the baseline visit there was an NR-characteristic delay in the maximum activation in the second half of the positive block. By the second scan, this early-late difference was abolished, only to return back in by the time of the final scan.

Table 17 – Inter-Scan Changes in BOLD Signal Evoked by Repeat Presentation of Positive Affective Visual Stimuli in NR

Gyrus/Region BA Side F-Score Size MNI TAL (voxels) X y z x y z DECREASES Inf Temporal 37 L12.92 62 -48 -50 -16 -48 -49 -11

Table 18 – Inter-Scan Changes in Intra-Scan BOLD Signal Habituation Evoked by Positive Affective Visual Stimuli in NR (Time x Intra-Scan Habituation)

Gyrus/Region BA Side F-Score Size MNI TAL (voxels) x y z x y z Caudate Head R 16.04 29 16 22 20 16 22 17

146 Figure 20 – Inter-Scan Changes in BOLD Signal Evoked by Repeat Presentation of Positive Affective Visual Stimuli in NR and Inter-Scan Changes in Intra-Scan BOLD Signal Habituation (Time x Intra-Scan Habituation)

LEGEND: Changes in BOLD signal meeting height threshold p<0.05 (FDR-corrected), and extent threshold k>15 voxels (120 mm3) overlay on representative structural MRI image, with approximate MNI z-axis coordinate in bottom right corner. Green-yellow colour flare represents F-score of a statistical test evaluating the main effect of time in BOLD signal in the RS group from Table 17.The asterisk is placed midway between the two time-points where a Bonferroni post-hoc test identified statistically significant changes in BOLD signal. Pink-yellow colour flare represents F-score of a statistical test evaluating the interaction between time and intra-scan habituation in BOLD signal in the RS group from Table 18.

147 In order to identify statistically significant between-group-difference in inter-scan changes in BOLD signal, a group x time interaction contrast was evaluated (Table 19). There was a statistically significant inter-scan increase in BOLD signal in the left OPT between the first and final scan in the RS group and between the second and final scan in the HC group, while the NR group was characterized by decreased BOLD signal during that same time frame. There was also a statistically significant increase in BOLD signal in the medial temporal lobe (BA 37) in the HC between the first and final scan, whereas, the opposite pattern was again identified in the NR (Figure 21).

Table 19 – Group by Time Interaction in BOLD Signal Evoked by Positive Affective Visual Stimuli in (Group x Time)

Gyrus/Region BA Side F-Score Size MNI TAL (voxels) X y z x y z ↑RS, ↑HC, ↓NR Inf Temporal 37 L 7.84 195 -58 -58 -10 -57 -57 -6 ↑HC, ↓NR Fusiform 37 L6.70 110 -36 -48 -18 -36 -47 -13

In summary, there were limited changes associated with repeat scans and the induction of positive affect between the baseline and final visit in the study participants. In the HC group, there were moderate increases in BOLD signal in the left OPT (BA 19/37) and in the substantia nigra. These time-dependent changes can be modeled as BOLD signal amplification in response to repeat presentation of affective visual stimuli in psychiatrically unaffected study participants. In the RS group, there was also a time-dependent increase in the left OPT (BA 19/37), along with a decrease in BOLD signal in the right dorsolateral PFC (BA 9). Conversely, in the NR group who did not experience significant therapeutic benefit, we found a decrease in BOLD signal in the left OPT (BA 37).

148 Figure 21 – Inter-Scan Changes in BOLD Signal Evoked by Repeat Presentation of Positive Affective Visual Stimuli – Between Group Differences

LEGEND: Changes in BOLD signal meeting height threshold p<0.05 (FDR-corrected), and extent threshold k>15 voxels (120 mm3) overlay on representative structural MRI image, with approximate MNI z-axis coordinate in bottom right corner. Green- yellow colour flare represents F-score of a statistical test evaluating the interaction between main effect of time and main effect of group in BOLD signal from Table 19 .The colour-coded asterisk is placed midway between the two time-points where a Bonferroni post-hoc test identified statistically significant changes in BOLD signal..

149 Evaluation of inter-scan changes in intra-block habituation revealed a statistically significant effect in the sACC of the RS group. At the baseline visit, there was a robust activation during the first (early) half of the positive block that subsequently decayed in the second (late) half. With successive scans, there was a decrease in BOLD signal in the early component, and an overall decrease in the difference between the early and late components.

Repeat Induction of Negative Affect – Inter Scan Changes in BOLD Signal

Comparison of inter-scan changes during the induction of negative affect in HC revealed a number of cortical clusters with decreased BOLD signal between the first and final scan (Figure 22, Table 20). Decreased BOLD signal was identified in the frontal and limbic lobes in the dorsomedial PFC (BA 10), and a cluster encompassing the rCMA and dCMA (32), and the vPCC (BA 23/30). Additional decreases were also noted in the medial temporal lobe (BA 30), the precuneus (BA7), and the secondary somatosensory cortex (BA 5/40).

Table 20 – Inter-Scan Changes in BOLD Signal Evoked by Repeat Presentation of Negative Affective Visual Stimuli in HC

Gyrus/Region BA Side F-Score Size MNI TAL (voxels) X y z x y z DECREASES Med Frontal 32 18.90 2311 -2 18 40 -2 19 36 Parahippocampal 30 14.61 1076 -8 -48 2 -8 -46 4 Postcentral 5 13.30 328 -38 -20 46 -38 -17 43 Mid Frontal 10 13.06 110 20 60 22 20 59 17 Vent Post Cingulate 23/30 12.18 127 20 -58 6 20 -56 8 Inf Parietal Lobule 40 11.84 145 56 -32 26 55 -30 25

150 Figure 22 – Inter-Scan Changes in BOLD Signal Evoked by Repeat Presentation of Negative Affective Visual Stimuli in HC

LEGEND: Changes in BOLD signal meeting height threshold p<0.05 (FDR-corrected), and extent threshold k>15 voxels (120 mm3) overlay on representative structural MRI image, with approximate MNI z-axis coordinate in bottom right corner. Green-yellow colour flare represents F-score of a statistical test evaluating the main effect of time in BOLD signal in the HC group from Table 21.The asterisk is placed midway between the two time-points where a Bonferroni post-hoc test identified statistically significant changes in BOLD signal.

151 Inter-scan changes in intra-scan habituation of the BOLD signal were also identified between the first and final scans in the HC group in the pACC (BA 32) and in the precuneus (BA 7) (Figure 23, Table 21). In the first two scans, the maximum activation in these regions was achieved in the second half (late) of the negative block. During the final scan, however, the maximum activation was identified in the early component (Figure 21).

Table 21 – Inter-Scan Changes in Intra-Scan BOLD Signal Habituation Evoked by Negative Affective Visual Stimuli in HC

Gyrus/Region BA Side F-Score Size MNI TAL (voxels) x y z x y z DECREASES Pregenual Cingulate 32 L 9.75 29 -18 42 -2 -18 41 -4 Precuneus 7 R 8.56 125 2 -68 38 2 -64 38

Following the fist baseline scan, both MDD groups began antidepressant treatment. Repeat induction of negative affect at follow-up visits evoked statistically significant differences in BOLD signal (Figure 24, Table 22). Comparable to the HC group, widespread decreases in BOLD signal were localized in the dorsomedial PFC (BA 9), and a cluster encompassing the rCMA and dCMA (32), preSMA, and SMA and in the precuneus (BA 7). Other time-dependent changes that were not observed in the HC group included decreased BOLD signal in the anterior nuclei of the thalamus, the ventromedial PFC (BA 10), and the left dorsolateral PFC (BA 8).

152 Figure 23 – Inter-Scan Changes in BOLD Signal Evoked by Repeat Presentation of Positive Affective Visual Stimuli in NR and Inter-Scan Changes in Intra-Scan BOLD Signal Habituation (Time x Intra-Scan Habituation)

LEGEND: Changes in BOLD signal meeting height threshold p<0.05 (FDR-corrected), and extent threshold k>15 voxels (120 mm3) overlay on representative structural MRI image, with approximate MNI z-axis coordinate in bottom right corner. Pink-yellow colour flare represents F-score of a statistical test evaluating the interaction between time and intra-scan habituation in BOLD signal in the HC group from Table 21, orange-yellow colour flare for the NR group from Table 25, and the green-yellow colour flare for the RS group from Table 23.

153 Table 22 – Inter-Scan Changes in BOLD Signal Evoked by Repeat Presentation of Negative Affective Visual Stimuli in RS

Gyrus/Region BA Side F-Score Size MNI TAL (voxels) X y z x y z DECREASES Angular 39 L12.66 411 -46 -76 34 -46 -72 35 Ant Nuclei Thalamus - - 12.31 578 -4 -10 18 -4 -9 17 Mid/Sup Frontal 8 L 10.74 310 -28 24 54 -28 26 48 Med Frontal 9 - 10.65 821 -4 48 26 -4 48 22 Precuneus 7 - 10.17 550 -4 -60 32 -4 -57 32 Angular 39 R9.90 171 54 -66 34 53 -62 34 Med Frontal 6/32 - 8.26 158 2 12 50 2 14 45 Med Frontal 10 - 7.94 184 0 52 -8 0 50 -9

Inter-scan changes in intra-scan habituation of the BOLD signal in the RS group were identified in the right putamen and the left anterior insula. Before treatment initiation, maximal activation in the anterior insula was identified in the late component of the negative block, after the first scan, however, maximum activation shifted to early component. A similar pattern was discovered in the right putamen, however, the switch from late to early, was only transient, reversing back by the time of the final scan.

Table 23 – Inter-Scan Changes in Intra-Scan BOLD Signal Habituation Evoked by Negative Affective Visual Stimuli in RS

Gyrus/Region BA Side F-Score Size MNI TAL (mm3) x y z x y z Putamen - R 6.84 27 16 8 10 16 8 9 Ant Insula - L 6.67 20 -34 20 10 -34 20 8

154 Figure 24 – Inter-Scan Changes in BOLD Signal Evoked by Repeat Presentation of Negative Affective Visual Stimuli in RS

LEGEND: Changes in BOLD signal meeting height threshold p<0.05 (FDR-corrected), and extent threshold k>15 voxels (120 mm3) overlay on representative structural MRI image, with approximate MNI z-axis coordinate in bottom right corner. Green-yellow colour flare represents F-score of a statistical test evaluating the main effect of time in BOLD signal in the RS group from Table 22.The asterisk is placed midway between the two time-points where a Bonferroni post-hoc test identified statistically significant changes in BOLD signal.

155 Following the fist baseline scan, the NR groups also began antidepressant treatment, unfortunately, with significantly less therapeutic benefit than the RS group. Repeat induction of negative affect at follow-up visit also evoked differences in BOLD signal (Figure 25, Table 24). Comparable to the RS and HC group, decreases in BOLD signal were localized in the aMCC (BA 24), however, the cluster that met statistical significance did not encompass the surrounding CMAs. Other time-dependent changes that were not observed in either the HC or RS groups included decreased BOLD signal in the right ventrolateral and dorsolateral PFC (BA 9/46) and the right anterior insula. Inter-scan changes in intra-scan habituation were limited to the ventromedial PFC (BA 10), which was characterized by increased activation between the early and late components during the first two visits, before reversing at the time of the final scan (Figure 21, Table 25).

Table 24 – Inter-Scan Changes in BOLD Signal Evoked by Repeat Presentation of Negative Affective Visual Stimuli in NR

Gyrus/Region BA Side F-Score Size MNI TAL (mm3) X y z x y z DECREASES Inf Frontal; 46 R16.17 1154 46 46 16 46 45 12 Ant Insula - R 12.57 97 46 18 -2 46 17 -3 Mid Frontal 9/46 L 10.26 43 -48 36 26 -48 36 22 Ant Mid Cingulate 24 L 8.27 108 -4 16 30 -4 17 27

Table 25 – Inter-Scan Changes in Intra-Scan BOLD Signal Habituation Evoked by Negative Affective Visual Stimuli in NR

Gyrus/Region BA Side F-Score Size MNI TAL (mm3) x y z x y z Med Frontal 10 R 7.37 23 16 44 16 16 44 13

156 Figure 25 – Inter-Scan Changes in BOLD Signal Evoked by Repeat Presentation of Negative Affective Visual Stimuli in NR

LEGEND: Changes in BOLD signal meeting height threshold p<0.05 (FDR-corrected), and extent threshold k>15 voxels (120 mm3) overlay on representative structural MRI image, with approximate MNI z-axis coordinate in bottom right corner. Green-yellow colour flare represents F-score of a statistical test evaluating the main effect of time in BOLD signal in the RS group from Table 22.The asterisk is placed midway between the two time-points where a Bonferroni post-hoc test identified statistically significant changes in BOLD signal.

157 In order to identify statistically significant between-group-difference in inter-scan changes in BOLD signal, a group x time interaction contrast was evaluated (Table 26). In the prefrontal lobe, the RS group was characterized by decreased BOLD signal in the ventromedial (BA 10/11) and ventrolateral (BA 45), while the opposite pattern was seen in NR, with increased BOLD signal between the first and final scan (Figure 21). The exception to this was in the right dorsolateral PFC (BA 9/46) where decreased BOLD signal was identified between the first and last scan. Analysis of time-dependent changes in BOLD-signal also identified a transient increase in the anterior nuclei of the thalamus between the first and second scan. This, increase, however, returned to baseline levels by the time of the final scan.

Table 26 – Group by Time Interaction in BOLD Signal Evoked by Negative Affective Visual Stimuli in (Group x Time)

Gyrus/Region BA Side F-Score Size MNI TAL (voxels) X y z x y z ↑ NR ↓ RS Mid/Med Frontal 10 R 14.44 77 26 42 -6 26 42 -6 Med Frontal 10/11 - 8.53 89 -2 56 -12 -1 50 -10 Inf Frontal 45 R 6.1 37 60 28 16 58 27 16 ↓ NR Mid Frontal 46 R 6.56 436 46 44 18 46 44 18 ↑HC↓NR Ant Nuclei Thalamus - R 5.39 34 4 4 12 1 2 10

In summary, decreases in BOLD signal were associated with repeat scans between the baseline and final visit in the study participants. In the HC and RS groups, decreased BOLD signal was located in midline cortical structures, in the dorsal frontal, limbic, and parietal lobes extending into the supplementary motor areas. Differential activity between RS and NR was identified in the ventromedial and ventrolateral PFC, where decreases in BOLD signal were unique to the RS group, while the NR group was characterized by decreases in the right dorsolateral PFC.

158 Figure 26 – Inter-Scan Changes in BOLD Signal Evoked by Repeat Presentation of Negative Affective Visual Stimuli – Between Group Differences

LEGEND: Changes in BOLD signal meeting height threshold p<0.05 (FDR-corrected), and extent threshold k>15 voxels (120 mm3) overlay on representative structural MRI image, with approximate MNI z-axis coordinate in bottom right corner. Green-yellow colour flare represents F-score of a statistical test evaluating the interaction between main effect of time and main effect of group in BOLD signal from Table 19 .The colour-coded asterisk is placed midway between the two time-points where a Bonferroni post-hoc test identified statistically significant changes in BOLD signal.

159 SECONDARY AIM A:

EVALUATE THE ASSOCIATION BETWEEN DEPRESSIVE SYMPTOM REDUCTION AND BRAIN ACTIVATION EVOKED BY REPEAT AFFECTIVE PROCESSING IN MDD

Influence of Depressive Symptoms Severity on Positive Affective Processing

To evaluate the effect of depressive symptom severity on the BOLD signal evoked by positive affective processing, a correlative analysis was constructed between self-rated (BDI) and clinician-rated (HRSD-17) severity of depressive symptoms (Table 27). The two MDD groups were combined to investigate changes in BOLD signal irrespective of ultimate responder status, and to control for the effect of inter-scan habituation.

Table 27 – Correlations between BOLD Signal Evoked by Positive Affective Visual Stimuli and Measures of Depressive Symptom Severity in MDD

R Gyrus/Region BA Side F-Score Size MNI TAL (voxels) x y z x y z (-) HRSD-17 -0.482 Putamen - R 11.61 29 22 -10 8 22 -9 8 -0.446 Inf Frontal 44 R 11.43 25 42 10 28 42 11 25 -0.352 Ant Insula - R 10.07 17 -30 14 -4 -30 13 -4 -0.305 Ant Insula - L 8.92 18 32 24 -4 32 23 -5 (+) HRSD-17 0.409 Precentral 6 R 20.34 38 -54 0 44 -53 2 40 (-) BDI -0.540 Fusiform 20/37 R 17.14 242 32 -42 -16 32 -41 -11 -0.484 Mid Frontal 8 R 14.29 94 44 22 52 44 24 47 (+) BDI 0.428 Post Insula - L 14.21 53 -40 -34 22 -40 -32 22 0.327 Post Insula - R 13.40 41 40 -24 24 40 -22 23

LEGEND: Correlations in BOLD signal meeting height threshold p<0.05 (FDR-corrected), and extent threshold k>15 voxels (120 mm3). Anatomical localization based on peak voxel coordinates in Talairach space. Cluster size corresponds to number of (8 mm3) voxels.

160

Decrease in clinician-rated severity of depressive symptoms was associated with increased activation bilaterally in the anterior insula, the right putamen, and in Broca’s area (Figure 27). Decreases in self-rated depressive symptom severity were associated with increased activation in the medial temporal lobe (BA 20/37) and in the dorsolateral PFC (BA 8). Increases in depressive symptom severity, on the other hand, were associated with increased activation in the posterior insula and the premotor cortex (BA 6). Discordance in brain region – depressive symptom severity scale correlations are likely due to the different psychometric properties between clinican-rated and self-rated psychiatric screening instruments. The clinical rated HRSD-17 may be more sensitive to pharmacotherapeutic antidepressant effects of the olanzapine–fluoxetine combination, while the BDI may reflect changes in depressive symptom severity sensitive to psychosocial interventions (Demyttenaere and De Fruyt, 2003).

An alternative method of evaluating the effects of depressive symptoms reduction on brain activation during affective processing is to perform an analysis on MDD subjects who meet the clinical criteria of depressive symptom remission. At the time of the final scan, ten of the 21 MDD subjects who completed at least one neuroimaging session were categorized as remitters (RM; subjects who scored < 7 on the HDRS-17 during the final visit). The clinical measures obtained during the final scanning visit were compared to values obtained from HC during the final scan (Table 28). By the end of the study, RM still had higher self-rated depressive symptoms, and less positive dispositional affect than HC (p=0.005). There was no statistically significant between-group difference between RM and HC in the amount of general negative affect reported.

161 Figure 27 – Correlations between BOLD Signal Evoked by Positive Affective Visual Stimuli and Measures of Depressive Symptom Severity in MDD

LEGEND: Correlation between BOLD signal and clinical measure meeting height threshold p<0.05 (FDR-corrected), and extent threshold k>15 voxels (120 mm3) overlay on representative structural MRI image, with approximate MNI z-axis coordinate in bottom right corner. Hot colour flare represents F-score of a statistical test evaluating positive correlation between BOLD signal and clinical measure, while the cold colour flare represents a negative correlation.

162 Table 28 – Between Group Comparison of Study Participants’ Measures of Depression Severity, Dispositional Affect, and Induced Affect at the Final Visit

Scale HC RM t17 P APRS-0N 1.9 +/- 1.6 3.8 +/- 1.4 0.88 0.905 APRS-0P 0.1 +/- 1.1 1.1 +/- 1.1 0.63 0.539 APRS-N -14.0 +/- 1.5 -14.3 +/- 1.8 0.28 0.787 APRS-P 10.7 +/- 1.5 10.1 +/- 1.6 0.15 0.886 BDI 1.5 +/- 1.0 12.7 +/- 2.1 4.69 <0.001 PANAS-P 36.6 +/- 2.0 25.8 +/- 2.5 3.30 0.005 PANAS-N 14.0 +/- 1.8 17.3 +/- 1.9 1.27 0.224

Means are presented first, followed by +/- standard error of measurement. The final two columns represent the t value from a Student’s t-test evaluating difference between means and the resultant p-value.

A direct contrast between the BOLD signal evoked by RM and HC subjects during positive affective processing at the final visit revealed differences in brain activation and deactivation between the two groups even in the absence of clinically significant depressive symptoms (Figure 28, Table 29). Decreased activation was identified in the RM group in the medial temporal lobe in the fusiform gyrus (BA 37) and in the extrastriate visual cortex (BA 19). Moreover, the two groups demonstrated opposite changes in BOLD signal in the ventromedial PFC (BA 11). Whereas, statistically significant deactivation was noted in HC, mild activation was identified in RM.

Table 29 – Between Group Differences in BOLD Signal Evoked by Positive Affective Stimuli during the Final Visit

Gyrus/Region BA Side F-Score Size MNI TAL (voxels) x y z x y z RM > HC Sup/Med Frontal 11 L 17.98 80 -20 42 -10 -20 40 -10 DECREASES Mid Occipital 19 17.10 -40 -70 -6 -40 -70 -6 Fusiform 37 L15.58 34 -28 -58 -10 -21 -45 -8

LEGEND: RM – Remitted MDD subject, HC – Healthy control. Changes in BOLD signal meeting height threshold p<0.05 (FDR-corrected), and extent threshold k>15 voxels (120 mm3). Anatomical localization based on peak voxel coordinates in Talairach space. Cluster size corresponds to number of (8 mm3) voxels. 163 In summary, decreases in depressive symptom severity were associated with increased BOLD signal during positive affect processing in the anterior insula, putamen, and activation in the right dorsolateral PFC (BA 8/44), and decreased BOLD signal in the posterior insula and the premotor cortex. In the absence of clinically significant depressive symptoms, RM were characterized by decreased activation in the OPT cortex, and activation, versus, deactivation in the ventromedial PFC compared to psychiatrically unaffected control subjects who underwent repeat induction of positive affect.

164 Figure 28 – Main Effect of Group: Between Group Differences in BOLD Signal Evoked by Positive Affective Stimuli during the Final Visit

LEGEND: HC – healthy control subject, RM – remitted MDD subject. Differences in BOLD signal meeting height threshold p<0.05 (FDR-corrected), and extent threshold k>15 voxels (120 mm3) overlay on representative structural MRI image, with approximate MNI z-axis coordinate in bottom right corner. Colour flare represents F-score of a statistical test evaluating the effect of group in BOLD signal during the presentation of the positive affective visual stimuli from Table 29.

165 Influence of Depressive Symptoms Severity on Negative Affective Processing

To evaluate the effect of depressive symptom severity on the brain activation pattern evoked by negative affective processing, a correlative analysis was also constructed between self-rated and clinician-rated indices of depressive symptoms severity and the BOLD signal (Figure 29, Table 30). As was done for the positive affective processing run, the two MDD groups were again combined to control for the effect of inter-scan habituation.

Table 30 – Correlations between BOLD Signal Evoked by Negative Affective Visual Stimuli and Measures of Depressive Symptom Severity in MDD

R Gyrus/Region BA Side F-Score Size MNI TAL (voxels) x y z x y z (+) HRSD-17 0.379 Med Frontal 10 L 13.69 44 -12 60 0 -12 59 0 (-) HRSD-17 - -0.399 VL Nuclei Thalamus - L 9.25 64 -16 -8 2 -17 -6 2 (-) BDI -0.485 Mid Frontal 10 L 21.44 22 -34 46 -10 -28 43 -8

Confirming our finding in the group x time interaction analysis, decrease in clinician-rated severity of depressive symptoms was associated with decreased activation in the ventromedial PFC (BA 10). Conversely, increased activation in the ventrolateral PFC (BA 10) was associated with decreases in self-rated depression symptom severity. Increases in ventral lateral nuclei of the thalamus were also associated with decreased clinician-rated depressive symptom severity. This finding also parallels our results obtained from the group x time interaction analysis, where a transient increase in BOLD signal was recorded during the second scanning session (where sharp decreases in HRSD-17 scale were also recorded).

166 Figure 29 – Correlations between BOLD Signal Evoked by Negative Affective Visual Stimuli and Measures of Depressive Symptom Severity in MDD

167 A direct contrast between the BOLD signal evoked by RM and HC subjects during negative affective processing at the final visit revealed differences in brain activation and deactivation between the two groups even in the absence of clinically significant depressive symptoms (Figure 30, Table 31). Activation was identified in the RM group bilaterally in the premotor cortex (BA 6) and in the right dorsolateral PFC (BA 45/46). Activation was also identified in the extrastriate occipital cortex (BA 18/19) and in the medial temporal lobe (BA 37). Conversely, deactivation in these same brain regions was identified in HC.

Table 31 – Between Group Differences in BOLD Signal Evoked by Negative Affective Stimuli during the Final Visit

Gyrus/Region BA Side F-Score Size MNI TAL (voxels) x y z x y z Inf/Mid Frontal 45/46 R 17.49 72 46 28 18 46 28 15 Inf Parietal Lobule 40 R 17.25 50 58 -38 34 57 -35 33 Mid Occipital 18/19 L 16.88 70 -42 -86 10 -42 -83 13 Mid Temporal 19/39 R 16.51 91 30 -70 12 30 -67 14 Precentral 6 R 15.73 45 58 12 8 57 12 7 Precentral 6 L 14.94 40 -54 6 8 -53 6 7 Mid Occipital 19 R 14.65 38 36 -84 20 36 -80 22 Inf Parietal Lobule 40 R 13.63 41 48 -26 32 48 -24 31 Fusiform 19/37 L 12.52 27 -34 -52 -14 -34 -51 -9

In summary, decreases in depressive symptom severity were associated with decreased BOLD signal during negative affect processing in the ventromedial PFC (BA 10). In the absence of clinically significant depressive symptoms, RM were characterized by activation in the premotor and extrastriate occipital cortex, in the medial temporal lobe, and the right dorsolateral PFC (BA 9), compared to HC in whom BOLD signal deactivations were identified in these regions.

168 Figure 30 – Main Effect of Group: Between Group Differences in BOLD Signal Evoked by Negative Affective Stimuli during the Final Visit

LEGEND: HC – healthy control, RM – remitted MDD subject. Differences in BOLD signal meeting height threshold p<0.05 (FDR-corrected), and extent threshold k>15 voxels (120 mm3) overlay on representative structural MRI image, with approximate MNI z-axis coordinate in bottom right corner. Colour flare represents F-score of a statistical test evaluating the effect of group in BOLD signal during the presentation of the positive affective visual stimuli from Table 31.

169 SECONDARY AIM B:

EVALUATE THE ASSOCIATION BETWEEN DISPOSITIONAL AND INDUCED AFFECT AND BRAIN ACTIVATION EVOKED BY AFFECTIVE PROCESSING

Influence of Dispositional Affect on Positive Affective Processing

Between the first and final scan MDD subjects experienced increases in positive affect and decreases in negative dispositional affect in addition to decreases in depressive symptom severity. To evaluate the association between dispositional affect normalization and brain activation during positive affective processing in MDD, a correlative analysis was constructed between positive dispositional affect (PANAS-P & PANAS-N) and the BOLD signal (Figure 31, Table 32). The two MDD groups were again combined to investigate changes in BOLD signal irrespective of ultimate responder status, and to control for the effect of inter-scan habituation.

Table 32 – Correlations between BOLD Signal Evoked by Positive Affective Visual Stimuli and Measures of Dispositional Affect in MDD

r Gyrus/Region BA Side F-Score Size MNI TAL (voxels) x y z x y z (-) PANAS-P -0.395 Inf Occipital 18 L 26.01 41 -40 -80 -10 -40 -78 -5 (+) PANAS-P 0.492 Mid Temporal 39 L 13.61 78 -46 -82 20 -45 -79 22 0.367 Hippocampus - R 10.11 18 36 -34 -6 36 -33 -3 0.363 Hippocampus - R 10.11 43 28 -18 -12 28 -18 -9 (-) PANAS-N -0.504 Mid Occipital 19 R 13.03 87 42 -78 10 42 -75 13 -0.455 Parahippocampal 30 R 12.96 24 12 -42 2 12 -41 4 (+) PANAS-N 0.446 Postcentral 2 L 13.76 31 -62 -16 24 -62 -14 23

170 Increases in positive dispositional affect (PANAS-P) were associated with increased BOLD signal in the head and tail of the right hippocampus and the left OPT (BA 39). Intriguingly, increases in positive dispositional affect were also associated with decreased BOLD signal in the extrastriate visual cortex (BA 18), suggesting that increases in positive affect were associated with a switch from the ventral visual pathway to the dorsal visual pathway.

By the time of the final scan, the majority of the RS group met criteria for remission, and reported levels of negative dispositional affect that were comparable to HC. Decreases in negative dispositional affect (PANAS-N) were associated with increased BOLD signal evoked during positive affective processing in the right OPT (BA 19) and medial temporal lobe (BA 30). Decreases in negative affect were also associated with decreased signal in the premotor cortex.

Influence of Dispositional Affect on Negative Affective Processing

Decreases in negative dispositional affect were associated with widespread decreases in BOLD signal evoked by negative affective processing. There was a positive correlation between the PANAS-N score and BOLD signal in the right amygdala, the right anterior insula, right putamen, left ventrolateral PFC (BA 47), and right dorsolateral PFC (BA 9/46) (Figure 32, Table 33). Decreases in PANAS-N were also associated with increased BOLD signal in the somatosensory association cortex (BA 5) and the left temporal pole.

Increases in positive dispositional affect were also associated with increased BOLD signal in the right ventrolateral PFC (BA 10/44) and right putamen, and decreased BOLD signal in the left caudate tail. As the PANAS-P subscale evaluates interest, excitement, determination, and alertness, and attention, these positive correlation between PANAS-P and BOLD signal evoked during negative affective processing may reflect increases attention to the presented visual stimuli.

171 Figure 31 – Correlations between BOLD Signal Evoked by Positive Affective Visual Stimuli and Measures of Dispositional Affect in MDD

172 Table 33 – Correlations between BOLD Signal Evoked by Negative Affective Visual Stimuli and Measures of Dispositional Affect in MDD

r Gyrus/Region BA Side F-Score Size MNI TAL (voxels) x y z x y z PANAS-P (+) 0.605 Inf Frontal 44 R 18.91 142 48 12 16 48 12 14 0.549 Inf Frontal 45 R 16.08 74 52 26 4 51 25 2 0.539 Putamen - R 13.47 109 30 -4 6 30 -4 6 0.523 Inf Frontal 10 R 12.39 61 46 44 0 46 43 -2 0.377 Inf Parietal Lobule 40 R 11.16 40 62 -32 42 61 -29 40 PANAS-P(-) -0.374 Caudate tail - L 10.30 101 -22 -38 16 -22 -36 17 PANAS-N(+) 0.430 Putamen - R 12.96 68 34 6 -2 34 6 -2 0.479 Mid Frontal 9/46 R 10.50 73 38 46 26 38 46 22 0.417 Inf Frontal 47 L 10.37 50 -34 28 -8 -34 27 -8 0.491 Sup Frontal 8 L 10.37 31 -16 24 60 -16 26 54 0.481 Amygdala - R 10.18 44 24 -6 -14 24 -6 -11 0.358 Ant Insula - R 7.95 54 48 16 4 48 16 3 PANAS-N(-) -0.391 Sup Parietal Lobule 5 L 14.52 173 -42 -42 64 -42 -38 61 -0.508 Inf Temporal 21 L 8.88 86 -48 -2 -32 -48 -3 -27

173 Figure 32 – Correlations between BOLD Signal Evoked by Negative Affective Visual Stimuli and Measures of Dispositional Affect in MDD

174 Correlation between Induced Affect and Positive Affective Processing

There were no statistically significant differences between self-rated levels of induced affect during both neutral-positive and neutral-negative runs, either within- or between-group (Table 3). A correlative analysis was prepared to evaluate the association between self-rated positive affect and BOLD signal evoked by positive affect processing (Figure 33, Table 34). In the MDD group there was a positive correlation between BOLD signal bilaterally in the ventrolateral PFC (BA 10/11/47), the left anterior insula, and the left premotor cortex (BA 6) and self-reported evoked positive affect. There was also a negative correlation between BOLD signal in the dPCC and the APRS-P.

Table 34 – Correlations between BOLD Signal Evoked by Positive Affective Visual Stimuli and Measures of Induced Affect in MDD While Viewing Positive Affective Visual Stimuli r Gyrus/Region BA Side F-Score Size MNI TAL voxels) x y z x y z (-) APRS-P -0.405 Dor Postcingulate 31 L 12.53 18 -14 -48 38 -14 -45 37 (+) APRS-P 0.520 Inf/Mid Frontal 11/47 16.89 45 48 44 -10 48 42 -11 0.621 Precentral/Mid Front 6 16.40 46 -48 2 52 -48 4 48 0.515 Mid Frontal 10 14.21 40 -34 58 14 -34 57 10 0.422 Ant Insula - 13.47 38 -42 0 2 -42 0 2 0.449 Mid Frontal 10 13.32 32 42 52 16 42 51 12

The amount of evoked positive affect in the HC group was also associated with increased BOLD signal in the premotor cortex (Table 35). In addition, increased BOLD signal in the left putamen, dorsomedial PFC (BA 9), right and left OPT (BA 22/37) and somatosensory association cortex were associated with increased rating of evoked positive affect, suggesting that the evaluation of the stimuli was dependent on multiple sensory modalities.

175 Figure 33 – Correlations between BOLD Signal Evoked by Positive Affective Visual Stimuli and Measures of Induced Affect in MDD While Viewing Positive Affective Visual Stimuli

176 Table 35 – Correlations between BOLD Signal Evoked by Positive Affective Visual Stimuli and Measures of Induced Affect in HC While Viewing Positive Affective Visual Stimuli r Gyrus/Region BA Side F-Score Size MNI TAL (voxels) x y z x y z (-) APRS-P -0.537 Angular 39 R 15.76 25 42 -68 30 42 -65 31 -0.532 Post Insula - R 15.44 33 42 -40 32 42 -37 31 (+) APRS-P 0.525 Mid Temporal 37 R 14.82 54 58 -62 2 57 -60 5 0.514 Sup Temporal 22 L 13.99 90 -46 -50 14 -45 -48 15 0.503 Med Frontal 9 R 13.19 59 18 44 18 18 43 14 0.486 Precentral 6 L 12.04 77 -28 -4 48 -28 -2 44 0.471 Inf Parietal Lobule 40 L 11.09 40 -32 -60 46 -32 -56 45 0.451 Putamen - L 9.96 38 -12 10 4 -12 10 3

Correlation between Induced Affect and Negative Affective Processing

In the MDD group, increased rating of evoked negative affect were associated with increased BOLD signal in head and tail of the caudate, the left OPT (BA 19/37), and the inferior medial temporal lobe (BA 36).

Table 36 – Correlations between BOLD Signal Evoked by Negative Affective Visual Stimuli and Measures of Induced Affect in MDD While Viewing Negative Affective Visual Stimuli r Gyrus/Region BA Side F-Score Size MNI TAL (voxels) x y z x y z APRS-N 0.416 Parahippocampal 36 L 17.06 49 -42 -20 -24 -42 -20 -19 0.410 Inf Temporal 19/37 R 16.56 30 44 -52 -8 44 -51 -4 0.462 Caudate Body - L 15.76 21 -16 -18 28 -16 -16 27 0.463 Mid Occipital 19/37 R 14.98 78 38 -66 4 38 -64 7 0.506 Caudate Head - L 14.29 35 -12 20 16 -12 20 14 0.479 Caudate Tail - R 13.84 43 16 -32 20 16 -30 20

177 Figure 34 – Correlations between BOLD Signal Evoked by Negative Affective Visual Stimuli and Measures of Induced Affect in MDD While Viewing Negative Affective Visual Stimuli

178 By comparison, rating of evoked negative affect by HC were not associated with activation in the basal ganglia or the OPT cortex (Table 37). On the contrary, ratings of evoked negative affect were associated with increased BOLD signal in the left anterior, and right posterior, insula, the left ventrolateral (BA 11) and dorsolateral (BA 9/44) PFC, bilaterally in the SMA, and in the precuneus (BA 7). These finding again suggest that HC and RS employ different brain regions in the evaluation of the own affect.

Table 37 – Correlations between BOLD Signal Evoked by Negative Affective Visual Stimuli and Measures of Induced Affect in HC While Viewing Negative Affective Visual Stimuli

R Gyrus/Region BA Side F-Score Size MNI TAL (voxels) x y z x y z

(+) APRS-B 0.5454 Ant Insula - L 16.89 123 -38 -16 4 -38 -15 4 0.522 Post Insula - R 14.99 66 42 -18 8 42 -17 8 0.495 Mid Frontal 11 L 12.96 23 -38 50 -14 -38 48 -14 0.495 Inf Frontal 9/44 L 12.95 140 -42 18 30 -42 19 27 0.494 Paracentral Lobule 7 - 12.93 100 0 -30 56 0 -26 53 0.484 Med Frontal 6 R 12.25 37 10 2 60 10 5 55 0.480 Tran Temporal 41 R 12.01 45 42 -34 12 42 -32 13 0.471 Med Frontal 6 R 11.78 20 8 -14 54 8 -11 50 0.466 Cuneus 19 L 11.09 36 -14 -88 26 -14 -84 28 0.465 Postcentral 40 R 11.04 21 48 -24 24 48 -22 23

179 In summary, increases in positive dispositional affect were associated with increased BOLD signal in the right hippocampus during positive affective processing, and increased BOLD signal in the right ventrolateral PFC (BA 10) and right putamen during negative affect processing. Decreases in negative dispositional affect were associated with increased activation in the right hippocampus during positive affective processing, and decreased activation in the right amygdala, insula, putamen, dorsolateral PFC (9/46), and left ventrolateral PFC (47) during negative affective processing. In the MDD group, ratings of induced positive affect were associated with increased BOLD signal in the ventrolateral PFC (BA 10/11/47) and left insula, while rating of induced negative affect were associated with increased BOLD signal in the caudate and OPT.

180 DISCUSSION

Summary of Findings

The presented clinical and neuroimaging results confirmed our hypothesis of differential brain activation evoked by affective processing. Study participants reported increases in positive affect during PAI and increases in negative affect during NAI with increased activation in most of the brain regions proposed to regulate the identification of emotional salience of a stimulus including bilateral amygdala, ventral PFC, ventral striatum, and anterior insula (Table 38). Furthermore, we also observed activation of brain regions involved in the regulation of the affective state during NAI, including the dorsomedial and dorsolateral PFC and the aMCC (Table 40).

Consistent with our hypothesis, PAI was characterized by decreased BOLD signal in the NR group in the OPT cortex and delayed activation in the ventral striatum. Contrary to our hypothesis, there were no statistically significant difference in brain activation between HC and RS (Table 39). Consistent with our hypothesis, NAI was characterized by increased activation in the MDD group in the dorsolateral PFC, and increased activation by NR compared to the RS was noted in the pACC and ventrolateral PFC. Delayed times to maximum activation were noted in RS in dorsolateral PFC (Table 41).

Repeat PAI was associated with increased activation in brain regions responsible for evaluation of the presented stimuli, in the OPT cortex and production of the affective state, in the substantia nigra. Inter-scan changes in RS, however, were limited to the OPT cortex, with decreased activation characterizing the NR group. Repeat NAI were associated with decreased activation in located in midline cortical structures, in the dorsal frontal, limbic, and parietal lobes extending into the supplementary motor areas in HC and RS. Decreases in the ventromedial and ventrolateral PFC differentiated RS from NR.

Decreases in depressive symptom severity were associated with increased activation in brain regions involved in the production of the affective state, with increased BOLD signal during PAI in the anterior insula, putamen. Decreases in depressive symptom severity were also associated with decreased brain activation in the ventromedial PFC during NAI. In the absence of clinically significant depressive symptoms, RM could be distinguished from HC based on their brain

181 activation during NAI with increased activation of brain regions involved in the regulation of the affective state, including the dorsolateral PFC. In spite of a clinical remission status, RM continued to manifest increased brain activation in the extrastriate visual cortex during NAI and decreased activation during PAI compared to HC (Table 42).

The remaining differences in BOLD activation between RM and HC during PAI and NAI may have been the result of decreased positive dispositional affect, which was significantly lower in the RM group. Indeed, increases in positive dispositional affect were associated with increased BOLD signal in the hippocampus during PAI in the MDD group. Furthermore, reductions in negative dispositional affect were also associated with decreased brain activation in the amygdala, putamen, anterior insula and dorsolateral PFC. These results suggest that measurement of dispositional affect may be more accurate in predicting MDD patients’ vulnerability to relapse rather than depressive symptom severity.

The results of the analyses confirmed the a-prior hypotheses and additionally demonstrated differential activation of the insula, medial temporal, and premotor cortex during repeat PAI and NAI between HC, RS, and NR. These findings provide: i) a neuroanatomical target of successful antidepressant therapy during PAI/NAI; ii) a differential effect of depressive symptoms and dispositional affect on brain activation during PAI/NAI; and iii) an a-prior method to differentiate RS from NR, and iv) demonstrate the need for additional treatment to prevent relapse in the remitted state.

182 Table 38 – Summary of Findings for Positive Affect – All Participants

Contrast All Study Participants

Activation Deactivation

ventromedial PFC (10) ventrolateral PFC (11/46) dorsomedial PFC (10) dorsolateral PFC (9) L hippocampus aMCC (24) Positive L:& R OPT (19/37) ant insula Affect L & R Temporal Pole (21/38) lateral temporal (21/22) Processing medial temporal (35/37) thalamus – lateral dorsal precuneus (7) premotor cortex (6) hypothalamus – Cm cerebellum – hemisphere cerebellum – vermis

Decreases (Early – Late) Increases(Early – Late)

Positive L & R OPT (19/37) R ventrolateral PFC (45/46) Intra-Scan L ventrolateral PFC (10) cerebellum - hemisphere Change precuneus (7) superior parietal lobule (7)

Table 39 – Summary of Findings for Positive Affect – Between Group Comparisons

Contrast HC vs RS vs HC vs MDD NR RM ↑ vmPFC (10) ↓ vmPFC (11) ↑ dmPFC (9/10) ↑ med temporal Positive ↑ R dlPFC (9) ↑ L OPT Affect ↑ dPCC (31) Processing ↑ precuneus (7) at Baseline

↑ L OPT (19/37) ↑ med temporal (37)

Positive

Inter-Scan

Change

LEGEND: Increased and decreased BOLD activation is indicated with up and down arrows, respectively. aMCC – anterior midcingulate cortex, ant – anterior, dm – dorsomedial , dPCC – dorsal posterior cingulate cortex, med – medial, OPT –occipitotemporal, PFC – prefrontal cortex, vm – ventromedial 183 Table 40 – Summary of Findings for Negative Affect – All Participants

Contrast All Study Participants

Activation Deactivation

dorsomedial PFC (10) L & R Posterior Insula aMCC (24) dPCC (31) thalamus Negative L & R amygdala Affect hippocampus Processing L:& R OPT (19/37) L & R temporal pole (20/38) L & R parietal lobule (7/40) precuneus (7) cerebellum – vermis

Decreases (Early – Late) Increases(Early – Late)

Negative substantia nigra pACC (24/32) Intra-Scan L extrastriate visual cortex (18) precuneus (7) Change L & R sup parietal lobule (7)

Table 41 – Summary of Findings for Negative Affect – Between Group Comparisons

Contrast HC vs RS vs HC vs MDD NR RM ↓ L lat temporal (21) ↓ pACC (24) ↓ R vlPFC (45/46) ↓ R vlPFC (10/46) ↓ R dlPFC (10/46) ↓ L & R OPT (19/39) Negative ↓ L dlPFC (9) ↓ L dlPFC (9) ↓ L & R premotor (6) Affect ↑ L OPT (39) ↓ L med temporal (37) Processing ↑ vmPFC (11) at Baseline

↓ R vlPFC (10) ↓ vmPFC (10/11) Negative Inter-Scan Change

LEGEND: Increased and decreased BOLD activation is indicated with up and down arrows, respectively. aMCC – anterior midcingulate cortex, ant – anterior, dm – dorsomedial , dPCC – dorsal posterior cingulate cortex, med – medial, OPT –occipitotemporal, pACC – pregenual anterior cingulate cortex, PFC – prefrontal cortex, vm – ventromedial, vl – ventrolateral 184 Table 42 – Summary of Clinical-Neuroimaging Correlative Analyses in MDD

Contrast PANAS-P PANAS-N HRSD-17 BDI

(+) Correlation  R OPT (39)  L somatosensory (2)  R premotor (6)  R & L post insula with Positive  R hippocampus Affect Processing BOLD Signal

(-) Correlation  L extrastriate visual  R OPT (19)  R putamen  R med temporal with Positive (18)  R parahippocampal  L & R ant insula (20/37) Affect (30)  R dlPFC (44)  R dlPFC (8) Processing BOLD Signal (+) Correlation  R vlPFC (10/44/45)  R amygdala  vmPFC (10)  - with Negative  R putamen  R putamen Affect  R parietal lobule (40)  R dlPFC (9/46) Processing  L vlPFC (47) BOLD Signal  R ant insula

(-) Correlation  L caudate tail  L parietal lobule (5)  thalamus – vl nuclei  vlPFC (10) with Negative  L temporal pole (21) Affect Processing BOLD Signal

LEGEND: aMCC – anterior midcingulate cortex, ant – anterior, dm – dorsomedial , dPCC – dorsal posterior cingulate cortex, med – medial, OPT –occipitotemporal, pACC – pregenual anterior cingulate cortex, PFC – prefrontal cortex, vm – ventromedial, vl – ventrolateral

185 Table 43 – Summary of Induced Affect Correlations in HC and MDD

Contrast APRS-P APRS-N

(+) Correlation  R & L vlPFC (10/11/47) with Positive  L ant insula - Affect  L premotor (6) Processing  BOLD Signal in MDD (+) Correlation  R & L OPT (37) with Positive  dm PFC (9) Affect  L premotor (6) Processing  L inf parietal lobule (40) BOLD Signal in  L putamen - HC (+) Correlation  L parahippocampal (36) with Negative  R OPT (19/37) Affect  L & R caudate Processing BOLD Signal in MDD (-) Correlation  L ant insula with Negative  R post insula Affect  L vlPFC (11) Processing  L dlPFC (9/44) BOLD Signal in  Precuneus (7) HC  SMA (6)  Cuneus (19)

186 Convergence with Previous Investigations

Differences in Positive Affective Processing: Depressed vs. Control Subjects

The IAPS positive stimuli that were employed in this study were selected on the basis of a conjunction of the highest reported valence and arousal, and included photographs of happy facial expressions, positive social interactions (e.g. loving family), and erotic imagery.

There is strong evidence to suggest that use of the IAPS reliably elicits affective responses (Lang et al., 1993) in both random (Bradley et al., 1996;Sutton et al., 1997) and blocked (Greenwald et al., 1989) presentations. Moreover, electromyography evidence indicates that use of the IAPS in conjunction with self report of affective state elicits activity in zygomaticus and corrugater muscle regions of the face (Davidson et al., 2003;Ekman et al., 1980). The IAPS has also been validated in neuroimaging investigations of affective processes in psychiatrically unaffected HC (Britton et al., 2006;Grimm et al., 2006;Hariri et al., 2002;Hariri et al., 2003;Lee et al., 2004;Northoff et al., 2000;Baumgartner et al., 2006) and in MDD subjects (Kalin et al., 1997;Anand et al., 2005a;Irwin et al., 2004;Tremblay et al., 2005;Wagner et al., 2004;Davidson et al., 2003).

Although many studies have examined the effect of positive affect using the presentation of affective facial expressions (Fu et al., 2007;Canli et al., 2005;Surguladze et al., 2005;Roberson- Nay et al., 2006) and affective imagery (Anand et al., 2005a;Lawrence et al., 2004;Kalin et al., 1997;Schaefer et al., 2006;Tremblay et al., 2005;Anand et al., 2005b;Wagner et al., 2004), several of these investigations have used positive affect as a control contrast to negative stimuli, while others have grouped positive and negative stimuli as affective stimuli or have failed to detect statistically significant differences. This approach obfuscates the possibility of identifying brain regions responsible for decreased positive dispositional affect.

Activation of the ventral caudate and putamen have been consistently reported during the presentation of happy facial expression in HC groups (Phan et al., 2002;Morris et al., 1996;Phillips et al., 1998b;Critchley et al., 2000), however, decreased BOLD signal has been reported in MDD subjects (Lawrence et al., 2004;Surguladze et al., 2005;Fu et al., 2007). In a comparison to age-matched HC subjects, Fu and colleagues have reported decreased BOLD

187 signal in the precuneus (BA 7) during activation to happy facial expressions in MDD subjects before and after 8-weeks of treatment with fluoxetine (Fu et al., 2007). Compared to HC groups, decreased activation to happy facial expression in MDD subjects has also been reported with decreases in the OPT areas; right inferior temporal gyrus (BA 20) (Schaefer et al., 2006), and bilateral fusiform gyrus (BA 19).

While neuroimaging of facial affective expressions may highlight areas involved in emotional recognition, evocative pictures may elicit a more direct experience of the emotion; production of an affective state. In an effort to compare differential brain regions involved in the processing of these two types of stimuli, Britton and colleagues compared BOLD activation between affective facial expression and pictures from the IAPS (Britton et al., 2006). Despite activating similar brain regions during the presentation of affective stimuli (amygdala, occipital cortex, and vmPFC), there were also differences in activation. While facial expressions of all affective stimuli recruited the amygdalae more reliably, IAPS stimuli were noted to recruit a larger portion of the OPT cortex. In regards to positive IAPS stimuli specifically, increases in activation were limited to the right OPT junction and the left cerebellum.

In a preliminary analysis of affective processing between treatment resistant MDD and HC, Mitterschiffthaler and colleagues reported that the processing of positive affective stimuli was also associated with activation of the left medial occipital gyrus in the HC, but not the MDD group (Mitterschiffthaler et al., 2003). Moreover, in another comparison between MDD and HC, decreased activations in medial occipital gyrus (BA 19) were reported in the MDD group in response to erotic IAPS, compared to age matched HC (Schaefer et al., 2006). Our findings are in agreement with previous reports in that we have noted reliable activation of the OPT cortex, without noting activation of the caudate or putamen across the three groups. It’s worth noting, however, that we found that delayed maximal activation of the right putamen in the NR group, compared to RS and HC.

Using a combination of electroencephalography, functional magnetic resonance imaging, and volumetric analyses, Wacker and colleagues report that e found that anhedonia, but not other symptoms of depression or anxiety, was correlated with reduced nucleus accumbens activation towards positive affect (Wacker et al., 2009). It is tempting to speculate that the presence of anhedonia might have prevented us from identifying differences in nucleus accumbens activation

188 during positive affect induction between HC and MDD>

Taken together, the induction of positive affect through visual stimuli in HC is associated with activations in the OPT cortex (BA 18, 19, 37). Although we did not find a difference in activation between HC and MDD, we report that activation in the right OPT was actually higher in the RS group compared to the HC group. Indeed, during PAI, there were no regions identified where the BOLD signal was higher in HC compared to both RS and NR. Evidence presented in this thesis suggests that previously described decreases may be primarily attributed to the lack of activation by NR. Thus, a separate analysis of the NR and R groups allows one to uncover differential activations by these two 'MDD' groups that would have otherwise been lost.

The temporal analysis of late and early blocks permitted differential inter- and intra- block habituation of BOLD signal between the three groups. Specifically, whereas activation in brain regions mediating memory processing in the parahippocampal region demonstrated maximal activation early in the block in the HC group, there was a delayed activation in both MDD groups.

Differences in Negative Affective Processing: Depressed vs. Control Subjects

Compared to HC groups, differences in the metabolism, blood flow, and reactivity of the dlPFC have been one of the most consistent findings in MDD (Dolan et al., 1992;Bonte et al., 2001;Ito et al., 1996;Oda et al., 2003). The majority of resting state PET and SPECT investigations that have compared activity in the dlPFC between HC and MDD populations have noted decreased CBF (Dolan et al., 1992;Bonte et al., 2001;Ito et al., 1996;Oda et al., 2003), with less consistent differences in glucose metabolism (Drevets et al., 1992;Kimbrell et al., 2002;Pizzagalli et al., 2004;Nofzinger et al., 2005). Part of the discrepancy between blood flow and metabolism may be explained by impaired perfusion-metabolism coupling in the prefrontal lobe of MDD subjects (Dunn et al., 2005).

15 18 In a combination O-H2O – FDG PET analysis of glucose metabolism and cerebral blood flow, a HC group was compared to unmedicated BD and MDD participants (Dunn et al., 2005). While global and regional correlation coefficients for blood flow and metabolism were positive in the HC and BD groups, global and regional measures of flow and metabolism were not

189 positively related in the MDD group. Significantly fewer positive regional correlation coefficients were reported in the MDD group compared to HC and BD subjects, particularly throughout the prefrontal lobe. Post-hoc analyses concluded that the degree of flow-metabolism uncoupling was inversely correlated with the severity of depressive symptoms, with greater depressive symptom severity associated with increased uncoupling between glucose metabolism and CBF (Dunn et al., 2005).

Differential activation of the dlPFC during the processing of negative affect has also been a consistent finding in MDD populations when compared to age-matched HC groups. When the induction of negative affect or mood is accompanied by concomitant execution of a cognitive task, decreased activation of the dlPFC by MDD groups has been reported (Kumari et al., 2003;Siegle et al., 2002). Conversely, when the induction of negative affect is passive, or without the concurrent cognitive task administration, increased activation of the dlPFC has been reported in MDD populations (Beauregard et al., 1998;Wagner et al., 2004;Fu et al., 2004;Liotti et al., 2002). Reduced activation of the dlPFC during changes in affect accompanied by cognitive tasks may reflect competition between the cognitive and affective components.

Compared to HC participants, hyperactivation of the dlPFC during the performance of cognitive tasks has been described in MDD during the performance of cognitive tasks (George et al., 1997;Hugdahl et al., 2004).Alternatively, hyperactivation of the dlPFC during the induction of negative affect without a cognitive component may reflect attempts by MDD subjects to regulate their affective states. Neuroimaging and clinical data suggest that MDD subjects may be impaired in the ability to down-regulate induced negative affect (Beauregard et al., 2006).

In this thesis, increased activation of the left dlPFC during NAI by both RS and NR groups was noted relative to HC. Interestingly, the NR group was characterized by a statistically significant increase in BOLD signal in the left dlPFC relative to the RS group. These results suggest that NR group may be more impaired in the regulation of negative affect during an MDE. Preliminary evidence indicates that SSRI responders may outperform nonresponders across multiple cognitive domains, with pronounced differences in executive, language and working memory functions (Gorlyn et al., 2008).

The dmPFC projects to cortical regions, thalamic nuclei, and the monoamine producing nuclei in

190 the brainstem(Tanaka, Jr., 1976;Arnsten and Goldman-Rakic, 1984). Preclinical evidence has implicated this region in addiction (Porrino et al., 2002;Fuchs et al., 2005), conflict monitoring (de Wit et al., 2006), and motor inhibition (Narayanan and Laubach, 2006). Functional neuroimaging literature implicates this region in self-referential stimuli processing.

In an evaluation of ten HC performing a self-referential task (judging whether visually presented traits described them) under fMRI scanning conditions, bilateral activation in the dmPFC cortex was reported (Fossati et al., 2003). The other-referential condition (whether presented traits are desirable) induced activation in lateral prefrontal areas. Although, activation in the right dmPFC cortex was unique to the self-referential condition regardless of the valence of the words, positive words produced a more robust activation. Similar results have been observed with 15O-

H2O PET investigation of thirteen HC who were asked to either rest or reflect about themselves, another person or social issues. Increases in CBF were noted in comparison to the rest conditions (D'Argembeau et al., 2005).

The dmPFC has also been frequently activated by manipulations of affect, primarily using the 15 IAPS. Using O-H2O PET, Lane and colleagues demonstrated that the induction of positive and negative affect was distinguished from neutral visual stimuli by significantly increased CBF in the vicinity of the dmPFC (Lane et al., 1997c). Shortly thereafter, similar increases in the dmPFC were reported using a different affect induction technique (affective video) and a different imaging modality (fMRI) (Beauregard et al., 1998). Using the IAPS picture set, Northoff and colleagues demonstrated the interaction between cognitive and affective processes by evaluating 13 HC viewing IAPS stimuli while making either a) emotional or b) non-emotional judgments (Northoff et al., 2004).

As a general finding, investigators noted that both emotional and non-emotional judgments of IAPS pictures were characterized by signal increases in vlPFC and dlPFC along with concurrent signal decreases in vmPFC and dmPFC. A further comparison of emotional to non-emotional judgments indicated relative signal increases in vmPFC and dmPFC while, relative signal increases were reported in vlPFC and dlPFC in the comparison of non-emotional to emotional judgments (Northoff et al., 2004). The close coupling between the valence of stimuli and dmPFC activation was further supported by a parametric, instead of a categorical, data analysis. When valence ratings of each stimuli presented were used as a regressor, the investigators noted a

191 valence-dependent modulation of the BOLD signal in the dmPFC, with positive affective stimuli eliciting an activation, and a deactivation for negatively valenced stimuli (Heinzel et al., 2005). In a comparable analysis, Grimm and colleagues reported that the valence of IAPS stimuli was actually correlated with the inferior vmPFC, while the level of arousal correlated with dmPFC (Grimm et al., 2006).

The middle temporal gyrus has traditionally been associated with language processing (Dronkers et al., 2004). The initial perception in the auditory fields of the superior temporal gyrus, is followed by two divergent ventral and dorsal neural processing pathways. The ventral stream is believed to project toward the posterior middle temporal gyrus and serves in mapping sound on to meaning, while the dorsal stream ultimately projects to frontal regions and is believed to be involved in mapping sound on to articulatory-based representations (Hickok and Poeppel, 2004). Evidence from fMRI investigations, however, has also implicated this brain region in other non- speech related affective functions, including affective facial recognition (Phillips et al., 1998a;Rosen et al., 2006), processing (Iidaka et al., 2001;Schilbach et al., 2006), ethical decision making (Heekeren et al., 2003), comprehension of humour (Bartolo et al., 2006), and retrieval of emotionally valenced memory (Piefke et al., 2003). Although we did not find a difference in BOLD activation in the mid temporal gyrus between HC and MDD, we report higher BOLD signal in the NR group compared to both RS and HC.

Evidence from HC groups reveals deactivation of the precentral and premotor gyri during the performance of emotionally arousing memory tasks (Mitchell et al., 2006;Mather et al., 2006). The induction of negative affect in HC has also been associated with decreased activation of the precentral gyrus, compared to positive affect inductions (Baumgartner et al., 2006;Teasdale et al., 1999;Keedwell et al., 2005a;Kumari et al., 2003;Tremblay et al., 2005). Previously, Fu and colleagues reported depressive state-independent increased activations in the left precentral gyrus (compared to HC subject), in response to sad facial expressions (Fu et al., 2004). A similar pattern of precentral gyrus hyperactivity in a psychiatric cohort was reported in subjects with obsessive-compulsive disorder during a visual symptom provocation paradigm (Mataix-Cols et al., 2004). Similar to these reports we have found that both MDD groups exhibited hyperactivity in the precentral gyri relative to HC during negative affective processing however, only after they met criteria for clinical remission.

192 In summary, dlPFC and dmPFC activations have been associated with dispositional affective processing in HC populations. Previous reports from MDD populations report increased activation of these brain regions during negative affect processing, consistent with a negative affective bias. The data presented herein suggest that this hyperactivity may be more predominant in MDD subjects not responsive to conventional antidepressants in combination with an atypical antipsychotic. However, the use of only a single treatment modality precludes a generalization that this hyperactivity may also characterize non-responders to alternative antidepressant treatment modalities. In support of a treatment-independent non response endophenotype, however, resting state glucose metabolism data suggest that response to either pharmacotherapy or psychotherapy is associated with time-dependent decreases in activity in the left medial prefrontal cortex (Kennedy et al., 2007b).

Effect of Exposure to Antidepressant

Early neuroimaging investigations of changes in glucose metabolism following administration of tricyclic antidepressants and monoamine oxidase inhibitors reported increased metabolism in the left dlPFC (Baxter, Jr. et al., 1985;Baxter, Jr. et al., 1989;Martinot et al., 1990;Serra et al., 2006). Subsequent reports have similarly reported increased resting state metabolism in dlPFC and decreased metabolism in vlPFC and amygdala following antidepressant treatment with selective serotonin reuptake inhibitors (Brody et al., 1999;Buchsbaum et al., 1997;Mayberg et al., 2000;Kennedy et al., 2001;Saxena et al., 2002;Davidson et al., 2003). The Neuroimaging investigations of atypical antipsychotics have been limited to the study of psychotic disorders (Davis et al., 2005) The limitation of these studies is that they have not differentiated between the therapeutic effects of the medication versus the non-selective, transient change in pharmacology that accompanies their administration.

The effect of serotonergic and noradrenergic antidepressant exposure on enhancing positive affective processing and down regulating negative affect processing has been described in HC populations using neuropsychological and neuroimaging analyses (Serra et al., 2006). In neuroimaging analyses evaluating affective processing at two time points in MDD and HC, an effect of antidepressant treatment is inferred from a group x time interaction (Davidson et al., 2003).

193 In a preliminary analysis of 2 HC and 2 MDD subjects before and after antidepressant treatment initiation, enhanced activation to positive affective processing (repeat presentation of positive IAPS stimuli) in V2 (Kalin et al., 1997) was reported in the MDD group. With a larger sample size and an event-related design employing a variety of positive IAPS stimuli (facial expression, social interactions, erotic imagery), increased activation in the OPT cortex was also been reported in MDD subjects receiving antidepressant treatment during the repeat presentation of positive stimuli (Schaefer et al., 2006). Similarly, in response to the presentations of facial expressions of happiness, the attenuated dynamic range reported at baseline by MDD subjects (compared to HC) was also increased following eight weeks of treatment with fluoxetine.

Repeat presentations of negative affective stimuli accompanied by antidepressant treatment have been frequently accompanied by decreased activation in the amygdala and sACC. During the repeat presentation of facial expressions of sadness, antidepressant treatment was associated with decreased activation in the amygdalae (Sheline et al., 2001). Davidson and colleagues reported decreased activation of the aMCC following antidepressant treatment with venlafaxine (Davidson et al., 2003). Reduced activation of the amygdala in response to negative affective IAPS distractors has also been reported following antidepressant treatment with bupropion, alongside decreases in the dmPFC, vmPFC, and PCC (Robertson et al., 2007).

Consistent with previous reports (Mayberg et al., 2000), exposure to pharmacotherapy in our investigation was also associated with decreased activation in brain regions mediating the identification of affective value of a stimulus, in the OFC (BA 11). and the amygdala. Time- dependent decreases in the OFC were reported in both patient groups and were not associated with inter-scan habituation in HC. Decreased activation in these regions can thus be modeled as a “depressive symptom severity independent effect” of pharmacotherapy on negative affective processing, or alternatively, as a strategy employed by both NR and RS when presented with multiple presentations of negative affect. In this thesis, we report decreased activation during NAI in the left and right dlPFC as an effect of exposure to antidepressant treatment, as this effect was common to both NR and RS groups, but was absent in the HC group. Interestingly, these decreases in dlPFC activations during NAI, had the net effect of eliminating the baseline differences in dlPFC activation during NAI between HC and MDD.

Effect of Response to Antidepressant

194

To identify the neural correlates of response to antidepressant treatment, or a decrease in depressive symptom severity, longitudinal comparisons were made directly between RS and NR. Decreased activation in the ventromedial PFC (BA 10/11) and ventrolateral PFC (BA 10/45), represented time-dependent changes differentiating RS from NR during NAI. Recent evidence suggests that subjects with MDD may have an increased functional connectivity between the vmPFC, the aMCC and the dorsolateral PFC (Lemogne et al., 2009).

The majority of neuroimaging investigations that have evaluated MDD subjects before and after treatment have frequently combined RS and NR into a single group, with post-hoc correlations that change in specific brain regions correlated with decreases in depressive symptom severity. To our knowledge, this is the first fMRI neuroimaging investigation that has separately evaluated RS and NR throughout the duration of the antidepressant trial.

18 15 Several neuroimaging investigations using resting state FDG and O-H2O PET however, have compared pre-post changes associated with the differential response between RS and NR to antidepressant pharmacotherapy. In an analysis of MDD outpatients evaluated with 18FDG PET before and after eight weeks of antidepressant treatment with paroxetine, Brody and colleagues reported changes in brain metabolism between RS and NR (Brody et al., 1999). Comparable to the decreased activation to negative stimuli that we have described in RS in the vmPFC, Brody and colleagues reported that RS had significantly more pronounced decreases in resting state vlPFC and OFC metabolism than NR.

Mayberg and colleagues analyzed the time course of changes in 18FDG PET in a group of MDD inpatients treated with fluoxetine, at baseline, and again at 1 and 6 weeks following treatment initiation (Mayberg et al., 2000). Time-specific and response-specific effects were examined at 1 and 6 weeks of treatment. Three distinct patterns of change were recorded; i) changes common to both RS and NR at 1 and 6 weeks, ii) reversal of the 1-week pattern at 6 weeks, and iii) unique changes seen only after chronic treatment. Analogous to the results in this study, similar neuroimaging changes in fluoxetine RS and NR were noted at 1 week after treatment initiation with different metabolism by week 6.

The investigators also reported decreased metabolism in limbic regions in the responders

195 (decreased activation to negative stimuli in the posterior cingulate in this study), and dorsal cortical increases (increased activation in the dlPFC and vlPFC to positive stimuli in this study). Failed response in the PET study (Mayberg et al., 2000) was also associated with an absence of metabolic changes in the dPCC (compared to consistent activation of (BA 31) to negative stimuli in NR) and prefrontal cortex (compared to persistently high activation to negative stimuli in NR in vmPFC). Results from this thesis also describe decreased activation of the dPCC (BA 31) during NAI only in the RS group. Evidence from other neuroimaging investigations supports the hypothesis that decreases in metabolism in the subgenual cingulate (BA 25) reflect a treatment response effect (Drevets et al., 2002;Kennedy et al., 2007b;Mayberg et al., 2000).

In one of the largest 18FDG pet studies conducted using an MDD sample, Holthoff and colleagues performed repeat scans on 41 inpatient MDD subjects during the depressed phase (moderate-severe), and during a fully remitted phase (mean between scan duration was 6.6 months) (Holthoff et al., 2004). The study had the advantage of evaluating subjects who were receiving different forms of pharmacotherapy, (thereby negating the effect of any single pharmacological agent) and remained asymptomatic for 3 months after meeting remission criteria. In addition to time-dependent decreases in metabolism in left prefrontal, anterior temporal and aMCC, decreased metabolism was also reported in the vmPFC (BA 10).

Decreases in the vmPFC have also been reported following response to non-pharmacological treatments of MDD. After 15 to 20 sessions of cognitive behavioural therapy, significant changes in 18FDG metabolism were reported, including decreases in vmPFC (BA 10) specifically noted in RS (Goldapple et al., 2004). Decreased CBF flow into the vmPFC have also been preliminarily reported in treatment resistant MDD subject responding to deep-brain stimulation (Mayberg et al., 2005).

Although there are no published fMRI investigations comparing RS and NR in a longitudinal (follow-up) study design, some fMRI investigations that have administered antidepressant treatments concomitantly with pre-post neuroimaging data acquisition have provided correlations between changes in depressive symptoms and corresponding changes in activation during affective stimuli. Although preliminary, reports indicate decreased activation to negative affect processing, and increased activation to positive stimuli as being associated with symptomatic improvement.

196

Fu and colleagues reported that symptomatic improvement in MDD subjects treated with fluoxetine was associated with a reduction of dynamic range in pregenual cingulate cortex in response to the presentation of sad facial affect (Fu et al., 2004). A correlation between decreased activation in the amygdala during negative IAPS stimuli and symptomatic improvement after 8 weeks of treatment with bupropion has also been reported (Robertson et al., 2007). In response to the repeated presentation of positive affect in MDD subjects, increased activation correlating with symptomatic improvement was dependent on the type of stimuli presented. In response to the presentation of happy facial expression, correlations were noted in the hippocampal and extrastriate regions (Fu et al., 2007). Repeat presentations of positive affective imagery have noted that symptomatic improvement was correlated with increased activation in the right caudate, thalamus, and medial and lateral portions of the dorsal dlPFC (BA 8), while repeat presentations of erotic imagery were associated with correlations in the hypothalamus, insula, and pACC (BA 32).

Affective Processing in Euthymia

Despite a significant number of fMRI investigations characterizing euthymic BD subjects (Caligiuri et al., 2006;Kronhaus et al., 2006;Monks et al., 2004;Lawrence et al., 2004;Blumberg et al., 2003b;Liotti et al., 2002), there has been a paucity of neuroimaging investigations evaluating affective processing in MDD subjects who have met objectively-defined remission criteria (i.e. HRSD-17 < 7) (Frank et al., 1991). In our analysis, in the absence of clinically significant depressive symptoms, RM were characterized by activation in the premotor and extrastriate occipital cortex, in the medial temporal lobe, and the right dorsolateral PFC (BA 9), compared to HC in whom BOLD signal deactivations were identified in these regions.

In an aforementioned investigation (Holthoff et al., 2004), 18FDG PET measures of glucose metabolism were obtained in MDD subjects at two time points, during an MDE, and later in remission. Although significant pre-post changes were observed between the depressed and remitted phase, the absence of HC group, did not permit an evaluation of residual differences in glucose metabolism between remitted MDD and HC.

In an effort to investigate the neural correlates of vulnerability to relapse in MDD subjects, Liotti

197 and colleagues induced a negative mood through self-recall in remitted MDD, acutely depressed MDD, and HC (Liotti et al., 2002). In the HC group this paradigm produced increased CBF in the subgenual cingulate (BA 25) and decreased CBF in the right dlPFC (BA 9); findings that were absent from both groups. In contrast, the induction of a sad mood in both MDD groups was associated with decreased CBF in the medial OFC (BA 10/11). In the remitted MDD group, unique increases were noted in the rACC (BA 24) and the vlPFC (BA 10, 45) that were absent in HC or acutely depressed MDD group.

Recently, these findings have been replicated with a sample of MDD subjects who not only met remission criteria, but were also medication free (Gemar et al., 2006). In this relatively small group of MDD subjects, decreased activation during induced sadness was again noted in the medial OFC (BA 32) and rostral ACC (BA 24). These findings suggest that in the remitted phase of illness, the presence of medications may not have a detectable neuroimaging footprint. Consequently, the observed differences between remitted MDD and HC groups may represent a neural correlate of evidence of relapse vulnerability.

Indeed, recent neuropsychological evidence indicates that magnitude of mood-linked cognitive reactivity is a significant predictor of relapse over the subsequent 18 months (Segal et al., 2006). Intriguingly, the investigators noted that patients who met remission with antidepressant medication, also demonstrated greater cognitive reactivity following the mood provocation than patients who received cognitive behavior therapy. However, irrespective of treatment the absolute magnitude of mood-linked cognitive reactivity was found to be significant predictor of relapse over the subsequent relapse. The authors suggest that the vulnerability of remitted MDD patients may be conceptualized as a “(re)activation of depressive thinking styles triggered by temporary dysphoric states” (Segal et al., 2006).

Affective words have also been used to evaluate the depressive-state independent component of affective processing in a group of remitted MDD subjects and HC(Hooley et al., 2009). Both groups were scanned while they heard praising, critical, and neutral comments from their own mothers. Compared to HC, remitted MDD subjects responded to criticism with greater activation in the amygdala and less activation in the dorsolateral PFC, and aMCC. Similar to our results in remitted MDD subjects, there were no statistically significant differences between the groups during praise and neutral commentary.

198

Although decreases in medial OFC during the affective induction in remitted MDD subjects was not observed, a number of differences must be highlighted. The induction of negative affect through visual and self-recall stimuli has been documented to recruit different brain regions (Reiman et al., 1997;Phan et al., 2002). Moreover, the detection of signal change in the orbitofrontal cortex may be compromised by static magnetic field gradients at air-tissue interfaces (Cordes et al., 2000). Finally, the reported differences in this comparison, represent not only trait-related differences in affective processing, but also the cumulative load of previous affective presentations, and as such, may actually reflect failure to down-regulate affective processing at the neuronal level. This is especially critical, as the RS group was characterized by statistically significant decreases in BOLD signal during NAI in the neighbouring vmPFC.

Response Prediction

Early evidence from resting state glucose metabolism and cerebral perfusion studies indicates that differences in baseline metabolism within subdivisions of the ACC may be used to predict ultimate response to SSRI and SNRI treatments. Non-response to treatment has been associated with hyperactivity in the subgenual (ventral) portions of the ACC, and hypoactivity of the pregenual (rostral) ACC. These results are supported by findings from previous studies.

Higher pretreatment metabolic activity in the pregenual and subgenual ACC identified subjects who would ultimately not respond to a trial of venlafaxine or bupropion (Little et al., 1996). In an investigation of glucose metabolic activity in outpatients with MDD before and after treatment with paroxetine, Brody and colleagues reported that NR displayed higher metabolism in the left ventral ACC (Brody et al., 1999). Conversely, higher pretreatment metabolism in the rostral ACC was associated with improvement in depressive symptoms with paroxetine treatment in a sample of subjects who met criteria for MDD with comorbid obsessive compulsive disorder (Saxena et al., 2003). Similarly, in an evaluation of subjects treated with various forms of antidepressant pharmacotherapy (n=18), rostral ACC hypoactivity characterized NR when compared with controls, in contrast to hyperactivity in the responders (Mayberg et al., 1997;Pizzagalli et al., 2001). Recently, it has been demonstrated that increased metabolism in the sACC has also been implicated in predicting a poor therapeutic outcome to CBT (Konarski et al., 2009).

199

These findings raise the possibility that hypermetabolism in the subgenual cingulate may be a marker of depression that is resistant to treatment with either pharmacotherapy or psychotherapy. Indeed, in an evaluation of MDD subjects (n=5) who failed to respond to a minimum of four different antidepressant treatments, administered at adequate doses and duration during the current episode, a unique pattern of elevated subgenual cingulate blood flow was noted relative to a well matched healthy control group (Mayberg et al., 1997;Pizzagalli et al., 2001). Hyperactivity in the subgenual cingulate (Dougherty et al., 2003) and ventral ACC (Wu et al., 1999) have also been associated with medication-resistant populations that respond to alternative antidepressant treatments. However, as the subgenual cingulate is implicated in other functions, including pain (Bingel et al., 2007;Godinho et al., 2008;Taylor et al., 2008) and guilt (Zahn et al., 2009), this prediction should be approached with caution. It is interesting to note, however, that patients who are non-responsive to treatment, also report higher levels of pain and that antidepressants have therapeutic value in the treatment of pain (Perahia et al., 2009;Verdu et al., 2008).

Hyper activation of the subgenual cingulate in response to emotional stimuli in MDD subjects (n=14) was associated with poor response to 16 sessions of CBT (Siegle et al., 2006). Meanwhile, in a separate study, greater activation of the subgenual cingulate during visual presentation of faces representing different intensities of sadness was associated with quicker antidepressant improvement with fluoxetine (Chen et al., 2007). Moreover, increased amygdalar activation in response to facial expressions of sadness (Canli et al., 2005) and affective words (Siegle et al., 2006) have also been noted in RS. Using comparable negative stimuli from the IAPS, Davidson and colleagues demonstrated that magnitude of deactivation in the dorsal ACC (BA 24) at baseline was predicted treatment response (Lee et al., 2004). In this analysis increased activation to negative affective stimuli was associated with treatment non-response in the pACC (BA 24) and right ventrolateral PFC (BA 10/45). Moreover, the NR group were also characterized by increased BOLD signal in the left amygdala during NAI, but this difference was not statistically significant.

Effect of Laterality

Perception of emotion has been associated with right hemisphere superiority (Heller,

200 1993;Borod, 1993;Borod, 1992). The valence model of emotion further proposes that positive, approach-related emotions are associated with the functions of left cerebral hemisphere regions, whereas negative, withdrawal-related emotions appear to be more aligned with right hemisphere mechanisms (Lee et al., 1990;Davidson, 1995). Anecdotal evidence supporting the model has recently been validated with functional neuroimaging. Using 20 positive and 20 negative affect inducing stimuli from the IAPS, Lee and colleagues compared hemispheric activation to both types of stimuli (Adolphs, 2002b). Although the presentation of affective stimuli, irrespective of valence, produced activation in the vmPFC, ACC, dlPFC, amygdala, and cerebellum, negative affective stimuli preferentially engaged the right hemisphere, while positive affective stimuli preferentially engaged the left hemisphere. Although not a primary hypothesis in this investigation, laterality to affective stimuli was compared post-hoc separately in HC and MDD groups. At baseline, activation to positive and negative stimuli did not appear to preferentially activate the left cerebral hemisphere in either group.

The difficulty in obtaining lateralization differences in response to positive affect processing may be related to the different levels of arousal generated by negative and positive valence stimuli. When a group of 14 HC female subjects were presented alternating blocks of positive and negative affective visual stimuli, brain reactivity was only lateralized towards the left hemisphere for positive pictures and towards the right hemisphere for negative pictures after the experience of valence was equated for arousal (Canli et al., 1998). Although the positive affect IAPS stimuli selected for this investigation were characterized by high ratings of valence and arousal, the ratings of arousal for the selected negative stimuli were nevertheless higher.

Pharmacodynamics of Olanzapine Fluoxetine Combination

Chronic fluoxetine administration has been reported to enhance extracellular serotonin in the PFC without a significant effect on dopamine concentration (Gartside et al., 2003;Hervas et al., 2001;Tanda et al., 1996). The atypical antipsychotics, particularly olanzapine are also potent 5-

HT2A/2C antagonists in addition to their D2 antagonism (Wettstein et al., 1999). Additionally, olanzapine also binds to the histamine H1 receptor, the α1-adrenoreceptor, and the metabotropic muscarinic receptors (M1-M5) (Zhang and Bymaster, 1999).

Evidence from preclinical animal data indicate that acute administration olanzapine is associated

201 with an increase in extracellular dopamine in the PFC of freely-moving rats (Rowley et al., 2000). The addition of fluoxetine with olanzapine has been reported to result in sustained increases in extracellular dopamine and norepinephrine levels significantly greater than either drug alone (Zhang et al., 2000). The authors also noted that the combination of haloperidol with fluoxetine did not increase the monoamines more than fluoxetine alone, while olanzapine plus sertraline combination only increased only extracellular dopamine levels. These finding les the authors to suggest that the pronounced increase in extracellular concentration of both catecholamines may be unique to the olanzapine fluoxetine combination and may contribute to antidepressive effect (Zhang et al., 2000).The acute administration of olanzapine and fluoxetine has also been reported to increase extracellular monoamine concentrations in subcortical areas (Koch et al., 2004). Synergistic increases in catecholamine concentrations were reported response to the combination of the two medications in comparison to either drug alone in the rat hypothalamus in addition to the PFC.

The effect of chronic (14-day) administration of fluoxetine, with a single dose of olanzapine on dialysate concentration of noradrenaline, dopamine and serotonin was also investigated in the rat medial PFC (margos-Bosch et al., 2005). While chronic fluoxetine only increased serotonin levels, single olanzapine administration also increased the concentration of noradrenaline and dopamine as well. Ironically, a single dose of olanzapine without chronic fluoxetine treatment decreased extracellular levels of serotonin, leading to authors to conclude that the effect was likely mediated by the stimulating action of noradrenergic neurons from the locus coeruleus and dopaminergic neurons in the ventral tegmental area that project to the PFC (margos-Bosch et al., 2005). The combination of both compounds has also been reported to lead to greater hippocampal pregnenolone and serum deoxycorticosterone concentratons in both higher- and lower-dose experiments, and elevated hippocampal allopregnanolone in higher-dose conditions (Marx et al., 2006).

Accompanying changes in extracellular neurotransmitter concentrations, direct changes in neuronal activity have also been recorded following olanzapine fluoxetine combination administration in rats. Using extracellular in vivo recordings, Gronier and colleagues reported that while acute or 5-day administration of olanzapine did not modulate the firing rate of PFC neurons, 21-day exposure to olanzapine significantly increased their firing rate and responsivity to the GABA-A antagonist, bicuculline (Gronier and Rasmussen, 2003) Similarly, acute

202 administration of fluoxetine also did not alter the firing rate of PFC neurons, however 21-day exposure actually decreased their firing rate and responsiveness to bicuculline. The authors noted that co-administration of olanzapine during the last 5 days of a 21-day fluoxetine treatment prevented the suppression of firing and decreased responsiveness to bicuculline, suggesting that short-term olanzapine treatment can attenuate the suppression of firing and excitability of PFC neurons induced by chronic fluoxetine treatment.

Similar electrophysiological findings were also noted in the locus coeruleus (LC) where acute administration of olanzapine produced a significant increase in the firing rate and burst firing of LC cells (Seager et al., 2005). Chronic administration of fluoxetine, on the other hand, decreased baseline and burst firing of LC cells. When these two compounds, however, were given in combination, olanzapine administration resulted in significantly greater increase in LC firing rate after both acute and chronic administration of fluoxetine, leading to enhanced norepinephrine release in projection areas including the PFC (Seager et al., 2004).

Changes in cell proliferation rates have also been reported following administration of the olanzapine fluoxetine combination (Kodama et al., 2004). Chronic (3-weeks) olanzapine administration resulted in an increase in the number of newborn cells in the hippocampus to the same extent as fluoxetine. Olanzapine or fluoxetine treatment additionally increased the number of proliferating cells in the prelimbic cortex. Moreover, changes in gene-expression have also been reported to follow administration of both drugs (Maragnoli et al., 2004). After single injection, mRNA levels of fibroblast growth factor were selectively upregulated in the PFC, hippocampus, and striatum only when the two drugs were coadministered. Conversely, decreased levels of two immediate-early gene transcription factors (e.g., pCREB and FOS) were also noted in rat PFC following olanzapine-fluoxetine combination for 7 days (Horowitz et al., 2003).

The behavioural effects of olanzapine fluoxetine combination are beginning to emerge from preclinical animal research. Their interpretation, however, is associated with a number of limitations that include tremendous methodological heterogeneity among investigation, obfuscating a clear understanding of the behavioural effect of chronic antidepressant and atypical antipsychotic treatment. These limitation notwithstanding, preliminary evidence indicates that the combination may increase the motivation state of marmosets(Cilia et al., 2001), diminishes spontaneous behaviors in rats (Horowitz et al., 2003), and increased swimming time in the

203 behavioural despair test in albino mice (Pawar et al., 2009). These biochemical, electrophysiological, and behavioural preclinical data support the antidepressant qualities of the combination of olanzapine and fluoxetine.

Affective Processing Models

Phillips and colleagues propose that a core structural and functional abnormality in the amygdala may underlie differences in affective processing. They propose that volumetric reductions within regions such as the amygdala, important for the identification of the emotional significance of a stimulus and production of affective states, may result in a restriction of the range of emotions available for experience in patients with major depression. Moreover, concomitant hyperactivity during depressive episodes may additionally predispose a bias toward the predominant role of the amygdala; the identification of threatening cues instead of neutral or positive stimuli. The continuous negative tone may portend depressed mood, while the restricted emotional range anhedonia. Moreover, structural and functional impairments in dorsolateral and dorsomedial regions of the PFC are believed to reflect impaired regulation of affective states and behaviour, and further promote depressed mood and anhedonia (Phillips et al., 2003b).

Additional evidence presented here reinforces the importance of the previously discussed nodes in the neural network subserving affective processing. Findings of persistent hyper reactivity even in remitted MDD subjects, underscores the trait component of negative bias in affective processing. This repeat affective-stimulus presentation investigation has additionally suggested a number of other brain regions that may serve to modulate the identification, processing, and regulation of affective stimuli. These include the OPT cortex, the premotor cortex, the precuneus, and the cerebellum.

These additional regions are commensurate with the spatio-temporal network proposed for the analysis of facial expression of affect. Three temporally-distinct neural networks have been proposed including initial evaluation with the core system, subsequent relay to the extended systems (150 ms later), followed by engagement of the cognitive system (300 ms) (Adolphs, 2002a). The core system comprises the initial response and involves the preliminary perceptual processing of highly salient stimuli by the amygdala, the thalamus, and the primary visual cortex.

204 It is worth noting that visual information enters the amygdala not only from the visual cortex, but also from the superior colliculus and the pulvinar thalamic nucleus. The extended system represents the second stage of visual processing and involves relaying visual information from the primary visual cortex to posterior temporal regions, back to the amygdala, and the OFC. Additional projections are sent by the amygdala and OFC to the brain stem in preparation for the autonomic effects of the affective state. The activation of the cognitive system results in conscious knowledge of the emotion being presented on the face. This final stage involves projections sent back from the OFC and amygdala to somatosensory cortices, and posterior temporal areas (Adolphs, 2002a).

Using this model as a reference, the identified differences in negative affective processing at baseline seem to reflect a dysfunction of both the extended and cognitive systems, whereas abnormalities in positive affect processing are limited to the extended systems. The downregulation of the BOLD signal to repeat presentations of negative affective is seen primarily in the extended system in HC, while the administration of pharmacotherapy appears to decrease the activation of both the extended and cognitive system in MDD subjects. The decrease in activation is relative, however, as hyperactivation of the extended and cognitive system persists even in euthymic MDD subjects when they are compared to their HC counterparts.

An evaluation of the longitudinal differences in activation during both positive and negative affective IAPS stimuli between RS and NR reinforces the assertion that pharmacotherapy administration operates on substrates comprising the cognitive system. In a between-group comparison persistent normalization of activation (increases to positive stimuli and decreases to negative stimuli) is seen exclusively in RS in regions believed to mediate the conscious knowledge of emotion.

Occipitotemporal Cortex

The term occipitotemporal cortex describes a region broadly corresponding to BA 37 at the junction of the temporal and occipital lobe. In the majority of neuroimaging investigations, this term has also been applied to a region also encompassing anterior BA 19, and posterior BA 21, and is inclusive of the middle and inferior temporal gyri, and the anterior portion of the medial

205 occipital gyrus.

A large portion of the occipitotemporal cortex is connected with other cortical and subcortical targets through the inferior longitudinal fasciculus which occupies the lateral aspect of the inferior fronto-occipital fasciculus (Nieuwenhuys et al., 1988;Burgel et al., 1999;Makris et al., 1999). These white matter tracts arise from the GTs, GTm, GTi, and GF and go on to project to the cuneus, and the lateral surface of the occipital lobe (Catani et al., 2002). While longer fibers connect parahippocampal and visual association areas, shorter fibers connect to the hippocampus and extend into the most inferior level of the external capsule, and radiate around the parahippocampal gyrus and uncus (Makris et al., 1999). Short fibers between V1, V2, V3, and the ventral occipitotemporal visual association cortex promote analysis of face, shape, form, color, and place. Short fibers also project dorsally to the inferior parietal lobe and intraparietal sulcus where the parietal higher visual association takes place (Nieuwenhuys et al., 1988;Aralasmak et al., 2006).

The influence of the occipitotemporal cortex on affective processing was noted before the advent of functional neuroimaging. In a patient admitted with bilateral occipitotemporal hematomas secondary to head trauma, a number of neuropsychological impairments were noted. Along with profound prosopagnosia, topographical memory disturbance, a modality-specific inability to become emotionally aroused by visual cues was also reported. Continued analysis of neurological, neuropsychological, and psychophysiological assessment of the patient's deficits were interpreted as indicating a visual-limbic disconnection syndrome (Bauer, 1982).

This region has been described in functional neuroimaging investigations as being at the cross- roads of an extended facial expression processing and an object recognition centre (Bly and Kosslyn, 1997;Barton, 2003). Affective neuroscience investigations have confirmed the activation of the area in response to the viewing of emotionally-salient stimuli, including infants by mothers (Ranote et al., 2004). A comparison between the anticipation and processing of affective IAPS stimuli, has recently confirmed activation of the occipitotemporal cortex in the processing, as opposed to the anticipation of affect (Bermpohl et al., 2006b).

A possible explanation for the consistent activation of the occipitotemporal area in response to affective processing is that it is a reflection of increased visual processing of affective imagery or

206 facial expressions. Evidence against this hypothesis, however, comes from 18FGF PET investigation of the neural correlates of the empathetic response. Shamay-Tsoory and colleagues obtained measures of glucose metabolism from 6 HC subjects during an interview about neutral story themes, and again during an empathic response eliciting interview about a character in distress. In addition to metabolic increases in the medial and superior frontal gyri, thalamus, and cerebellum, increased activity in the occipitotemporal cortex was also noted during the empathic response compared to the neutral theme interview (Shamay-Tsoory et al., 2005).

The strongest support for the role of the occipitotemporal cortex in modulating the presentation of processing, however, comes from studies of evoked response potentials (ERPs). A repetition paradigm was used to assess the nature of affective modulation of early and late components of the ERP during picture viewing. In one investigation, high-density ERPs were measured while HC subjects passively viewed affective or neutral pictures that were repeated up to 90 times each. Both early (occipitotemporal) and late (centroparietal) components of the ERP were modulated by emotional arousal, with differential ERPs being elicited during the viewing of affective stimuli compared to neutral visual stimuli. When the effect of repetition was assessed, the early occipitotemporal component showed a decrease within a block of repetitions. The late centroparietal component additionally displayed both within- and between- block decreases with greater deactivation towards neutral stimuli (Codispoti et al., 2007)

In a separate investigation, the processing resources allocated to the presentation of affective stimuli from the IAPS were evaluated with steady-state visual evoked brain potentials (ssVEP). Visual stimuli were divided into six picture categories: threat and mutilation (unpleasant), families and erotica (pleasant), and household objects and persons (neutral), and were presented in a flickering mode at 10 Hz in order to elicit ssVEPs (Morgan et al., 1996). While the presentation of all affective stimuli was associated with enhanced ssVEP amplitude at parieto- occipital recording sites, increased coactivation of right occipitotemporal sources was only reported during the processing of affectively arousing stimuli (Keil et al., 2003).

In accordance with ERP findings and ratings of general positive affect, we have identified decreased activation to positive stimuli in the cortex in NR and RS compared to HC. Moreover, in contrast to a time-dependent increase in activation to positive stimuli observed in the HC group, there were attenuated increases in the RS group, and no significant baseline-endpoint

207 change in the processing of positive IAPS stimuli in the NR group. Interestingly, baseline activation during positive stimuli in the right OPT regions was able to predict eventual response to treatment, with higher activation noted in eventual responders.

The opposite time-dependent pattern was observed for negative affect processing. Although initially high bilateral activation of the OPT cortex was recorded in response to negative IAPS stimuli in the HC and RS groups, this activation decayed 2-3 fold over time. In contrast, persistent hyperactivation of the OPT region was noted in the NR group throughout the trial.

Premotor Cortex

Based on cytoarchitectural differences between the agranular cortex with large pyramidal cells in the anterior bank of the precentral sulcus, and the agranular cortex in the precentral gyrus and the posterior portion of the superior frontal gyrus, Brodmann differentiated BA 4 and 6 respectively (Brodmann, 1909). Evidence from cortical lesions experiments in non-human primates and clinical observations led to the division of the motor cortex into the primary motor area (BA 4) and the premotor area (BA 6) (Fulton, 1935). Based on the basis of the presence of large pyramidal cells, the dorsal premotor area can be further subdivided into rostral and caudal components (Barbas and Pandya, 1987). Whereas the caudal the dorsal premotor cortex has strong connections with motor cortex and is positioned to influence the generation of movements, the rostral PMd has strong connections with the prefrontal cortex and selects responses based on arbitrary and spatial cues (Lu et al., 1994;Chouinard and Paus, 2006).

Neuroimaging investigations have additionally highlighted the involvement of the dorsal precentral gyrus in a number of non-motor cognitive tasks, including information retrieval from short-term memory (Koch et al., 2006), working memory maintenance (Zarahn et al., 2005), and manipulation of working memory (Tsukiura et al., 2001). Through the analysis of brain activity during the execution of two fundamental cognitive tasks - same-different discrimination, and integer computation, Hirsch and colleagues demonstrated that both of these two tasks activated the right precentral gyrus (BA 6).

In regards to the processing of negative affect, the rostral border of area 6 separates the "motor domain" of the supplementary motor/premotor cortex from the "cognitive domain" of the

208 prefrontal cortex, and as such, may be activated in the preparation of negative-affect induced withdrawal (Davidson et al., 1990). Stimulation of the polysensory zone (PLZ) in the precentral gyrus of the monkey brain evokes defensive-like withdrawing or blocking movements. It has been suggested that the PLZ in conjunction with the intraparietal area (VIP) participates in a range of functions including the construction of a margin of safety around the body and the selection and coordination of defensive behaviour (Graziano and Cooke, 2006). In support of this hypothesis, are reports that attention to threatening stimuli, where the self is viewed as an active intentional agent, involves activation of the left precentral gyrus (BA 6).

Neuroimaging reports from HC groups further report decreases in activation of the precentral gyrus during the performance of emotionally arousing memory tasks (Mather et al., 2006;Mitchell et al., 2006). The induction of negative affect in HC has been associated with decreased activation of the precentral gyrus, compared to positive affect inductions (Teasdale et al., 1999;Kumari et al., 2003;Tremblay et al., 2005;Keedwell et al., 2005a;Baumgartner et al., 2006). Previously, Fu and colleagues reported depressive state-independent increased activations in the left precentral gyrus (compared to HC subject), in response to sad facial expressions (Fu et al., 2004). A similar pattern of precentral gyrus hyperactivity in a psychiatric cohort was observed in subjects with obsessive-compulsive disorder during a visual symptom provocation paradigm (Mataix-Cols et al., 2004).

Changes in activation in the premotor cortex may also reflect subjects subconsciously planning an appropriate facial expression commensurate with the affective state. Recently Kim and colleagues examined the neural correlates of facial expression planning under fMRI scanning conditions (Blum et al., 1950;Pribram and Barry, 1956;Mesulam et al., 1977;Pandya and Seltzer, 1982;Petrides and Pandya, 1984;Goldman-Rakic, 1988;Leichnetz, 2001). The investigators noted that imagining of emotional facial affect versus the imagining of neutral facial affects was associated with increased activation in the amygdala, the dlPFC, the lingual gyrus, the parahippocampal gyrus, and the ventral premotor cortex.

In summary, increased activation within the precentral gyrus in response to an affective induction may be conceptualized as a future movement plan; towards positive stimuli in HC and away from negative stimuli in MDD populations. As the frontal eye fields are located in the precentral gyrus, and are involved in the preparation and the triggering of purposive saccades

209 (Gaymard et al., 1998), an alternative explanation may be that visual presentations of an affect recruit greater visual scrutiny.

Precuneus

As a consequence of its deep location and the absence of well characterized lesion studies, the precuneus has not traditionally received attention by the affective neuroscience community (Cavanna and Trimble, 2006). The precuneus is well defined with an anterior boundary defined by the cingulate sulcus, a posterior border delimited by the parieto-occipital fissure, and an inferior border by the subparietal sulcus (Critchley, 1953). A large number of historical and recent investigations have characterized the connectivity of the precuneus in non-human primates (Blum et al., 1950;Pribram and Barry, 1956;Mesulam et al., 1977;Pandya and Seltzer, 1982;Petrides and Pandya, 1984;Goldman-Rakic, 1988;Leichnetz, 2001). The precuneus has extensive connections with a number of other parietal cortical targets, the extraparietal cortex and subcortical targets.

Within the parietal cortex the precuneus has established bilateral connections with the posterior cingulate and retrosplenial cortex. The primary extraparietal cortical projections of the precuneus are directed towards the frontal lobe, with projections concentrating in the dlPFC, corresponding to BA 8, 9 and 46. Additional connections between the precuneus and the frontal eye fields in the supplementary motor area may be related to the involvement of the precuneus in guiding eye- movements (Thier and Andersen, 1993;Thier and Andersen, 1998). Additional extraparietal connections with portions of the extrastriate cortex and the temporo-parieto-occipital cortex may be responsible for a heteromodal higher associative network involved in the integration of auditory, somatosensory and visual information (Cavanna and Trimble, 2006). Subcortical targets and origins of precuneus connectivity include the dorsal thalamic nuclei that are connected with higher association cortices, but not the sensory thalamic nuclei (Yeterian and Pandya, 1985;Yeterian and Pandya, 1988). Efferent connections with the caudate nucleus and putamen, and oculomotor structures of the brainstem have also been described (Yeterian and Pandya, 1993).

Functional neuroimaging investigations have implicated the precuneus in four related processes; i) visuo-spatial imagery (Satoh et al., 2001;Wenderoth et al., 2005), ii) episodic memory retrieval

210 (Gilboa et al., 2004;Addis et al., 2004), and iii) self-referential processing (den Ouden et al., 2005;Vogeley et al., 2004;Ochsner et al., 2004), iv). All of these processes are also measure of consciousness. Indeed, the precuneus exhibits one of the highest resting metabolic rates, consuming approximately 35% more glucose than any other area of the cerebral cortex (Gusnard et al., 2001b). Moreover, activity of the precuneus is decreased during altered states of consciousness including sleep (Maquet et al., 1996;Maquet et al., 1997;Braun et al., 1997) and hypnosis (Rainville et al., 1999).

The presentation of affective stimuli may indeed increase levels of general and visual attention, induce the recall of appropriately valenced memory, and self-referential comparisons. Evidence from neuroimaging investigations evaluating the processing of affect induced by the presentation of affective imagery alone (Beauregard et al., 1998;Grimm et al., 2006;Kumari et al., 2003), or in combination with music (Baumgartner et al., 2006) supports activation of the precuneus. Furthermore, preliminary evidence suggests that a negative bias in affective processing is represented in the activation profile of the precuneus.

Results of hyperactivation of the precuneus during negative affect processing in MDD compared to HC, its sensitivity to antidepressant treatment, and persistence into the euthymic state have all been previously reported. Fu and colleagues noted that the presentation of facial expressions of negative affect leads to a hyperactivation of the precuneus in MDD groups compared to HC, while the presentation of happy facial expression elicits the opposite pattern (Fu et al., 2004;Fu et al., 2007). Similar patterns of hyper/hypo activation have also been noted when changes in affect are induced using the IAPS (Robertson et al., 2007;Tremblay et al., 2005). Moreover, a normalization of the aberrant affective processing in the precuneus has been reported following antidepressant treatment (Fu et al., 2004;Fu et al., 2007;Robertson et al., 2007).

Cerebellum

In the late 19th century, Babinski observed that patients with cerebellar lesions could not properly execute complex motor tasks and named the resulting condition “dysmetria.” Andreasen and colleagues (Andreasen et al., 1996) have reported that disruption of neural circuits linking the cortex, thalamus and cerebellum (the cortico-thalamic-cerebellar-cortical circuit, or CTCCC) may presage the complex psychopathology of schizophrenia. They hypothesized that the CTCCC

211 monitors and coordinates the fluid execution of mental activity, a process that appears to be aberrant in schizophrenia. Structural and functional cerebellar abnormalities have also been described in psychiatric disorders other than schizophrenia, including anxiety depressive and bipolar disorders (Phillips et al., 2003b).

The cerebellum overlies the posterior aspect of the pons and projects bidirectional fibres to brainstem structures via 3 paired peduncles. The midline (vermis) and lateral hemispheres are demarcated by fissures into smaller lobes and lobules. Inside, pairs of intrinsic nuclei — dentate (most lateral), emboliform, globose and fastigial — can be found under the grey cortical mantle within a medullary core of white matter (Gilman, 1996). The quintessential role of the cerebellar cortex is essentially to modulate information flowing through the deep nuclei. Sensorimotor information is carried into the deep nuclei by mossy and climbing fibers, from there it travels to premotor areas, allowing a fine-regulation of motor control. In addition, the mossy and climbing fibers also feed some of this information in parallel to the cerebellar cortex for additional computation. The output of the cerebellar cortex in turn regulates Purkinje cell firing, which have an inhibitory role on the deep nuclei (Kingsley, 2000).

Traditionally, the emphasis of studies on cerebellar function has been on the coordination of somatic motor function, control of muscle tone and equilibrium. The cerebellum, however, receives input directly and indirectly (via projections from cortical association areas and the midbrain) from nearly all sensory receptors; its output systems emanate from the cerebellar nuclei, and their influences upon cortical function are mediated primarily through brainstem nuclei at multiple levels (Schmahmann, 2000). Connections between the cerebellum and the nonmotor cortical and subcortical areas have been documented through both electrophysiologic studies and anatomic tracing techniques (Schmahmann, 2000). The cerebellum shares bidirectional connections with a large portion of the limbic lobe and the associated subcortical nuclei, the amygdaloid complex, the septal nuclei, and various hypothalamic and thalamic nuclei, regions of interest to psychiatry through their association with emotional processing (Papez, 1937). Furthermore, the cerebellum also communicates with the monoamine-producing brainstem nuclei, which supply the limbic system and the cerebrum with serotonin, norepinephrine and dopamine.

In animal studies, axonal transport mechanisms have been used to document synaptic contact

212 between the fastigial nucleus of the cerebellum and the ventral tegmental area, the periaqueductal gray, the locus ceruleus and the pontine raphe (Snider and Maiti, 1976;Noda et al., 1990). Conversely, the ventral tegmental area and the mammillary body have been shown to project back to the monkey cerebellum (Haines and Dietrichs, 1984;Oades and Halliday, 1987). Functionally, electrical stimulation of the cerebellum, particularly stimulation of the vermis area 3 (V3) and the fastigial nucleus, modulates the physiology of limbic lobe structures. Evoked responses in the orbitomesial cortex, sACC, amygdala, hippocampal and dentate gyri, pyriform and preamygdaloid cortical regions and hypothalamus have been recorded through stimulation of the cerebellum (Anand et al., 1959;Heath et al., 1978;Vilensky and van Hoesen, 1981). Thus, the cerebellum earns its Latin name of “miniature brain,” as it appears armed with the connections to “instruct” and “report” to a large proportion of the regions of the brain, including those involved in cognition, affect and mood regulation.

The first accounts of cerebellar atrophy and agenesis, which appeared early in the 19th century, described patients with intellectual, emotional, social and other behavioural responses that were distorted to the point of irreversible character and personality alteration (Schmahmann, 1991). By the mid 20th century, descriptions of dementia and psychosis in patients suffering from cerebellar degeneration began to appear (Schut, 1950), leading Snider to hypothesize that cerebellar activity must also influence the “non-motor centres of the cerebrum.” (Snider, 1950)

Therapeutic interventions involving the cerebellum began appearing in the 1970s, in particular the implanting of electrodes over the superior surface of the cerebellum in an attempt to control epilepsy (Riklan et al., 1974). In addition to improvements in seizure control, the procedure resulted in emotional improvements in aggression, anxiety and depression. Conversely, when recording from the cerebellar fastigial nucleus of an emotionally disturbed patient, Heath and colleagues (Heath, 1977) described a direct relationship between increased neuronal discharges and the patient’s experience of fear and anger. Similarly, when Nashold and Slaughter (Nashold, Jr. and Slaughter, 1969) stimulated the cerebellar dentate nucleus and superior cerebellar peduncle, the subject experienced an unpleasant sensation of fear. In 11 patients with severe emotional dyscontrol, the implantation of bilateral electrodes for stimulation over the superior aspect of the cerebellar cortex resulted in “extraordinary” improvements in behaviour (Heath, 1977). Although these studies were limited in number and were not controlled, they provide evidence that direct stimulation of the cerebellum has the potential to alter moods or induce

213 different moods in humans.

When Schmahmann and Sherman (Schmahmann and Sherman, 1998) subsequently analyzed the behavioural impairment of 20 adult patients with lesions confined to the cerebellum, they noted that behavioural changes were most prominent in patients whose lesions were localized to the posterior lobe of the cerebellum and vermis. The clinical presentation included a combination of passivity and flattening or blunting of emotion, sometimes occurring simultaneously with disinhibited, inappropriately jocular, silly or child-like behaviour. The authors named this newly defined clinical entity “cerebellar cognitive affective syndrome” and postulated that it resulted from impaired cerebellar modulation of neural circuits that link prefrontal, posterior parietal, superior temporal and limbic cortices with the cerebellum.

Interestingly, these displays of passivity and flattening or blunting of emotion and a disinhibition of restraint are phenotypically similar to the depressed and manic states in mood disorders. This presentation may also resemble some of the classical symptoms of schizophrenia, although schizophrenia is also associated with more severe cognitive impairments. Similarly, children with mutism secondary to cerebellar tumours also display altered moods (Levisohn et al., 2000). In particular, lesions of the vermis produced behavioural changes that extended beyond the cognitive domain, including a flattening of affect and a silly, disinhibited, regressive quality to the children’s interactions, with some exhibiting a reduced tolerance of others and a general tendency to avoid physical and eye contact. Disruption of the cerebellar circuitry may thus impair the processing of emotional responses to challenging stimuli. Furthermore, the finding that single lesions of the cerebellum can impart such a marked change in the personality of affected individuals highlights the role of cerebellar interconnectivity in affective and cognitive processing.

In a previously discussed PET investigation evaluating neuroanatomic correlates of externally generated emotions (Reiman et al., 1997), increased CBF into the limbic and paralimbic areas was accompanied by activation of the cerebellar hemispheres. Using a similar technique, Lane et al (Lane et al., 1997b) extended these results by demonstrating that sadness, but not happiness, increased activation of the anterior cerebellar vermis. Use of fMRI and visual-inductions of affect has also reliably induced activation of the cerebellum (Baumgartner et al., 2006;Bermpohl et al., 2006b;Eugene et al., 2003;Grimm et al., 2006;Lee et al., 2004).

214

In contradistinction to the relevant body of morphometric studies examining cerebellar structure in schizophrenia, there are very few studies examining cerebellar size in unipolar depression. Early MRI studies (Shah et al., 1992;Escalona et al., 1993) showed reduced cerebellar size in patients with unipolar depression, whereas a recent quantitative MRI investigation failed to find any statistically significant differences (Pillay et al., 1997). This MRI investigation (Pillay et al., 1997) did reveal, however, that in patients who did not respond to fluoxetine treatment, there was an inverse correlation between total cerebellar tissue volume and baseline depression scores.

In an effort to discern the differential brain activation pattern that results from evoking sadness in healthy control subjects and patients with unipolar depression, Beauregard and colleagues (Beauregard et al., 1998) performed fMRI scans while both groups viewed an emotionally laden film. Transient sadness produced significant activation in both groups, not only in the medial and inferior prefrontal cortices and the middle temporal cortex, but also in the cerebellum. Furthermore, the patients with depression displayed greater activation of the left medial prefrontal cortex and the right ACC, but less activation of the cerebellum. Similar differences in the activation of the cerebellum have been reported in comparisons of HC and MDD subjects (Fu et al., 2004;Fu et al., 2007;Canli et al., 2004;Keedwell et al., 2005a)

In a previously reviewed PET study, cerebral blood flow was compared in subjects with acute depression and healthy controls, before and after a transient mood challenge. In line with the results obtained with fMRI, the patients with depression displayed less activation of the cerebellum and thalamus (Liotti et al., 2002). To test whether subjects who recover from depression show abnormal brain activity, Smith et al (Smith et al., 2002) acquired fMRI data during a conditioning paradigm with a noxious pain stimulus. Although similar patterns of brain activation during painful stimulation were found for both patients and healthy controls, subjects who had recovered from depression displayed less cerebellar activation than the control subjects during anticipation of the noxious stimulus. These findings suggest that depression may impart a permanent and irreversible change in cerebellar function.

More recent data suggest that this tonic increase in cerebellar activity is characteristic of major depression, regardless of mood state or medication history (Videbech et al., 2002;Kimbrell et al., 2002). Similar to the findings in schizophrenia, in which cerebellar blood flow decreases

215 following antipsychotic treatment, (Loeber et al., 1999;Miller et al., 2001) an association between successful treatment of depression and a decrease in cerebellar perfusion has been reported (Davies et al., 2003;Fu et al., 2004;Fu et al., 2007).

Limitations

A fundamental limitation of any fMRI experimental design is that the results obtained cannot validate the necessity of a brain region’s involvement in the execution of a desired cognitive task (Brown et al., 2007). Preclinical animal data of classical conditioning indicate that even though some neurons may be activated by a particular task, damage to these cells, does not alter the conditioning (Berger, 1984). Moreover, the relationship between antidepressant use and the coupling between oxygen tissue demands and cerebrovascular response has not been explored. Additional limitations may relate to a smaller sample size, a blocked experimental design, and the use of univariate statistics.

Sample Size

The primary limitations of this study relate to the sample size, the attrition of subjects during the study, the absence of parametric modulation of affective stimuli, and the use of an unvalidated scale to assess magnitude of affective induction outside the scanning environment. Additional limitations are inherent to the fundamental principles of statistical parametric mapping, classical inference, and univariate statistics.

For an epoch design fMRI experiment, analysis of power curves, suggests that a minimum of 12 subjects is required to detect signal changes of 0.5% with 80% power at the p<0.05 significance level (Desmond and Glover, 2002). However, when more stringent thresholds are applied (i.e. multiple comparison correction), more subjects may be required to yield similar power. In a comparison of power estimates in event-related fMRI, Murphy and Garavan concluded that even though power was poor in fMRI studies evaluating less than 20 subjects, the majority of activated clusters were true positives with satisfactory voxelwise correlations with the gold standard (0.60–0.75) (Berger, 1999) . Although our total samples size exceeds this minimum, subject attrition particularly in the NR group, prevented us from conducting all our analyses with 80% power, particularly Primary Aim B.

216

To address this limitation, a fixed-effects design was also considered as, it would have increased the ability to detect statistically significant changes in brain activation, however, this would limits our inference to only those subjects who participated in this study and increases the chance of false-positives. To address the possibility that we used too conservative extent and height thresholds, we tried to reduce our height and extent threshold, but did not see any additional activation/deactivation, until we began to see artifacts outside the brain and in white matter.

During negative affect induction, it was possible that the presentation of the first negative block to study participants elicited maximal and persistent activation throughout the entire neutral- negative run. To evaluate this possibility, a specific contrast was setup to identify any brain regions where activation peaked at the first block, and then remained elevated for the duration of the negative-neutral run. Although there were no brain regions that met statistical significance criteria, it is possible that we were underpowered to detect these early ‘ceiling’ effects.

In spite of identifying differences in brain activation and deactivation during NAI and PAI, our comparison of HC and remitted MDD subjects was also underpowered. Moreover, the comparison, may not give an accurate portrait of the trait component of disturbances in negative affect, as both groups were previously primed by prior exposure to the same stimuli. This was particularly valid in NAI in the HC, where a number of prior activations became deactivations during the final scan. An ideal comparison, would involve NAI and PAI in ‘virgin’ HC and remitted MDD subjects, who would undergo the affective induction for the first time. We believe, however, that our approach has ecological validity, as it represents repeat exposure to affectvive stressors that are frequently encountered in MDD.

Finally, we did not monitor the intensity of induced affect ‘real-time’ during the scanning session, to ensure that the BOLD signal reflected the passive induction of an affect, rather than an admixture of the perception, and the self-referential cognitive process of affective evaluation. Although the APRS instrument employed has not been validated, it has successfully been used in previous neuroimaging investigations evaluating MDD subjects.

Statistical Parametric Mapping

217 Statistical parametric mapping has a number of fundamental limitations; most are centered on the general linear model and classical inference. For example, the calculated P value of an effect does not describe the likelihood that the effect is present, but simply the probability of obtaining the observed data in the effect’s absence (Friston et al., 2002a). If sufficiently small, the P value can be used as evidence to reject the null hypothesis that the effect is negligible. The challenge arises that one can never reject the alternate hypothesis: that activation due to a factor has not occurred. As a result, if multiple factors are being evaluated, one could not claim that a brain region responds to factor A, but not factor B.

Moreover, due to the very definition of the student’s t or F statistic, given enough scans or subjects one can theoretically always demonstrate a significant effect at every voxel by reducing the variability of the activation’s estimate. Finally, in an effort to resolve the problem of multiple comparisons, P values are corrected based on the size, or number, of brain regions being evaluated. Thus, the inference about one region of the brain is contingent on whether another part is being examined (Friston et al., 2002b;Friston, 2005).

One approach to overcoming these limitations of classical inference, is to evaluate the probability that a voxel had activated, or indeed its activation was greater than some threshold (Berger, 1999). The probability distribution of the activation given the data, is based on Bayesian inference where evidence or observations are used to iteratively update the probability that a hypothesis is true (Friston et al., 2002a). This posterior probability requires the likelihood, based on assumptions about the distribution of errors, and the prior probability of activation (Labatut et al., 2004;Descombes et al., 1998;Everitt and Bullmore, 1999;Woolrich et al., 2004). A number of investigators have proposed means of calculating these or estimating them from the data, provided there are sufficient observed multiple instances of the effect (Petersson et al., 1999).

Univariate Statistics

Although neuroscience has established functional specialization as a principle of brain organization, the integration of specialized areas has proven difficult to assess with conventional neuroimaging univariate approaches (Petersson et al., 1999). Functional integration thus must rely on multivariate analyses to model interactions among different brain regions (Sharma and Sharma, 2004). The modelling of both functional connectivity (correlations among remote

218 neurophysiological events) and effective connectivity (the influence that one neural system exerts over another) has received increasing attention during the last decade (Friston, 2005).

Although a description of multivariate neuroimaging analytical approaches is beyond the scope of this thesis, they can be broadly divided into those that are inferential, (probing effective connectivity), and those that are exploratory (based on the concept of functional connectivity)(Friston et al., 1993). Examples of exploratory multivariate analyses include principal component analysis (PCA) (McKeown et al., 1998), independent component analysis (ICA) (Calhoun et al., 2001), temporal –ICA (Baumgartner et al., 1997), and cluster analysis (Friston, 2005). Overcoming the limitations of exploratory multivariate functional connectivity, structurally bound inferential multivariate analyses aim to evaluate the coupling among brain areas and how that coupling is influenced by changes in experimental context (Penny et al., 2004). Variation of dynamic causal modelling (DCM) and structural equation modelling (SEM) have been the most popular analytical methods used in inferential multivariate neuroimaging analyses.

In order to enable comparison with extant investigations, a decision was made to employ statistical parametric mapping with a univariate data analysis plan. A number of implementations of multivariate statistics in the study of mood disorders, specifically the neurocircuitry mediating affective processing, have been recently competed. An overview of these findings is presented in the next section, along with possibilities for follow-up investigations.

Multivariate Analysis

Functional connectivity is a measure of spatiotemporal correlations between spatially distinct regions of the cerebral cortex (Lowe et al., 1998;Friston et al., 1993). In an evaluation of more than 100 depressed patients, differences in connectivity between the dlPFC cortex (BA9), the subgenual cingulate (BA25), and the orbitofrontal cortex (BA11) were found between pharmacotherapy responders and non-responders (Seminowicz et al., 2004). Meanwhile, limited limbic-cortical and cortical-cortical connectivity differentiated CBT responders from pharmacotherapy responders. Moreover, the finding that MDD-NRS demonstrated abnormalities in limbic-subcortical pathways involving the rostral (BA24) and subgenual (BA25) ACC may account for some divergent prediction response finding in the ACC (i.e. hypo/hyper metabolism)

219 (Evans et al., 2006;Chen et al., 2007;Siegle et al., 2006;Wu et al., 1999;Mayberg et al., 1997;Little et al., 1996;Brody et al., 2001a).

Evidence from fMRI investigations has provided confirmation of PET reports of increased subgenual cingulate hyperactivity in MDD. Using fMRI functional connectivity of default-mode network in depressed and control subjects, Greicius and colleagues noted greater subgenual cingulate-thalamic functional connectivity in the depressed cohort (Greicius et al., 2007). Moreover, the authors noted a positive correlation between the length of the current depressive episode and the functional connectivity in the subgenual cingulate.

The effect of antidepressant exposure and repeat affective processing on functional connectivity was recently evaluated using a conventional block-design experiment using correlations of low- frequency BOLD fluctuations (LFBF) during continuous exposure to neutral, positive, and negative pictures in HC and MDS subjects. At baseline, decreased corticolimbic LFBF correlations compared to HC subjects were noted in MDD subjects during the resting state and with exposure to affective stimuli. Following treatment, LFBF correlation between the ACC and limbic regions was significantly increased in MDD subjects, on exposure to neutral and positive pictures and at rest, while upon exposure to negative affective stimuli, corticolimbic LFBF correlations remained decreased (Anand et al., 2005b).

Event-Related Design

Two common experimental designs in fMRI investigations are blocked (or epoch) and event- related designs. While in an event-related design, stimuli of different types are intermixed, in a blocked design stimuli belonging to the same category are presented in a continuous block. Differential analyses are also used in the evaluation of data obtained with each experimental design. While effects of interest in blocked designs are modeled as boxcar (ON-OFF-ON-OFF) regressor convolved with a predicted hemodynamic response function, effects of interest in event-related designs are modeled as an impulse following the trial onset. One of the limitations of the blocked design is the inherent assumption that the BOLD signal reflects steady-state hemodynamic response resulting from neuronal activity in each block (Mechelli et al., 2003).

It has been proposed that for cognitive or perceptual experiments event-related designs may be

220 preferable due to the fact that they limit predictability and habituation (Liu et al., 2001), as the presented stimuli are short in duration and presented in random order with short interstimulus intervals (Friston et al., 1999). This may be particularly relevant in the induction of changes in affect, as time-dependent habituation of response has been reported (Kalin et al., 1997;Davidson et al., 2003;Schaefer et al., 2006;Adolphs et al., 1996;Fu et al., 2004;Fu et al., 2007;Strauss et al., 2005b;Siegle et al., 2002). Nevertheless, we elected to employ a blocked design to take advantage of the increased power to detect regional activity (sensitivity) that blocked designs offer (Liu et al., 2001;Friston et al., 1999;Buxton et al., 2000) and because of the experimental objectives. Our goals of characterizing affective processing necessitated the prolonged exposure to different affective stimuli.

Recent developments have combined the advantages of event-related and epoch designs (Amaro E Jr and Barker, 2006;Laurienti et al., 2003). Use of these designs may be able to address some of the problems comparing HC and MDD groups with fMRI, where persistent activation in MDD subjects to negative stimuli may persist into the neutral conditions (Siegle et al., 2002).

Parametric Design

A subtraction paradigm, where the stimulus is on at one point, and off at another, may not optimally isolate a specific stage of cognitive, or affective, processing, as this assumes that a cognitive process can be inserted into a series of cognitive stages, without altering the function of any processes involved (Townsend and Ashby, 1983;Friston et al., 1996). Parametric designs offer another approach to increase the power to detect statistically significant differences in brain activation through the presentation of different levels (or intensities) of the experimental factor. The conversion of the regressor from a binary to continuous variable permits the evaluation of the hypothesis that a linear association may exist between activation in a particular brain region, and the intensity of the presented stimulus. More specific to the evaluation of affective processing in MDD, use of this experimental design permits the evaluation of not only maximal activation (capacity), but also the range of activations corresponding to different stimulus intensities (dynamic range).

Through computer manipulation of a standardised series of facial expressions of happiness, sadness, fear, and disgust (Young et al., 2002), a number of investigators have probed the

221 dynamic range of affective processing through the presentation of different intensities of the target affective state in the HC population (Surguladze et al., 2003b) and MDD populations (Chen et al., 2007;Fu et al., 2004;Fu et al., 2007;Surguladze et al., 2005;Surguladze et al., 2004). Results from these investigations emphasize that delimiting evaluations to stimuli that elicit maximal activation, may obscure the delineation of finer-scale changes in the available response range. For example, evaluation of dynamic range, in addition to capacity, uncovered additional brain regions affected by antidepressant treatment (Fu et al., 2004;Fu et al., 2007).

As each visual IAPS stimulus is rated based on valence and arousal, a similar implementation may be possible with the IAPS set. Unfortunately, as our efforts were focused on maximizing the ability to detect regional brain activation, the stimuli employed did not cover the entire range of positive and negative affective spectrum. A longer scanning session may have permitted the use of the fuller IAPS affective spectrum; however, this would have come at the expense of subject fatigue and increased intra-scanner movement.

BOLD Signal Interpretation

The presented neuroimaging data are interpreted in the context of affective processing functionality based on preclinical animal data, human lesion case studies, and a rich database of neuroimaging studies in psychiatrically unaffected subjects. Nevertheless, it is possible that the observed changes in BOLD signal may represent alternative neuropsychological phenomena. Moreover, in the absence of accompanying blood perfusion and resting-state glucose metabolism data, additional caution in the interpretation of HC- MDD group differences is warranted (Dunn et al., 2005;Conca et al., 2000). Recent fMRI investigations in MDD have combined the use of BOLD fMRI along with arterial spin labelling (ASL) to distinguish between differences in oxygen consumption and differences in overt blood flow and volume (Roiser et al., 2009).

We also did not evaluate changes in monoaminergic receptor or transporter availability before and after treatment initiation. Statistically significant changes in serotonin transporter and receptor binding potential have been reported to accompany SSRI treatment (Meyer, 2008). Finally, the presented data represent group means, and may be unable to adequately reflect within group variability in BOLD signal and/or changes in small brain regions with large

222 anatomical variations (Eugene et al., 2003). These limitations, however, may also assist in the interpretation of activation clusters with a relatively low spatial extent as representing functional and/or anatomical variability.

Clinical Limitations

In addition to the aforementioned methodological limitations, a number of clinical limitations must also be acknowledged that limit the generalizability of the results. First, the clinical sample that was evaluated was derived from a tertiary medical centre, whereas most MDD patients are treated in primary care settings. Therefore, the results may reflect a more severe, or treatment refractory MDD population. Second, only one modality of antidepressant treatment was employed for all MDD subjects. Therefore, we cannot generalize the effects of therapeutic improvement to other forms of pharmacotherapy, or alternative psychosocial, or neurostimulatory interventions. Whereas, there is a growing research interest in the neuropsychological and neuroimaging effects of SSRI and other second generation antidepressants on affective processing, there is limited data on the effect of atypical antipsychotics on affective processing in non schizophrenic patients. Third, the pharmacotherapy that was employed in this thesis is not standard first-line treatment for non-treatment resistant MDD. Moreover, the use of a combination therapy for all MDD subjects precluded an analysis dissecting the neuromodulatory potential of each compound separately and in concert. Nevertheless, based on the clinical profile of the patients that are treated at the MDPU (treatment refractory with psychiatric comorbidity), the positive clinical data for non treatment refractory MDD patients with the olanzapine-fluoxetine combination, and routine laboratory assays, the treatment was effective in providing treatment response to more than half of the MDD subjects.

Finally, our results for MDD subjects meeting remission criteria must also be interpreted with caution. Previous neuroimaging investigations that have evaluated remitted MDD subjects had frequently observed asymptomatic status for an extended period prior to the neuroimaging appointment. In contrast, the remitted MDD subjects in this thesis were ‘newly-remitted.’ It is quite likely that continuous and prolonged reduction of depressive symptoms may be necessary for optimal remission.

223 CONCLUSIONS

PRIMARY AIM A: EVALUATE DIFFERENCES IN BRAIN ACTIVATION EVOKED BY AFFECTIVE PROCESSING BETWEEN HC AND MDD, AND BETWEEN RS AND NR

Both PAI and NAI experimental protocol evoked statistically significant changes in brain activation between the affective and neutral conditions. Consistent with our hypothesis, the induction of positive affect was associated with increased activation by RS compared to NR in brain regions associated with the identification of positive affect stimuli including activation in the vmPFC (BA 10/11), and in monitoring of the affective state including the dmPFC (BA 9). The NR group was also characterized by decreased visual attention as indicated by decreased activation in both the ventral (BA 19/37) and dorsal visual streams (BA 7/31/39). Consistent with out hypotheses, there was a delay in activation of the ventral striatum in the NR group. Contrary to our hypothesis, there were no differences in BOLD signal between HC and RS during PAI, suggesting that even in the depressed state, RS, versus NR, have the capacity to experience positive affect.

Higher levels of negative dispositional affect were noted in RS and NR versus HC, with NR being characterized by even higher levels than RS. Consistent with our hypothesis, both MDD groups were characterized by increased activation in brain regions responsible for the identification and production of the affective state, with increased activation in the amygdala and the ventrolateral PFC; with higher BOLD signal in the NR (versus RS) group. Consistent with previous investigations and our hypotheses, NR were distinguished from RS with increased BOLD signal in the pACC (BA 24), suggesting that activity in this region can predict treatment response before treatment initiation.

224 PRIMARY AIM B: EVALUATE DIFFERENCES IN INTER-SCAN HABITUATION ON BRAIN ACTIVATION EVOKED BY REPEAT AFFECTIVE PROCESSING BETWEEN HC, RS, AND NR

Repeat scanning sessions were associated with statistically significant changes in brain activation. Comparable levels of induced negative and positive affect were generated within- and between- groups. Changes in brain activation in the RS group were the result of i) inter-scan habituation, ii) dispositional affect normalization (increases in positive affect and decreases in negative affect), iii) decreases in symptom severity, iv) and pharmacology of the medications.

Consistent with our hypothesis, repeat inductions of positive affect was associated with increased visual evaluation of the presented stimuli by the HC group. Contrary to our hypothesis, we did not see increases in BOLD signal in brain regions responsible for the identification of affective value and the production of the affective state in RS. Instead, inter-scan changes during PAI were limited to increased visual evaluation of the presented stimuli.

Just as the initial induction of negative affect was associated with significantly more regions of activation compared to positive affect, the repeat induction of negative affect was also associated with significantly more decreases in activation. Consistent with our hypotheses based on resting state metabolic studies SSRI treatment studies and fMRI neuroimaging investigations using repeat presentation of IAPS, decreased activation was noted in brain regions involved in the identification of the affective state including the ventromedial PFC (BA 10) and production of the affective state in the ventrolateral PFC (BA 10). Consistent with our hypotheses, inter-scan habituation in the NR group was associated with decreased activation of brain regions responsible for the identification of affective state including the anterior insula.

225 SECONDAY AIM A: EVALUATE THE ASSOCIATION BETWEEN DEPRESSIVE SYMPTOM REDUCTION AND BRAIN ACTIVATION EVOKED BY REPEAT AFFECTIVE PROCESSING IN MDD

At study endpoint, significant decreases in depressive symptom severity were recorded in the MDD group. Consistent with our hypothesis, decreases in depressive symptom severity were associated with increased activation evoked by PAI of brain regions responsible for the identification and production of the affective state, with increased activation in the ventral striatum and anterior insula. In addition, we report that decreases in depressive symptom severity are also associated with increased activation of brain regions in the retrieval of memory (medial temporal lobe), thereby enabling the MDD subjects to relate the presented visual stimuli with their own past experiences. Unfortunately, even in a state of clinical remission, activation of medial temporal lobe was still not as robust as activity in the HC group.

Consistent with our hypothesis, decreases in depressive symptom severity were associated with decreased activation during NAI of brain regions involved in the identification of the affective state, with decreased activation in the ventromedial PFC (BA 10). Contrary to our hypothesis, we did not observe decreased activation of the amygdala during NAI with decreased depressive symptoms. In a state of clinical remission, depressed subjects still, however, continued to manifest increased activation during NAI in brain regions involved in the production of the affective state, they also spent more time evaluating the negative visual stimuli. These two neuropsychological processes may help to explain the vulnerability to relapse in remitted MDD patients, and should be considered as integral targets of psychosocial and pharmacological treatment interventions. Future investigations should compare brain activation during affective processing in patients treated with both treatment modalities.

226 SECONDARY AIM B: EVALUATE THE ASSOCIATION BETWEEN DISPOSITIONAL AND INDUCED AFFECT AND BRAIN ACTIVATION EVOKED BY AFFECTIVE PROCESSING

At study endpoint, significant decreases in negative dispositional affect and increases in positive dispositional affect were recorded in the MDD group, while constant scores were maintained by the HC group. Consistent with our hypothesis, increases in positive dispositional affect were correlated with brain activation during PAI and NAI. We found that increases in positive affect were associated with increased visual evaluation of the positive visual stimuli, and greater activation of autobiographical memory with increased activation in the hippocampus. These findings suggest that depressive symptoms and positive affect may complementarily modulate brain activation during PAI. Whereas decreases in depressive symptom severity are associated with increased activation of brain regions responsible for the identification of positive affect, increases in positive dispositional affect may be associated with additional activation in brain regions that allow depressed subjects to evaluate positive stimuli and compare it to their own previous experiences.

Consistent with our hypothesis, decreases in negative dispositional affect were also found to modulate brain activation during both NAI and PAI. Decreases in negative dispositional affect were found to have the same effect on brain activation during PAI as increases in positive dispositional affect, with increased visual evaluation of the stimuli and greater activation of autobiographical memory. Decreases in negative dispositional affect were also associated with decreased brain activation during NAI in regions that modulate the identification, production, and regulation of the affective state. These findings imply that the measurement of negative dispositional affect may be a better marker of abnormal brain activation during NAI in MDD. Moreover, these results strongly suggest that monitoring dispositional affect in depressed patients maybe a more accurate measure of therapeutic benefit and protection from future relapse.

227 FUTURE DIRECTIONS

The opportunity to follow-up with future investigation allows the opportunity to address some of the previously listed limitations of the original study, and to further pursue questions brought about by current research findings. From a clinical perspective, the research findings provide a neurophysiological rationale for the effects of antidepressant, and a neurobiological basis for continued vulnerability to relapse in an asymptomatic individual. As such, they provide a future starting point for relapse prevention.

Although indices of glucose metabolism and cerebral perfusion have been quantified following the administration of manual-based psychosocial interventions, their effect on the neural correlates of affective processing have yet to be elucidated. Based on the top-down antidepressant mode of action of psychotherapy (Mayberg, 2003b;Goldapple et al., 2004;Seminowicz et al., 2004), it may be hypothesized that administration of psychotherapy may be associated with decreased residual hyper-reactivity to negative stimuli in a euthymic MDD subject. Indeed, extant evidence suggests that the use of cognitive therapy may have a more powerful relapse prevention profile when compared to pharmacotherapy (Hollon et al., 2005), and that this protective effect may be mediated by decreased affective lability (Segal et al., 2006).

The combination of a documented negative bias in MDD and the difficulty of reliably eliciting changes in positive affect in the HC population has presented affective neuroscience with a difficult task of studying the anhedonic basis of MDD. Although the three reviewed visual induction strategies summarized earlier - happy facial expressions, positive words, and pleasant photographs from the IAPS - have been used to induce mood in MDD, alternative forms have also been investigated. The use of gambling or reward tasks may prove effective in the study of reward circuitry abnormalities in anhedonia (Forbes et al., 2006). Alternatively pharmacological manipulations of dopaminergic neurotransmission under neuroimaging conditions represent another possible avenue (Tremblay et al., 2005).

The combination of multimedia presentations may enhance the induction of positive affect. In a series of investigations, Keedwell and colleagues have used a visual and a verbal cue to trigger changes in positive and negative affect (Keedwell et al., 2005b;Keedwell et al., 2005a) in an

228 MDD population. Similarly, the combined presentation of visual and auditory stimuli was found to enhance the conscious and neural response to changes in affect (Baumgartner et al., 2006).

Based on results from the recently completed The Sequenced Treatment Alternatives to Relieve Depression (STAR*D) trial, response and remission rates to first-line pharmacotherapy are suboptimal (Rush, 2007). Similarly, the heterogeneous clinical portrait of MDD suggests that differential symptomatic profiles that may have differential response to different classes of antidepressant and other psychotropic augmenting agents. For example, subjects meeting MDD criteria may meet additional specifier criteria for MDE with catatatonic features; melancholic features; atypical features; and psychotic features (American Psychiatric Association, 1994).

It may be hypothesized that medications that preferentially target one neurotransmitter system may offer enhanced efficacy in controlling a specific symptom domain. Future evaluations should strive to define the neural correlates of not only aberrant affective processing, but the neural correlates of MDD-specific cognitive impairment (Ebmeier et al., 2006) and somatic complaints (Mee et al., 2006). The elucidation of specific neural pathways mediating the specific symptom clusters provides an organizational framework against which pharmacological hypotheses can be evaluated. In other words, the differential efficacy of a specific antidepressant treatment on a specific symptom can be established not only with clinical rating scales, but with functional neuroimaging evidence.

Using a cost-effectiveness model it was recently concluded that although effective treatment of depression is cost effective, the evidence of a medical or productivity cost offset for depression treatment remains equivocal (Donohue and Pincus, 2007). In other words, although current treatment algorithms are effective in reducing symptom severity, they may not be effective enough to restore social and employment functionality. In light of these socioeconomic trends, future investigations of biological correlates of impaired cognitive, affective, or somatic processing in MDD need to be conducted with an emphasis on how they affect the restoration of functionality in subjects with MDD.

A number of recent neuroimaging investigations in MDD have begun to make this transition. For example, two recent investigations have investigated the neural correlates of affective regulation in MDD groups. In the first study, Beauregard and colleagues provided explicit instruction that

229 MDD and HC subjects voluntarily down-regulate sad feelings (Beauregard et al., 2006). In another study, the effect of emotional distractors on the performance of an attentional executive function task was investigated before and after antidepressant treatment (Robertson et al., 2007). Both of these investigations present MDD subjects with situations they are likely to encounter on a daily basis that demand cognitive control over affective content.

The neural correlates of the BOLD signal have been well characterized in the non-human primate with compelling evidence in the human brain. As the BOLD signal ultimately depends on cerebrovascular redistribution, one of the tacit assumptions in neuroimaging investigations comparing HC and MDD samples is that cerebrovascular dynamics are comparable between the two populations. Recent evidence suggests an uncoupling between glucose metabolism and cerebral perfusion in MDD groups, that was not apparent in either HC or BD age-matched comparisons (Dunn et al., 2005). Adding additional interpretative caution are findings that functional neuroimaging changes may have to be corrected for regional differences in gray matter content (Drevets et al., 1997).

The macrocellular and subcellular elements putatively mediating this uncoupling of metabolism and perfusion deserve further attention, as they form the neurophysiological basis for most modern neuroimaging techniques. The coupling of perfusion and metabolism is believed to be regulated by both short-term, dynamic mechanisms involving vasoactive factors, and long-term, static mechanisms involving the distribution and density of capillary beds (Duelli and Kuschinsky, 2001). At the cell membrane level, changes in ionic gradients, particularly extracellular potassium (Dietzel et al., 1989), and intracellular calcium (Mohamed et al., 1985) may be important in the coupling. It is interesting that a number of antidepressant and psychotropic augmenting agents that are used in the treatment of MDD have a direct effect on cerebrovasculature (Parsons, 1991;Baserga et al., 2005;Luiten et al., 1996). At the subcellular level, changes in mitochondrial coupling may additionally impair coupling, as changes in mitochondrial function have been reported in MDD (Gardner et al., 2003;Vawter et al., 2006).

In summary, future neuroimaging investigations in MDD should employ activation paradigms that address specific impairments and barriers to full functional restoration. An attempt should be made to explore individual symptom profiles, the use of different antidepressant treatment modalities, and the delineation of the physiological basis of perfusion-metabolism uncoupling,

230 and the role of gray matter density differences. Continued refinement of multi-modality scanning systems may be particularly instrumental in the realization of these long-term goals (Raylman et al., 2006).

231 REFERENCES

Reference List

Abercrombie HC, Schaefer SM, Larson CL, Oakes TR, Lindgren KA, Holden JE, Perlman SB, Turski PA, Krahn DD, Benca RM, Davidson RJ (1998) Metabolic rate in the right amygdala predicts negative affect in depressed patients. Neuroreport 9:3301-3307.

Addis DR, McIntosh AR, Moscovitch M, Crawley AP, McAndrews MP (2004) Characterizing spatial and temporal features of autobiographical memory retrieval networks: a partial least squares approach. Neuroimage 23:1460-1471.

Adler RJ (1981) The Geometry of Random Fields. New York: Wiley.

Adolphs R (2002a) Neural systems for recognizing emotion. Curr Opin Neurobiol 12:169-177.

Adolphs R (2002b) Recognizing emotions from facial expressions: psychological and neurological mechanisms. Behav Cognit Neurosci Rev 1:21-61.

Adolphs R, Damasio H, Tranel D, Damasio AR (1996) Cortical systems for the recognition of emotion in facial expressions. J Neurosci 16:7678-7687.

Adolphs R, Tranel D, Damasio H, Damasio A (1994) Impaired recognition of emotion in facial expressions following bilateral damage to the human amygdala. Nature 372:669-672.

Aggleton JP, Brown MW (1999) Episodic memory, amnesia, and the hippocampal-anterior thalamic axis. Behav Brain Sci 22:425-444.

Aggleton JP, Mishkin M (1990) Visual impairments in macaques following inferior temporal lesions are exacerbated selectively by additional damage to superior temporal sulcus. Behav Brain Res 39:262-274.

Aikens JE, Kroenke K, Nease DE, Jr., Klinkman MS, Sen A (2008) Trajectories of improvement for six depression-related outcomes. Gen Hosp Psychiatry 30:26-31.

Alexander GE, Crutcher MD, DeLong MR (1990) Basal ganglia-thalamocortical circuits: parallel substrates for motor, oculomotor, "prefrontal" and "limbic" functions. Prog Brain Res 85:119-146.

Alexander GE, DeLong MR, Strick PL (1986) Parallel organization of functionally segregated circuits linking basal ganglia and cortex. Annu Rev Neurosci 9:357-381.

Alexopoulos GS, Katz IR, Bruce ML, Heo M, Ten Have T, Raue P, Bogner HR, Schulberg HC, Mulsant BH, Reynolds CF, III (2005) Remission in depressed geriatric primary care patients: a report from the PROSPECT study. Am J Psychiatry 162:718-724.

Alheid GF, Heimer L (1988) New perspectives in basal forebrain organization of special relevance for neuropsychiatric disorders: the striatopallidal, amygdaloid, and corticopetal components of substantia innominata. Neuroscience 27:1-39.

232 Amaral DG, Price JL (1984) Amygdalo-cortical projections in the monkey (Macaca fascicularis). J Comp Neurol 230:465-496.

Amaro E Jr, Barker GJ (2006) Study design in fMRI: basic principles. Brain Cogn 60:220-232.

American Psychiatric Association (1994) Diagnostic and Statistical Manual of Mental Disorders. Washington, DC: American Psychiatric Press Inc.

Ames A, III (2000) CNS energy metabolism as related to function. Brain Res Brain Res Rev 34:42-68.

Amsterdam JD, Garcia-Espana F, Fawcett J, Quitkin FM, Reimherr FW, Rosenbaum JF, Schweizer E, Beasley C (1998) Efficacy and safety of fluoxetine in treating bipolar II major depressive episode. J Clin Psychopharmacol 18:435-440.

Amsterdam JD, Shults J, Brunswick DJ, Hundert M (2004) Short-term fluoxetine monotherapy for bipolar type II or bipolar NOS major depression - low manic switch rate. Bipolar Disord 6:75-81.

Amunts K, Schleicher A, Burgel U, Mohlberg H, Uylings HB, Zilles K (1999) Broca's region revisited: cytoarchitecture and intersubject variability. J Comp Neurol 412:319-341.

Anand A, Li Y, Wang Y, Wu J, Gao S, Bukhari L, Mathews VP, Kalnin A, Lowe MJ (2005a) Activity and connectivity of brain mood regulating circuit in depression: a functional magnetic resonance study. Biol Psychiatry 57:1079-1088.

Anand A, Li Y, Wang Y, Wu J, Gao S, Bukhari L, Mathews VP, Kalnin A, Lowe MJ (2005b) Antidepressant effect on connectivity of the mood-regulating circuit: an FMRI study. Neuropsychopharmacology 30:1334-1344.

Anand BK, Malhorta CL, Singh B, Dua S (1959) Cerebellar projections to the limbic system. J Neurophysiol 22:451-458.

Ances BM, Liang CL, Leontiev O, Perthen JE, Fleisher AS, Lansing AE, Buxton RB (2009) Effects of aging on cerebral blood flow, oxygen metabolism, and blood oxygenation level dependent responses to visual stimulation. Hum Brain Mapp 30:1120-1132.

Anders S, Birbaumer N, Sadowski B, Erb M, Mader I, Grodd W, Lotze M (2004) Parietal somatosensory association cortex mediates affective blindsight. Nat Neurosci 7:339-340.

Anderson AK, Christoff K, Panitz D, De RE, Gabrieli JD (2003) Neural correlates of the automatic processing of threat facial signals. J Neurosci 23:5627-5633.

Anderson AK, Phelps EA (2002) Is the human amygdala critical for the subjective experience of emotion? Evidence of intact dispositional affect in patients with amygdala lesions. J Cogn Neurosci 14:709-720.

Anderson IM, Del-Ben CM, McKie S, Richardson P, Williams SR, Elliott R, Deakin JF (2007) Citalopram modulation of neuronal responses to aversive face emotions: a functional MRI study. Neuroreport 18:1351-1355.

233 Andreasen NC, O'Leary DS, Cizadlo T, Arndt S, Rezai K, Ponto LL, Watkins GL, Hichwa RD (1996) Schizophrenia and cognitive dysmetria: a positron-emission tomography study of dysfunctional prefrontal-thalamic-cerebellar circuitry. Proc Natl Acad Sci U S A 93:9985-9990.

Andreoli V, Caillard V, Deo RS, Rybakowski JK, Versiani M (2002) Reboxetine, a new noradrenaline selective antidepressant, is at least as effective as fluoxetine in the treatment of depression. J Clin Psychopharmacol 22:393-399.

Angelini L, Mazzucchi A, Picciotto F, Nardocci N, Broggi G (1981) Focal lesion of the right cingulum: a case report in a child. J Neurol Neurosurg Psychiatry 44:355-357.

Anisman H, Merali Z (2003) Cytokines, stress and depressive illness: brain-immune interactions. Ann Med 35:2-11.

Aouizerate B, Guehl D, Cuny E, Rougier A, Bioulac B, Tignol J, Burbaud P (2004) Pathophysiology of obsessive-compulsive disorder: a necessary link between phenomenology, neuropsychology, imagery and physiology. Prog Neurobiol 72:195-221.

Applegate CD, Kapp BS, Underwood MD, McNall CL (1983) Autonomic and somatomotor effects of amygdala central N. stimulation in awake rabbits. Physiol Behav 31:353-360.

Aralasmak A, Ulmer JL, Kocak M, Salvan CV, Hillis AE, Yousem DM (2006) Association, commissural, and projection pathways and their functional deficit reported in literature. J Comput Assist Tomogr 30:695-715.

Arce E, Simmons AN, Lovero KL, Stein MB, Paulus MP (2008) Escitalopram effects on insula and amygdala BOLD activation during emotional processing. Psychopharmacology (Berl) 196:661-672.

Arnsten AF, Goldman-Rakic PS (1984) Selective prefrontal cortical projections to the region of the locus coeruleus and raphe nuclei in the rhesus monkey. Brain Res 306:9-18.

Aron A, Fisher H, Mashek DJ, Strong G, Li H, Brown LL (2005) Reward, motivation, and emotion systems associated with early-stage intense romantic love. J Neurophysiol 94:327-337.

Ashburner J, Friston KJ (1999) Nonlinear spatial normalization using basis functions. Hum Brain Mapp 7:254-266.

Ashburner J, Neelin P, Collins DL, Evans A, Friston K (1997) Incorporating prior knowledge into image registration. Neuroimage 6:344-352.

Ashtari M, Greenwald BS, Kramer-Ginsberg E, Hu J, Wu H, Patel M, Aupperle P, Pollack S (1999) Hippocampal/amygdala volumes in geriatric depression. Psychol Med 29:629-638.

Attenburrow MJ, Williams C, Odontiadis J, Reed A, Powell J, Cowen PJ, Harmer CJ (2003) Acute administration of nutritionally sourced tryptophan increases fear recognition. Psychopharmacology (Berl) 169:104-107.

Attwell D, Iadecola C (2002) The neural basis of functional brain imaging signals. Trends Neurosci 25:621-625.

234 Attwell D, Laughlin SB (2001) An energy budget for signaling in the grey matter of the brain. J Cereb Blood Flow Metab 21:1133-1145.

Augustine JR (1996) Circuitry and functional aspects of the insular lobe in primates including humans. Brain Res Brain Res Rev 22:229-244.

Austin MP, Dougall N, Ross M, Murray C, O'Carroll RE, Moffoot A, Ebmeier KP, Goodwin GM (1992) Single photon emission tomography with 99mTc-exametazime in major depression and the pattern of brain activity underlying the psychotic/neurotic continuum. J Affect Disord 26:31-43.

Axelson DA, Doraiswamy PM, McDonald WM, Boyko OB, Tupler LA, Patterson LJ, Nemeroff CB, Ellinwood EH, Jr., Krishnan KR (1993) Hypercortisolemia and hippocampal changes in depression. Psychiatry Res 47:163-173.

Bagby RM, Ryder AG, Schuller DR, Marshall MB (2004) The hamilton depression rating scale: has the gold standard become a lead weight? Am J Psychiatry 161:2163-2177.

Bagshaw MH, Kimble DP, Pribram KH (1965) The GSR of monkeys during orienting and habituation and after ablation of the amygdala, hippocampus and inferotemporal cortex. pp 381- 414.

Baillarger J (1840) Recherches sur la structure de la couche corticale des circonvolutions du cerveau. Mem Acad R Med 8:149-183.

Baird J, Lewis J (1964) THE EFFECTS OF COCAINE, AMPHETAMINE AND SOME AMPHETAMINE-LIKE COMPOUNDS ON THE IN VIVO LEVELS OF NORADRENALINE AND DOPAMINE IN THE RAT BRAIN. Biochem Pharmacol 13:1475-1482.

Ballmaier M, Sowell ER, Thompson PM, Kumar A, Narr KL, Lavretsky H, Welcome SE, DeLuca H, Toga AW (2004a) Mapping brain size and cortical gray matter changes in elderly depression. Biol Psychiatry 55:382-389.

Ballmaier M, Toga AW, Blanton RE, Sowell ER, Lavretsky H, Peterson J, Pham D, Kumar A (2004b) Anterior cingulate, gyrus rectus, and orbitofrontal abnormalities in elderly depressed patients: an MRI-based parcellation of the prefrontal cortex. Am J Psychiatry 161:99-108.

Bancaud J, Talairach J (1992) Clinical semiology of frontal lobe seizures. Adv Neurol 57:3-58.

Bandettini PA, Wong EC, Hinks RS, Tikofsky RS, Hyde JS (1992) Time course EPI of human brain function during task activation. Magn Reson Med 25:390-397.

Bar M (2003) A cortical mechanism for triggering top-down facilitation in visual object recognition. J Cogn Neurosci 15:600-609.

Barbee JG, Conrad EJ, Jamhour NJ (2004) Aripiprazole augmentation in treatment-resistant depression. Ann Clin Psychiatry 16:189-194.

Barrett SL, Bell R, Watson D, King DJ (2004) Effects of amisulpride, risperidone and chlorpromazine on auditory and visual latent inhibition, prepulse inhibition, executive function and eye movements in healthy volunteers. J Psychopharmacol 18:156-172. 235 Bartels A, Zeki S (2000) The neural basis of romantic love. Neuroreport 11:3829-3834.

Bartolo A, Benuzzi F, Nocetti L, Baraldi P, Nichelli P (2006) Humor comprehension and appreciation: an FMRI study. J Cogn Neurosci 18:1789-1798.

Barton JJ (2003) Disorders of face perception and recognition. Neurol Clin 21:521-548.

Baserga MC, Bertolotto C, Sola A (2005) Different doses of dopamine have heterogeneous effects on cerebral hemodynamics and dopamine receptors in young rabbits as measured with near infrared spectroscopy. Biol Neonate 87:229-235.

Bauer RM (1982) Visual hypoemotionality as a symptom of visual-limbic disconnection in man. Arch Neurol 39:702-708.

Baumgartner R, Scarth G, Teichtmeister C, Somorjai R, Moser E (1997) Fuzzy clustering of gradient-echo functional MRI in the human visual cortex. Part I: reproducibility. J Magn Reson Imaging 7:1094-1101.

Baumgartner T, Lutz K, Schmidt CF, Jancke L (2006) The emotional power of music: how music enhances the feeling of affective pictures. Brain Res 1075:151-164.

Baxter LR, Jr., Phelps ME, Mazziotta JC, Schwartz JM, Gerner RH, Selin CE, Sumida RM (1985) Cerebral metabolic rates for glucose in mood disorders. Studies with positron emission tomography and fluorodeoxyglucose F 18. Arch Gen Psychiatry 42:441-447.

Baxter LR, Jr., Schwartz JM, Phelps ME, Mazziotta JC, Guze BH, Selin CE, Gerner RH, Sumida RM (1989) Reduction of prefrontal cortex glucose metabolism common to three types of depression. Arch Gen Psychiatry 46:243-250.

Beauregard M, Leroux JM, Bergman S, Arzoumanian Y, Beaudoin G, Bourgouin P, Stip E (1998) The functional neuroanatomy of major depression: an fMRI study using an emotional activation paradigm. Neuroreport 9:3253-3258.

Beauregard M, Levesque J, Bourgouin P (2001) Neural correlates of conscious self-regulation of emotion. J Neurosci 21:RC165.

Beauregard M, Paquette V, Levesque J (2006) Dysfunction in the neural circuitry of emotional self-regulation in major depressive disorder. Neuroreport 17:843-846.

Bechara A, Damasio H, Tranel D, Anderson SW (1998) Dissociation Of working memory from decision making within the human prefrontal cortex. J Neurosci 18:428-437.

Bechara A, Tranel D, Damasio H, Adolphs R, Rockland C, Damasio AR (1995) Double dissociation of conditioning and declarative knowledge relative to the amygdala and hippocampus in humans. Science 269:1115-1118.

Bechara A, Tranel D, Damasio H, Damasio AR (1996) Failure to respond autonomically to anticipated future outcomes following damage to prefrontal cortex. Cereb Cortex 6:215-225.

Beck AT, Ward CH, Mendelson M, Mock J, Erbaugh J (1961) An inventory for measuring depression. Arch Gen Psychiatry 4:561-571.

236 Bell-McGinty S, Butters MA, Meltzer CC, Greer PJ, Reynolds CF, III, Becker JT (2002) Brain morphometric abnormalities in geriatric depression: long-term neurobiological effects of illness duration. Am J Psychiatry 159:1424-1427.

Belmaker RH (2008) The future of depression psychopharmacology. CNS Spectr 13:682-687.

Bench CJ, Frackowiak RS, Dolan RJ (1995) Changes in regional cerebral blood flow on recovery from depression. Psychol Med 25:247-261.

Bench CJ, Friston KJ, Brown RG, Frackowiak RS, Dolan RJ (1993) Regional cerebral blood flow in depression measured by positron emission tomography: the relationship with clinical dimensions. Psychol Med 23:579-590.

Benuzzi F, Lui F, Duzzi D, Nichelli PF, Porro CA (2008) Does it look painful or disgusting? Ask your parietal and cingulate cortex. J Neurosci 28:923-931.

Berger JO (1999) Statistical Decision Theory and Bayesian Statistics. New York: Springer Verlag.

Berger TW (1984) Neural representation of associative learning in the hippocampus. In: Neuropsychology of memory (Squire LR, Butters N, eds), pp 443-461. New York: The Guilford Press.

Bermpohl F, Pascual-Leone A, Amedi A, Merabet LB, Fregni F, Gaab N, Alsop D, Schlaug G, Northoff G (2006a) Attentional modulation of emotional stimulus processing: an fMRI study using emotional expectancy. Hum Brain Mapp 27:662-677.

Bermpohl F, Pascual-Leone A, Amedi A, Merabet LB, Fregni F, Gaab N, Alsop D, Schlaug G, Northoff G (2006b) Dissociable networks for the expectancy and perception of emotional stimuli in the human brain. Neuroimage 30:588-600.

Bernstein HG, Krell D, Baumann B, Danos P, Falkai P, Diekmann S, Henning H, Bogerts B (1998) Morphometric studies of the entorhinal cortex in neuropsychiatric patients and controls: clusters of heterotopically displaced lamina II neurons are not indicative of schizophrenia. Schizophr Res 33:125-132.

Besag J (1974) Spatial interaction and the statistical analysis of lattice systems. J Royal Statistical Society B 36:192-225.

Betz W (1874) Anatomischer Nachweis zweier Gehirncentra. In: Centralblatt fur die medicinischen Wissenschaften pp 578-580.

Biber MP, Kneisley LW, LaVail JH (1978) Cortical neurons projecting to the cervical and lumbar enlargements of the spinal cord in young and adult rhesus monkeys. Exp Neurol 59:492- 508.

Bingel U, Quante M, Knab R, Bromm B, Weiller C, Buchel C (2003) Single trial fMRI reveals significant contralateral bias in responses to laser pain within thalamus and somatosensory cortices. Neuroimage 18:740-748.

Bingel U, Schoell E, Herken W, Buchel C, May A (2007) Habituation to painful stimulation 237 involves the antinociceptive system. Pain 131:21-30.

Birkhoff GD (1943) SIR JOSEPH LARMOR AND MODERN MATHEMATICAL PHYSICS. Science 97:77-79.

Birn RM, Murphy K, Handwerker DA, Bandettini PA (2009) fMRI in the presence of task- correlated breathing variations. Neuroimage.

Bishop GH, O'Leary JL (1942) Factors determining the form of the potential record in the vicinity of the synapses of the dorsal nucleus of the lateral geniculate body. J Cell Comp Physiol 19:315-331.

Biswal BB, Kannurpatti SS, Rypma B (2007) Hemodynamic scaling of fMRI-BOLD signal: validation of low-frequency spectral amplitude as a scalability factor. Magn Reson Imaging 25:1358-1369.

Biver F, Goldman S, Delvenne V, Luxen A, De M, V, Hubain P, Mendlewicz J, Lotstra F (1994) Frontal and parietal metabolic disturbances in unipolar depression. Biol Psychiatry 36:381-388.

Blair RJ, Morris JS, Frith CD, Perrett DI, Dolan RJ (1999) Dissociable neural responses to facial expressions of sadness and anger. Brain 122 ( Pt 5):883-893.

Blanchard DC, Blanchard RJ (1972) Innate and conditioned reactions to threat in rats with amygdaloid lesions. J Comp Physiol Psychol 81:281-290.

Blier P, de Montigny C (1994) Current advances and trends in the treatment of depression. Trends Pharmacol Sci 15:220-226.

Blier P, Ward NM (2003) Is there a role for 5-HT1A agonists in the treatment of depression? Biol Psychiatry 53:193-203.

Blum J, CHOW KL, Pribram KH (1950) A behavioral analysis of the organization of the parieto- temporo-preoccipital cortex. J Comp Neurol 93:53-100.

Blumberg HP, Kaufman J, Martin A, Whiteman R, Zhang JH, Gore JC, Charney DS, Krystal JH, Peterson BS (2003a) Amygdala and hippocampal volumes in adolescents and adults with bipolar disorder. Arch Gen Psychiatry 60:1201-1208.

Blumberg HP, Leung HC, Skudlarski P, Lacadie CM, Fredericks CA, Harris BC, Charney DS, Gore JC, Krystal JH, Peterson BS (2003b) A functional magnetic resonance imaging study of bipolar disorder: state- and trait-related dysfunction in ventral prefrontal cortices. Arch Gen Psychiatry 60:601-609.

Bly BM, Kosslyn SM (1997) Functional anatomy of object recognition in humans: evidence from positron emission tomography and functional magnetic resonance imaging. Curr Opin Neurol 10:5-9.

Bobo WV, Shelton RC (2009) Fluoxetine and olanzapine combination therapy in treatment- resistant major depression: review of efficacy and safety data. Expert Opin Pharmacother 10:2145-2159.

238 Bonte FJ, Trivedi MH, Devous MD, Sr., Harris TS, Payne JK, Weinberg WA, Haley RW (2001) Occipital brain perfusion deficits in children with major depressive disorder. J Nucl Med 42:1059-1061.

Borkowska A, Rybakowski JK (2001) Neuropsychological frontal lobe tests indicate that bipolar depressed patients are more impaired than unipolar. Bipolar Disord 3:88-94.

Borod J (1993) Cerebral mechanisms underlying facial, prosodic and lexical emotional expression : a review of neuropsychological studies and methodological issues. Neuropsychology 7:445-463.

Borod JC (1992) Interhemispheric and intrahemispheric control of emotion: a focus on unilateral brain damage. J Consult Clin Psychol 60:339-348.

Botteron KN, Raichle ME, Drevets WC, Heath AC, Todd RD (2002) Volumetric reduction in left subgenual prefrontal cortex in early onset depression. Biol Psychiatry 51:342-344.

Botvinick M, Nystrom LE, Fissell K, Carter CS, Cohen JD (1999) Conflict monitoring versus selection-for-action in anterior cingulate cortex. Nature 402:179-181.

Bowley MP, Drevets WC, Ongur D, Price JL (2002) Low glial numbers in the amygdala in major depressive disorder. Biol Psychiatry 52:404-412.

Boxerman JL, Bandettini PA, Kwong KK, Baker JR, Davis TL, Rosen BR, Weisskoff RM (1995) The intravascular contribution to fMRI signal change: Monte Carlo modeling and diffusion-weighted studies in vivo. Magn Reson Med 34:4-10.

Braak H (1976) A primitive gigantopyramidal field buried in the depth of the cingulate sulcus of the human brain. Brain Res 109:219-223.

Bradley MM, Cuthbert BN, Lang PJ (1996) Picture media and emotion: effects of a sustained affective context. Psychophysiology 33:662-670.

Braun AR, Balkin TJ, Wesenten NJ, Carson RE, Varga M, Baldwin P, Selbie S, Belenky G, Herscovitch P (1997) Regional cerebral blood flow throughout the sleep-wake cycle. An H2(15)O PET study. Brain 120 ( Pt 7):1173-1197.

Brazdil M, Roman R, Urbanek T, Chladek J, Spok D, Marecek R, Mikl M, Jurak P, Halamek J, Daniel P, Rektor I (2009) Neural correlates of affective picture processing--a depth ERP study. Neuroimage 47:376-383.

Breiter HC, Etcoff NL, Whalen PJ, Kennedy WA, Rauch SL, Buckner RL, Strauss MM, Hyman SE, Rosen BR (1996) Response and habituation of the human amygdala during visual processing of facial expression. Neuron 17:875-887.

Breiter HC, Gollub RL, Weisskoff RM, Kennedy DN, Makris N, Berke JD, Goodman JM, Kantor HL, Gastfriend DR, Riorden JP, Mathew RT, Rosen BR, Hyman SE (1997) Acute effects of cocaine on human brain activity and emotion. Neuron 19:591-611.

Bremmer F, Schlack A, Duhamel JR, Graf W, Fink GR (2001) Space coding in primate posterior parietal cortex. Neuroimage 14:S46-S51. 239 Bremner JD (1984) Fluoxetine in depressed patients: a comparison with imipramine. J Clin Psychiatry 45:414-419.

Bremner JD (2002) Structural changes in the brain in depression and relationship to symptom recurrence. CNS Spectr 7:129-130.

Bremner JD, Innis RB, Salomon RM, Staib LH, Ng CK, Miller HL, Bronen RA, Krystal JH, Duncan J, Rich D, Price LH, Malison R, Dey H, Soufer R, Charney DS (1997) Positron emission tomography measurement of cerebral metabolic correlates of tryptophan depletion-induced depressive relapse. Arch Gen Psychiatry 54:364-374.

Bremner JD, Narayan M, Anderson ER, Staib LH, Miller HL, Charney DS (2000) Hippocampal volume reduction in major depression. Am J Psychiatry 157:115-118.

Bremner JD, Vythilingam M, Ng CK, Vermetten E, Nazeer A, Oren DA, Berman RM, Charney DS (2003) Regional brain metabolic correlates of alpha-methylparatyrosine-induced depressive symptoms: implications for the neural circuitry of depression. JAMA 289:3125-3134.

Bremner JD, Vythilingam M, Vermetten E, Nazeer A, Adil J, Khan S, Staib LH, Charney DS (2002) Reduced volume of orbitofrontal cortex in major depression. Biol Psychiatry 51:273-279.

Brett M (1999) The MNI brain and the Talairach atlas.

Britton JC, Taylor SF, Sudheimer KD, Liberzon I (2006) Facial expressions and complex IAPS pictures: common and differential networks. Neuroimage 31:906-919.

Broca P (1878) Anatomie comparee des circonvolutions cerebrales. Le grand lobe limbique et la scissure limbique dans la serie des mammiferes. Rev Antrhop 1:385-498.

Brodal A (1969) Neurological Anatomy in Relation to Clinical Medicine. New York: Oxford University Press.

Brodmann K (1905) Beiträge zur histologischen Lokalisation der Grosshirnrinde: dritte Mitteilung: Die Rindenfelder der niederen Affen. Journal für Psychologie und Neurologie 4:177- 206.

Brodmann K (1909) Vergleichende Lokalisationslehre der Grosshirnrinde in ihren Prinzipien dargestellt auf Grund des Zellenbaues. Leipzig: JA Barth.

Brody AL, Saxena S, Mandelkern MA, Fairbanks LA, Ho ML, Baxter LR (2001a) Brain metabolic changes associated with symptom factor improvement in major depressive disorder. Biol Psychiatry 50:171-178.

Brody AL, Saxena S, Silverman DH, Alborzian S, Fairbanks LA, Phelps ME, Huang SC, Wu HM, Maidment K, Baxter LR, Jr. (1999) Brain metabolic changes in major depressive disorder from pre- to post-treatment with paroxetine. Psychiatry Res 91:127-139.

Brody AL, Saxena S, Stoessel P, Gillies LA, Fairbanks LA, Alborzian S, Phelps ME, Huang SC, Wu HM, Ho ML, Ho MK, Au SC, Maidment K, Baxter LR, Jr. (2001b) Regional brain metabolic changes in patients with major depression treated with either paroxetine or interpersonal therapy: preliminary findings. Arch Gen Psychiatry 58:631-640. 240 Brooks JC, Nurmikko TJ, Bimson WE, Singh KD, Roberts N (2002) fMRI of thermal pain: effects of stimulus laterality and attention. Neuroimage 15:293-301.

Brothers L, Ring B (1993) Mesial temporal neurons in the macaque monkey with responses selective for aspects of social stimuli. Behav Brain Res 57:53-61.

Brown ES, Varghese FP, McEwen BS (2004) Association of depression with medical illness: does cortisol play a role? Biol Psychiatry 55:1-9.

Brown GG, Perthen JE, Liu TT, Buxton RB (2007) A primer on functional magnetic resonance imaging. Neuropsychol Rev 17:107-125.

Brugue E, Vieta E (2007) Atypical antipsychotics in bipolar depression: neurobiological basis and clinical implications. Prog Neuropsychopharmacol Biol Psychiatry 31:275-282.

Buchanan SL, Powell DA (1993) Cingulothalamic and prefrontal control of autonomic function. In: Neurobiology of Cingulate Cortex and Limbic Thalamus: A Comprehensive Handbook (Vogt B, Gabriel M, eds), pp 381-414. Boston, Mass.: Birkhauser.

Buchel C, Dolan RJ, Armony JL, Friston KJ (1999) Amygdala-hippocampal involvement in human aversive trace conditioning revealed through event-related functional magnetic resonance imaging. J Neurosci 19:10869-10876.

Buchsbaum MS, Cappelletti J, Ball R, Hazlett E, King AC, Johnson J, Wu J, DeLisi LE (1984a) Positron emission tomographic image measurement in schizophrenia and affective disorders. Ann Neurol 15 Suppl:S157-S165.

Buchsbaum MS, DeLisi LE, Holcomb HH, Cappelletti J, King AC, Johnson J, Hazlett E, Dowling-Zimmerman S, Post RM, Morihisa J, . (1984b) Anteroposterior gradients in cerebral glucose use in schizophrenia and affective disorders. Arch Gen Psychiatry 41:1159-1166.

Buchsbaum MS, Wu J, Siegel BV, Hackett E, Trenary M, Abel L, Reynolds C (1997) Effect of sertraline on regional metabolic rate in patients with affective disorder. Biol Psychiatry 41:15-22.

Bullier J (2001) Integrated model of visual processing. Brain Res Brain Res Rev 36:96-107.

Bunney WE, Jr., Brodie HK, Murphy DL, Goodwin FK (1971) Studies of alpha-methyl-para- tyrosine, L-dopa, and L-tryptophan in depression and mania. Am J Psychiatry 127:872-881.

Burgel U, Schormann T, Schleicher A, Zilles K (1999) Mapping of histologically identified long fiber tracts in human cerebral hemispheres to the MRI volume of a reference brain: position and spatial variability of the optic radiation. Neuroimage 10:489-499.

Burt DB, Zembar MJ, Niederehe G (1995) Depression and memory impairment: a meta-analysis of the association, its pattern, and specificity. Psychol Bull 117:285-305.

Bush G, Luu P, Posner MI (2000) Cognitive and emotional influences in anterior cingulate cortex. Trends Cogn Sci 4:215-222.

Buxton RB (2002) Introduction to functional magnetic resonance imaging. Cambridge: Cambridge University Press.

241 Buxton RB, Liu TT, Martinez A, Frank LR, Luh WM, Wong EC (2000) Sorting out event- related paradigms in fMRI: the distinction between detecting an activation and estimating the hemodynamic response. Neuroimage 11:S457.

Cacioppo JT, Gardner WL (1999) Emotion. Annu Rev Psychol 50:191-214.

Caetano SC, Hatch JP, Brambilla P, Sassi RB, Nicoletti M, Mallinger AG, Frank E, Kupfer DJ, Keshavan MS, Soares JC (2004) Anatomical MRI study of hippocampus and amygdala in patients with current and remitted major depression. Psychiatry Res 132:141-147.

Cahill L, Haier RJ, Fallon J, Alkire MT, Tang C, Keator D, Wu J, McGaugh JL (1996) Amygdala activity at encoding correlated with long-term, free recall of emotional information. Proc Natl Acad Sci U S A 93:8016-8021.

Cahill L, Prins B, Weber M, McGaugh JL (1994) Beta-adrenergic activation and memory for emotional events. Nature 371:702-704.

Calder AJ, Keane J, Manes F, Antoun N, Young AW (2000) Impaired recognition and experience of disgust following brain injury. Nat Neurosci 3:1077-1078.

Calder AJ, Lawrence AD, Young AW (2001) Neuropsychology of fear and loathing. Nat Rev Neurosci 2:352-363.

Calhoun VD, Adali T, Pearlson GD, Pekar JJ (2001) Spatial and temporal independent component analysis of functional MRI data containing a pair of task-related waveforms. Hum Brain Mapp 13:43-53.

Caligiuri MP, Brown GG, Meloy MJ, Eberson S, Niculescu AB, Lohr JB (2006) Striatopallidal regulation of affect in bipolar disorder. J Affect Disord 91:235-242.

Caligiuri MP, Brown GG, Meloy MJ, Eberson SC, Kindermann SS, Frank LR, Zorrilla LE, Lohr JB (2003) An fMRI study of affective state and medication on cortical and subcortical brain regions during motor performance in bipolar disorder. Psychiatry Res 123:171-182.

Campbell AW (1905) Histological studies on the localisation of cerebral function. Cambridge: Cambridge University Press.

Campbell DW, Sareen J, Paulus MP, Goldin PR, Stein MB, Reiss JP (2006) Time-Varying Amygdala Response to Emotional Faces in Generalized Social Phobia. Biol Psychiatry.

Canli T, Cooney RE, Goldin P, Shah M, Sivers H, Thomason ME, Whitfield-Gabrieli S, Gabrieli JD, Gotlib IH (2005) Amygdala reactivity to emotional faces predicts improvement in major depression. Neuroreport 16:1267-1270.

Canli T, Desmond JE, Zhao Z, Glover G, Gabrieli JD (1998) Hemispheric asymmetry for emotional stimuli detected with fMRI. Neuroreport 9:3233-3239.

Canli T, Sivers H, Thomason ME, Whitfield-Gabrieli S, Gabrieli JD, Gotlib IH (2004) Brain activation to emotional words in depressed vs healthy subjects. Neuroreport 15:2585-2588.

Canli T, Zhao Z, Brewer J, Gabrieli JD, Cahill L (2000) Event-related activation in the human

242 amygdala associates with later memory for individual emotional experience. J Neurosci 20:RC99.

Cannon WB (1927) The James–Lange theory of emotions: a critical examination and an alternative theory. Am J Psychol 39:106-124.

Cannon WB (1929) Bodily Changes in Pain, Hunger, Fear and Rage. Appleton.

Cardoso dA, Jr., Phillips ML, Cerqueira CT, Zilberman M, Lobo D, Henna E, Tavares H, Amaro E, Gorenstein C, Gentil V, Busatto GF (2009) Neural activity changes to emotional stimuli in healthy individuals under chronic use of clomipramine. J Psychopharmacol.

Carmichael ST, Price JL (1995a) Limbic connections of the orbital and medial prefrontal cortex in macaque monkeys. J Comp Neurol 363:615-641.

Carmichael ST, Price JL (1995b) Sensory and premotor connections of the orbital and medial prefrontal cortex of macaque monkeys. J Comp Neurol 363:642-664.

Carter CS, Braver TS, Barch DM, Botvinick MM, Noll D, Cohen JD (1998) Anterior cingulate cortex, error detection, and the online monitoring of performance. Science 280:747-749.

Casey KL, Minoshima S, Berger KL, Koeppe RA, Morrow TJ, Frey KA (1994) Positron emission tomographic analysis of cerebral structures activated specifically by repetitive noxious heat stimuli. J Neurophysiol 71:802-807.

Catani M, Howard RJ, Pajevic S, Jones DK (2002) Virtual in vivo interactive dissection of white matter fasciculi in the human brain. Neuroimage 17:77-94.

Cavanna AE, Trimble MR (2006) The precuneus: a review of its functional anatomy and behavioural correlates. Brain 129:564-583.

Chen BK, Sassi R, Axelson D, Hatch JP, Sanches M, Nicoletti M, Brambilla P, Keshavan MS, Ryan ND, Birmaher B, Soares JC (2004a) Cross-sectional study of abnormal amygdala development in adolescents and young adults with bipolar disorder. Biol Psychiatry 56:399-405.

Chen CH, Ridler K, Suckling J, Williams S, Fu CH, Merlo-Pich E, Bullmore E (2007) Brain Imaging Correlates of Depressive Symptom Severity and Predictors of Symptom Improvement After Antidepressant Treatment. Biol Psychiatry.

Chen HH, Nicoletti M, Sanches M, Hatch JP, Sassi RB, Axelson D, Brambilla P, Keshavan MS, Ryan N, Birmaher B, Soares JC (2004b) Normal pituitary volumes in children and adolescents with bipolar disorder: a magnetic resonance imaging study. Depress Anxiety 20:182-186.

Chikama M, McFarland NR, Amaral DG, Haber SN (1997) Insular cortical projections to functional regions of the striatum correlate with cortical cytoarchitectonic organization in the primate. J Neurosci 17:9686-9705.

Chouinard G (1985) A double-blind controlled clinical trial of fluoxetine and amitriptyline in the treatment of outpatients with major depressive disorder. J Clin Psychiatry 46:32-37.

Chouinard PA, Paus T (2006) The primary motor and premotor areas of the human cerebral 243 cortex. Neuroscientist 12:143-152.

Chow TW, Mamo DC, Uchida H, Graff-Guerrero A, Houle S, Smith GS, Pollock BG, Mulsant BH (2009) Test-retest variability of high resolution positron emission tomography (PET) imaging of cortical serotonin (5HT2A) receptors in older, healthy adults. BMC Med Imaging 9:12.

Cilia J, Piper DC, Upton N, Hagan JJ (2001) Clozapine enhances breakpoint in common marmosets responding on a progressive ratio schedule. Psychopharmacology (Berl) 155:135- 143.

Clark LA, Watson D (1991) Tripartite model of anxiety and depression: psychometric evidence and taxonomic implications. J Abnorm Psychol 100:316-336.

Clore GL, Ortony A (2000) Cognition in emotion: Always, sometimes or never. In: Cognitive Neuroscience of Emotion (Lane R, Nadel G, hern J, llen A, aszniak S, apcsak S, chwartz G, eds), pp 24-61. New York: Oxford University Press.

Coccaro EF, McCloskey MS, Fitzgerald DA, Phan KL (2007) Amygdala and orbitofrontal reactivity to social threat in individuals with impulsive aggression. Biol Psychiatry 62:168-178.

Codispoti M, Ferrari V, Bradley MM (2007) Repetition and event-related potentials: distinguishing early and late processes in affective picture perception. J Cogn Neurosci 19:577- 586.

Coffey CE, Wilkinson WE, Weiner RD, Parashos IA, Djang WT, Webb MC, Figiel GS, Spritzer CE (1993) Quantitative cerebral anatomy in depression. A controlled magnetic resonance imaging study. Arch Gen Psychiatry 50:7-16.

Coffman JA, Bornstein RA, Olson SC, Schwarzkopf SB, Nasrallah HA (1990) Cognitive impairment and cerebral structure by MRI in bipolar disorder. Biol Psychiatry 27:1188-1196.

Coghill RC, Sang CN, Maisog JM, Iadarola MJ (1999) Pain intensity processing within the human brain: a bilateral, distributed mechanism. J Neurophysiol 82:1934-1943.

Cohn JB, Collins G, Ashbrook E, Wernicke JF (1989) A comparison of fluoxetine imipramine and placebo in patients with bipolar depressive disorder. Int Clin Psychopharmacol 4:313-322.

Cohn JB, Wilcox C (1985) A comparison of fluoxetine, imipramine, and placebo in patients with major depressive disorder. J Clin Psychiatry 46:26-31.

Conca A, Fritzsche H, Peschina W, Konig P, Swoboda E, Wiederin H, Haas C (2000) Preliminary findings of simultaneous 18F-FDG and 99mTc-HMPAO SPECT in patients with depressive disorders at rest: differential correlates with ratings of anxiety. Psychiatry Res 98:43- 54.

Cook JW, Spring B, McChargue DE, Borrelli B, Hitsman B, Niaura R, Keuthen NJ, Kristeller J (2004) Influence of fluoxetine on positive and negative affect in a clinic-based smoking cessation trial. Psychopharmacology (Berl) 173:153-159.

Corbetta M, Miezin FM, Shulman GL, Petersen SE (1993) A PET study of visuospatial attention. 244 J Neurosci 13:1202-1226.

Cordes D, Turski PA, Sorenson JA (2000) Compensation of susceptibility-induced signal loss in echo-planar imaging for functional applications. Magn Reson Imaging 18:1055-1068.

Corruble E, Guelfi JD (2000) Pain complaints in depressed inpatients. Psychopathology 33:307- 309.

Corya SA, Andersen SW, Detke HC, Kelly LS, Van Campen LE, Sanger TM, Williamson DJ, Dube S (2003) Long-term antidepressant efficacy and safety of olanzapine/fluoxetine combination: a 76-week open-label study. J Clin Psychiatry 64:1349-1356.

Corya SA, Williamson D, Sanger TM, Briggs SD, Case M, Tollefson G (2006) A randomized, double-blind comparison of olanzapine/fluoxetine combination, olanzapine, fluoxetine, and venlafaxine in treatment-resistant depression. Depress Anxiety 23:364-372.

Costa DC, Verhoeff NP, Cullum ID, Ell PJ, Syed GM, Barrett J, Palazidou E, Toone B, Van RE, Bobeldijk M (1990) In vivo characterisation of 3-iodo-6-methoxybenzamide 123I in humans. Eur J Nucl Med 16:813-816.

Cotter D, Mackay D, Landau S, Kerwin R, Everall I (2001) Reduced glial cell density and neuronal size in the anterior cingulate cortex in major depressive disorder. Arch Gen Psychiatry 58:545-553.

Cowey A, Stoerig P (1991) The neurobiology of blindsight. Trends Neurosci 14:140-145.

Craig AD (2004) Human feelings: why are some more aware than others? Trends Cogn Sci 8:239-241.

Craig AD, Bushnell MC, Zhang ET, Blomqvist A (1994) A thalamic nucleus specific for pain and temperature sensation. Nature 372:770-773.

Craig AD, Chen K, Bandy D, Reiman EM (2000) Thermosensory activation of insular cortex. Nat Neurosci 3:184-190.

Crawford JR, Henry JD (2004) The positive and negative affect schedule (PANAS): construct validity, measurement properties and normative data in a large non-clinical sample. Br J Clin Psychol 43:245-265.

Critchley H, Daly E, Phillips M, Brammer M, Bullmore E, Williams S, Van AT, Robertson D, David A, Murphy D (2000) Explicit and implicit neural mechanisms for processing of social information from facial expressions: a functional magnetic resonance imaging study. Hum Brain Mapp 9:93-105.

Critchley HD, Mathias CJ, Dolan RJ (2001a) Neural activity in the human brain relating to uncertainty and arousal during anticipation. Neuron 29:537-545.

Critchley HD, Melmed RN, Featherstone E, Mathias CJ, Dolan RJ (2001b) Brain activity during biofeedback relaxation: a functional neuroimaging investigation. Brain 124:1003-1012.

Critchley HD, Rolls ET (1996) Olfactory neuronal responses in the primate orbitofrontal cortex:

245 analysis in an olfactory discrimination task. J Neurophysiol 75:1659-1672.

Critchley HD, Rotshtein P, Nagai Y, O'Doherty J, Mathias CJ, Dolan RJ (2005) Activity in the human brain predicting differential heart rate responses to emotional facial expressions. Neuroimage 24:751-762.

Critchley M (1953) The Parietal Lobes. London: Edward Arnold.

Crockford D, Quickfall J, Currie S, Furtado S, Suchowersky O, El-Guebaly N (2008) Prevalence of problem and pathological gambling in Parkinson's disease. J Gambl Stud 24:411-422.

CSEA-NIMH (1999) The International Affective Picture System (Photographic Slides). Gainesville, FL: Center for the Study of Emotion and Attention The Center for Research in Psychophysiology, University of Florida.

D'Argembeau A, Collette F, van der LM, Laureys S, Del FG, Degueldre C, Luxen A, Salmon E (2005) Self-referential reflective activity and its relationship with rest: a PET study. Neuroimage 25:616-624.

Damasio AR (1994) Descartes' error and the future of human life. Sci Am 271:144.

Damasio AR (1995) Towards a neurobiology of emotion and feeling: operational concepts and hypotheses. The Neuroscientist 1:19-25.

Damasio AR, Grabowski TJ, Bechara A, Damasio H, Ponto LL, Parvizi J, Hichwa RD (2000) Subcortical and cortical brain activity during the feeling of self-generated emotions. Nat Neurosci 3:1049-1056.

Damasio AR, Tranel D, Damasio H (1990) Individuals with sociopathic behavior caused by frontal damage fail to respond autonomically to social stimuli. Behav Brain Res 41:81-94.

Danbolt NC (1994) The high affinity uptake system for excitatory amino acids in the brain. Prog Neurobiol 44:377-396.

Danion JM, Willard-Schroeder D, Zimmermann MA, Grange D, Schlienger JL, Singer L (1991) Explicit memory and repetition priming in depression. Preliminary findings. Arch Gen Psychiatry 48:707-711.

Dannlowski U, Ohrmann P, Bauer J, Kugel H, Arolt V, Heindel W, Kersting A, Baune BT, Suslow T (2007) Amygdala reactivity to masked negative faces is associated with automatic judgmental bias in major depression: a 3 T fMRI study. J Psychiatry Neurosci 32:423-429.

Darwin C (1872) The Expression of the Emotions in Man and Animals. London, England: J Murray.

David AS, Cutting JC (1990) Affect, affective disorder and schizophrenia. A neuropsychological investigation of right hemisphere function. Br J Psychiatry 156:491-495.

Davidson RJ (1995) Cerebal asymmetry, emotion and affective style. In: Brain Asymmetry (Davidson RJ, Hugdahl K, eds), pp 361-387. MIT press.

246 Davidson RJ (1998) Affective styles and affective disorders : prespectives from affective neuroscience. Cognition Emotion 12:307-312.

Davidson RJ (2002) Anxiety and affective style: role of prefrontal cortex and amygdala. Biol Psychiatry 51:68-80.

Davidson RJ, Ekman P, Saron CD, Senulis JA, Friesen WV (1990) Approach-withdrawal and cerebral asymmetry: emotional expression and brain physiology. I. J Pers Soc Psychol 58:330- 341.

Davidson RJ, Fox NA (1982) Asymmetrical brain activity discriminates between positive and negative affective stimuli in human infants. Science 218:1235-1237.

Davidson RJ, Irwin W (1999) The functional neuroanatomy of emotion and affective style. Trends Cogn Sci 3:11-21.

Davidson RJ, Irwin W, Anderle MJ, Kalin NH (2003) The neural substrates of affective processing in depressed patients treated with venlafaxine. Am J Psychiatry 160:64-75.

Davies J, Lloyd KR, Jones IK, Barnes A, Pilowsky LS (2003) Changes in regional cerebral blood flow with venlafaxine in the treatment of major depression. Am J Psychiatry 160:374-376.

Davis CE, Jeste DV, Eyler LT (2005) Review of longitudinal functional neuroimaging studies of drug treatments in patients with schizophrenia. Schizophr Res 78:45-60.

Davis KA, Kwon A, Cardenas VA, Deicken RF (2004) Decreased cortical gray and cerebral white matter in male patients with familial bipolar I disorder. J Affect Disord 82:475-485.

Davis KD (2003) Neurophysiological and anatomical considerations in functional imaging of pain. Pain 105:1-3.

Davis KD, Lozano AM, Tasker RR, Dostrovsky JO (1998) Brain targets for pain control. Stereotact Funct Neurosurg 71:173-179.

Davis M, Whalen PJ (2001) The amygdala: vigilance and emotion. Mol Psychiatry 6:13-34.

De Araujo IE, Rolls ET, Kringelbach ML, McGlone F, Phillips N (2003) Taste-olfactory convergence, and the representation of the pleasantness of flavour, in the human brain. Eur J Neurosci 18:2059-2068.

De Lacalle S, Saper CB (1997) The Cholinergic System in the Primate Brain: Basil forebrain and pontine-tegmental cell groups. In: Handbook of Chemical Neuroanatomy The Primate Nervous System, Part 1 (Bloom FE, Bjorklund A, Hokflet T, eds), Amsterdam: Elsevier Science B.V.

De Marte ML, Enesco HE (1985) Influence of diet on plasma tryptophan and brain serotonin levels in mice. Experientia 41:48-50. de Olmos JS (2004) Amygdala. In: The Human Nervous System (McGinty JF, ed), pp 739-868. Amsterdam: Elsevier. de Olmos JS, Ingram WR (1972) The projection field of the stria terminalis in the rat brain. An

247 experimental study. J Comp Neurol 146:303-334. de Wit S, Kosaki Y, Balleine BW, Dickinson A (2006) Dorsomedial prefrontal cortex resolves response conflict in rats. J Neurosci 26:5224-5229. de A, I, Rolls ET, Kringelbach ML, McGlone F, Phillips N (2003) Taste-olfactory convergence, and the representation of the pleasantness of flavour, in the human brain. Eur J Neurosci 18:2059-2068. de GB, Vroomen J, Pourtois G, WEISKRANTZ L (2000) Affective blindsight: are we blindly led by emotions?Response to Heywood and Kentridge (2000). Trends Cogn Sci 4:126-127.

DeBattista C (2005) Executive dysfunction in major depressive disorder. Expert Rev Neurother 5:79-83.

DeBattista C, Hawkins J (2009) Utility of atypical antipsychotics in the treatment of resistant unipolar depression. CNS Drugs 23:369-377.

Dehaene S, Posner MI, Tucker DM (1994) Localization of a neural system for error detection and compensation. Psychol Sci 5:302-305.

DelBello MP, Zimmerman ME, Mills NP, Getz GE, Strakowski SM (2004) Magnetic resonance imaging analysis of amygdala and other subcortical brain regions in adolescents with bipolar disorder. Bipolar Disord 6:43-52.

Delgado PL (2000) Depression: the case for a monoamine deficiency. J Clin Psychiatry 61 Suppl 6:7-11.

Delgado PL (2006) Monoamine depletion studies: implications for antidepressant discontinuation syndrome. J Clin Psychiatry 67 Suppl 4:22-26.

Delgado PL, Charney DS, Price LH, Aghajanian GK, Landis H, Heninger GR (1990) Serotonin function and the mechanism of antidepressant action. Reversal of antidepressant-induced remission by rapid depletion of plasma tryptophan. Arch Gen Psychiatry 47:411-418.

Delgado PL, Charney DS, Price LH, Landis H, Heninger GR (1989) Neuroendocrine and behavioral effects of dietary tryptophan restriction in healthy subjects. Life Sci 45:2323-2332.

Delgado PL, Miller HL, Salomon RM, Licinio J, Heninger GR, Gelenberg AJ, Charney DS (1993) Monoamines and the mechanism of antidepressant action: effects of catecholamine depletion on mood of patients treated with antidepressants. Psychopharmacol Bull 29:389-396.

Delgado PL, Miller HL, Salomon RM, Licinio J, Krystal JH, Moreno FA, Heninger GR, Charney DS (1999) Tryptophan-depletion challenge in depressed patients treated with desipramine or fluoxetine: implications for the role of serotonin in the mechanism of antidepressant action. Biol Psychiatry 46:212-220.

Delgado PL, Price LH, Miller HL, Salomon RM, Aghajanian GK, Heninger GR, Charney DS (1994) Serotonin and the neurobiology of depression. Effects of tryptophan depletion in drug- free depressed patients. Arch Gen Psychiatry 51:865-874.

248 Delgado PL, Price LH, Miller HL, Salomon RM, Licinio J, Krystal JH, Heninger GR, Charney DS (1991) Rapid serotonin depletion as a provocative challenge test for patients with major depression: relevance to antidepressant action and the neurobiology of depression. Psychopharmacol Bull 27:321-330.

Demyttenaere K, et al. (2004) Prevalence, severity, and unmet need for treatment of mental disorders in the World Health Organization World Mental Health Surveys. JAMA 291:2581- 2590.

Demyttenaere K, De Fruyt J (2003) Getting what you ask for: on the selectivity of depression rating scales. Psychother Psychosom 72:61-70. den Ouden HE, Frith U, Frith C, Blakemore SJ (2005) Thinking about intentions. Neuroimage 28:787-796.

DeRubeis RJ, Hollon SD, Amsterdam JD, Shelton RC, Young PR, Salomon RM, O'Reardon JP, Lovett ML, Gladis MM, Brown LL, Gallop R (2005) Cognitive therapy vs medications in the treatment of moderate to severe depression. Arch Gen Psychiatry 62:409-416.

Descombes X, Kruggel F, von Cramon DY (1998) fMRI signal restoration using a spatiotemporal Markov Random Field preserving transitions. Neuroimage 8:340-349.

Desmond JE, Glover GH (2002) Estimating sample size in functional MRI (fMRI) neuroimaging studies: statistical power analyses. J Neurosci Methods 118:115-128.

Devinsky O, Morrell MJ, Vogt BA (1995) Contributions of anterior cingulate cortex to behaviour. Brain 118 ( Pt 1):279-306.

Dichter GS, Tomarken AJ, Freid CM, Addington S, Shelton RC (2005) Do venlafaxine XR and paroxetine equally influence negative and positive affect? J Affect Disord 85:333-339.

Diener E, Larsen RJ, Levine S, Emmons RA (1985) Intensity and frequency: dimensions underlying positive and negative affect. J Pers Soc Psychol 48:1253-1265.

Dietzel I, Heinemann U, Lux HD (1989) Relations between slow extracellular potential changes, glial potassium buffering, and electrolyte and cellular volume changes during neuronal hyperactivity in cat brain. Glia 2:25-44.

Dolan RJ, Bench CJ, Brown RG, Scott LC, Friston KJ, Frackowiak RS (1992) Regional cerebral blood flow abnormalities in depressed patients with cognitive impairment. J Neurol Neurosurg Psychiatry 55:768-773.

Donohue JM, Pincus HA (2007) Reducing the societal burden of depression: a review of economic costs, quality of care and effects of treatment. Pharmacoeconomics 25:7-24.

Dougherty DD, Shin LM, Alpert NM, Pitman RK, Orr SP, Lasko M, Macklin ML, Fischman AJ, Rauch SL (1999) Anger in healthy men: a PET study using script-driven imagery. Biol Psychiatry 46:466-472.

Dougherty DD, Weiss AP, Cosgrove GR, Alpert NM, Cassem EH, Nierenberg AA, Price BH, Mayberg HS, Fischman AJ, Rauch SL (2003) Cerebral metabolic correlates as potential 249 predictors of response to anterior cingulotomy for treatment of major depression. J Neurosurg 99:1010-1017.

Downar J, Crawley AP, Mikulis DJ, Davis KD (2000) A multimodal cortical network for the detection of changes in the sensory environment. Nat Neurosci 3:277-283.

Downar J, Crawley AP, Mikulis DJ, Davis KD (2002) A cortical network sensitive to stimulus salience in a neutral behavioral context across multiple sensory modalities. J Neurophysiol 87:615-620.

Downar J, Mikulis DJ, Davis KD (2003) Neural correlates of the prolonged salience of painful stimulation. Neuroimage 20:1540-1551.

Drevets W, Raichle ME (1998) Reciprocal suppression of regional cerebral blood flow during emotional versus higher cognitive processes: Implications for interactions between emotion and cognition. Cognition Emotion 12:353-385.

Drevets WC (1999) Prefrontal cortical-amygdalar metabolism in major depression. Ann N Y Acad Sci 877:614-637.

Drevets WC (2000) Functional anatomical abnormalities in limbic and prefrontal cortical structures in major depression. Prog Brain Res 126:413-431.

Drevets WC, Bogers W, Raichle ME (2002) Functional anatomical correlates of antidepressant drug treatment assessed using PET measures of regional glucose metabolism. Eur Neuropsychopharmacol 12:527-544.

Drevets WC, Gautier C, Price JC, Kupfer DJ, Kinahan PE, Grace AA, Price JL, Mathis CA (2001) Amphetamine-induced dopamine release in human ventral striatum correlates with euphoria. Biol Psychiatry 49:81-96.

Drevets WC, Price JL, Simpson JR, Jr., Todd RD, Reich T, Vannier M, Raichle ME (1997) Subgenual prefrontal cortex abnormalities in mood disorders. Nature 386:824-827.

Drevets WC, Videen TO, Price JL, Preskorn SH, Carmichael ST, Raichle ME (1992) A functional anatomical study of unipolar depression. J Neurosci 12:3628-3641.

Dronkers NF, Wilkins DP, Van VR, Jr., Redfern BB, Jaeger JJ (2004) Lesion analysis of the brain areas involved in language comprehension. Cognition 92:145-177.

Dubois S, Rossion B, Schiltz C, Bodart JM, Michel C, Bruyer R, Crommelinck M (1999) Effect of familiarity on the processing of human faces. Neuroimage 9:278-289.

Duelli R, Kuschinsky W (2001) Brain glucose transporters: relationship to local energy demand. News Physiol Sci 16:71-76.

Duncan J, Owen AM (2000) Common regions of the human frontal lobe recruited by diverse cognitive demands. Trends Neurosci 23:475-483.

Dunkin JJ, Leuchter AF, Cook IA, Kasl-Godley JE, Abrams M, Rosenberg-Thompson S (2000) Executive dysfunction predicts nonresponse to fluoxetine in major depression. J Affect Disord

250 60:13-23.

Dunn LT, Everitt BJ (1988) Double dissociations of the effects of amygdala and insular cortex lesions on conditioned taste aversion, passive avoidance, and neophobia in the rat using the excitotoxin ibotenic acid. Behav Neurosci 102:3-23.

Dunn RT, Kimbrell TA, Ketter TA, Frye MA, Willis MW, Luckenbaugh DA, Post RM (2002) Principal components of the Beck Depression Inventory and regional cerebral metabolism in unipolar and bipolar depression. Biol Psychiatry 51:387-399.

Dunn RT, Willis MW, Benson BE, Repella JD, Kimbrell TA, Ketter TA, Speer AM, Osuch EA, Post RM (2005) Preliminary findings of uncoupling of flow and metabolism in unipolar compared with bipolar affective illness and normal controls. Psychiatry Res 140:181-198.

Dupont RM, Jernigan TL, Heindel W, Butters N, Shafer K, Wilson T, Hesselink J, Gillin JC (1995) Magnetic resonance imaging and mood disorders. Localization of white matter and other subcortical abnormalities. Arch Gen Psychiatry 52:747-755.

Duvernoy HM, Delon S, Vannson JL (1981) Cortical blood vessels of the human brain. Brain Res Bull 7:519-579.

Ebert D, Feistel H, Barocka A (1991) Effects of sleep deprivation on the limbic system and the frontal lobes in affective disorders: a study with Tc-99m-HMPAO SPECT. Psychiatry Res 40:247-251.

Ebmeier K, Rose E, Steele D (2006) Cognitive impairment and fMRI in major depression. Neurotox Res 10:87-92.

Eckelman WC, Reba RC, Rzeszotarski WJ, Gibson RE, Hill T, Holman BL, Budinger T, Conklin JJ, Eng R, Grissom MP (1984) External imaging of cerebral muscarinic acetylcholine receptors. Science 223:291-293.

Economo C, Koskinas GN (1925) Die cytoarchitektonik der hirnrinde des erwachsenen menschen. Wien: Springer.

Edwards BC, Lambert MJ, Moran PW, McCully T, Smith KC, Ellingson AG (1984) A meta- analytic comparison of the Beck Depression Inventory and the Hamilton Rating Scale for Depression as measures of treatment outcome. Br J Clin Psychol 23 ( Pt 2):93-99.

Eggers C, Hilker R, Burghaus L, Schumacher B, Heiss WD (2009) High resolution positron emission tomography demonstrates basal ganglia dysfunction in early Parkinson's disease. J Neurol Sci 276:27-30.

Eippert F, Veit R, Weiskopf N, Erb M, Birbaumer N, Anders S (2007) Regulation of emotional responses elicited by threat-related stimuli. Hum Brain Mapp 28:409-423.

Ekman P, Freisen WV, Ancoli S (1980) Facial signs of emotional experience. J Pers Soc Psychol 39:1125-1134.

Elliott R, Friston KJ, Dolan RJ (2000) Dissociable neural responses in human reward systems. J Neurosci 20:6159-6165. 251 Elliott R, Sahakian BJ, Michael A, Paykel ES, Dolan RJ (1998) Abnormal neural response to feedback on planning and guessing tasks in patients with unipolar depression. Psychol Med 28:559-571.

Elster AD, Burdette JH (2001) Questions & answers in magnetic resonance imaging. St. Louis: Mosby.

Engelman K, Horwitz D, Jequier E, Sjoerdsma A (1968) Biochemical and pharmacologic effects of alpha-methyltyrosine in man. J Clin Invest 47:577-594.

Erlandsson K, Bressan RA, Mulligan RS, Gunn RN, Cunningham VJ, Owens J, Wyper D, Ell PJ, Pilowsky LS (2003) Kinetic modelling of [123I]CNS 1261--a potential SPET tracer for the NMDA receptor. Nucl Med Biol 30:441-454.

Escalona PR, Early B, McDonald WM (1993) Reduction of cerebellar volume in major depression: a controlled magnetic resonance imaging study. Depression 1:156-158.

Etzel JA, Gazzola V, Keysers C (2009) An introduction to anatomical ROI-based fMRI classification analysis. Brain Res 1282:114-125.

Eugene F, Levesque J, Mensour B, Leroux JM, Beaudoin G, Bourgouin P, Beauregard M (2003) The impact of individual differences on the neural circuitry underlying sadness. Neuroimage 19:354-364.

Evans KC, Dougherty DD, Pollack MH, Rauch SL (2006) Using neuroimaging to predict treatment response in mood and anxiety disorders. Ann Clin Psychiatry 18:33-42.

Everitt BS, Bullmore ET (1999) Mixture model mapping of the brain activation in functional magnetic resonance images. Hum Brain Mapp 7:1-14.

Fales CL, Barch DM, Rundle MM, Mintun MA, Snyder AZ, Cohen JD, Mathews J, Sheline YI (2008) Altered emotional interference processing in affective and cognitive-control brain circuitry in major depression. Biol Psychiatry 63:377-384.

Farde L, Hall H, Ehrin E, Sedvall G (1986) Quantitative analysis of D2 dopamine receptor binding in the living human brain by PET. Science 231:258-261.

Faries D, Herrera J, Rayamajhi J, DeBrota D, Demitrack M, Potter WZ (2000) The responsiveness of the Hamilton Depression Rating Scale. J Psychiatr Res 34:3-10.

Fava M (2002) Somatic symptoms, depression, and antidepressant treatment. J Clin Psychiatry 63:305-307.

Fava M, Rush AJ, Trivedi MH, Nierenberg AA, Thase ME, Sackeim HA, Quitkin FM, Wisniewski S, Lavori PW, Rosenbaum JF, Kupfer DJ (2003) Background and rationale for the sequenced treatment alternatives to relieve depression (STAR*D) study. Psychiatr Clin North Am 26:457-94, x.

Ferry AT, Ongur D, An X, Price JL (2000) Prefrontal cortical projections to the striatum in macaque monkeys: evidence for an organization related to prefrontal networks. J Comp Neurol 425:447-470. 252 Finger EC, Marsh AA, Kamel N, Mitchell DG, Blair JR (2006) Caught in the act: the impact of audience on the neural response to morally and socially inappropriate behavior. Neuroimage 33:414-421.

Fink GR, Markowitsch HJ, Reinkemeier M, Bruckbauer T, Kessler J, Heiss WD (1996) Cerebral representation of one's own past: neural networks involved in autobiographical memory. J Neurosci 16:4275-4282.

Fisher H, Aron A, Brown LL (2005) Romantic love: an fMRI study of a neural mechanism for mate choice. J Comp Neurol 493:58-62.

Fogassi L, Luppino G (2005) Motor functions of the parietal lobe. Curr Opin Neurobiol 15:626- 631.

Foote SL, Ashton-Jones GS (1995) Pharmacology and physiology of central noradrenergic systems. In: Psychopharmacology: The Fourth Generation of Progress (Bloom FE, Kupfer DJ, eds), pp 335-345. New York, NY: Raven Press Ltd.

Forbes EE, Christopher MJ, Siegle GJ, Ladouceur CD, Ryan ND, Carter CS, Birmaher B, Axelson DA, Dahl RE (2006) Reward-related decision-making in pediatric major depressive disorder: an fMRI study. J Child Psychol Psychiatry 47:1031-1040.

Fossati P, Hevenor SJ, Graham SJ, Grady C, Keightley ML, Craik F, Mayberg H (2003) In search of the emotional self: an FMRI study using positive and negative emotional words. Am J Psychiatry 160:1938-1945.

Francis S, Rolls ET, Bowtell R, McGlone F, O'Doherty J, Browning A, Clare S, Smith E (1999) The representation of pleasant touch in the brain and its relationship with taste and olfactory areas. Neuroreport 10:453-459.

Frank E, Prien RF, Jarrett RB, Keller MB, Kupfer DJ, Lavori PW, Rush AJ, Weissman MM (1991) Conceptualization and rationale for consensus definitions of terms in major depressive disorder. Remission, recovery, relapse, and recurrence. Arch Gen Psychiatry 48:851-855.

Frank GK, Kaye WH, Carter CS, Brooks S, May C, Fissell K, Stenger VA (2003) The evaluation of brain activity in response to taste stimuli--a pilot study and method for central taste activation as assessed by event-related fMRI. J Neurosci Methods 131:99-105.

Franklin TR, Wang Z, Wang J, Sciortino N, Harper D, Li Y, Ehrman R, Kampman K, O'Brien CP, Detre JA, Childress AR (2007) Limbic Activation to Cigarette Smoking Cues Independent of Nicotine Withdrawal: A Perfusion fMRI Study. Neuropsychopharmacology.

Friston K, Holmes A, Worsley K, Poline J, Frith C, Frackowiak R (1995) Statistical parametric maps in functional imaging: a general linear approach. Hum Brain Mapp 2:189-210.

Friston KJ (1994) Statistical parametric mapping. In: Functional Neuroimaging: Technical Foundations (Thatcher RW, Hallett M, et al, eds), pp 79-83. San Diego, California: Academic Press.

Friston KJ, Holmes AP, Worsley JP, Poline JP, Frith CD, Frackowiak RSJ (1999) Statistical Parametric Maps in Functional Imaging: A General Linear Approach. pp 189-210. 253 Friston KJ (2005) Models of brain function in neuroimaging. Annu Rev Psychol 56:57-87.

Friston KJ, Ashburner J, Frith CD, Poline JB, Heather JD, Frackowiak RS (1995) Spatial registration and normalization of images. Hum Brain Map 2:165-189.

Friston KJ, Frith CD, Liddle PF, Frackowiak RS (1991) Comparing functional (PET) images: the assessment of significant change. J Cereb Blood Flow Metab 11:690-699.

Friston KJ, Frith CD, Liddle PF, Frackowiak RS (1993) Functional connectivity: the principal- component analysis of large (PET) data sets. J Cereb Blood Flow Metab 13:5-14.

Friston KJ, Glaser DE, Henson RN, Kiebel S, Phillips C, Ashburner J (2002a) Classical and Bayesian inference in neuroimaging: applications. Neuroimage 16:484-512.

Friston KJ, Penny W, Phillips C, Kiebel S, Hinton G, Ashburner J (2002b) Classical and Bayesian inference in neuroimaging: theory. Neuroimage 16:465-483.

Friston KJ, Price CJ, Fletcher P, Moore C, Frackowiak RS, Dolan RJ (1996) The trouble with cognitive subtraction. Neuroimage 4:97-104.

Friston KJ, Zarahn E, Josephs O, Henson RN, Dale AM (1999) Stochastic designs in event- related fMRI. Neuroimage 10:607-619.

Frodl T, Meisenzahl E, Zetzsche T, Bottlender R, Born C, Groll C, Jager M, Leinsinger G, Hahn K, Moller HJ (2002) Enlargement of the amygdala in patients with a first episode of major depression. Biol Psychiatry 51:708-714.

Frodl T, Meisenzahl EM, Zetzsche T, Hohne T, Banac S, Schorr C, Jager M, Leinsinger G, Bottlender R, Reiser M, Moller HJ (2004a) Hippocampal and amygdala changes in patients with major depressive disorder and healthy controls during a 1-year follow-up. J Clin Psychiatry 65:492-499.

Frodl T, Meisenzahl EM, Zill P, Baghai T, Rujescu D, Leinsinger G, Bottlender R, Schule C, Zwanzger P, Engel RR, Rupprecht R, Bondy B, Reiser M, Moller HJ (2004b) Reduced hippocampal volumes associated with the long variant of the serotonin transporter polymorphism in major depression. Arch Gen Psychiatry 61:177-183.

Fromm G, Bond H (1964) SLOW CHANGES IN THE ELECTROCORTICOGRAM AND THE ACTIVITY OF CORTICAL NEURONS. Electroencephalogr Clin Neurophysiol 17:520-523.

Frot M, Mauguiere F (2003) Dual representation of pain in the operculo-insular cortex in humans. Brain 126:438-450.

Frysztak RJ, Neafsey EJ (1991) The effect of medial frontal cortex lesions on respiration, "freezing," and ultrasonic vocalizations during conditioned emotional responses in rats. Cereb Cortex 1:418-425.

Frysztak RJ, Neafsey EJ (1994) The effect of medial frontal cortex lesions on cardiovascular conditioned emotional responses in the rat. Brain Res 643:181-193.

Fu CH, Williams SC, Brammer MJ, Suckling J, Kim J, Cleare AJ, Walsh ND, Mitterschiffthaler

254 MT, Andrew CM, Pich EM, Bullmore ET (2007) Neural responses to happy facial expressions in major depression following antidepressant treatment. Am J Psychiatry 164:599-607.

Fu CH, Williams SC, Cleare AJ, Brammer MJ, Walsh ND, Kim J, Andrew CM, Pich EM, Williams PM, Reed LJ, Mitterschiffthaler MT, Suckling J, Bullmore ET (2004) Attenuation of the neural response to sad faces in major depression by antidepressant treatment: a prospective, event-related functional magnetic resonance imaging study. Arch Gen Psychiatry 61:877-889.

Fuchs RA, Evans KA, Ledford CC, Parker MP, Case JM, Mehta RH, See RE (2005) The role of the dorsomedial prefrontal cortex, basolateral amygdala, and dorsal hippocampus in contextual reinstatement of cocaine seeking in rats. Neuropsychopharmacology 30:296-309.

Fusar-Poli P, Broome MR, Matthiasson P, Williams SC, Brammer M, McGuire PK (2007) Effects of acute antipsychotic treatment on brain activation in first episode psychosis: an fMRI study. Eur Neuropsychopharmacol 17:492-500.

Fuster JM (1997) The Prefrontal Cortex. New York: Raven Press.

Gabriel M (1990) Functions of anterior and posterior cingulate cortex during avoidance learning in rabbits. Prog Brain Res 85:467-482.

Gallagher M, Graham PW, Holland PC (1990) The amygdala central nucleus and appetitive Pavlovian conditioning: lesions impair one class of conditioned behavior. J Neurosci 10:1906- 1911.

Galvan A, Hare TA, Davidson M, Spicer J, Glover G, Casey BJ (2005) The role of ventral frontostriatal circuitry in reward-based learning in humans. J Neurosci 25:8650-8656.

Gardner A, Johansson A, Wibom R, Nennesmo I, von DU, Hagenfeldt L, Hallstrom T (2003) Alterations of mitochondrial function and correlations with personality traits in selected major depressive disorder patients. J Affect Disord 76:55-68.

Gartside SE, Leitch MM, Young AH (2003) Altered glucocorticoid rhythm attenuates the ability of a chronic SSRI to elevate forebrain 5-HT: implications for the treatment of depression. Neuropsychopharmacology 28:1572-1578.

Gemar MC, Kapur S, Segal ZV, Brown GM, Houle S (1996) Effects of self-generated sad mood on regional cerebral activity: a PET study in normal subjects. Depression 4:81-88.

Gemar MC, Segal ZV, Mayberg HS, Goldapple K, Carney C (2006) Changes in regional cerebral blood flow following mood challenge in drug-free, remitted patients with unipolar depression. Depress Anxiety Epub ahead of print.

Gemba H, Sasaki K, Brooks VB (1986) 'Error' potentials in limbic cortex (anterior cingulate area 24) of monkeys during motor learning. Neurosci Lett 70:223-227.

George MS, Ketter TA, Parekh PI, Herscovitch P, Post RM (1996) Gender differences in regional cerebral blood flow during transient self-induced sadness or happiness. Biol Psychiatry 40:859-871.

George MS, Ketter TA, Parekh PI, Horwitz B, Herscovitch P, Post RM (1995) Brain activity 255 during transient sadness and happiness in healthy women. Am J Psychiatry 152:341-351.

George MS, Ketter TA, Parekh PI, Rosinsky N, Ring HA, Pazzaglia PJ, Marangell LB, Callahan AM, Post RM (1997) Blunted left cingulate activation in mood disorder subjects during a response interference task (the Stroop). J Neuropsychiatry Clin Neurosci 9:55-63.

Giacobbe P, Kennedy SH, Fulton KA, Lozano AM, Mayberg HS (2006) Predictors of Response to Deep Brain Stimulation for Treatment Resistant Depression.

Gilbert AR, Moore GJ, Keshavan MS, Paulson LA, Narula V, Mac Master FP, Stewart CM, Rosenberg DR (2000) Decrease in thalamic volumes of pediatric patients with obsessive- compulsive disorder who are taking paroxetine. Arch Gen Psychiatry 57:449-456.

Gilboa A, Winocur G, Grady CL, Hevenor SJ, Moscovitch M (2004) Remembering our past: functional neuroanatomy of recollection of recent and very remote personal events. Cereb Cortex 14:1214-1225.

Gilman S (1996) Manter and Gatz's essentials of clinical neuroanatomy and neurophysiology. Philadephia: F.A. Davis.

Gimenez S, Clos S, Romero S, Grasa E, Morte A, Barbanoj MJ (2007) Effects of olanzapine, risperidone and haloperidol on sleep after a single oral morning dose in healthy volunteers. Psychopharmacology (Berl) 190:507-516.

Glassman AH, Shapiro PA (1998) Depression and the course of coronary artery disease. Am J Psychiatry 155:4-11.

Gloor P (1992) Role of the amygdala in temporal lobe epilepsy. In: The Amygdala: Neurobiological Aspects of Emotion, Memory, and Mental Dysfunction (Aggleton JP, ed), pp 505-538. New York, NY: Wiley-Liss.

Glover GH, Lee AT (1995) Motion artifacts in fMRI: comparison of 2DFT with PR and spiral scan methods. Magn Reson Med 33:624-635.

Godinho F, Frot M, Perchet C, Magnin M, Garcia-Larrea L (2008) Pain influences hedonic assessment of visual inputs. Eur J Neurosci 27:2219-2228.

Goldapple K, Segal Z, Garson C, Lau M, Bieling P, Kennedy S, Mayberg H (2004) Modulation of cortical-limbic pathways in major depression: treatment-specific effects of cognitive behavior therapy. Arch Gen Psychiatry 61:34-41.

Goldberg TE, Gold JM, Greenberg R, Griffin S, Schulz SC, Pickar D, Kleinman JE, Weinberger DR (1993) Contrasts between patients with affective disorders and patients with schizophrenia on a neuropsychological test battery. Am J Psychiatry 150:1355-1362.

Goldman S, Dethy S, Lotstra F, Biver F, Stanus E, Wikler D, Hildebrand J, Mendlewicz J, Luxen A (1995) Basal ganglia and frontal lobe glucose metabolism. A reproducibility positron emission tomography study. J Neuroimaging 5:219-226.

Goldman-Rakic PS (1988) Topography of cognition: parallel distributed networks in primate association cortex. Annu Rev Neurosci 11:137-156. 256 Goldman-Rakic PS (1995) Architecture of the prefrontal cortex and the central executive. Ann N Y Acad Sci 769:71-83.

Goodale MA, Milner AD (1992) Separate visual pathways for perception and action. Trends Neurosci 15:20-25.

Goodwin GM (2006) Depression and associated physical diseases and symptoms. Dialogues Clin Neurosci 8:259-265.

Gorlyn M, Keilp JG, Grunebaum MF, Taylor BP, Oquendo MA, Bruder GE, Stewart JW, Zalsman G, Mann JJ (2008) Neuropsychological characteristics as predictors of SSRI treatment response in depressed subjects. J Neural Transm 115:1213-1219.

Gotlib IH, Sivers H, Gabrieli JD, Whitfield-Gabrieli S, Goldin P, Minor KL, Canli T (2005) Subgenual anterior cingulate activation to valenced emotional stimuli in major depression. Neuroreport 16:1731-1734.

Gottfries CF, Oreland L, Wiberg A, Winblad B (1974) Letter: Brain-levels of monoamine oxidase in depression. Lancet 2:360-361.

Gray H (1918) Gray's Anatomy. New York: Lea & Febiger.

Gray JA, McNaughton N (2000) The Neuropsychology of Anxiety: An Enquiry into the Functions of the Septohippocampal System. Oxford, UK: Oxford University Press.

Graziano MS, Aflalo TN (2007) Mapping behavioral repertoire onto the cortex. Neuron 56:239- 251.

Greene JD, Sommerville RB, Nystrom LE, Darley JM, Cohen JD (2001) An fMRI investigation of emotional engagement in moral judgment. Science 293:2105-2108.

Greenwald BS, Kramer-Ginsberg E, Bogerts B, Ashtari M, Aupperle P, Wu H, Allen L, Zeman D, Patel M (1997) Qualitative magnetic resonance imaging findings in geriatric depression. Possible link between later-onset depression and Alzheimer's disease? Psychol Med 27:421-431.

Greenwald MK, Cook EW, Lang PJ (1989) Affective judgment and psychophysiological response:Dimensional covariation in the evaluation of pictorial stimuli. Journal of Psychophysiology 3:51-64.

Greicius MD, Flores BH, Menon V, Glover GH, Solvason HB, Kenna H, Reiss AL, Schatzberg AF (2007) Resting-State Functional Connectivity in Major Depression: Abnormally Increased Contributions from Subgenual Cingulate Cortex and Thalamus. Biol Psychiatry.

Grimm S, Beck J, Schuepbach D, Hell D, Boesiger P, Bermpohl F, Niehaus L, Boeker H, Northoff G (2008a) Imbalance between left and right dorsolateral prefrontal cortex in major depression is linked to negative emotional judgment: an fMRI study in severe major depressive disorder. Biol Psychiatry 63:369-376.

Grimm S, Boesiger P, Beck J, Schuepbach D, Bermpohl F, Walter M, Ernst J, Hell D, Boeker H, Northoff G (2008b) Altered Negative BOLD Responses in the Default-Mode Network during Emotion Processing in Depressed Subjects. Neuropsychopharmacology. 257 Grimm S, Schmidt CF, Bermpohl F, Heinzel A, Dahlem Y, Wyss M, Hell D, Boesiger P, Boeker H, Northoff G (2006) Segregated neural representation of distinct emotion dimensions in the prefrontal cortex-an fMRI study. Neuroimage 30:325-340.

Grinvald A, Lieke E, Frostig RD, Gilbert CD, Wiesel TN (1986) Functional architecture of cortex revealed by optical imaging of intrinsic signals. Nature 324:361-364.

Gronier BS, Rasmussen K (2003) Electrophysiological effects of acute and chronic olanzapine and fluoxetine in the rat prefrontal cortex. Neurosci Lett 349:196-200.

Gur RC, Erwin RJ, Gur RE, Zwil AS, Heimberg C, Kraemer HC (1992) Facial emotion discrimination: II. Behavioral findings in depression. Psychiatry Res 42:241-251.

Gusnard DA, Akbudak E, Shulman GL, Raichle ME (2001a) Medial prefrontal cortex and self- referential mental activity: relation to a default mode of brain function. Proc Natl Acad Sci U S A 98:4259-4264.

Gusnard DA, Raichle ME, Raichle ME (2001b) Searching for a baseline: functional imaging and the resting human brain. Nat Rev Neurosci 2:685-694.

Gustafsson B (1984) Afterpotentials and transduction properties in different types of central neurones. Arch Ital Biol 122:17-30.

Guy W (1976) Clinical Global Impression. Rockville, MD.

Habel U, Klein M, Kellermann T, Shah NJ, Schneider F (2005) Same or different? Neural correlates of happy and sad mood in healthy males. Neuroimage 26:206-214.

Haber SN, Groenewegen HJ, Grove EA, Nauta WJ (1985) Efferent connections of the ventral pallidum: evidence of a dual striato pallidofugal pathway. J Comp Neurol 235:322-335.

Haber SN, Kunishio K, Mizobuchi M, Lynd-Balta E (1995) The orbital and medial prefrontal circuit through the primate basal ganglia. J Neurosci 15:4851-4867.

Haines DE, Dietrichs E (1984) An HRP study of hypothalamo-cerebellar and cerebello- hypothalamic connections in squirrel monkey (Saimiri sciureus). J Comp Neurol 229:559-575.

Hajos M, Fleishaker JC, Filipiak-Reisner JK, Brown MT, Wong EH (2004) The selective norepinephrine reuptake inhibitor antidepressant reboxetine: pharmacological and clinical profile. CNS Drug Rev 10:23-44.

Halliday GM, Cullen K, Cairns MJ (1993) Quantitation and three-dimensional reconstruction of Ch4 nucleus in the human basal forebrain. Synapse 15:1-16.

Hamann S, Mao H (2002) Positive and negative emotional verbal stimuli elicit activity in the left amygdala. Neuroreport 13:15-19.

Hamann SB, Ely TD, Grafton ST, Kilts CD (1999) Amygdala activity related to enhanced memory for pleasant and aversive stimuli. Nat Neurosci 2:289-293.

Hamilton M (1960) A rating scale for depression. J Neurol Neurosurg Psychiatry 23:56-62.

258 Hamm AO, Weike AI, Schupp HT, Treig T, Dressel A, Kessler C (2003) Affective blindsight: intact fear conditioning to a visual cue in a cortically blind patient. Brain 126:267-275.

Hariri AR, Bookheimer SY, Mazziotta JC (2000) Modulating emotional responses: effects of a neocortical network on the limbic system. Neuroreport 11:43-48.

Hariri AR, Mattay VS, Tessitore A, Fera F, Weinberger DR (2003) Neocortical modulation of the amygdala response to fearful stimuli. Biol Psychiatry 53:494-501.

Hariri AR, Tessitore A, Mattay VS, Fera F, Weinberger DR (2002) The amygdala response to emotional stimuli: a comparison of faces and scenes. Neuroimage 17:317-323.

Harmer CJ, Bhagwagar Z, Perrett DI, Vollm BA, Cowen PJ, Goodwin GM (2003a) Acute SSRI administration affects the processing of social cues in healthy volunteers. Neuropsychopharmacology 28:148-152.

Harmer CJ, Hill SA, Taylor MJ, Cowen PJ, Goodwin GM (2003b) Toward a neuropsychological theory of antidepressant drug action: increase in positive emotional bias after potentiation of norepinephrine activity. Am J Psychiatry 160:990-992.

Harmer CJ, Shelley NC, Cowen PJ (2002) Repeated administration of citalopram or reboxetine reduces the perception of negative emotions from facial expression. J Psychopharmacol S16:A79.

Harmer CJ, Shelley NC, Cowen PJ, Goodwin GM (2004) Increased positive versus negative affective perception and memory in healthy volunteers following selective serotonin and norepinephrine reuptake inhibition. Am J Psychiatry 161:1256-1263.

Harrison PJ (2000) Neuropathological studies of bipolar disorder. Acta Neuropsychiatr 12:96-98.

Harrison RV, Harel N, Panesar J, Mount RJ (2002) Blood capillary distribution correlates with hemodynamic-based functional imaging in cerebral cortex. Cereb Cortex 12:225-233.

Hart CL, Ward AS, Haney M, Foltin RW, Fischman MW (2001) Methamphetamine self- administration by humans. Psychopharmacology (Berl) 157:75-81.

Hasin DS, Goodwin RD, Stinson FS, Grant BF (2005) Epidemiology of major depressive disorder: results from the National Epidemiologic Survey on Alcoholism and Related Conditions. Arch Gen Psychiatry 62:1097-1106.

Hastings RS, Parsey RV, Oquendo MA, Arango V, Mann JJ (2004) Volumetric analysis of the prefrontal cortex, amygdala, and hippocampus in major depression. Neuropsychopharmacology 29:952-959.

Hauser P, Matochik J, Altshuler LL, Denicoff KD, Conrad A, Li X, Post RM (2000) MRI-based measurements of temporal lobe and ventricular structures in patients with bipolar I and bipolar II disorders. J Affect Disord 60:25-32.

He SQ, Dum RP, Strick PL (1995) Topographic organization of corticospinal projections from the frontal lobe: motor areas on the medial surface of the hemisphere. J Neurosci 15:3284-3306.

259 Heath RG (1977) Modulation of emotion with a brain pacemamer. Treatment for intractable psychiatric illness. J Nerv Ment Dis 165:300-317.

Heath RG, Dempesy CW, Fontana CJ, Myers WA (1978) Cerebellar stimulation: effects on septal region, hippocampus, and amygdala of cats and rats. Biol Psychiatry 13:501-529.

Heekeren HR, Wartenburger I, Schmidt H, Schwintowski HP, Villringer A (2003) An fMRI study of simple ethical decision-making. Neuroreport 14:1215-1219.

Heimer L, Alheid GF (1991) Piecing together the puzzle of basal forebrain anatomy. In: The Basal Forebrain (Napier TC, Kalivas PW, Hanin I, eds), pp 1-42. New York: Plenum Press.

Heimer L, de Olmos J, Alheid GF, Pearson J, Sakamoto M, Marksteiner J, Switzer RC (1999) The Human Basal Forebrain, Part 2, Handbook of Chemical Neuroanatomy. Amsterdam: Elsevier.

Heimer L, van Hoesen GW (2006) The limbic lobe and its output channels: implications for emotional functions and adaptive behavior. Neurosci Biobehav Rev 30:126-147.

Heimer L, Wilson RD (1975) The subcortical projections of allocortex: Similarities in the neuronal associations of the hippocampus, the piriform cortex and the neocortex. In: Golgi Centennial Symposium Proceedings (Santini M, ed), pp 173-193. New York: Raven Press.

Heinzel A, Bermpohl F, Niese R, Pfennig A, Pascual-Leone A, Schlaug G, Northoff G (2005) How do we modulate our emotions? Parametric fMRI reveals cortical midline structures as regions specifically involved in the processing of emotional valences. Brain Res Cogn Brain Res 25:348-358.

Heit G, Smith ME, Halgren E (1988) Neural encoding of individual words and faces by the human hippocampus and amygdala. Nature 333:773-775.

Heller W (1993) Neuropsychological mechanism of individual differences in emotion, personality, and arousal. Neuropsychology 7:476-489.

Helms G, Draganski B, Frackowiak R, Ashburner J, Weiskopf N (2009) Improved segmentation of deep brain grey matter structures using magnetization transfer (MT) parameter maps. Neuroimage 47:194-198.

Herman JP, Cullinan WE (1997) Neurocircuitry of stress: central control of the hypothalamo- pituitary-adrenocortical axis. Trends Neurosci 20:78-84.

Hervas I, Vilaro MT, Romero L, Scorza MC, Mengod G, Artigas F (2001) Desensitization of 5- HT(1A) autoreceptors by a low chronic fluoxetine dose effect of the concurrent administration of WAY-100635. Neuropsychopharmacology 24:11-20.

Hickok G, Poeppel D (2004) Dorsal and ventral streams: a framework for understanding aspects of the functional anatomy of language. Cognition 92:67-99.

Ho AP, Gillin JC, Buchsbaum MS, Wu JC, Abel L, Bunney WE, Jr. (1996) Brain glucose metabolism during non-rapid eye movement sleep in major depression. A positron emission tomography study. Arch Gen Psychiatry 53:645-652. 260 Hollon SD, Jarrett RB, Nierenberg AA, Thase ME, Trivedi M, Rush AJ (2005) Psychotherapy and medication in the treatment of adult and geriatric depression: which monotherapy or combined treatment? J Clin Psychiatry 66:455-468.

Holmes AP, Friston KJ (1998) Generalisability, Random Effects & Population Inference.

Holthoff VA, Beuthien-Baumann B, Zundorf G, Triemer A, Ludecke S, Winiecki P, Koch R, Fuchtner F, Herholz K (2004) Changes in brain metabolism associated with remission in unipolar major depression. Acta Psychiatr Scand 110:184-194.

Hooker CI, Paller KA, Gitelman DR, Parrish TB, Mesulam MM, Reber PJ (2003) Brain networks for analyzing eye gaze. Brain Res Cogn Brain Res 17:406-418.

Hooley JM, Gruber SA, Parker HA, Guillaumot J, Rogowska J, Yurgelun-Todd DA (2009) Cortico-limbic response to personally challenging emotional stimuli after complete recovery from depression. Psychiatry Res 172:83-91.

Hornak J, Rolls ET, Wade D (1996) Face and voice expression identification in patients with emotional and behavioural changes following ventral frontal lobe damage. Neuropsychologia 34:247-261.

Horowitz JM, Goyal A, Ramdeen N, Hallas BH, Horowitz AT, Torres G (2003) Characterization of fluoxetine plus olanzapine treatment in rats: a behavior, endocrine, and immediate-early gene expression analysis. Synapse 50:353-364.

Houle S, Ginovart N, Hussey D, Meyer JH, Wilson AA (2000) Imaging the serotonin transporter with positron emission tomography: initial human studies with [11C]DAPP and [11C]DASB. Eur J Nucl Med 27:1719-1722.

Hounsfield GN (1973) Computerized transverse axial scanning (tomography). 1. Description of system. Br J Radiol 46:1016-1022.

Hsu M, Bhatt M, Adolphs R, Tranel D, Camerer CF (2005) Neural systems responding to degrees of uncertainty in human decision-making. Science 310:1680-1683.

Hu S, Bu Y, Song Y, Zhen Z, Liu J (2009) Dissociation of attention and intention in human posterior parietal cortex: an fMRI study. Eur J Neurosci 29:2083-2091.

Hua lH, Strigo IA, Baxter LC, Johnson SC, Craig AD (2005) Anteroposterior somatotopy of innocuous cooling activation focus in human dorsal posterior insular cortex. Am J Physiol Regul Integr Comp Physiol 289:R319-R325.

Hugdahl K, Rund BR, Lund A, Asbjornsen A, Egeland J, Ersland L, Landro NI, Roness A, Stordal KI, Sundet K, Thomsen T (2004) Brain activation measured with fMRI during a mental arithmetic task in schizophrenia and major depression. Am J Psychiatry 161:286-293.

Husain MM, McDonald WM, Doraiswamy PM, Figiel GS, Na C, Escalona PR, Boyko OB, Nemeroff CB, Krishnan KR (1991) A magnetic resonance imaging study of putamen nuclei in major depression. Psychiatry Res 40:95-99.

Hutchison WD, Davis KD, Lozano AM, Tasker RR, Dostrovsky JO (1999) Pain-related neurons 261 in the human cingulate cortex. Nat Neurosci 2:403-405.

Iidaka T, Omori M, Murata T, Kosaka H, Yonekura Y, Okada T, Sadato N (2001) Neural interaction of the amygdala with the prefrontal and temporal cortices in the processing of facial expressions as revealed by fMRI. J Cogn Neurosci 13:1035-1047.

Irwin W, Anderle MJ, Abercrombie HC, Schaefer SM, Kalin NH, Davidson RJ (2004) Amygdalar interhemispheric functional connectivity differs between the non-depressed and depressed human brain. Neuroimage 21:674-686.

Isaacson RL (1972) The Limbic System. New York: Plenum.

Isenberg N, Silbersweig D, Engelien A, Emmerich S, Malavade K, Beattie B, Leon AC, Stern E (1999) Linguistic threat activates the human amygdala. Proc Natl Acad Sci U S A 96:10456- 10459.

Ito H, Kawashima R, Awata S, Ono S, Sato K, Goto R, Koyama M, Sato M, Fukuda H (1996) Hypoperfusion in the limbic system and prefrontal cortex in depression: SPECT with anatomic standardization technique. J Nucl Med 37:410-414.

Iwai E, Yukie M, Suyama H, Shirakawa S (1987) Amygdalar connections with middle and inferior temporal gyri of the monkey. Neurosci Lett 83:25-29.

Iyo M, Namba H, Fukushi K, Shinotoh H, Nagatsuka S, Suhara T, Sudo Y, Suzuki K, Irie T (1997) Measurement of acetylcholinesterase by positron emission tomography in the brains of healthy controls and patients with Alzheimer's disease. Lancet 349:1805-1809.

Jabbi M, Bastiaansen J, Keysers C (2008) A common anterior insula representation of disgust observation, experience and imagination shows divergent functional connectivity pathways. PLoS One 3:e2939.

Jabbi M, Keysers C (2008) Inferior frontal gyrus activity triggers anterior insula response to emotional facial expressions. Emotion 8:775-780.

James W (1884) What is an emotion? Mind 9:188-205.

Jann MW, Hon YY, Shamsi SA, Zheng J, Awad EA, Spratlin V (2006) Lack of pharmacokinetic interaction between lamotrigine and olanzapine in healthy volunteers. Pharmacotherapy 26:627- 633.

Janssen J, Hulshoff Pol HE, Lampe IK, Schnack HG, de Leeuw FE, Kahn RS, Heeren TJ (2004) Hippocampal changes and white matter lesions in early-onset depression. Biol Psychiatry 56:825-831.

Javitt DC (2001) Management of negative symptoms of schizophrenia. Curr Psychiatry Rep 3:413-417.

Jensen J, McIntosh AR, Crawley AP, Mikulis DJ, Remington G, Kapur S (2003) Direct activation of the ventral striatum in anticipation of aversive stimuli. Neuron 40:1251-1257.

Jensen J, Smith AJ, Willeit M, Crawley AP, Mikulis DJ, Vitcu I, Kapur S (2007) Separate brain

262 regions code for salience vs. valence during reward prediction in humans. Hum Brain Mapp 28:294-302.

Johnston JB (1923) Further contribution to the study of the evolution of the forebrain. J Comp Neurol 35:337-481.

Juckel G, Schlagenhauf F, Koslowski M, Filonov D, Wustenberg T, Villringer A, Knutson B, Kienast T, Gallinat J, Wrase J, Heinz A (2006) Dysfunction of ventral striatal reward prediction in schizophrenic patients treated with typical, not atypical, neuroleptics. Psychopharmacology (Berl) 187:222-228.

Judd LL, Akiskal HS, Maser JD, Zeller PJ, Endicott J, Coryell W, Paulus MP, Kunovac JL, Leon AC, Mueller TI, Rice JA, Keller MB (1998a) A prospective 12-year study of subsyndromal and syndromal depressive symptoms in unipolar major depressive disorders. Arch Gen Psychiatry 55:694-700.

Judd LL, Akiskal HS, Maser JD, Zeller PJ, Endicott J, Coryell W, Paulus MP, Kunovac JL, Leon AC, Mueller TI, Rice JA, Keller MB (1998b) Major depressive disorder: a prospective study of residual subthreshold depressive symptoms as predictor of rapid relapse. J Affect Disord 50:97- 108.

Judd LL, Akiskal HS, Zeller PJ, Paulus M, Leon AC, Maser JD, Endicott J, Coryell W, Kunovac JL, Mueller TI, Rice JP, Keller MB (2000a) Psychosocial disability during the long-term course of unipolar major depressive disorder. Arch Gen Psychiatry 57:375-380.

Judd LL, Paulus MJ, Schettler PJ, Akiskal HS, Endicott J, Leon AC, Maser JD, Mueller T, Solomon DA, Keller MB (2000b) Does incomplete recovery from first lifetime major depressive episode herald a chronic course of illness? Am J Psychiatry 157:1501-1504.

Jueptner M, Weiller C (1995) Review: does measurement of regional cerebral blood flow reflect synaptic activity? Implications for PET and fMRI. Neuroimage 2:148-156.

Kaada B (1960) Cingulate, posterior orbital, anterior insular and temporal pole cortex. In: Handbook of physiology, Section 1. Neurophysiology (Field J, Magoun HW, Hall VE, eds), pp 1345-1372. Washington, D.C.: American Physiological Society.

Kaas JH (1996) Theories of visual cortex organization in primates: areas of the third level. Prog Brain Res 112:213-221.

Kalin NH, Davidson RJ, Irwin W, Warner G, Orendi JL, Sutton SK, Mock BJ, Sorenson JA, Lowe M, Turski PA (1997) Functional magnetic resonance imaging studies of emotional processing in normal and depressed patients: effects of venlafaxine. J Clin Psychiatry 58 Suppl 16:32-39.

Kamondi A, Acsady L, Buzsaki G (1998) Dendritic spikes are enhanced by cooperative network activity in the intact hippocampus. J Neurosci 18:3919-3928.

Kapp BS, Supple WF, Jr., Whalen PJ (1994) Effects of electrical stimulation of the amygdaloid central nucleus on neocortical arousal in the rabbit. Behav Neurosci 108:81-93.

Karnath HO, Baier B, Nagele T (2005) Awareness of the functioning of one's own limbs 263 mediated by the insular cortex? J Neurosci 25:7134-7138.

Kasai K, Shenton ME, Salisbury DF, Onitsuka T, Toner SK, Yurgelun-Todd D, Kikinis R, Jolesz FA, McCarley RW (2003) Differences and similarities in insular and temporal pole MRI gray matter volume abnormalities in first-episode schizophrenia and affective psychosis. Arch Gen Psychiatry 60:1069-1077.

Kaufman J, Plotsky PM, Nemeroff CB, Charney DS (2000) Effects of early adverse experiences on brain structure and function: clinical implications. Biol Psychiatry 48:778-790.

Kawasaki H, Kaufman O, Damasio H, Damasio AR, Granner M, Bakken H, Hori T, Howard MA, III, Adolphs R (2001) Single-neuron responses to emotional visual stimuli recorded in human ventral prefrontal cortex. Nat Neurosci 4:15-16.

Kawashima R, Sugiura M, Kato T, Nakamura A, Hatano K, Ito K, Fukuda H, Kojima S, Nakamura K (1999) The human amygdala plays an important role in gaze monitoring. A PET study. Brain 122 ( Pt 4):779-783.

Keedwell PA, Andrew C, Williams SC, Brammer MJ, Phillips ML (2005a) A double dissociation of ventromedial prefrontal cortical responses to sad and happy stimuli in depressed and healthy individuals. Biol Psychiatry 58:495-503.

Keedwell PA, Andrew C, Williams SC, Brammer MJ, Phillips ML (2005b) The neural correlates of anhedonia in major depressive disorder. Biol Psychiatry 58:843-853.

Keightley ML, Seminowicz DA, Bagby RM, Costa PT, Fossati P, Mayberg HS (2003) Personality influences limbic-cortical interactions during sad mood induction. Neuroimage 20:2031-2039.

Keil A, Gruber T, Muller MM, Moratti S, Stolarova M, Bradley MM, Lang PJ (2003) Early modulation of visual perception by emotional arousal: evidence from steady-state visual evoked brain potentials. Cogn Affect Behav Neurosci 3:195-206.

Keller MB (2003) Past, present, and future directions for defining optimal treatment outcome in depression: remission and beyond. JAMA 289:3152-3160.

Keller MB (2004) Remission versus response: the new gold standard of antidepressant care. J Clin Psychiatry 65 Suppl 4:53-59.

Keller MB, Lavori PW, Kane JM, Gelenberg AJ, Rosenbaum JF, Walzer EA, Baker LA (1992a) Subsyndromal symptoms in bipolar disorder. A comparison of standard and low serum levels of lithium. Arch Gen Psychiatry 49:371-376.

Keller MB, Lavori PW, Mueller TI, Endicott J, Coryell W, Hirschfeld RM, Shea T (1992b) Time to recovery, chronicity, and levels of psychopathology in major depression. A 5-year prospective follow-up of 431 subjects. Arch Gen Psychiatry 49:809-816.

Kemp AH, Gray MA, Silberstein RB, Armstrong SM, Nathan PJ (2004) Augmentation of serotonin enhances pleasant and suppresses unpleasant cortical electrophysiological responses to visual emotional stimuli in humans. Neuroimage 22:1084-1096.

264 Kennedy N, Foy K, Sherazi R, McDonough M, McKeon P (2007a) Long-term social functioning after depression treated by psychiatrists: a review. Bipolar Disord 9:25-37.

Kennedy SH, Dickens SE, Eisfeld BS, Bagby RM (1999) Sexual dysfunction before antidepressant therapy in major depression. J Affect Disord 56:201-208.

Kennedy SH, Evans KR, Kruger S, Mayberg HS, Meyer JH, McCann S, Arifuzzman AI, Houle S, Vaccarino FJ (2001) Changes in regional brain glucose metabolism measured with positron emission tomography after paroxetine treatment of major depression. Am J Psychiatry 158:899- 905.

Kennedy SH, Konarski JZ, Segal ZV, Lau MA, Bieling PJ, McIntyre RS, Mayberg HS (2007b) Differences in Brain Glucose Metabolism Between Responders to CBT and Venlafaxine in a 16- Week Randomized Controlled Trial. Am J Psychiatry 164:778-788.

Kennedy SH, Lam RW, Parikh SV, Patten SB, Ravindran AV (2009a) Canadian Network for Mood and Anxiety Treatments (CANMAT) clinical guidelines for the management of major depressive disorder in adults. J Affect Disord.

Kennedy SH, Milev R, Giacobbe P, Ramasubbu R, Lam RW, Parikh SV, Patten SB, Ravindran AV (2009b) Canadian Network for Mood and Anxiety Treatments (CANMAT) Clinical guidelines for the management of major depressive disorder in adults. IV. Neurostimulation therapies. J Affect Disord 117 Suppl 1:S44-S53.

Kessing LV (1998) Cognitive impairment in the euthymic phase of affective disorder. Psychol Med 28:1027-1038.

Kessing LV (2007) Epidemiology of subtypes of depression. Acta Psychiatr Scand Suppl85-89.

Kessler RC, Berglund P, Demler O, Jin R, Koretz D, Merikangas KR, Rush AJ, Walters EE, Wang PS (2003) The epidemiology of major depressive disorder: results from the National Comorbidity Survey Replication (NCS-R). JAMA 289:3095-3105.

Ketter TA (2005) Neuroimaging in Mood Disorders (Book Chapter). (Drevets WC, ed).

Ketter TA, Kimbrell TA, George MS, Dunn RT, Speer AM, Benson BE, Willis MW, Danielson A, Frye MA, Herscovitch P, Post RM (2001) Effects of mood and subtype on cerebral glucose metabolism in treatment-resistant bipolar disorder. Biol Psychiatry 49:97-109.

Kety SS, Schmidt CF (1948) THE NITROUS OXIDE METHOD FOR THE QUANTITATIVE DETERMINATION OF CEREBRAL BLOOD FLOW IN MAN: THEORY, PROCEDURE AND NORMAL VALUES. J Clin Invest 27:476-483.

Khan A, Brodhead AE, Kolts RL (2004) Relative sensitivity of the Montgomery-Asberg depression rating scale, the Hamilton depression rating scale and the Clinical Global Impressions rating scale in antidepressant clinical trials: a replication analysis. Int Clin Psychopharmacol 19:157-160.

Khan A, Khan SR, Shankles EB, Polissar NL (2002) Relative sensitivity of the Montgomery- Asberg Depression Rating Scale, the Hamilton Depression rating scale and the Clinical Global Impressions rating scale in antidepressant clinical trials. Int Clin Psychopharmacol 17:281-285. 265 Kiebel SJ, Poline JB, Friston KJ, Holmes AP, Worsley KJ (1999) Robust smoothness estimation in statistical parametric maps using standardized residuals from the general linear model. Neuroimage 10:756-766.

Kilgard MP, Merzenich MM (1998) Cortical map reorganization enabled by nucleus basalis activity. Science 279:1714-1718.

Killgore WD, Oki M, Yurgelun-Todd DA (2001) Sex-specific developmental changes in amygdala responses to affective faces. Neuroreport 12:427-433.

Kilts CD, Gross RE, Ely TD, Drexler KP (2004) The neural correlates of cue-induced craving in cocaine-dependent women. Am J Psychiatry 161:233-241.

Kimbrell TA, George MS, Parekh PI, Ketter TA, Podell DM, Danielson AL, Repella JD, Benson BE, Willis MW, Herscovitch P, Post RM (1999) Regional brain activity during transient self- induced anxiety and anger in healthy adults. Biol Psychiatry 46:454-465.

Kimbrell TA, Ketter TA, George MS, Little JT, Benson BE, Willis MW, Herscovitch P, Post RM (2002) Regional cerebral glucose utilization in patients with a range of severities of unipolar depression. Biol Psychiatry 51:237-252.

Kingsley RE (2000) Concise Text of Neuroscience (2nd ed.). New York: Lippincott Williams and Wilkins.

Klein DF (1974) Endogenomorphic depression. A conceptual and terminological revision. Arch Gen Psychiatry 31:447-454.

Kleiser R, Konen CS, Seitz RJ, Bremmer F (2009) I know where you'll look: an fMRI study of oculomotor intention and a change of motor plan. Behav Brain Funct 5:27.

Kluver H, Bucy PC (1939) Preliminary analysis of functions of the temporal lobes in monkeys. Arch Neurol Psychiat 42:979-1000.

Knight DC, Nguyen HT, Bandettini PA (2005) The role of the human amygdala in the production of conditioned fear responses. Neuroimage 26:1193-1200.

Knops A, Thirion B, Hubbard EM, Michel V, Dehaene S (2009) Recruitment of an area involved in eye movements during mental arithmetic. Science 324:1583-1585.

Knutson B, Bhanji JP, Cooney RE, Atlas LY, Gotlib IH (2008) Neural responses to monetary incentives in major depression. Biol Psychiatry 63:686-692.

Knutson B, Fong GW, Adams CM, Varner JL, Hommer D (2001) Dissociation of reward anticipation and outcome with event-related fMRI. Neuroreport 12:3683-3687.

Kobayashi M, Inoue T, Matsuo R, Masuda Y, Hidaka O, Kang Y, Morimoto T (1997) Role of calcium conductances on spike afterpotentials in rat trigeminal motoneurons. J Neurophysiol 77:3273-3283.

Koch S, Perry KW, Bymaster FP (2004) Brain region and dose effects of an olanzapine/fluoxetine combination on extracellular monoamine concentrations in the rat.

266 Neuropharmacology 46:232-242.

Kodama M, Fujioka T, Duman RS (2004) Chronic olanzapine or fluoxetine administration increases cell proliferation in hippocampus and prefrontal cortex of adult rat. Biol Psychiatry 56:570-580.

Koenigs M, Huey ED, Calamia M, Raymont V, Tranel D, Grafman J (2008) Distinct regions of prefrontal cortex mediate resistance and vulnerability to depression. J Neurosci 28:12341-12348.

Kollias SS, Alkadhi H, Jaermann T, Crelier G, Hepp-Reymond MC (2001) Identification of multiple nonprimary motor cortical areas with simple movements. Brain Res Brain Res Rev 36:185-195.

Konarski JZ, Kennedy SH, Segal ZV, Lau MA, Bieling PJ, McIntyre RS, Mayberg HS (2009) Predictors of nonresponse to cognitive behavioural therapy or venlafaxine using glucose metabolism in major depressive disorder. J Psychiatry Neurosci 34:175-180.

Konarski JZ, McIntyre RS, Grupp LA, Kennedy SH (2004) Is the cerebellum relevant in the circuitry of neuropsychiatric disorders.

Konarski JZ, McIntyre RS, Kennedy SH, Rafi-Tari S, Soczynska JK, Ketter TA (2008) Volumetric neuroimaging investigations in mood disorders: bipolar disorder versus major depressive disorder. Bipolar Disord 10:1-37.

Kondo HM, Kashino M (2007) Neural mechanisms of auditory awareness underlying verbal transformations. Neuroimage 36:123-130.

Kring AM, Persons JB, Thomas C (2007) Changes in affect during treatment for depression and anxiety. Behav Res Ther 45:1753-1764.

Kringelbach ML (2005) The human orbitofrontal cortex: linking reward to hedonic experience. Nat Rev Neurosci 6:691-702.

Krishnan KR, McDonald WM, Escalona PR, Doraiswamy PM, Na C, Husain MM, Figiel GS, Boyko OB, Ellinwood EH, Nemeroff CB (1992) Magnetic resonance imaging of the caudate nuclei in depression. Preliminary observations. Arch Gen Psychiatry 49:553-557.

Kronhaus DM, Lawrence NS, Williams AM, Frangou S, Brammer MJ, Williams SC, Andrew CM, Phillips ML (2006) Stroop performance in bipolar disorder: further evidence for abnormalities in the ventral prefrontal cortex. Bipolar Disord 8:28-39.

Kruger S, Seminowicz D, Goldapple K, Kennedy SH, Mayberg HS (2003) State and trait influences on mood regulation in bipolar disorder: blood flow differences with an acute mood challenge. Biol Psychiatry 54:1274-1283.

Kuchinke L, Jacobs AM, Vo ML, Conrad M, Grubich C, Herrmann M (2006) Modulation of prefrontal cortex activation by emotional words in recognition memory. Neuroreport 17:1037- 1041.

Kuhl DE, Koeppe RA, Fessler JA, Minoshima S, Ackermann RJ, Carey JE, Gildersleeve DL, Frey KA, Wieland DM (1994) In vivo mapping of cholinergic neurons in the human brain using 267 SPECT and IBVM. J Nucl Med 35:405-410.

Kumar A, Jin Z, Bilker W, Udupa J, Gottlieb G (1998) Late-onset minor and major depression: early evidence for common neuroanatomical substrates detected by using MRI. Proc Natl Acad Sci U S A 95:7654-7658.

Kumar A, Newberg A, Alavi A, Berlin J, Smith R, Reivich M (1993) Regional cerebral glucose metabolism in late-life depression and Alzheimer disease: a preliminary positron emission tomography study. Proc Natl Acad Sci U S A 90:7019-7023.

Kumari V, Antonova E, Geyer MA, Ffytche D, Williams SC, Sharma T (2006) A fMRI investigation of startle gating deficits in schizophrenia patients treated with typical or atypical antipsychotics. Int J Neuropsychopharmacol1-15.

Kumari V, Mitterschiffthaler MT, Teasdale JD, Malhi GS, Brown RG, Giampietro V, Brammer MJ, Poon L, Simmons A, Williams SC, Checkley SA, Sharma T (2003) Neural abnormalities during cognitive generation of affect in treatment-resistant depression. Biol Psychiatry 54:777- 791.

LaBar KS, Gatenby JC, Gore JC, LeDoux JE, Phelps EA (1998) Human amygdala activation during conditioned fear acquisition and extinction: a mixed-trial fMRI study. Neuron 20:937- 945.

Labatut V, Pastor J, Ruff S, Demonet JF, Celsis P (2004) Cerebral modeling and dynamic Bayesian networks. Artif Intell Med 30:119-139.

Lacerda AL, Keshavan MS, Hardan AY, Yorbik O, Brambilla P, Sassi RB, Nicoletti M, Mallinger AG, Frank E, Kupfer DJ, Soares JC (2004) Anatomic evaluation of the orbitofrontal cortex in major depressive disorder. Biol Psychiatry 55:353-358.

Ladouceur CD, Dahl RE, Williamson DE, Birmaher B, Ryan ND, Casey BJ (2005) Altered emotional processing in pediatric anxiety, depression, and comorbid anxiety-depression. J Abnorm Child Psychol 33:165-177.

Lai T, Payne ME, Byrum CE, Steffens DC, Krishnan KR (2000) Reduction of orbital frontal cortex volume in geriatric depression. Biol Psychiatry 48:971-975.

Lam RW, Kennedy SH, Grigoriadis S, McIntyre RS, Milev R, Ramasubbu R, Parikh SV, Patten SB, Ravindran AV (2009) Canadian Network for Mood and Anxiety Treatments (CANMAT) clinical guidelines for the management of major depressive disorder in adults. III. Pharmacotherapy. J Affect Disord 117 Suppl 1:S26-S43.

Lambert MJ, Masters KS, Astle D (1988) An effect-size comparison of the Beck, Zung, and Hamilton rating scales for depression: a three-week and twelve-week analysis. Psychol Rep 63:467-470.

Lane P (2007) Handling drop-out in longitudinal clinical trials: a comparison of the LOCF and MMRM approaches. Pharm Stat.

Lane RD, Fink GR, Chau PM, Dolan RJ (1997a) Neural activation during selective attention to subjective emotional responses. Neuroreport 8:3969-3972. 268 Lane RD, Reiman EM, Ahern GL, Schwartz GE, Davidson RJ (1997b) Neuroanatomical correlates of happiness, sadness, and disgust. Am J Psychiatry 154:926-933.

Lane RD, Reiman EM, Axelrod B, Yun LS, Holmes A, Schwartz GE (1998) Neural correlates of levels of emotional awareness. Evidence of an interaction between emotion and attention in the anterior cingulate cortex. J Cogn Neurosci 10:525-535.

Lane RD, Reiman EM, Bradley MM, Lang PJ, Ahern GL, Davidson RJ, Schwartz GE (1997c) Neuroanatomical correlates of pleasant and unpleasant emotion. Neuropsychologia 35:1437- 1444.

Lang PJ, Bradley MM, Cuthbert BN (1990) Emotion, attention, and the startle reflex. Psychol Rev 97:377-395.

Lang PJ, Greenwald MK (1988a) The International Affective Picture System standardization procedure and initial group results for affective judgments: Technical report I A. Gainesville, FL: The Center for Research in Psychophysiology: University of Florida.

Lang PJ, Greenwald MK (1988b) The International Affective Picture System standardization procedure and initial group results for affective judgments: Technical report l B. Gainesville, FL: The Center for Research in Psychophysiology: University of Florida.

Lang PJ, Greenwald MK, Bradley MM, Hamm AO (1993) Looking at pictures: affective, facial, visceral, and behavioral reactions. Psychophysiology 30:261-273.

Lange K, Williams LM, Young AW, Bullmore ET, Brammer MJ, Williams SC, Gray JA, Phillips ML (2003) Task instructions modulate neural responses to fearful facial expressions. Biol Psychiatry 53:226-232.

Laurienti PJ, Burdette JH, Maldjian JA (2003) Separating neural processes using mixed event- related and epoch-based fMRI paradigms. J Neurosci Methods 131:41-50.

Lavretsky H, Kurbanyan K, Ballmaier M, Mintz J, Toga A, Kumar A (2004) Sex differences in brain structure in geriatric depression. Am J Geriatr Psychiatry 12:653-657.

Lawrence NS, Williams AM, Surguladze S, Giampietro V, Brammer MJ, Andrew C, Frangou S, Ecker C, Phillips ML (2004) Subcortical and ventral prefrontal cortical neural responses to facial expressions distinguish patients with bipolar disorder and major depression. Biol Psychiatry 55:578-587.

Lazarus RS (1991a) Cognition and motivation in emotion. Am Psychol 46:352-367.

Lazarus RS (1991b) Cognition and motivation in emotion. Am Psychol 46:352-367.

LeDoux JE (1993) Emotional memory systems in the brain. Behav Brain Res 58:69-79.

LeDoux JE (1996) The Emotional Brain. New York: Simon & Schuster.

LeDoux JE (2000) Emotion circuits in the brain. Annu Rev Neurosci 23:155-184.

Lee BT, Seok JH, Lee BC, Cho SW, Yoon BJ, Lee KU, Chae JH, Choi IG, Ham BJ (2008)

269 Neural correlates of affective processing in response to sad and angry facial stimuli in patients with major depressive disorder. Prog Neuropsychopharmacol Biol Psychiatry 32:778-785.

Lee BT, Seong WC, Hyung SK, Lee BC, Choi IG, In KL, Ham BJ (2007) The neural substrates of affective processing toward positive and negative affective pictures in patients with major depressive disorder. Prog Neuropsychopharmacol Biol Psychiatry.

Lee GP, Loring DW, Meader KJ, Brooks BB (1990) Hemispheric specialization for emotional expression: a reexamination of results from intracarotid administration of sodium amobarbital. Brain Cogn 12:267-280.

Lee GP, Meador KJ, Loring DW, Allison JD, Brown WS, Paul LK, Pillai JJ, Lavin TB (2004) Neural substrates of emotion as revealed by functional magnetic resonance imaging. Cogn Behav Neurol 17:9-17.

Lee SH, Payne ME, Steffens DC, McQuoid DR, Lai TJ, Provenzale JM, Krishnan KR (2003) Subcortical lesion severity and orbitofrontal cortex volume in geriatric depression. Biol Psychiatry 54:529-533.

Legatt AD, Arezzo J, Vaughan HG, Jr. (1980) Averaged multiple unit activity as an estimate of phasic changes in local neuronal activity: effects of volume-conducted potentials. J Neurosci Methods 2:203-217.

Leichnetz GR (2001) Connections of the medial posterior parietal cortex (area 7m) in the monkey. Anat Rec 263:215-236.

Lemogne C, le BG, Mayberg H, Volle E, Bergouignan L, Lehericy S, Allilaire JF, Fossati P (2009) In search of the depressive self: extended medial prefrontal network during self- referential processing in major depression. Soc Cogn Affect Neurosci.

Lethbridge R, Allen NB (2008) Mood induced cognitive and emotional reactivity, life stress, and the prediction of depressive relapse. Behav Res Ther.

Leventhal AG, Rodieck RW, Dreher B (1985) Central projections of cat retinal ganglion cells. J Comp Neurol 237:216-226.

Levesque J, Eugene F, Joanette Y, Paquette V, Mensour B, Beaudoin G, Leroux JM, Bourgouin P, Beauregard M (2003a) Neural circuitry underlying voluntary suppression of sadness. Biol Psychiatry 53:502-510.

Levesque J, Joanette Y, Mensour B, Beaudoin G, Leroux JM, Bourgouin P, Beauregard M (2003b) Neural correlates of sad feelings in healthy girls. Neuroscience 121:545-551.

Levesque J, Joanette Y, Mensour B, Beaudoin G, Leroux JM, Bourgouin P, Beauregard M (2004) Neural basis of emotional self-regulation in childhood. Neuroscience 129:361-369.

Levin B, Duchowny M (1991) Childhood obsessive-compulsive disorder and cingulate epilepsy. Biol Psychiatry 30:1049-1055.

Levin CS (2005) Primer on molecular imaging technology. Eur J Nucl Med Mol Imaging 32 Suppl 2:S325-S345. 270 Levisohn L, Cronin-Golomb A, Schmahmann JD (2000) Neuropsychological consequences of cerebellar tumour resection in children: cerebellar cognitive affective syndrome in a paediatric population. Brain 123 ( Pt 5):1041-1050.

Lewis BW, Clarke H (1878) The cortical lamination of the motor area of the brain. Proc R Soc Lond 27:49.

Leyton M, Young SN, Blier P, Ellenbogen MA, Palmour RM, Ghadirian AM, Benkelfat C (1997) The effect of tryptophan depletion on mood in medication-free, former patients with major affective disorder. Neuropsychopharmacology 16:294-297.

Lim KO, Rosenbloom MJ, Faustman WO, Sullivan EV, Pfefferbaum A (1999) Cortical gray matter deficit in patients with bipolar disorder. Schizophr Res 40:219-227.

Lindberg N, Virkkunen M, Tani P, Appelberg B, Virkkala J, Rimon R, Porkka-Heiskanen T (2002) Effect of a single-dose of olanzapine on sleep in healthy females and males. Int Clin Psychopharmacol 17:177-184.

Linke R, De Lima AD, Schwegler H, Pape HC (1999) Direct synaptic connections of axons from superior colliculus with identified thalamo-amygdaloid projection neurons in the rat: possible substrates of a subcortical visual pathway to the amygdala. J Comp Neurol 403:158-170.

Liotti M, Mayberg HS, McGinnis S, Brannan SL, Jerabek P (2002) Unmasking disease-specific cerebral blood flow abnormalities: mood challenge in patients with remitted unipolar depression. Am J Psychiatry 159:1830-1840.

Lisanby SH (2007) Electroconvulsive therapy for depression. N Engl J Med 357:1939-1945.

Little JT, Ketter TA, Kimbrell TA, Danielson A, Benson B, Willis MW, Post RM (1996) Venlafaxine or bupropion responders but not nonresponders show baseline prefrontal and paralimbic hypometabolism compared with controls. Psychopharmacol Bull 32:629-635.

Liu TT, Frank LR, Wong EC, Buxton RB (2001) Detection power, estimation efficiency, and predictability in event-related fMRI. Neuroimage 13:759-773.

Liu X, Powell DK, Wang H, Gold BT, Corbly CR, Joseph JE (2007) Functional dissociation in frontal and striatal areas for processing of positive and negative reward information. J Neurosci 27:4587-4597.

Lochhead RA, Parsey RV, Oquendo MA, Mann JJ (2004) Regional brain gray matter volume differences in patients with bipolar disorder as assessed by optimized voxel-based morphometry. Biol Psychiatry 55:1154-1162.

Loeber RT, Sherwood AR, Renshaw PF, Cohen BM, Yurgelun-Todd DA (1999) Differences in cerebellar blood volume in schizophrenia and bipolar disorder. Schizophr Res 37:81-89.

Logothetis NK (2002) The neural basis of the blood-oxygen-level-dependent functional magnetic resonance imaging signal. Philos Trans R Soc Lond B Biol Sci 357:1003-1037.

Logothetis NK, Pauls J, Augath M, Trinath T, Oeltermann A (2001) Neurophysiological investigation of the basis of the fMRI signal. Nature 412:150-157. 271 Logothetis NK, Pfeuffer J (2004) On the nature of the BOLD fMRI contrast mechanism. Magn Reson Imaging 22:1517-1531.

Logothetis NK, Wandell BA (2004) Interpreting the BOLD signal. Annu Rev Physiol 66:735- 769.

Lopez JF, Akil H, Watson SJ (1999) Neural circuits mediating stress. Biol Psychiatry 46:1461- 1471.

Lowe MJ, Mock BJ, Sorenson JA (1998) Functional connectivity in single and multislice echoplanar imaging using resting-state fluctuations. Neuroimage 7:119-132.

Lowen SB, Nickerson LD, Levin JM (2009) Differential effects of acute cocaine and placebo administration on visual cortical activation in healthy subjects measured using BOLD fMRI. Pharmacol Biochem Behav 92:277-282.

Lozano AM, Mayberg HS, Giacobbe P, Hamani C, Craddock RC, Kennedy SH (2008) Subcallosal Cingulate Gyrus Deep Brain Stimulation for Treatment-Resistant Depression. Biol Psychiatry.

Lu H, Yezhuvath US, Xiao G (2009) Improving fMRI sensitivity by normalization of basal physiologic state. Hum Brain Mapp.

Luiten PG, de Jong GI, Van der Zee EA, van DH (1996) Ultrastructural localization of cholinergic muscarinic receptors in rat brain cortical capillaries. Brain Res 720:225-229.

Lyoo IK, Hwang J, Sim M, Dunn BJ, Renshaw PF (2006) Advances in magnetic resonance imaging methods for the evaluation of bipolar disorder. CNS Spectr 11:269-280.

Lyoo IK, Kim MJ, Stoll AL, Demopulos CM, Parow AM, Dager SR, Friedman SD, Dunner DL, Renshaw PF (2004) Frontal lobe gray matter density decreases in bipolar I disorder. Biol Psychiatry 55:648-651.

Macias WL, Bergstrom RF, Cerimele BJ, Kassahun K, Tatum DE, Callaghan JT (1998) Lack of effect of olanzapine on the pharmacokinetics of a single aminophylline dose in healthy men. Pharmacotherapy 18:1237-1248.

MacLean PD (1990) The Triune Brain in Evolution. New York: Plenum.

MacLean PD (1952) Some psychiatric implications of physiological studies on frontotemporal portion of limbic system (visceral brain). Electroencephalogr Clin Neurophysiol Suppl 4:407- 418.

MacLean PD, Newman JD (1988) Role of midline frontolimbic cortex in production of the isolation call of squirrel monkeys. Brain Res 450:111-123.

MacMaster FP, Kusumakar V (2004) Hippocampal volume in early onset depression. BMC Med 2:2.

MacMillan S, Szeszko PR, Moore GJ, Madden R, Lorch E, Ivey J, Banerjee SP, Rosenberg DR (2003) Increased amygdala: hippocampal volume ratios associated with severity of anxiety in

272 pediatric major depression. J Child Adolesc Psychopharmacol 13:65-73.

Macpherson JM, Marangoz C, Miles TS, Wiesendanger M (1982) Microstimulation of the supplementary motor area (SMA) in the awake monkey. Exp Brain Res 45:410-416.

MacQueen GM, Campbell S, McEwen BS, Macdonald K, Amano S, Joffe RT, Nahmias C, Young LT (2003) Course of illness, hippocampal function, and hippocampal volume in major depression. Proc Natl Acad Sci U S A 100:1387-1392.

Magen H, Emmanouil TA, McMains SA, Kastner S, Treisman A (2009) Attentional demands predict short-term memory load response in posterior parietal cortex. Neuropsychologia 47:1790-1798.

Magistretti PJ, Pellerin L (1999) Cellular mechanisms of brain energy metabolism and their relevance to functional brain imaging. Philos Trans R Soc Lond B Biol Sci 354:1155-1163.

Maguire EA, Henson RN, Mummery CJ, Frith CD (2001) Activity in prefrontal cortex, not hippocampus, varies parametrically with the increasing remoteness of memories. Neuroreport 12:441-444.

Maier W, Philipp M, Heuser I, Schlegel S, Buller R, Wetzel H (1988) Improving depression severity assessment--I. Reliability, internal validity and sensitivity to change of three observer depression scales. J Psychiatr Res 22:3-12.

Makris N, Meyer JW, Bates JF, Yeterian EH, Kennedy DN, Caviness VS (1999) MRI-Based topographic parcellation of human cerebral white matter and nuclei II. Rationale and applications with systematics of cerebral connectivity. Neuroimage 9:18-45.

Malhi GS, Lagopoulos J, Sachdev P, Mitchell PB, Ivanovski B, Parker GB (2004a) Cognitive generation of affect in hypomania: an fMRI study. Bipolar Disord 6:271-285.

Malhi GS, Lagopoulos J, Ward PB, Kumari V, Mitchell PB, Parker GB, Ivanovski B, Sachdev P (2004b) Cognitive generation of affect in bipolar depression: an fMRI study. Eur J Neurosci 19:741-754.

Mallinckrodt CH, Kaiser CJ, Watkin JG, Molenberghs G, Carroll RJ (2004) The effect of correlation structure on treatment contrasts estimated from incomplete clinical trial data with likelihood-based repeated measures compared with last observation carried forward ANOVA. Clin Trials 1:477-489.

Manji HK (2009) Neuroimaging in developmental clinical neuroscience. New York: Cambridge University Press.

Maquet P, Degueldre C, Delfiore G, Aerts J, Peters JM, Luxen A, Franck G (1997) Functional neuroanatomy of human slow wave sleep. J Neurosci 17:2807-2812.

Maquet P, Peters J, Aerts J, Delfiore G, Degueldre C, Luxen A, Franck G (1996) Functional neuroanatomy of human rapid-eye-movement sleep and dreaming. Nature 383:163-166.

Maragnoli ME, Fumagalli F, Gennarelli M, Racagni G, Riva MA (2004) Fluoxetine and olanzapine have synergistic effects in the modulation of fibroblast growth factor 2 expression 273 within the rat brain. Biol Psychiatry 55:1095-1102.

Maravilla KR, Sory WC (1986) Magnetic resonance imaging of brain tumors. Semin Neurol 6:33-42.

Marciani L, Pfeiffer JC, Hort J, Head K, Bush D, Taylor AJ, Spiller RC, Francis S, Gowland PA (2006) Improved methods for fMRI studies of combined taste and aroma stimuli. J Neurosci Methods 158:186-194. margos-Bosch M, Artigas F, Adell A (2005) Effects of acute olanzapine after sustained fluoxetine on extracellular monoamine levels in the rat medial prefrontal cortex. Eur J Pharmacol 516:235-238.

Markowitsch HJ (1997) The functional neuroanatomy of episodic memory retrieval. Trends Neurosci 20:557-558.

Markowitsch HJ, Vandekerckhove MM, Lanfermann H, Russ MO (2003) Engagement of lateral and medial prefrontal areas in the ecphory of sad and happy autobiographical memories. Cortex 39:643-665.

Martin LJ, Powers RE, Dellovade TL, Price DL (1991) The bed nucleus-amygdala continuum in human and monkey. J Comp Neurol 309:445-485.

Martin SD, Martin E, Rai SS, Richardson MA, Royall R (2001) Brain blood flow changes in depressed patients treated with interpersonal psychotherapy or venlafaxine hydrochloride: preliminary findings. Arch Gen Psychiatry 58:641-648.

Martinot JL, Hardy P, Feline A, Huret JD, Mazoyer B, ttar-Levy D, Pappata S, Syrota A (1990) Left prefrontal glucose hypometabolism in the depressed state: a confirmation. Am J Psychiatry 147:1313-1317.

Marx CE, Shampine LJ, Khisti RT, Trost WT, Bradford DW, Grobin AC, Massing MW, Madison RD, Butterfield MI, Lieberman JA, Morrow AL (2006) Olanzapine and fluoxetine administration and coadministration increase rat hippocampal pregnenolone, allopregnanolone and peripheral deoxycorticosterone: implications for therapeutic actions. Pharmacol Biochem Behav 84:609-617.

Mataix-Cols D, Wooderson S, Lawrence N, Brammer MJ, Speckens A, Phillips ML (2004) Distinct neural correlates of washing, checking, and hoarding symptom dimensions in obsessive- compulsive disorder. Arch Gen Psychiatry 61:564-576.

Mather M, Mitchell KJ, Raye CL, Novak DL, Greene EJ, Johnson MK (2006) Emotional arousal can impair feature binding in working memory. J Cogn Neurosci 18:614-625.

Mathew RJ, Meyer JS, Semchuk KM, Francis D, Mortel K, Claghorn JL (1980) Regional cerebral blood flow in depression: a preliminary report. J Clin Psychiatry 41:71-72.

Mathiesen C, Caesar K, Akgoren N, Lauritzen M (1998) Modification of activity-dependent increases of cerebral blood flow by excitatory synaptic activity and spikes in rat cerebellar cortex. J Physiol 512 ( Pt 2):555-566.

274 Matthews JD, Bottonari KA, Polania LM, Mischoulon D, Dording CM, Irvin R, Fava M (2002) An open study of olanzapine and fluoxetine for psychotic major depressive disorder: interim analyses. J Clin Psychiatry 63:1164-1170.

Mayberg HS (1997) Limbic-cortical dysregulation: a proposed model of depression. J Neuropsychiatry Clin Neurosci 9:471-481.

Mayberg HS (2003a) Modulating dysfunctional limbic-cortical circuits in depression: towards development of brain-based algorithms for diagnosis and optimised treatment. Br Med Bull 65:193-207.

Mayberg HS (2003b) Positron emission tomography imaging in depression: a neural systems perspective. Neuroimaging Clin N Am 13:805-815.

Mayberg HS, Brannan SK, Mahurin RK, Jerabek PA, Brickman JS, Tekell JL, Silva JA, McGinnis S, Glass TG, Martin CC, Fox PT (1997) Cingulate function in depression: a potential predictor of treatment response. Neuroreport 8:1057-1061.

Mayberg HS, Brannan SK, Tekell JL, Silva JA, Mahurin RK, McGinnis S, Jerabek PA (2000) Regional metabolic effects of fluoxetine in major depression: serial changes and relationship to clinical response. Biol Psychiatry 48:830-843.

Mayberg HS, Lewis PJ, Regenold W, Wagner HN, Jr. (1994) Paralimbic hypoperfusion in unipolar depression. J Nucl Med 35:929-934.

Mayberg HS, Liotti M, Brannan SK, McGinnis S, Mahurin RK, Jerabek PA, Silva JA, Tekell JL, Martin CC, Lancaster JL, Fox PT (1999) Reciprocal limbic-cortical function and negative mood: converging PET findings in depression and normal sadness. Am J Psychiatry 156:675-682.

Mayberg HS, Lozano AM, Voon V, McNeely HE, Seminowicz D, Hamani C, Schwalb JM, Kennedy SH (2005) Deep brain stimulation for treatment-resistant depression. Neuron 45:651- 660.

Maziere M (1995) Cholinergic neurotransmission studied in vivo using positron emission tomography or single photon emission computerized tomography. Pharmacol Ther 66:83-101.

Mazziotta JC, Toga AW, Evans A, Fox P, Lancaster J (1995) A probabilistic atlas of the human brain: theory and rationale for its development. The International Consortium for Brain Mapping (ICBM). Neuroimage 2:89-101.

McCabe C, Cowen PJ, Harmer CJ (2009) Neural representation of reward in recovered depressed patients. Psychopharmacology (Berl).

McDonald AJ, Shammah-Lagnado SJ, Shi C, Davis M (1999) Cortical afferents to the extended amygdala. In: Advancing from the ventral striatum to the extended amygdala (McGinty JF, ed), pp 309-338. New York: New York Academy of Sciences.

McElroy SL, Altshuler LL, Suppes T, Keck PE, Jr., Frye MA, Denicoff KD, Nolen WA, Kupka RW, Leverich GS, Rochussen JR, Rush AJ, Post RM (2001) Axis I psychiatric comorbidity and its relationship to historical illness variables in 288 patients with bipolar disorder. Am J Psychiatry 158:420-426. 275 McGaugh JL (2004) The amygdala modulates the consolidation of memories of emotionally arousing experiences. Annu Rev Neurosci 27:1-28.

McIntosh AM, Forrester A, Lawrie SM, Byrne M, Harper A, Kestelman JN, Best JJ, Johnstone EC, Owens DG (2001) A factor model of the functional psychoses and the relationship of factors to clinical variables and brain morphology. Psychol Med 31:159-171.

McIntyre RS, Konarski JZ, Kennedy SH, Dickens SE, Bagby RM (2007) Measuring depressive symptoms in the naturalistic primary-care setting. Int J Clin Pract 61:1278-1282.

McIntyre RS, Konarski JZ, Mancini DA, Fulton KA, Parikh SV, Grigoriadis S, Grupp LA, Bakish D, Filteau MJ, Gorman C, Nemeroff CB, Kennedy SH (2005) Measuring the severity of depression and remission in primary care: validation of the HAMD-7 scale. CMAJ 173:1327- 1334.

McIntyre RS, Konarski JZ, Wilkins K, Soczynska JK, Kennedy SH (2006) Obesity in bipolar disorder and major depressive disorder: results from a national community health survey on mental health and well-being. Can J Psychiatry 51:274-280.

McIntyre RS, Konarski JZ, Yatham LN (2004) Comorbidity in bipolar disorder: a framework for rational treatment selection. Human Psychopharmacology 19:369-386.

McIntyre RS, Muzina DJ, Adams A, Lourenco MT, Law CW, Soczynska JK, Woldeyohannes HO, Nathanson J, Kennedy SH (2009) Quetiapine XR efficacy and tolerability as monotherapy and as adjunctive treatment to conventional antidepressants in the acute and maintenance treatment of major depressive disorder: a review of registration trials. Expert Opin Pharmacother 10:3061-3075.

McKeown MJ, Makeig S, Brown GG, Jung TP, Kindermann SS, Bell AJ, Sejnowski TJ (1998) Analysis of fMRI data by blind separation into independent spatial components. Hum Brain Mapp 6:160-188.

McNaughton N, Wickens J (2003) Hebb, pandemonium and catastrophic hypermnesia: the hippocampus as a suppressor of inappropriate associations. Cortex 39:1139-1163.

Mechelli A, Henson RN, Price CJ, Friston KJ (2003) Comparing event-related and epoch analysis in blocked design fMRI. Neuroimage 18:806-810.

Mee S, Bunney BG, Reist C, Potkin SG, Bunney WE (2006) Psychological pain: a review of evidence. J Psychiatr Res 40:680-690.

Mega MS, Cummings JL, Salloway S, Malloy P (1997) The limbic system: an anatomic, phylogenetic, and clinical perspective. J Neuropsychiatry Clin Neurosci 9:315-330.

Meisenzahl EM, Scheuerecker J, Zipse M, Ufer S, Wiesmann M, Frodl T, Koutsouleris N, Zetzsche T, Schmitt G, Riedel M, Spellmann I, Dehning S, Linn J, Bruckmann H, Moller HJ (2006) Effects of treatment with the atypical neuroleptic quetiapine on working memory function: a functional MRI follow-up investigation. Eur Arch Psychiatry Clin Neurosci 256:522- 531.

Mervaala E, Fohr J, Kononen M, Valkonen-Korhonen M, Vainio P, Partanen K, Partanen J, 276 Tiihonen J, Viinamaki H, Karjalainen AK, Lehtonen J (2000) Quantitative MRI of the hippocampus and amygdala in severe depression. Psychol Med 30:117-125.

Meseguer V, Romero MJ, Barros-Loscertales A, Belloch V, Bosch-Morell F, Romero J, Avila C (2007) Mapping the apetitive and aversive systems with emotional pictures using a block-design fMRI procedure. Psicothema 19:483-488.

Mesulam MM (2000) Principles of Behavioral and Cognitive Neurology. Oxford: Oxford University Press.

Mesulam MM, Geula C (1988) Nucleus basalis (Ch4) and cortical cholinergic innervation in the human brain: observations based on the distribution of acetylcholinesterase and choline acetyltransferase. J Comp Neurol 275:216-240.

Mesulam MM, Mufson EJ (1984) Neural inputs into the nucleus basalis of the substantia innominata (Ch4) in the rhesus monkey. Brain 107 ( Pt 1):253-274.

Mesulam MM, van Hoesen GW (1976) Acetylcholinesterase-rich projections from the basal forebrain of the rhesus monkey to neocortex. Brain Res 109:152-157.

Mesulam MM, van Hoesen GW, Pandya DN, Geschwind N (1977) Limbic and sensory connections of the inferior parietal lobule (area PG) in the rhesus monkey: a study with a new method for horseradish peroxidase histochemistry. Brain Res 136:393-414.

Meunier M, Bachevalier J, Mishkin M (1997) Effects of orbital frontal and anterior cingulate lesions on object and spatial memory in rhesus monkeys. Neuropsychologia 35:999-1015.

Meyer C (2004) Depressive disorders were the fourth leading cause of global disease burden in the year 2000. Evid Based Ment Health 7:123.

Meyer JH (2008) Applying neuroimaging ligands to study major depressive disorder. Semin Nucl Med 38:287-304.

Meyer JH, Wilson AA, Ginovart N, Goulding V, Hussey D, Hood K, Houle S (2001) Occupancy of serotonin transporters by paroxetine and citalopram during treatment of depression: a [(11)C]DASB PET imaging study. Am J Psychiatry 158:1843-1849.

Meynert T (1867) Der Bau der Grosshirnrinde und seine ortlichen Verschiedenheiten, nebst einem pathologischanatomischen Corollarium. In: Vierteljahresschrift fur Psychiattrie pp 77-93.

Meynert T (1885) A clinical treatise on diseases of the fore-brain based upon a study of its structure, functions, and nutrition. Part 1. The anatomy, physiology, and chemistry of the brain. New York: G.P. Putnam's Sons.

Miller DD, Andreasen NC, O'Leary DS, Watkins GL, Boles Ponto LL, Hichwa RD (2001) Comparison of the effects of risperidone and haloperidol on regional cerebral blood flow in schizophrenia. Biol Psychiatry 49:704-715.

Miller HL, Delgado PL, Salomon RM, Berman R, Krystal JH, Heninger GR, Charney DS (1996a) Clinical and biochemical effects of catecholamine depletion on antidepressant-induced remission of depression. Arch Gen Psychiatry 53:117-128. 277 Miller HL, Delgado PL, Salomon RM, Heninger GR, Charney DS (1996b) Effects of alpha- methyl-para-tyrosine (AMPT) in drug-free depressed patients. Neuropsychopharmacology 14:151-157.

Mitchell KJ, Mather M, Johnson MK, Raye CL, Greene EJ (2006) A functional magnetic resonance imaging investigation of short-term source and item memory for negative pictures. Neuroreport 17:1543-1547.

Mitterschiffthaler MT, Kumari V, Malhi GS, Brown RG, Giampietro VP, Brammer MJ, Suckling J, Poon L, Simmons A, Andrew C, Sharma T (2003) Neural response to pleasant stimuli in anhedonia: an fMRI study. Neuroreport 14:177-182.

Mitterschiffthaler MT, Williams SC, Walsh ND, Cleare AJ, Donaldson C, Scott J, Fu CH (2008) Neural basis of the emotional Stroop interference effect in major depression. Psychol Med 38:247-256.

Mitzdorf U (1987) Properties of the evoked potential generators: current source-density analysis of visually evoked potentials in the cat cortex. Int J Neurosci 33:33-59.

Mobbs D, Hagan CC, Azim E, Menon V, Reiss AL (2005) Personality predicts activity in reward and emotional regions associated with humor. Proc Natl Acad Sci U S A 102:16502-16506.

Mochizuki H, Sadato N, Saito DN, Toyoda H, Tashiro M, Okamura N, Yanai K (2007) Neural correlates of perceptual difference between itching and pain: a human fMRI study. Neuroimage 36:706-717.

Mohamed AA, Gotoh O, Graham DI, Osborne KA, McCulloch J, Mendelow AD, Teasdale GM, Harper AM (1985) Effect of pretreatment with the calcium antagonist nimodipine on local cerebral blood flow and histopathology after middle cerebral artery occlusion. Ann Neurol 18:705-711.

Mohanty A, Engels AS, Herrington JD, Heller W, Ho MH, Banich MT, Webb AG, Warren SL, Miller GA (2007) Differential engagement of anterior cingulate cortex subdivisions for cognitive and emotional function. Psychophysiology 44:343-351.

Moller HJ (2009) Standardised rating scales in psychiatry: methodological basis, their possibilities and limitations and descriptions of important rating scales. World J Biol Psychiatry 10:6-26.

Moncrieff J (2006) Psychiatric drug promotion and the politics of neo-liberalism. Br J Psychiatry 188:301-302.

Monk CS, Nelson EE, McClure EB, Mogg K, Bradley BP, Leibenluft E, Blair RJ, Chen G, Charney DS, Ernst M, Pine DS (2006) Ventrolateral prefrontal cortex activation and attentional bias in response to angry faces in adolescents with generalized anxiety disorder. Am J Psychiatry 163:1091-1097.

Monks PJ, Thompson JM, Bullmore ET, Suckling J, Brammer MJ, Williams SC, Simmons A, Giles N, Lloyd AJ, Harrison CL, Seal M, Murray RM, Ferrier IN, Young AH, Curtis VA (2004) A functional MRI study of working memory task in euthymic bipolar disorder: evidence for task- specific dysfunction. Bipolar Disord 6:550-564. 278 Montgomery SA, Asberg M (1979) A new depression scale designed to be sensitive to change. Br J Psychiatry 134:382-389.

Moore GJ, Bebchuk JM, Wilds IB, Chen G, Manji HK, Menji HK (2000a) Lithium-induced increase in human brain grey matter. Lancet 356:1241-1242.

Moore P, Landolt HP, Seifritz E, Clark C, Bhatti T, Kelsoe J, Rapaport M, Gillin JC (2000b) Clinical and physiological consequences of rapid tryptophan depletion. Neuropsychopharmacology 23:601-622.

Morecraft RJ, Louie JL, Herrick JL, Stilwell-Morecraft KS (2001) Cortical innervation of the facial nucleus in the non-human primate: a new interpretation of the effects of stroke and related subtotal brain trauma on the muscles of facial expression. Brain 124:176-208.

Morecraft RJ, van Hoesen GW (1992) Cingulate input to the primary and supplementary motor cortices in the rhesus monkey: evidence for somatotopy in areas 24c and 23c. J Comp Neurol 322:471-489.

Morecraft RJ, van Hoesen GW (1998) Convergence of limbic input to the cingulate motor cortex in the rhesus monkey. Brain Res Bull 45:209-232.

Morgan MA, LeDoux JE (1999) Contribution of ventrolateral prefrontal cortex to the acquisition and extinction of conditioned fear in rats. Neurobiol Learn Mem 72:244-251.

Morgan MA, Romanski LM, LeDoux JE (1993) Extinction of emotional learning: contribution of medial prefrontal cortex. Neurosci Lett 163:109-113.

Morgan ST, Hansen JC, Hillyard SA (1996) Selective attention to stimulus location modulates the steady-state visual evoked potential. Proc Natl Acad Sci U S A 93:4770-4774.

Morgane PJ, McFarland WL, Jacobs MS (1982) The limbic lobe of the dolphin brain: a quantitative cytoarchitectonic study. J Hirnforsch 23:465-552.

Morilak DA, Frazer A (2004) Antidepressants and brain monoaminergic systems: a dimensional approach to understanding their behavioural effects in depression and anxiety disorders. Int J Neuropsychopharmacol 7:193-218.

Morosan P, Rademacher J, Schleicher A, Amunts K, Schormann T, Zilles K (2001) Human primary auditory cortex: cytoarchitectonic subdivisions and mapping into a spatial reference system. Neuroimage 13:684-701.

Morris JS, deBonis M, Dolan RJ (2002) Human amygdala responses to fearful eyes. Neuroimage 17:214-222.

Morris JS, DeGelder B, WEISKRANTZ L, Dolan RJ (2001) Differential extrageniculostriate and amygdala responses to presentation of emotional faces in a cortically blind field. Brain 124:1241-1252.

Morris JS, Friston KJ, Buchel C, Frith CD, Young AW, Calder AJ, Dolan RJ (1998a) A neuromodulatory role for the human amygdala in processing emotional facial expressions. Brain 121 ( Pt 1):47-57. 279 Morris JS, Friston KJ, Buchel C, Frith CD, Young AW, Calder AJ, Dolan RJ (1998b) A neuromodulatory role for the human amygdala in processing emotional facial expressions. Brain 121 ( Pt 1):47-57.

Morris JS, Frith CD, Perrett DI, Rowland D, Young AW, Calder AJ, Dolan RJ (1996) A differential neural response in the human amygdala to fearful and happy facial expressions. Nature 383:812-815.

Morris JS, Ohman A, Dolan RJ (1998c) Conscious and unconscious emotional learning in the human amygdala. Nature 393:467-470.

Morys JM, Bobek-Billewicz B, Dziewiatkowski J, Ratajczak I, Pankiewicz P, Narkiewicz O, Morys J (2003) A magnetic resonance volumetric study of the temporal lobe structures in depression. Folia Morphol (Warsz ) 62:347-352.

Muller MJ, Himmerich H, Kienzle B, Szegedi A (2003) Differentiating moderate and severe depression using the Montgomery-Asberg depression rating scale (MADRS). J Affect Disord 77:255-260.

Murphy FC, Rubinsztein JS, Michael A, Rogers RD, Robbins TW, Paykel ES, Sahakian BJ (2001) Decision-making cognition in mania and depression. Psychol Med 31:679-693.

Murphy FC, Sahakian BJ, Rubinsztein JS, Michael A, Rogers RD, Robbins TW, Paykel ES (1999) Emotional bias and inhibitory control processes in mania and depression. Psychol Med 29:1307-1321.

Murphy FC, Smith KA, Cowen PJ, Robbins TW, Sahakian BJ (2002) The effects of tryptophan depletion on cognitive and affective processing in healthy volunteers. Psychopharmacology (Berl) 163:42-53.

Murray CJ, Lopez AD (1997) Global mortality, disability, and the contribution of risk factors: Global Burden of Disease Study. Lancet 349:1436-1442.

Myin-Germeys I, Peeters F, Havermans R, Nicolson NA, DeVries MW, Delespaul P, Van OJ (2003) Emotional reactivity to daily life stress in psychosis and affective disorder: an experience sampling study. Acta Psychiatr Scand 107:124-131.

Nagai M, Kishi K, Kato S (2007) Insular cortex and neuropsychiatric disorders: a review of recent literature. Eur Psychiatry 22:387-394.

Narayanan NS, Laubach M (2006) Top-down control of motor cortex ensembles by dorsomedial prefrontal cortex. Neuron 52:921-931.

Nashold BS, Jr., Slaughter DG (1969) Effects of stimulating or destroying the deep cerebellar regions in man. J Neurosurg 31:172-186.

Nelson EE, McClure EB, Monk CS, Zarahn E, Leibenluft E, Pine DS, Ernst M (2003) Developmental differences in neuronal engagement during implicit encoding of emotional faces: an event-related fMRI study. J Child Psychol Psychiatry 44:1015-1024.

Nicolelis MA, Ribeiro S (2002) Multielectrode recordings: the next steps. Curr Opin Neurobiol 280 12:602-606.

Nieto-Castanon A, Ghosh SS, Tourville JA, Guenther FH (2003) Region of interest based analysis of functional imaging data. Neuroimage 19:1303-1316.

Nieuwenhuys R, Voogd J, Van Huizen C (1988) The Human Central Nervous System. Berlin: Springer-Verlag.

Nitschke JB, Sarinopoulos I, Mackiewicz KL, Schaefer HS, Davidson RJ (2006) Functional neuroanatomy of aversion and its anticipation. Neuroimage 29:106-116.

Nobler MS, Sackeim HA, Prohovnik I, Moeller JR, Mukherjee S, Schnur DB, Prudic J, Devanand DP (1994) Regional cerebral blood flow in mood disorders, III. Treatment and clinical response. Arch Gen Psychiatry 51:884-897.

Noda H, Sugita S, Ikeda Y (1990) Afferent and efferent connections of the oculomotor region of the fastigial nucleus in the macaque monkey. J Comp Neurol 302:330-348.

Nofzinger EA, Buysse DJ, Germain A, Price JC, Meltzer CC, Miewald JM, Kupfer DJ (2005) Alterations in regional cerebral glucose metabolism across waking and non-rapid eye movement sleep in depression. Arch Gen Psychiatry 62:387-396.

Nofzinger EA, Nichols TE, Meltzer CC, Price J, Steppe DA, Miewald JM, Kupfer DJ, Moore RY (1999) Changes in forebrain function from waking to REM sleep in depression: preliminary analyses of [18F]FDG PET studies. Psychiatry Res 91:59-78.

Norbury R, Taylor MJ, Selvaraj S, Murphy SE, Harmer CJ, Cowen PJ (2009) Short-term antidepressant treatment modulates amygdala response to happy faces. Psychopharmacology (Berl).

Northoff G, Heinzel A, Bermpohl F, Niese R, Pfennig A, Pascual-Leone A, Schlaug G (2004) Reciprocal modulation and attenuation in the prefrontal cortex: an fMRI study on emotional- cognitive interaction. Hum Brain Mapp 21:202-212.

Northoff G, Richter A, Gessner M, Schlagenhauf F, Fell J, Baumgart F, Kaulisch T, Kotter R, Stephan KE, Leschinger A, Hagner T, Bargel B, Witzel T, Hinrichs H, Bogerts B, Scheich H, Heinze HJ (2000) Functional dissociation between medial and lateral prefrontal cortical spatiotemporal activation in negative and positive emotions: a combined fMRI/MEG study. Cereb Cortex 10:93-107.

Nutt D, Demyttenaere K, Janka Z, Aarre T, Bourin M, Canonico PL, Carrasco JL, Stahl S (2007) The other face of depression, reduced positive affect: the role of catecholamines in causation and cure. J Psychopharmacol 21:461-471.

Nutt DJ (2008) Relationship of neurotransmitters to the symptoms of major depressive disorder. J Clin Psychiatry 69 Suppl E1:4-7.

Nutt DJ, Glue P (1992) The neurobiology of ECT: animal studies. In: ECT: From Research to Clinical Practice (Coffey CE, ed), pp 241-265. Washington, DC: American Psychiatric Association.

281 O'Carroll RE, Drysdale E, Cahill L, Shajahan P, Ebmeier KP (1999) Stimulation of the noradrenergic system enhances and blockade reduces memory for emotional material in man. Psychol Med 29:1083-1088.

O'Doherty J, Kringelbach ML, Rolls ET, Hornak J, Andrews C (2001a) Abstract reward and punishment representations in the human orbitofrontal cortex. Nat Neurosci 4:95-102.

O'Doherty J, Rolls ET, Francis S, Bowtell R, McGlone F (2001b) Representation of pleasant and aversive taste in the human brain. J Neurophysiol 85:1315-1321.

O'Doherty J, Rolls ET, Francis S, Bowtell R, McGlone F, Kobal G, Renner B, Ahne G (2000) Sensory-specific satiety-related olfactory activation of the human orbitofrontal cortex. Neuroreport 11:893-897.

O'Keefe J, Nadel L (1978) The Hippocampus as a Cognitive Map. Oxford: Clarendon Press.

Oades RD, Halliday GM (1987) Ventral tegmental (A10) system: neurobiology. 1. Anatomy and connectivity. Brain Res 434:117-165.

Ochsner KN, Knierim K, Ludlow DH, Hanelin J, Ramachandran T, Glover G, Mackey SC (2004) Reflecting upon feelings: an fMRI study of neural systems supporting the attribution of emotion to self and other. J Cogn Neurosci 16:1746-1772.

Ochsner KN, Kosslyn SM, Cosgrove GR, Cassem EH, Price BH, Nierenberg AA, Rauch SL (2001) Deficits in visual cognition and attention following bilateral anterior cingulotomy. Neuropsychologia 39:219-230.

Oda K, Okubo Y, Ishida R, Murata Y, Ohta K, Matsuda T, Matsushima E, Ichimiya T, Suhara T, Shibuya H, Nishikawa T (2003) Regional cerebral blood flow in depressed patients with white matter magnetic resonance hyperintensity. Biol Psychiatry 53:150-156.

Ogawa S, Lee TM, Kay AR, Tank DW (1990) Brain magnetic resonance imaging with contrast dependent on blood oxygenation. Proc Natl Acad Sci U S A 87:9868-9872.

Ogawa S, Menon RS, Tank DW, Kim SG, Merkle H, Ellermann JM, Ugurbil K (1993) Functional brain mapping by blood oxygenation level-dependent contrast magnetic resonance imaging. A comparison of signal characteristics with a biophysical model. Biophys J 64:803- 812.

Ogino Y, Nemoto H, Inui K, Saito S, Kakigi R, Goto F (2007) Inner experience of pain: imagination of pain while viewing images showing painful events forms subjective pain representation in human brain. Cereb Cortex 17:1139-1146.

Ohman A (2005) The role of the amygdala in human fear: automatic detection of threat. Psychoneuroendocrinology 30:953-958.

Ongur D, Drevets WC, Price JL (1998) Glial reduction in the subgenual prefrontal cortex in mood disorders. Proc Natl Acad Sci U S A 95:13290-13295.

Ongur D, Price JL (2000) The organization of networks within the orbital and medial prefrontal cortex of rats, monkeys and humans. Cereb Cortex 10:206-219. 282 Ostroff RB, Nelson JC (1999) Risperidone augmentation of selective serotonin reuptake inhibitors in major depression. J Clin Psychiatry 60:256-259.

Ostrowsky K, Desestret V, Ryvlin P, Coste S, Mauguiere F (2002) Direct electrical stimulations of the temporal pole in human. Epileptic Disord 4 Suppl 1:S23-S27.

Osuch EA, Ketter TA, Kimbrell TA, George MS, Benson BE, Willis MW, Herscovitch P, Post RM (2000) Regional cerebral metabolism associated with anxiety symptoms in affective disorder patients. Biol Psychiatry 48:1020-1023.

Ottowitz WE, Dougherty DD, Savage CR (2002) The neural network basis for abnormalities of attention and executive function in major depressive disorder: implications for application of the medical disease model to psychiatric disorders. Harv Rev Psychiatry 10:86-99.

Paelecke-Habermann Y, Pohl J, Leplow B (2005) Attention and executive functions in remitted major depression patients. J Affect Disord 89:125-135.

Pagnoni G, Zink CF, Montague PR, Berns GS (2002) Activity in human ventral striatum locked to errors of reward prediction. Nat Neurosci 5:97-98.

Pakhomov S (2001) MNI Space Utility.

Pandya DN, Seltzer B (1982) Intrinsic connections and architectonics of posterior parietal cortex in the rhesus monkey. J Comp Neurol 204:196-210.

Papakostas GI, Petersen TJ, Nierenberg AA, Murakami JL, Alpert JE, Rosenbaum JF, Fava M (2004) Ziprasidone augmentation of selective serotonin reuptake inhibitors (SSRIs) for SSRI- resistant major depressive disorder. J Clin Psychiatry 65:217-221.

Papez JW (1937) A proposed mechanism of emotion. Arch Neurol Psychiat 38:725-744.

Pardo JV, Pardo PJ, Raichle ME (1993) Neural correlates of self-induced dysphoria. Am J Psychiatry 150:713-719.

Parikh SV, Lam RW (2001) Clinical guidelines for the treatment of depressive disorders, I. Definitions, prevalence, and health burden. Can J Psychiatry 46 Suppl 1:13S-20S.

Parikh SV, Segal ZV, Grigoriadis S, Ravindran AV, Kennedy SH, Lam RW, Patten SB (2009) Canadian Network for Mood and Anxiety Treatments (CANMAT) Clinical guidelines for the management of major depressive disorder in adults. II. Psychotherapy alone or in combination with antidepressant medication. J Affect Disord 117 Suppl 1:S15-S25.

Parsons AA (1991) 5-HT receptors in human and animal cerebrovasculature. Trends Pharmacol Sci 12:310-315.

Partiot A, Grafman J, Sadato N, Wachs J, Hallett M (1995) Brain activation during the generation of non-emotional and emotional plans. Neuroreport 6:1397-1400.

Pasley BN, Mayes LC, Schultz RT (2004) Subcortical discrimination of unperceived objects during binocular rivalry. Neuron 42:163-172.

283 Passchier J, van WA, Pieterman RM, Elsinga PH, Pruim J, Hendrikse HN, Willemsen AT, Vaalburg W (2000) In vivo delineation of 5-HT1A receptors in human brain with [18F]MPPF. J Nucl Med 41:1830-1835.

Patten SB, Kennedy SH, Lam RW, O'Donovan C, Filteau MJ, Parikh SV, Ravindran AV (2009) Canadian Network for Mood and Anxiety Treatments (CANMAT) clinical guidelines for the management of major depressive disorder in adults. I. Classification, burden and principles of management. J Affect Disord 117 Suppl 1:S5-14.

Paulus MP, Feinstein JS, Simmons A, Stein MB (2004) Anterior cingulate activation in high trait anxious subjects is related to altered error processing during decision making. Biol Psychiatry 55:1179-1187.

Paus T (2001) Primate anterior cingulate cortex: where motor control, drive and cognition interface. Nat Rev Neurosci 2:417-424.

Paus T, Koski L, Caramanos Z, Westbury C (1998) Regional differences in the effects of task difficulty and motor output on blood flow response in the human anterior cingulate cortex: a review of 107 PET activation studies. Neuroreport 9:R37-R47.

Paushter DM, Modic MT, Borkowski GP, Weinstein MA, Zeman RK (1984) Magnetic resonance. Principles and applications. Med Clin North Am 68:1393-1421.

Pawar GR, Agrawal RP, Phadnis P, Paliwal A, Vyas S, Solanki P (2009) Evaluation of antidepressant like property of amisulpride per se and its comparison with fluoxetine and olanzapine using forced swimming test in albino mice. Acta Pol Pharm 66:327-331.

Paykel ES (1998) Remission and residual symptomatology in major depression. Psychopathology 31:5-14.

Payne BR (1993) Evidence for visual cortical area homologs in cat and macaque monkey. Cereb Cortex 3:1-25.

Pearlson GD, Barta PE, Powers RE (1997) Medial and superior temporal gyral volumes and cerebral asymmetry in schizophrenia versus bipolar disorder. Biol Psychiatry 41:1-14.

Penfield W, Faulk ME (1955) The insula; further observations on its function. Brain 78:445-470.

Penny WD, Stephan KE, Mechelli A, Friston KJ (2004) Modelling functional integration: a comparison of structural equation and dynamic causal models. Neuroimage 23 Suppl 1:S264- S274.

Perahia DG, Quail D, Desaiah D, Montejo AL, Schatzberg AF (2009) Switching to duloxetine in selective serotonin reuptake inhibitor non- and partial-responders: effects on painful physical symptoms of depression. J Psychiatr Res 43:512-518.

Petersson KM, Nichols TE, Poline JB, Holmes AP (1999) Statistical limitations in functional neuroimaging. I. Non-inferential methods and statistical models. Philos Trans R Soc Lond B Biol Sci 354:1239-1260.

Petrides M (1994) Frontal lobes and behaviour. Curr Opin Neurobiol 4:207-211. 284 Petrides M (2005) Lateral prefrontal cortex: architectonic and functional organization. Philos Trans R Soc Lond B Biol Sci 360:781-795.

Petrides M, Pandya DN (1984) Projections to the frontal cortex from the posterior parietal region in the rhesus monkey. J Comp Neurol 228:105-116.

Petroff OA (2002) GABA and glutamate in the human brain. Neuroscientist 8:562-573.

Phan KL, Wager T, Taylor SF, Liberzon I (2002) Functional neuroanatomy of emotion: a meta- analysis of emotion activation studies in PET and fMRI. Neuroimage 16:331-348.

Phelps EA, Anderson AK (1997) Emotional memory: what does the amygdala do? Curr Biol 7:R311-R314.

Phelps EA, O'Connor KJ, Gatenby JC, Gore JC, Grillon C, Davis M (2001) Activation of the left amygdala to a cognitive representation of fear. Nat Neurosci 4:437-441.

Philip NS, Carpenter LL, Tyrka AR, Price LH (2008) Augmentation of antidepressants with atypical antipsychotics: a review of the current literature. J Psychiatr Pract 14:34-44.

Phillips ML, Bullmore ET, Howard R, Woodruff PW, Wright IC, Williams SC, Simmons A, Andrew C, Brammer M, David AS (1998a) Investigation of facial recognition memory and happy and sad facial expression perception: an fMRI study. Psychiatry Res 83:127-138.

Phillips ML, Drevets WC, Rauch SL, Lane R (2003a) Neurobiology of emotion perception I: The neural basis of normal emotion perception. Biol Psychiatry 54:504-514.

Phillips ML, Drevets WC, Rauch SL, Lane R (2003b) Neurobiology of emotion perception II: Implications for major psychiatric disorders. Biol Psychiatry 54:515-528.

Phillips ML, Medford N, Young AW, Williams L, Williams SC, Bullmore ET, Gray JA, Brammer MJ (2001) Time courses of left and right amygdalar responses to fearful facial expressions. Hum Brain Mapp 12:193-202.

Phillips ML, Williams LM, Heining M, Herba CM, Russell T, Andrew C, Bullmore ET, Brammer MJ, Williams SC, Morgan M, Young AW, Gray JA (2004) Differential neural responses to overt and covert presentations of facial expressions of fear and disgust. Neuroimage 21:1484-1496.

Phillips ML, Young AW, Scott SK, Calder AJ, Andrew C, Giampietro V, Williams SC, Bullmore ET, Brammer M, Gray JA (1998b) Neural responses to facial and vocal expressions of fear and disgust. Proc R Soc Lond B Biol Sci 265:1809-1817.

Phillips ML, Young AW, Senior C, Brammer M, Andrew C, Calder AJ, Bullmore ET, Perrett DI, Rowland D, Williams SC, Gray JA, David AS (1997) A specific neural substrate for perceiving facial expressions of disgust. Nature 389:495-498.

Piefke M, Weiss PH, Zilles K, Markowitsch HJ, Fink GR (2003) Differential remoteness and emotional tone modulate the neural correlates of autobiographical memory. Brain 126:650-668.

Pietrini P, Guazzelli M, Basso G, Jaffe K, Grafman J (2000) Neural correlates of imaginal

285 aggressive behavior assessed by positron emission tomography in healthy subjects. Am J Psychiatry 157:1772-1781.

Pillay SS, Yurgelun-Todd DA, Bonello CM, Lafer B, Fava M, Renshaw PF (1997) A quantitative magnetic resonance imaging study of cerebral and cerebellar gray matter volume in primary unipolar major depression: relationship to treatment response and clinical severity. Biol Psychiatry 42:79-84.

Pine DS, Lissek S, Klein RG, Mannuzza S, Moulton JL, III, Guardino M, Woldehawariat G (2004) Face-memory and emotion: associations with major depression in children and adolescents. J Child Psychol Psychiatry 45:1199-1208.

Pizzagalli D, Pascual-Marqui RD, Nitschke JB, Oakes TR, Larson CL, Abercrombie HC, Schaefer SM, Koger JV, Benca RM, Davidson RJ (2001) Anterior cingulate activity as a predictor of degree of treatment response in major depression: evidence from brain electrical tomography analysis. Am J Psychiatry 158:405-415.

Pizzagalli DA, Oakes TR, Fox AS, Chung MK, Larson CL, Abercrombie HC, Schaefer SM, Benca RM, Davidson RJ (2004) Functional but not structural subgenual prefrontal cortex abnormalities in melancholia. Mol Psychiatry 9:325, 393-325, 405.

Ploghaus A, Tracey I, Gati JS, Clare S, Menon RS, Matthews PM, Rawlins JN (1999) Dissociating pain from its anticipation in the human brain. Science 284:1979-1981.

Plutchik R (1984) Emotions: A general psychoevolutionary theory. In: Approaches to Emotion (Scerer KC, Ekman P, eds), pp 197-219. Hillsdale, NJ: Lawrence Erlbaum.

Podruchny TA, Connolly C, Bokde A, Herscovitch P, Eckelman WC, Kiesewetter DO, Sunderland T, Carson RE, Cohen RM (2003) In vivo muscarinic 2 receptor imaging in cognitively normal young and older volunteers. Synapse 48:39-44.

Porrino LJ, Lyons D, Miller MD, Smith HR, Friedman DP, Daunais JB, Nader MA (2002) Metabolic mapping of the effects of cocaine during the initial phases of self-administration in the nonhuman primate. J Neurosci 22:7687-7694.

Porro CA, Baraldi P, Pagnoni G, Serafini M, Facchin P, Maieron M, Nichelli P (2002) Does anticipation of pain affect cortical nociceptive systems? J Neurosci 22:3206-3214.

Posener JA, Wang L, Price JL, Gado MH, Province MA, Miller MI, Babb CM, Csernansky JG (2003) High-dimensional mapping of the hippocampus in depression. Am J Psychiatry 160:83- 89.

Post RM, DeLisi LE, Holcomb HH, Uhde TW, Cohen R, Buchsbaum MS (1987) Glucose utilization in the temporal cortex of affectively ill patients: positron emission tomography. Biol Psychiatry 22:545-553.

Pribram H, Barry JE (1956) Further behavioral analysis of parieto-temporo-preoccipital cortex. J Neurophysiol 19:99-106.

Price JL (1999) Prefrontal cortical networks related to visceral function and mood. Ann N Y Acad Sci 877:383-396. 286 Protopopescu X, Pan H, Tuescher O, Cloitre M, Goldstein M, Engelien W, Epstein J, Yang Y, Gorman J, LeDoux J, Silbersweig D, Stern E (2005) Differential time courses and specificity of amygdala activity in posttraumatic stress disorder subjects and normal control subjects. Biol Psychiatry 57:464-473.

Quraishi S, Frangou S (2002) Neuropsychology of bipolar disorder: a review. J Affect Disord 72:209-226.

Raichle ME (2000) A brief history of human functional brain mapping. In: Brain Mapping: The Systems (Toga AW, Mazziotta JC, eds), pp 33-75. San Diego: Academic.

Raichle ME, Fiez JA, Videen TO, MacLeod AM, Pardo JV, Fox PT, Petersen SE (1994) Practice-related changes in human brain functional anatomy during nonmotor learning. Cereb Cortex 4:8-26.

Rainville P, Duncan GH, Price DD, Carrier B, Bushnell MC (1997) Pain affect encoded in human anterior cingulate but not somatosensory cortex. Science 277:968-971.

Rainville P, Hofbauer RK, Paus T, Duncan GH, Bushnell MC, Price DD (1999) Cerebral mechanisms of hypnotic induction and suggestion. J Cogn Neurosci 11:110-125.

Rajkowska G (2000) Postmortem studies in mood disorders indicate altered numbers of neurons and glial cells. Biol Psychiatry 48:766-777.

Rajkowska G, Miguel-Hidalgo JJ, Wei J, Dilley G, Pittman SD, Meltzer HY, Overholser JC, Roth BL, Stockmeier CA (1999) Morphometric evidence for neuronal and glial prefrontal cell pathology in major depression. Biol Psychiatry 45:1085-1098.

Ramel W, Goldin PR, Eyler LT, Brown GG, Gotlib IH, McQuaid JR (2007) Amygdala reactivity and mood-congruent memory in individuals at risk for depressive relapse. Biol Psychiatry 61:231-239.

Ranote S, Elliott R, Abel KM, Mitchell R, Deakin JF, Appleby L (2004) The neural basis of maternal responsiveness to infants: an fMRI study. Neuroreport 15:1825-1829.

Ravindran AV, Lam RW, Filteau MJ, Lesperance F, Kennedy SH, Parikh SV, Patten SB (2009a) Canadian Network for Mood and Anxiety Treatments (CANMAT) Clinical guidelines for the management of major depressive disorder in adults. V. Complementary and alternative medicine treatments. J Affect Disord 117 Suppl 1:S54-S64.

Ravindran AV, Smith A, Cameron C, Bhatla R, Cameron I, Georgescu TM, Hogan MJ (2009b) Toward a functional neuroanatomy of dysthymia: A functional magnetic resonance imaging study. J Affect Disord.

Raylman RR, Majewski S, Lemieux SK, Velan SS, Kross B, Popov V, Smith MF, Weisenberger AG, Zorn C, Marano GD (2006) Simultaneous MRI and PET imaging of a rat brain. Phys Med Biol 51:6371-6379.

Regier DA, Boyd JH, Burke JD, Jr., Rae DS, Myers JK, Kramer M, Robins LN, George LK, Karno M, Locke BZ (1988) One-month prevalence of mental disorders in the United States. Based on five Epidemiologic Catchment Area sites. Arch Gen Psychiatry 45:977-986. 287 Regier DA, Kaelber CT, Rae DS, Farmer ME, Knauper B, Kessler RC, Norquist GS (1998) Limitations of diagnostic criteria and assessment instruments for mental disorders. Implications for research and policy. Arch Gen Psychiatry 55:109-115.

Reiman EM (1997) The application of positron emission tomography to the study of normal and pathologic emotions. J Clin Psychiatry 58 Suppl 16:4-12.

Reiman EM, Lane RD, Ahern GL, Schwartz GE, Davidson RJ, Friston KJ, Yun LS, Chen K (1997) Neuroanatomical correlates of externally and internally generated human emotion. Am J Psychiatry 154:918-925.

Riccardi P, Li R, Ansari MS, Zald D, Park S, Dawant B, Anderson S, Doop M, Woodward N, Schoenberg E, Schmidt D, Baldwin R, Kessler R (2006) Amphetamine-induced displacement of [18F] fallypride in striatum and extrastriatal regions in humans. Neuropsychopharmacology 31:1016-1026.

Riklan M, Marisak I, Cooper IS (1974) Psychological studies of chronic cerebellar stimulation in man. In: The cerebellum epilepsy and behavior. (Cooper IS, Riklan M, Snider RS, eds), pp 285- 342. New York: Plenum Press.

Rilling JK, Goldsmith DR, Glenn AL, Jairam MR, Elfenbein HA, Dagenais JE, Murdock CD, Pagnoni G (2008) The neural correlates of the affective response to unreciprocated cooperation. Neuropsychologia 46:1256-1266.

Roberson-Nay R, McClure EB, Monk CS, Nelson EE, Guyer AE, Fromm SJ, Charney DS, Leibenluft E, Blair J, Ernst M, Pine DS (2006) Increased amygdala activity during successful memory encoding in adolescent major depressive disorder: An FMRI study. Biol Psychiatry 60:966-973.

Robertson B, Wang L, Diaz MT, Aiello M, Gersing K, Beyer J, Mukundan S Jr, McCarthy G, Doraiswamy PM (2007) Effect of bupropion extended release on negative emotion processing in major depressive disorder: a pilot functional magnetic resonance imaging study. J Clin Psychiatry 68:261-267.

Roiser JP, Levy J, Fromm SJ, Nugent AC, Talagala SL, Hasler G, Henn FA, Sahakian BJ, Drevets WC (2009) The Effects of Tryptophan Depletion on Neural Responses to Emotional Words in Remitted Depression. Biol Psychiatry.

Rolls ET (1994) Central taste anatomy and physiology. In: Handbook of Clinical Olfaction and Gustation (Doty RL, ed), New York: Dekker.

Rolls ET (2000) The orbitofrontal cortex and reward. Cereb Cortex 10:284-294.

Rolls ET (2004) The functions of the orbitofrontal cortex. Brain Cogn 55:11-29.

Rolls ET, McCabe C, Redoute J (2007) Expected Value, Reward Outcome, and Temporal Difference Error Representations in a Probabilistic Decision Task. Cereb Cortex.

Rolls ET, O'Doherty J, Kringelbach ML, Francis S, Bowtell R, McGlone F (2003) Representations of pleasant and painful touch in the human orbitofrontal and cingulate cortices. Cereb Cortex 13:308-317. 288 Rose EJ, Simonotto E, Spencer EP, Ebmeier KP (2006) The effects of escitalopram on working memory and brain activity in healthy adults during performance of the n-back task. Psychopharmacology (Berl) 185:339-347.

Rosen HJ, Wilson MR, Schauer GF, Allison S, Gorno-Tempini ML, Pace-Savitsky C, Kramer JH, Levenson RW, Weiner M, Miller BL (2006) Neuroanatomical correlates of impaired recognition of emotion in dementia. Neuropsychologia 44:365-373.

Rothemund Y, Preuschhof C, Bohner G, Bauknecht HC, Klingebiel R, Flor H, Klapp BF (2007) Differential activation of the dorsal striatum by high-calorie visual food stimuli in obese individuals. Neuroimage.

Rowe JB, Toni I, Josephs O, Frackowiak RS, Passingham RE (2000) The prefrontal cortex: response selection or maintenance within working memory? Science 288:1656-1660.

Rowley HL, Needham PL, Kilpatrick IC, Heal DJ (2000) A comparison of the acute effects of zotepine and other antipsychotics on rat cortical dopamine release, in vivo. Naunyn Schmiedebergs Arch Pharmacol 361:187-192.

Roy CS, Sherrington CS (1890) On the Regulation of the Blood-supply of the Brain. J Physiol 11:85-158.

Rubinow DR, Post RM (1992) Impaired recognition of affect in facial expression in depressed patients. Biol Psychiatry 31:947-953.

Rudolph RL (1999) Goal of antidepressant therapy: response or remission and recovery? J Clin Psychiatry 60 Suppl 6:3-4.

Rush AJ (2007) STAR*D: what have we learned? Am J Psychiatry 164:201-204.

Rush AJ, Trivedi MH, Wisniewski SR, Nierenberg AA, Stewart JW, Warden D, Niederehe G, Thase ME, Lavori PW, Lebowitz BD, McGrath PJ, Rosenbaum JF, Sackeim HA, Kupfer DJ, Luther J, Fava M (2006a) Acute and longer-term outcomes in depressed outpatients requiring one or several treatment steps: a STAR*D report. Am J Psychiatry 163:1905-1917.

Rush AJ, Trivedi MH, Wisniewski SR, Stewart JW, Nierenberg AA, Thase ME, Ritz L, Biggs MM, Warden D, Luther JF, Shores-Wilson K, Niederehe G, Fava M (2006b) Bupropion-SR, sertraline, or venlafaxine-XR after failure of SSRIs for depression. N Engl J Med 354:1231- 1242.

Russell JA (1980) A circumplex model of affect. Journal of Personality and Social Psychology 39:1161-1178.

Russell JA (1991) Culture and the categorization of emotions. Psychol Bull 110:426-450.

Salokangas RK, Cannon T, Van Erp T, Ilonen T, Taiminen T, Karlsson H, Lauerma H, Leinonen KM, Wallenius E, Kaljonen A, Syvalahti E, Vilkman H, Alanen A, Hietala J (2002) Structural magnetic resonance imaging in patients with first-episode schizophrenia, psychotic and severe non-psychotic depression and healthy controls. Results of the schizophrenia and affective psychoses (SAP) project. Br J Psychiatry Suppl 43:s58-s65.

289 Sambataro F, Dimalta S, Di GA, Taurisano P, Blasi G, Scarabino T, Giannatempo G, Nardini M, Bertolino A (2006) Preferential responses in amygdala and insula during presentation of facial contempt and disgust. Eur J Neurosci 24:2355-2362.

Sanches M, Sassi RB, Axelson D, Nicoletti M, Brambilla P, Hatch JP, Keshavan MS, Ryan ND, Birmaher B, Soares JC (2005) Subgenual prefrontal cortex of child and adolescent bipolar patients: a morphometric magnetic resonance imaging study. Psychiatry Res 138:43-49.

Sander D, Grandjean D, Pourtois G, Schwartz S, Seghier ML, Scherer KR, Vuilleumier P (2005) Emotion and attention interactions in social cognition: brain regions involved in processing anger prosody. Neuroimage 28:848-858.

Sanders SK, Shekhar A (1991) Blockade of GABAA receptors in the region of the anterior basolateral amygdala of rats elicits increases in heart rate and blood pressure. Brain Res 567:101- 110.

Sanides F (1970) Functional architecture of motor and sensory cortices in primates in the light of a new concept of neocortex evolution. In: The Primate Brain: Advances in Primatology (Noback CR, Montagna W, eds), pp 137-201. New York: Appleton-Century-Crofts.

Santel S, Baving L, Krauel K, Munte TF, Rotte M (2006) Hunger and satiety in anorexia nervosa: fMRI during cognitive processing of food pictures. Brain Res 1114:138-148.

Sarkissov SA, Filimonoff IN, Kononowa EP, Preobraschenkskaja IS, Kukuew LA (1955) Atlas of the cytoarchitectonics of the human cerebral cortex. Moscow: Medgiz.

Sassi RB, Nicoletti M, Brambilla P, Harenski K, Mallinger AG, Frank E, Kupfer DJ, Keshavan MS, Soares JC (2001) Decreased pituitary volume in patients with bipolar disorder. Biol Psychiatry 50:271-280.

Satoh M, Takeda K, Nagata K, Hatazawa J, Kuzuhara S (2001) Activated brain regions in musicians during an ensemble: a PET study. Brain Res Cogn Brain Res 12:101-108.

Sax KW, Strakowski SM, Zimmerman ME, DelBello MP, Keck PE, Jr., Hawkins JM (1999) Frontosubcortical neuroanatomy and the continuous performance test in mania. Am J Psychiatry 156:139-141.

Saxena S, Brody AL, Ho ML, Alborzian S, Ho MK, Maidment KM, Huang SC, Wu HM, Au SC, Baxter LR, Jr. (2001) Cerebral metabolism in major depression and obsessive-compulsive disorder occurring separately and concurrently. Biol Psychiatry 50:159-170.

Saxena S, Brody AL, Ho ML, Alborzian S, Maidment KM, Zohrabi N, Ho MK, Huang SC, Wu HM, Baxter LR, Jr. (2002) Differential cerebral metabolic changes with paroxetine treatment of obsessive-compulsive disorder vs major depression. Arch Gen Psychiatry 59:250-261.

Saxena S, Brody AL, Ho ML, Zohrabi N, Maidment KM, Baxter LR, Jr. (2003) Differential brain metabolic predictors of response to paroxetine in obsessive-compulsive disorder versus major depression. Am J Psychiatry 160:522-532.

Schachter SC, Singer J (1962) Cognitive, social, and physiological determinants of emotional state. Psychol Rev 69:379-399. 290 Schaefer HS, Putnam KM, Benca RM, Davidson RJ (2006) Event-related functional magnetic resonance imaging measures of neural activity to positive social stimuli in pre- and post- treatment depression. Biol Psychiatry 60:974-986.

Schilbach L, Wohlschlaeger AM, Kraemer NC, Newen A, Shah NJ, Fink GR, Vogeley K (2006) Being with virtual others: Neural correlates of social interaction. Neuropsychologia 44:718-730.

Schiller PH, Tehovnik EJ (2001) Look and see: how the brain moves your eyes about. Prog Brain Res 134:127-142.

Schmahmann JD (1991) An emerging concept: The cerebellar contribution to higher funciton. Arch Neurol 48:1178-1187.

Schmahmann JD (2000) The role of the cerebellum in affect and psychosis. J Neuroling 13:189- 214.

Schmahmann JD, Sherman JC (1998) The cerebellar cognitive affective syndrome. Brain 121 ( Pt 4):561-579.

Schmid SP, Freid CM, Hollon SD, DeRubeis RJ (2002) Negative affect, positive affect, and autonomic arousal over the course of treatment of major depression.

Schoenbaum G, Chiba AA, Gallagher M (1998) Orbitofrontal cortex and basolateral amygdala encode expected outcomes during learning. Nat Neurosci 1:155-159.

Schreckenberger M, Siessmeier T, Viertmann A, Landvogt C, Buchholz HG, Rolke R, Treede RD, Bartenstein P, Birklein F (2005) The unpleasantness of tonic pain is encoded by the insular cortex. Neurology 64:1175-1183.

Schultz W, Tremblay L, Hollerman JR (2000) Reward processing in primate orbitofrontal cortex and basal ganglia. Cereb Cortex 10:272-284.

Schut JW (1950) Hereditary ataxia. Arch Neurol Psychiat 63:535-568.

Schwartz CE, Wright CI, Shin LM, Kagan J, Whalen PJ, McMullin KG, Rauch SL (2003) Differential amygdalar response to novel versus newly familiar neutral faces: a functional MRI probe developed for studying inhibited temperament. Biol Psychiatry 53:854-862.

Scott SK, Young AW, Calder AJ, Hellawell DJ, Aggleton JP, Johnson M (1997) Impaired auditory recognition of fear and anger following bilateral amygdala lesions. Nature 385:254-257.

Seager MA, Barth VN, Phebus LA, Rasmussen K (2005) Chronic coadministration of olanzapine and fluoxetine activates locus coeruleus neurons in rats: implications for bipolar disorder. Psychopharmacology (Berl) 181:126-133.

Seager MA, Huff KD, Barth VN, Phebus LA, Rasmussen K (2004) Fluoxetine administration potentiates the effect of olanzapine on locus coeruleus neuronal activity. Biol Psychiatry 55:1103-1109.

Segal ZV, Gemar M, Truchon C, Guirguis M, Horowitz LM (1995) A priming methodology for studying self-representation in major depressive disorder. J Abnorm Psychol 104:205-213.

291 Segal ZV, Kennedy SH, Gemar M, Hood K, Pedersen R, Buis T (2006) Cognitive reactivity to sad mood provocation and the prediction of depressive relapse.

Selemon LD, Goldman-Rakic PS (1988) Common cortical and subcortical targets of the dorsolateral prefrontal and posterior parietal cortices in the rhesus monkey: evidence for a distributed neural network subserving spatially guided behavior. J Neurosci 8:4049-4068.

Seminowicz DA, Davis KD (2006) Cortical responses to pain in healthy individuals depends on pain catastrophizing. Pain 120:297-306.

Seminowicz DA, Davis KD (2007a) A re-examination of pain-cognition interactions: implications for neuroimaging. Pain 130:8-13.

Seminowicz DA, Davis KD (2007b) Interactions of pain intensity and cognitive load: the brain stays on task. Cereb Cortex 17:1412-1422.

Seminowicz DA, Mayberg HS, McIntosh AR, Goldapple K, Kennedy S, Segal Z, Rafi-Tari S (2004) Limbic-frontal circuitry in major depression: a path modeling metanalysis. Neuroimage 22:409-418.

Serra M, Salgado-Pineda P, Delaveau P, Fakra E, Gasto C, Blin O (2006) Effects of antidepressant drugs on emotion. Clin Neuropharmacol 29:170-185.

Sesack SR, Deutch AY, Roth RH, Bunney BS (1989) Topographical organization of the efferent projections of the medial prefrontal cortex in the rat: an anterograde tract-tracing study with Phaseolus vulgaris leucoagglutinin. J Comp Neurol 290:213-242.

Shah PJ, Ebmeier KP, Glabus MF, Goodwin GM (1998) Cortical grey matter reductions associated with treatment-resistant chronic unipolar depression. Controlled magnetic resonance imaging study. Br J Psychiatry 172:527-532.

Shah PJ, Glabus MF, Goodwin GM, Ebmeier KP (2002) Chronic, treatment-resistant depression and right fronto-striatal atrophy. Br J Psychiatry 180:434-440.

Shah SA, Doraiswamy PM, Husain MM, Escalona PR, Na C, Figiel GS, Patterson LJ, Ellinwood EH, Jr., McDonald WM, Boyko OB, . (1992) Posterior fossa abnormalities in major depression: a controlled magnetic resonance imaging study. Acta Psychiatr Scand 85:474-479.

Shamay-Tsoory SG, Lester H, Chisin R, Israel O, Bar-Shalom R, Peretz A, Tomer R, Tsitrinbaum Z, haron-Peretz J (2005) The neural correlates of understanding the other's distress: a positron emission tomography investigation of accurate empathy. Neuroimage 27:468-472.

Sharma R, Sharma A (2004) Physiological basis and image processing in functional magnetic resonance imaging: neuronal and motor activity in brain. Biomed Eng Online 3:13.

Sharpley AL, Bhagwagar Z, Hafizi S, Whale WR, Gijsman HJ, Cowen PJ (2003) Risperidone augmentation decreases rapid eye movement sleep and decreases wake in treatment-resistant depressed patients. J Clin Psychiatry 64:192-196.

Sharpley AL, Vassallo CM, Cowen PJ (2000) Olanzapine increases slow-wave sleep: evidence for blockade of central 5-HT(2C) receptors in vivo. Biol Psychiatry 47:468-470. 292 Sheline YI, Barch DM, Donnelly JM, Ollinger JM, Snyder AZ, Mintun MA (2001) Increased amygdala response to masked emotional faces in depressed subjects resolves with antidepressant treatment: an fMRI study. Biol Psychiatry 50:651-658.

Sheline YI, Gado MH, Price JL (1998) Amygdala core nuclei volumes are decreased in recurrent major depression. Neuroreport 9:2023-2028.

Sheline YI, Sanghavi M, Mintun MA, Gado MH (1999) Depression duration but not age predicts hippocampal volume loss in medically healthy women with recurrent major depression. J Neurosci 19:5034-5043.

Sheline YI, Wang PW, Gado MH, Csernansky JG, Vannier MW (1996) Hippocampal atrophy in recurrent major depression. Proc Natl Acad Sci U S A 93:3908-3913.

Shelton RC, Tollefson GD, Tohen M, Stahl S, Gannon KS, Jacobs TG, Buras WR, Bymaster FP, Zhang W, Spencer KA, Feldman PD, Meltzer HY (2001) A novel augmentation strategy for treating resistant major depression. Am J Psychiatry 158:131-134.

Shelton RC, Williamson DJ, Corya SA, Sanger TM, Van Campen LE, Case M, Briggs SD, Tollefson GD (2005) Olanzapine/fluoxetine combination for treatment-resistant depression: a controlled study of SSRI and nortriptyline resistance. J Clin Psychiatry 66:1289-1297.

Shin LM, Dougherty DD, Orr SP, Pitman RK, Lasko M, Macklin ML, Alpert NM, Fischman AJ, Rauch SL (2000) Activation of anterior paralimbic structures during guilt-related script-driven imagery. Biol Psychiatry 48:43-50.

Shulman RG, Rothman DL, Behar KL, Hyder F (2004) Energetic basis of brain activity: implications for neuroimaging. Trends Neurosci 27:489-495.

Siegle GJ, Carter CS, Thase ME (2006) Use of FMRI to predict recovery from unipolar depression with cognitive behavior therapy. Am J Psychiatry 163:735-738.

Siegle GJ, Konecky RO, Thase ME, Carter CS (2003) Relationships between amygdala volume and activity during emotional information processing tasks in depressed and never-depressed individuals: an fMRI investigation. Ann N Y Acad Sci 985:481-484.

Siegle GJ, Steinhauer SR, Thase ME, Stenger VA, Carter CS (2002) Can't shake that feeling: event-related fMRI assessment of sustained amygdala activity in response to emotional information in depressed individuals. Biol Psychiatry 51:693-707.

Siegle GJ, Thompson W, Carter CS, Steinhauer SR, Thase ME (2007) Increased amygdala and decreased dorsolateral prefrontal BOLD responses in unipolar depression: related and independent features. Biol Psychiatry 61:198-209.

Simon GE, VonKorff M, Piccinelli M, Fullerton C, Ormel J (1999) An international study of the relation between somatic symptoms and depression. N Engl J Med 341:1329-1335.

Simpson JR, Jr., Snyder AZ, Gusnard DA, Raichle ME (2001) Emotion-induced changes in human medial prefrontal cortex: I. During cognitive task performance. Proc Natl Acad Sci U S A 98:683-687.

293 Singer T, Kiebel SJ, Winston JS, Dolan RJ, Frith CD (2004a) Brain responses to the acquired moral status of faces. Neuron 41:653-662.

Singer T, Seymour B, O'Doherty J, Kaube H, Dolan RJ, Frith CD (2004b) Empathy for pain involves the affective but not sensory components of pain. Science 303:1157-1162.

Sjoerdsma A, Engelman K, Spector S, Udenfriend S (1965) Inhibition of catecholamine synthesis in man with alpha-methyl-tyrosine, an inhibitor of tyrosine hydroxylase. Lancet 2:1092-1094.

Small DM, Voss J, Mak YE, Simmons KB, Parrish T, Gitelman D (2004) Experience-dependent neural integration of taste and smell in the human brain. J Neurophysiol 92:1892-1903.

Small DM, Zald DH, Jones-Gotman M, Zatorre RJ, Pardo JV, Frey S, Petrides M (1999) Human cortical gustatory areas: a review of functional neuroimaging data. Neuroreport 10:7-14.

Smeets PA, de GC, Stafleu A, van Osch MJ, Nievelstein RA, van der GJ (2006) Effect of satiety on brain activation during chocolate tasting in men and women. Am J Clin Nutr 83:1297-1305.

Smith GS, Koppel J, Goldberg S (2003) Applications of neuroreceptor imaging to psychiatry research. Psychopharmacol Bull 37:26-65.

Smith GS, Reynolds CF, III, Pollock B, Derbyshire S, Nofzinger E, Dew MA, Houck PR, Milko D, Meltzer CC, Kupfer DJ (1999) Cerebral glucose metabolic response to combined total sleep deprivation and antidepressant treatment in geriatric depression. Am J Psychiatry 156:683-689.

Smith KA, Fairburn CG, Cowen PJ (1997) Relapse of depression after rapid depletion of tryptophan. Lancet 349:915-919.

Smith KA, Ploghaus A, Cowen PJ, McCleery JM, Goodwin GM, Smith S, Tracey I, Matthews PM (2002) Cerebellar responses during anticipation of noxious stimuli in subjects recovered from depression. Functional magnetic resonance imaging study. Br J Psychiatry 181:411-415.

Smith MJ, Brebion G, Banquet JP, Allilaire JF (1994) Experimental evidence for two dimensions of cognitive disorders in depressives. J Psychiatr Res 28:401-411.

Snider RS (1950) Recent contributions to the anatomy and physiology of the cerebellum. Arch Neurol Psychiat 64:196-219.

Snider RS, Maiti A (1976) Cerebellar contributions to the Papez circuit. J Neurosci Res 2:133- 146.

Snitz BE, MacDonald A, III, Cohen JD, Cho RY, Becker T, Carter CS (2005) Lateral and medial hypofrontality in first-episode schizophrenia: functional activity in a medication-naive state and effects of short-term atypical antipsychotic treatment. Am J Psychiatry 162:2322-2329.

Snyder LH, Batista AP, Andersen RA (1997) Coding of intention in the posterior parietal cortex. Nature 386:167-170.

Soares JC (2003) Contributions from brain imaging to the elucidation of pathophysiology of bipolar disorder. Int J Neuropsychopharmacol 6:171-180.

294 Sokoloff L (1981) The deoxyglucose method for the measurement of local glucose utilization and the mapping of local functional activity in the central nervous system. Int Rev Neurobiol 22:287-333.

Spanagel R, Weiss F (1999) The dopamine hypothesis of reward: past and current status. Trends Neurosci 22:521-527.

Spitzer RL, Kroenke K, Williams JB (1999) Validation and utility of a self-report version of PRIME-MD: the PHQ primary care study. Primary Care Evaluation of Mental Disorders. Patient Health Questionnaire. JAMA 282:1737-1744.

Sprengelmeyer R, Rausch M, Eysel UT, Przuntek H (1998) Neural structures associated with recognition of facial expressions of basic emotions. Proc R Soc Lond B Biol Sci 265:1927-1931.

Sprengelmeyer R, Young AW, Calder AJ, Karnat A, Lange H, Homberg V, Perrett DI, Rowland D (1996) Loss of disgust. Perception of faces and emotions in Huntington's disease. Brain 119 ( Pt 5):1647-1665.

Squire LR (1992) Memory and the hippocampus: a synthesis from findings with rats, monkeys, and humans. Psychol Rev 99:195-231.

Squire LR, Ojemann JG, Miezin FM, Petersen SE, Videen TO, Raichle ME (1992) Activation of the hippocampus in normal humans: a functional anatomical study of memory. Proc Natl Acad Sci U S A 89:1837-1841.

Stahl SM (2003) Deconstructing psychiatric disorders, part 2: An emerging, neurobiologically based therapeutic strategy for the modern psychopharmacologist. J Clin Psychiatry 64:1145- 1146.

Steele JD, Meyer M, Ebmeier KP (2004) Neural predictive error signal correlates with depressive illness severity in a game paradigm. Neuroimage 23:269-280.

Steer RA, Ball R, Ranieri WF, Beck AT (1999) Dimensions of the Beck Depression Inventory-II in clinically depressed outpatients. J Clin Psychol 55:117-128.

Steffens DC, McQuoid DR, Welsh-Bohmer KA, Krishnan KR (2003) Left orbital frontal cortex volume and performance on the benton visual retention test in older depressives and controls. Neuropsychopharmacology 28:2179-2183.

Stein EA, Pankiewicz J, Harsch HH, Cho JK, Fuller SA, Hoffmann RG, Hawkins M, Rao SM, Bandettini PA, Bloom AS (1998) Nicotine-induced limbic cortical activation in the human brain: a functional MRI study. Am J Psychiatry 155:1009-1015.

Stein NL, Trabasso T (1992) The organization of emotional experience: creating links among emotion, thinking, language and intentional action. Cognition Emotion 6:225-244.

Steingard RJ, Renshaw PF, Hennen J, Lenox M, Cintron CB, Young AD, Connor DF, Au TH, Yurgelun-Todd DA (2002) Smaller frontal lobe white matter volumes in depressed adolescents. Biol Psychiatry 52:413-417.

Sternberg DE, Jarvik ME (1976) Memory functions in depression. Arch Gen Psychiatry 33:219- 295 224.

Stone AA (1981) The association between perceptions of daily experiences and self- and spouse- rated mood. Journal of Research in Personality 15:510-522.

Strakowski SM, Adler CM, DelBello MP (2002a) Volumetric MRI studies of mood disorders: do they distinguish unipolar and bipolar disorder? Bipolar Disord 4:80-88.

Strakowski SM, DelBello MP, Adler C, Cecil DM, Sax KW (2000) Neuroimaging in bipolar disorder. Bipolar Disord 2:148-164.

Strakowski SM, DelBello MP, Zimmerman ME, Getz GE, Mills NP, Ret J, Shear P, Adler CM (2002b) Ventricular and periventricular structural volumes in first- versus multiple-episode bipolar disorder. Am J Psychiatry 159:1841-1847.

Strakowski SM, Wilson DR, Tohen M, Woods BT, Douglas AW, Stoll AL (1993) Structural brain abnormalities in first-episode mania. Biol Psychiatry 33:602-609.

Strauss MM, Makris N, Aharon I, Vangel MG, Goodman J, Kennedy DN, Gasic GP, Breiter HC (2005a) fMRI of sensitization to angry faces. Neuroimage 26:389-413.

Strauss MM, Makris N, Aharon I, Vangel MG, Goodman J, Kennedy DN, Gasic GP, Breiter HC (2005b) fMRI of sensitization to angry faces. Neuroimage 26:389-413.

Sullivan RM, Gratton A (1999) Lateralized effects of medial prefrontal cortex lesions on neuroendocrine and autonomic stress responses in rats. J Neurosci 19:2834-2840.

Surguladze S, Brammer MJ, Keedwell P, Giampietro V, Young AW, Travis MJ, Williams SC, Phillips ML (2005) A differential pattern of neural response toward sad versus happy facial expressions in major depressive disorder. Biol Psychiatry 57:201-209.

Surguladze SA, Brammer MJ, Young AW, Andrew C, Travis MJ, Williams SC, Phillips ML (2003a) A preferential increase in the extrastriate response to signals of danger. Neuroimage 19:1317-1328.

Surguladze SA, Brammer MJ, Young AW, Andrew C, Travis MJ, Williams SC, Phillips ML (2003b) A preferential increase in the extrastriate response to signals of danger. Neuroimage 19:1317-1328.

Surguladze SA, Young AW, Senior C, Brebion G, Travis MJ, Phillips ML (2004) Recognition accuracy and response bias to happy and sad facial expressions in patients with major depression. Neuropsychology 18:212-218.

Sutton SK, Davidson RJ, Donzella B, Irwin W, Dottl DA (1997) Manipulating affective state using extended picture presentations. Psychophysiology 34:217-226.

Swayze VW, Andreasen NC, Alliger RJ, Ehrhardt JC, Yuh WT (1990) Structural brain abnormalities in bipolar affective disorder. Ventricular enlargement and focal signal hyperintensities. Arch Gen Psychiatry 47:1054-1059.

Sweeney JA, Kmiec JA, Kupfer DJ (2000) Neuropsychologic impairments in bipolar and

296 unipolar mood disorders on the CANTAB neurocognitive battery. Biol Psychiatry 48:674-684.

Szabo Z, Kao PF, Scheffel U, Suehiro M, Mathews WB, Ravert HT, Musachio JL, Marenco S, Kim SE, Ricaurte GA, . (1995) Positron emission tomography imaging of serotonin transporters in the human brain using [11C](+)McN5652. Synapse 20:37-43.

Szeszko PR, MacMillan S, McMeniman M, Lorch E, Madden R, Ivey J, Banerjee SP, Moore GJ, Rosenberg DR (2004) Amygdala volume reductions in pediatric patients with obsessive- compulsive disorder treated with paroxetine: preliminary findings. Neuropsychopharmacology 29:826-832.

Takahashi S, Driscoll BF, Law MJ, Sokoloff L (1995) Role of sodium and potassium ions in regulation of glucose metabolism in cultured astroglia. Proc Natl Acad Sci U S A 92:4616-4620.

Talairach J, Tournoux P (1988) Coplanar Stereotaxic Atlas of the Human Brain. 3-Dimensional Proportional Systems: An Approach to Cerebral Imaging. New York: Tieme Medical Publishers.

Tanaka D, Jr. (1976) Thalamic projections of the dorsomedial prefrontal cortex in the rhesus monkey (Macaca mulatta). Brain Res 110:21-38.

Tanaka SC, Samejima K, Okada G, Ueda K, Okamoto Y, Yamawaki S, Doya K (2006) Brain mechanism of reward prediction under predictable and unpredictable environmental dynamics. Neural Netw 19:1233-1241.

Tanda G, Frau R, Di CG (1996) Chronic desipramine and fluoxetine differentially affect extracellular dopamine in the rat prefrontal cortex. Psychopharmacology (Berl) 127:83-87.

Tauscher J, Verhoeff NP, Christensen BK, Hussey D, Meyer JH, Kecojevic A, Javanmard M, Kasper S, Kapur S (2001) Serotonin 5-HT1A receptor binding potential declines with age as measured by [11C]WAY-100635 and PET. Neuropsychopharmacology 24:522-530.

Taylor BP, Bruder GE, Stewart JW, McGrath PJ, Halperin J, Ehrlichman H, Quitkin FM (2006a) Psychomotor slowing as a predictor of fluoxetine nonresponse in depressed outpatients. Am J Psychiatry 163:73-78.

Taylor KS, Seminowicz DA, Davis KD (2008) Two systems of resting state connectivity between the insula and cingulate cortex. Hum Brain Mapp.

Taylor SE, Eisenberger NI, Saxbe D, Lehman BJ, Lieberman MD (2006b) Neural responses to emotional stimuli are associated with childhood family stress. Biol Psychiatry 60:296-301.

Taylor SF, Liberzon I, Koeppe RA (2000) The effect of graded aversive stimuli on limbic and visual activation. Neuropsychologia 38:1415-1425.

Taylor WD, Steffens DC, McQuoid DR, Payne ME, Lee SH, Lai TJ, Krishnan KR (2003) Smaller orbital frontal cortex volumes associated with functional disability in depressed elders. Biol Psychiatry 53:144-149.

Teasdale JD, Fogarty SJ (1979) Differential effects of induced mood on retrieval of pleasant and unpleasant events from episodic memory. J Abnorm Psychol 88:248-257.

297 Teasdale JD, Howard RJ, Cox SG, Ha Y, Brammer MJ, Williams SC, Checkley SA (1999) Functional MRI study of the cognitive generation of affect. Am J Psychiatry 156:209-215.

Teasdale JD, Russell ML (1983) Differential effects of induced mood on the recall of positive, negative and neutral words. Br J Clin Psychol 22 (Pt 3):163-171.

Tellegen A (1985) Structure of mood and personality and their relevance to assessing anxiety, with an emphasis on self-report. In: Anxiety and the anxiety disorders. (Tuma AH, Maser JD, eds), pp 681-706. Hillsdale, NJ: Erlbaum.

Thase ME, Corya SA, Osuntokun O, Case M, Henley DB, Sanger TM, Watson SB, Dube S (2007a) A randomized, double-blind comparison of olanzapine/fluoxetine combination, olanzapine, and fluoxetine in treatment-resistant major depressive disorder. J Clin Psychiatry 68:224-236.

Thase ME, Denko T (2008) Pharmacotherapy of mood disorders. Annu Rev Clin Psychol 4:53- 91.

Thase ME, Friedman ES, Biggs MM, Wisniewski SR, Trivedi MH, Luther JF, Fava M, Nierenberg AA, McGrath PJ, Warden D, Niederehe G, Hollon SD, Rush AJ (2007b) Cognitive Therapy Versus Medication in Augmentation and Switch Strategies as Second-Step Treatments: A STAR*D Report. Am J Psychiatry 164:739-752.

Thayer JF, Lane RD (2000) A model of neurovisceral integration in emotion regulation and dysregulation. J Affect Disord 61:201-216.

Thier P, Andersen RA (1993) Electrophysiological evidence for a second medial "parietal eye field". Abstr Soc Neurosci 19:27.

Thier P, Andersen RA (1998) Electrical microstimulation distinguishes distinct saccade-related areas in the posterior parietal cortex. J Neurophysiol 80:1713-1735.

Thomas KM, Drevets WC, Dahl RE, Ryan ND, Birmaher B, Eccard CH, Axelson D, Whalen PJ, Casey BJ (2001a) Amygdala response to fearful faces in anxious and depressed children. Arch Gen Psychiatry 58:1057-1063.

Thomas KM, Drevets WC, Whalen PJ, Eccard CH, Dahl RE, Ryan ND, Casey BJ (2001b) Amygdala response to facial expressions in children and adults. Biol Psychiatry 49:309-316.

Thorpe SJ, Rolls ET, Maddison S (1983) The orbitofrontal cortex: neuronal activity in the behaving monkey. Exp Brain Res 49:93-115.

Thulborn KR, Waterton JC, Matthews PM, Radda GK (1982) Oxygenation dependence of the transverse relaxation time of water protons in whole blood at high field. Biochim Biophys Acta 714:265-270.

Thyrum PT, Wong YW, Yeh C (2000) Single-dose pharmacokinetics of quetiapine in subjects with renal or hepatic impairment. Prog Neuropsychopharmacol Biol Psychiatry 24:521-533.

Timms RJ (1977) Cortical inhibition and facilitation of the defence reaction [proceedings]. J Physiol 266:98P-99P. 298 Tobler PN, O'Doherty JP, Dolan RJ, Schultz W (2007) Reward value coding distinct from risk attitude-related uncertainty coding in human reward systems. J Neurophysiol 97:1621-1632.

Tohen M, Chengappa KN, Suppes T, Baker RW, Zarate CA, Bowden CL, Sachs GS, Kupfer DJ, Ghaemi SN, Feldman PD, Risser RC, Evans AR, Calabrese JR (2004) Relapse prevention in bipolar I disorder: 18-month comparison of olanzapine plus mood stabiliser v. mood stabiliser alone. Br J Psychiatry 184:337-345.

Tohen M, Jacobs TG, Grundy SL, McElroy SL, Banov MC, Janicak PG, Sanger T, Risser R, Zhang F, Toma V, Francis J, Tollefson GD, Breier A (2000) Efficacy of olanzapine in acute bipolar mania: a double-blind, placebo-controlled study. The Olanzipine HGGW Study Group. Arch Gen Psychiatry 57:841-849.

Tohen M, Ketter TA, Zarate CA, Suppes T, Frye M, Altshuler L, Zajecka J, Schuh LM, Risser RC, Brown E, Baker RW (2003a) Olanzapine versus divalproex sodium for the treatment of acute mania and maintenance of remission: a 47-week study. Am J Psychiatry 160:1263-1271.

Tohen M, Vieta E, Calabrese J, Ketter TA, Sachs G, Bowden C, Mitchell PB, Centorrino F, Risser R, Baker RW, Evans AR, Beymer K, Dube S, Tollefson GD, Breier A (2003b) Efficacy of olanzapine and olanzapine-fluoxetine combination in the treatment of bipolar I depression. Arch Gen Psychiatry 60:1079-1088.

Tomarken AJ, Davidson RJ, Henriques JB (1990) Resting frontal brain asymmetry predicts affective responses to films. J Pers Soc Psychol 59:791-801.

Tomarken AJ, Dichter GS, Freid C, Addington S, Shelton RC (2004) Assessing the effects of bupropion SR on mood dimensions of depression. J Affect Disord 78:235-241.

Townsend JT, Ashby FG (1983) Stochastic modeling of elementary psychological processes. New York: Cambridge University Press.

Tremblay LK, Naranjo CA, Cardenas L, Herrmann N, Busto UE (2002) Probing brain reward system function in major depressive disorder: altered response to dextroamphetamine. Arch Gen Psychiatry 59:409-416.

Tremblay LK, Naranjo CA, Graham SJ, Herrmann N, Mayberg HS, Hevenor S, Busto UE (2005) Functional neuroanatomical substrates of altered reward processing in major depressive disorder revealed by a dopaminergic probe. Arch Gen Psychiatry 62:1228-1236.

Tremblay P, Blier P (2006) Catecholaminergic strategies for the treatment of major depression. Curr Drug Targets 7:149-158.

Trivedi MH, Fava M, Wisniewski SR, Thase ME, Quitkin F, Warden D, Ritz L, Nierenberg AA, Lebowitz BD, Biggs MM, Luther JF, Shores-Wilson K, Rush AJ (2006a) Medication augmentation after the failure of SSRIs for depression. N Engl J Med 354:1243-1252.

Trivedi MH, Rush AJ, Wisniewski SR, Nierenberg AA, Warden D, Ritz L, Norquist G, Howland RH, Lebowitz B, McGrath PJ, Shores-Wilson K, Biggs MM, Balasubramani GK, Fava M (2006b) Evaluation of outcomes with citalopram for depression using measurement-based care in STAR*D: implications for clinical practice. Am J Psychiatry 163:28-40.

299 Tsang O, Gholipour A, Kehtarnavaz N, Gopinath K, Briggs R, Panahi I (2008) Comparison of tissue segmentation algorithms in neuroimage analysis software tools. Conf Proc IEEE Eng Med Biol Soc 2008:3924-3928.

Tseng WS (2001) Handbook of Cultural Psychiatry. San Diego: Academic Press.

Tsuji S, Uozumi T, Akamatsu N, Tamagawa A, Matsunaga K, Ishiguchi H, Hashimoto T, Kojima Y (2006) Mechanism of voluntary and involuntary movements in humans. Suppl Clin Neurophysiol 59:135-142.

Tucker DM, Luu P, Pribram KH (1995) Social and emotional self-regulation. Ann N Y Acad Sci 769:213-239.

Uludag K, Dubowitz DJ, Buxton RB (2006) Basic principles of functional MRI. In: Clinical magnetic resonance imaging. (Edelman RR, Hesselink JR, Zlatkin MB, Crues JV, eds), pp 249- 287. Philadelphia: Saunders Elsevier.

Urry HL, van Reekum CM, Johnstone T, Kalin NH, Thurow ME, Schaefer HS, Jackson CA, Frye CJ, Greischar LL, Alexander AL, Davidson RJ (2006) Amygdala and ventromedial prefrontal cortex are inversely coupled during regulation of negative affect and predict the diurnal pattern of cortisol secretion among older adults. J Neurosci 26:4415-4425.

Ustun TB, yuso-Mateos JL, Chatterji S, Mathers C, Murray CJ (2004) Global burden of depressive disorders in the year 2000. Br J Psychiatry 184:386-392.

Vakili K, Pillay SS, Lafer B, Fava M, Renshaw PF, Bonello-Cintron CM, Yurgelun-Todd DA (2000) Hippocampal volume in primary unipolar major depression: a magnetic resonance imaging study. Biol Psychiatry 47:1087-1090. van der Zwaag W, Francis S, Head K, Peters A, Gowland P, Morris P, Bowtell R (2009) fMRI at 1.5, 3 and 7 T: Characterising BOLD signal changes. Neuroimage. van der GC, Minderaa RB, Keysers C (2007) Facial expressions: what the mirror neuron system can and cannot tell us. Soc Neurosci 2:179-222. van Stegeren AH, Everaerd W, Cahill L, McGaugh JL, Gooren LJ (1998) Memory for emotional events: differential effects of centrally versus peripherally acting beta-blocking agents. Psychopharmacology (Berl) 138:305-310. van Stegeren AH, Goekoop R, Everaerd W, Scheltens P, Barkhof F, Kuijer JP, Rombouts SA (2005) Noradrenaline mediates amygdala activation in men and women during encoding of emotional material. Neuroimage 24:898-909. van Velden FH, Kloet RW, van Berckel BN, Buijs FL, Luurtsema G, Lammertsma AA, Boellaard R (2009) HRRT versus HR+ human brain PET studies: an interscanner test-retest study. J Nucl Med 50:693-702. van LL, Crone EA, Bunge SA (2006) Neural correlates of developmental differences in risk estimation and feedback processing. Neuropsychologia 44:2158-2170.

Varrone A, Sjoholm N, Eriksson L, Gulyas B, Halldin C, Farde L (2009) Advancement in PET 300 quantification using 3D-OP-OSEM point spread function reconstruction with the HRRT. Eur J Nucl Med Mol Imaging.

Vawter MP, Tomita H, Meng F, Bolstad B, Li J, Evans S, Choudary P, Atz M, Shao L, Neal C, Walsh DM, Burmeister M, Speed T, Myers R, Jones EG, Watson SJ, Akil H, Bunney WE (2006) Mitochondrial-related gene expression changes are sensitive to agonal-pH state: implications for brain disorders. Mol Psychiatry 11:615, 663-615, 679.

Verdu B, Decosterd I, Buclin T, Stiefel F, Berney A (2008) Antidepressants for the treatment of chronic pain. Drugs 68:2611-2632.

Vicq d'Azyr F (1786) Traite d'anatomie et de physiologie. Paris: F.A. Didot.

Videbech P, Ravnkilde B (2004) Hippocampal volume and depression: a meta-analysis of MRI studies. Am J Psychiatry 161:1957-1966.

Videbech P, Ravnkilde B, Pedersen TH, Hartvig H, Egander A, Clemmensen K, Rasmussen NA, Andersen F, Gjedde A, Rosenberg R (2002) The Danish PET/depression project: clinical symptoms and cerebral blood flow. A regions-of-interest analysis. Acta Psychiatr Scand 106:35- 44.

Vieta E, Calabrese J, Goikolea J, Raines S, Macfadden W (2007) Quetiapine monotherapy in the treatment of patients with bipolar I or II depression and a rapid-cycling disease course: a randomized, double-blind, placebo-controlled study. Bipolar Disord 9:413-425.

Vilensky JA, van Hoesen GW (1981) Corticopontine projections from the cingulate cortex in the rhesus monkey. Brain Res 205:391-395.

Vogeley K, May M, Ritzl A, Falkai P, Zilles K, Fink GR (2004) Neural correlates of first-person perspective as one constituent of human self-consciousness. J Cogn Neurosci 16:817-827.

Vogt BA (2005) Pain and emotion interactions in subregions of the cingulate gyrus. Nat Rev Neurosci 6:533-544.

Vogt BA, Finch DM, Olson CR (1992) Functional heterogeneity in cingulate cortex: the anterior executive and posterior evaluative regions. Cereb Cortex 2:435-443.

Volkow ND, Wang GJ, Fowler JS, Logan J, Gatley SJ, Wong C, Hitzemann R, Pappas NR (1999) Reinforcing effects of psychostimulants in humans are associated with increases in brain dopamine and occupancy of D(2) receptors. J Pharmacol Exp Ther 291:409-415.

Von Korff MR, Katon W, Unutzer J, Wells K, Wagner EH (2001) Improving depression care: barriers, solutions, and research needs. J Fam Pract 50:E1.

Vuilleumier P, Armony JL, Clarke K, Husain M, Driver J, Dolan RJ (2002) Neural response to emotional faces with and without awareness: event-related fMRI in a parietal patient with visual extinction and spatial neglect. Neuropsychologia 40:2156-2166.

Vuilleumier P, Armony JL, Driver J, Dolan RJ (2001) Effects of attention and emotion on face processing in the human brain: an event-related fMRI study. Neuron 30:829-841.

301 Vuilleumier P, Armony JL, Driver J, Dolan RJ (2003) Distinct spatial frequency sensitivities for processing faces and emotional expressions. Nat Neurosci 6:624-631.

Vuilleumier P, Pourtois G (2007) Distributed and interactive brain mechanisms during emotion face perception: evidence from functional neuroimaging. Neuropsychologia 45:174-194.

Wacker J, Dillon DG, Pizzagalli DA (2009) The role of the nucleus accumbens and rostral anterior cingulate cortex in anhedonia: integration of resting EEG, fMRI, and volumetric techniques. Neuroimage 46:327-337.

Wagner A, Aizenstein H, Mazurkewicz L, Fudge J, Frank GK, Putnam K, Bailer UF, Fischer L, Kaye WH (2007) Altered Insula Response to Taste Stimuli in Individuals Recovered from Restricting-Type Anorexia Nervosa. Neuropsychopharmacology.

Wagner HN, Jr., Burns HD, Dannals RF, Wong DF, Langstrom B, Duelfer T, Frost JJ, Ravert HT, Links JM, Rosenbloom SB, Lukas SE, Kramer AV, Kuhar MJ (1983) Imaging dopamine receptors in the human brain by positron tomography. Science 221:1264-1266.

Wagner V, Muller JL, Sommer M, Klein HE, Hajak G (2004) [Changes in the emotional processing in depressive patients: a study with functional magnetoresonance tomography under the employment of pictures with affective contents]. Psychiatr Prax 31 Suppl 1:S70-S72.

Wang L, LaBar KS, Smoski M, Rosenthal MZ, Dolcos F, Lynch TR, Krishnan RR, McCarthy G (2008) Prefrontal mechanisms for executive control over emotional distraction are altered in major depression. Psychiatry Res 163:143-155.

Watson D, Clark LA, Tellegen A (1984) Cross-cultural convergence in the structure of mood: A Japanese replication study and a comparison with U.S. findings. Journal of Personality and Social Psychology 47:127-144.

Watson D, Clark LA, Tellegen A (1988) Development and validation of brief measures of positive and negative affect: the PANAS scales. J Pers Soc Psychol 54:1063-1070.

Watson D, Clark LA, Weber K, Assenheimer JS, Strauss ME, McCormick RA (1995a) Testing a tripartite model: II. Exploring the symptom structure of anxiety and depression in student, adult, and patient samples. J Abnorm Psychol 104:15-25.

Watson D, Tellegen A (1985) Toward a consensual structure of mood. Psychol Bull 98:219-235.

Watson D, Weber K, Assenheimer JS, Clark LA, Strauss ME, McCormick RA (1995b) Testing a tripartite model: I. Evaluating the convergent and discriminant validity of anxiety and depression symptom scales. J Abnorm Psychol 104:3-14.

Weiskrantz LE (1956) Behavioral changes associated with ablation of the amygdaloid complex in monkeys. J Comp Physiol Psychol 49:381-391.

Wells KB, Stewart A, Hays RD, Burnam MA, Rogers W, Daniels M, Berry S, Greenfield S, Ware J (1989) The functioning and well-being of depressed patients. Results from the Medical Outcomes Study. JAMA 262:914-919.

Wendelken C, Bunge SA, Carter CS (2008) Maintaining structured information: an investigation 302 into functions of parietal and lateral prefrontal cortices. Neuropsychologia 46:665-678.

Wenderoth N, Debaere F, Sunaert S, Swinnen SP (2005) The role of anterior cingulate cortex and precuneus in the coordination of motor behaviour. Eur J Neurosci 22:235-246.

Wettstein JG, Host M, Hitchcock JM (1999) Selectivity of action of typical and atypical antipsychotic drugs as antagonists of the behavioral effects of 1-[2,5-dimethoxy-4-iodophenyl]-2- aminopropane (DOI). Prog Neuropsychopharmacol Biol Psychiatry 23:533-544.

Whalen PJ (1998) Fear, vigilance, and ambiguity: initial neuroimaging studies of the human amygdala. Curr Dir Psychol Sci 7:177-188.

Whalen PJ, Bush G, McNally RJ, Wilhelm S, McInerney SC, Jenike MA, Rauch SL (1998a) The emotional counting Stroop paradigm: a functional magnetic resonance imaging probe of the anterior cingulate affective division. Biol Psychiatry 44:1219-1228.

Whalen PJ, Kagan J, Cook RG, Davis FC, Kim H, Polis S, McLaren DG, Somerville LH, McLean AA, Maxwell JS, Johnstone T (2004) Human amygdala responsivity to masked fearful eye whites. Science 306:2061.

Whalen PJ, Rauch SL, Etcoff NL, McInerney SC, Lee MB, Jenike MA (1998b) Masked presentations of emotional facial expressions modulate amygdala activity without explicit knowledge. J Neurosci 18:411-418.

WHO (2001) World Health Report. Geneva: WHO.

Wilke M, Kowatch RA, DelBello MP, Mills NP, Holland SK (2004) Voxel-based morphometry in adolescents with bipolar disorder: first results. Psychiatry Res 131:57-69.

Wilkinson HA (2000) Bilateral anterior cingulotomy for chronic noncancer pain. Neurosurgery 46:1535-1536.

Willeit M, Ginovart N, Graff A, Rusjan P, Vitcu I, Houle S, Seeman P, Wilson AA, Kapur S (2007) First Human Evidence of d-Amphetamine Induced Displacement of a D(2/3) Agonist Radioligand: A [(11)C]-(+)-PHNO Positron Emission Tomography Study. Neuropsychopharmacology.

Williams DS, Detre JA, Leigh JS, Koretsky AP (1992) Magnetic resonance imaging of perfusion using spin inversion of arterial water. Proc Natl Acad Sci U S A 89:212-216.

Williams JM, Mathews A, MacLeod C (1996) The emotional Stroop task and psychopathology. Psychol Bull 120:3-24.

Williams MA, Morris AP, McGlone F, Abbott DF, Mattingley JB (2004) Amygdala responses to fearful and happy facial expressions under conditions of binocular suppression. J Neurosci 24:2898-2904.

Wilson JD, Foster JW, Kronenberg HM, Larsen PR (1998) Williams Textbook of Endocrinology. Philadelphia, PA: W.B. Saunders.

Winston JS, O'Doherty J, Dolan RJ (2003) Common and distinct neural responses during direct

303 and incidental processing of multiple facial emotions. Neuroimage 20:84-97.

Wolf RC, Vasic N, Hose A, Spitzer M, Walter H (2007) Changes over time in frontotemporal activation during a working memory task in patients with schizophrenia. Schizophr Res 91:141- 150.

Wolfensberger SP, Veltman DJ, Hoogendijk WJ, Boomsma DI, de Geus EJ (2008) Amygdala responses to emotional faces in twins discordant or concordant for the risk for anxiety and depression. Neuroimage 41:544-552.

Wong DF, Wagner HN, Jr., Dannals RF, Links JM, Frost JJ, Ravert HT, Wilson AA, Rosenbaum AE, Gjedde A, Douglass KH, . (1984) Effects of age on dopamine and serotonin receptors measured by positron tomography in the living human brain. Science 226:1393-1396.

Wood JN, Romero SG, Knutson KM, Grafman J (2005) Representation of attitudinal knowledge: role of prefrontal cortex, amygdala and parahippocampal gyrus. Neuropsychologia 43:249-259.

Woods RP (1996) Modeling for intergroup comparisons of imaging data. Neuroimage 4:S84- S94.

Woolrich MW, Behrens TE, Beckmann CF, Jenkinson M, Smith SM (2004) Multilevel linear modelling for FMRI group analysis using Bayesian inference. Neuroimage 21:1732-1747.

Woolsey C, Settlage P, Meyer D, Sencer WW, Pinto HT, Travis A (1952) Patterns of localization in precentral and "supplementary" motor areas and their relation to the concept of a premotor area. Res Publ Assoc Res Nerv Ment Dis 30:238-264.

World Medical Association (1964) WORLD MEDICAL ASSOCIATION DECLARATION OF HELSINKI Ethical Principles for Medical Research Involving Human Subjects. Helsinki.

Worsley KJ, Evans AC, Marrett S, Neelin P (1992) A three-dimensional statistical analysis for CBF activation studies in human brain. J Cereb Blood Flow Metab 12:900-918.

Worsley KJ, Friston KJ (1995) Analysis of fMRI time-series revisited--again. Neuroimage 2:173-181.

Worsley KJ, Marrett S, Neelin P, Evans AC (1996) Searching scale space for activation in PET images. Hum Brain Map 4:74-80.

Wright CI, Fischer H, Whalen PJ, McInerney SC, Shin LM, Rauch SL (2001) Differential prefrontal cortex and amygdala habituation to repeatedly presented emotional stimuli. Neuroreport 12:379-383.

Wright CI, Wedig MM, Williams D, Rauch SL, Albert MS (2006) Novel fearful faces activate the amygdala in healthy young and elderly adults. Neurobiol Aging 27:361-374.

Wu J, Buchsbaum MS, Gillin JC, Tang C, Cadwell S, Wiegand M, Najafi A, Klein E, Hazen K, Bunney WE, Jr., Fallon JH, Keator D (1999) Prediction of antidepressant effects of sleep deprivation by metabolic rates in the ventral anterior cingulate and medial prefrontal cortex. Am J Psychiatry 156:1149-1158.

304 Yakovlev PI (1972) A Proposed definition of the limbic system. In: Limbic System Mechanisms and Automatic Function (Hockman CH, ed), pp 241-283. Springfield, IL: Charles C. Thomas.

Yang JC, Park K, Eun SJ, Lee MS, Yoon JS, Shin IS, Kim YK, Chung TW, Kang HK, Jeong GW (2008) Assessment of cerebrocortical areas associated with sexual arousal in depressive women using functional MR imaging. J Sex Med 5:602-609.

Yen CP, Kung SS, Su YF, Lin WC, Howng SL, Kwan AL (2005) Stereotactic bilateral anterior cingulotomy for intractable pain. J Clin Neurosci 12:886-890.

Yeterian EH, Pandya DN (1985) Corticothalamic connections of the posterior parietal cortex in the rhesus monkey. J Comp Neurol 237:408-426.

Yeterian EH, Pandya DN (1988) Corticothalamic connections of paralimbic regions in the rhesus monkey. J Comp Neurol 269:130-146.

Yeterian EH, Pandya DN (1993) Striatal connections of the parietal association cortices in rhesus monkeys. J Comp Neurol 332:175-197.

Yoshimura S, Okamoto Y, Onoda K, Matsunaga M, Ueda K, Suzuki SI, Shigetoyamawaki (2009) Rostral anterior cingulate cortex activity mediates the relationship between the depressive symptoms and the medial prefrontal cortex activity. J Affect Disord.

Young AW, Aggleton JP, Hellawell DJ, Johnson M, Broks P, Hanley JR (1995) Face processing impairments after amygdalotomy. Brain 118 ( Pt 1):15-24.

Young AW, Perrett DI, Calder AJ, Sprengelmeyer R, Ekman P (2002) Facial Expressions of Emotion: Stimuli and Tests (FEEST). Bury St. Edmunds: Thames Valley Test Company.

Young SN, Smith SE, Pihl RO, Ervin FR (1985) Tryptophan depletion causes a rapid lowering of mood in normal males. Psychopharmacology (Berl) 87:173-177.

Zaborszky L, Cullinan WE, Luine VN (1993) Catecholaminergic-cholinergic interaction in the basal forebrain. Prog Brain Res 98:31-49.

Zahn R, de Oliveira-Souza R, Bramati I, Garrido G, Moll J (2009) Subgenual cingulate activity reflects individual differences in empathic concern. Neurosci Lett 457:107-110.

Zald DH, Pardo JV (1997) Emotion, olfaction, and the human amygdala: amygdala activation during aversive olfactory stimulation. Proc Natl Acad Sci U S A 94:4119-4124.

Zevon MA, Tellegen A (1982) The structure of mood change: An idiographic/nomothetic analysis. Journal of Personality and Social Psychology 43:111-122.

Zhang N, Yacoub E, Zhu XH, Ugurbil K, Chen W (2009) Linearity of Blood-Oxygenation-Level Dependent Signal at Microvasculature. Neuroimage.

Zhang W, Bymaster FP (1999) The in vivo effects of olanzapine and other antipsychotic agents on receptor occupancy and antagonism of dopamine D1, D2, D3, 5HT2A and muscarinic receptors. Psychopharmacology (Berl) 141:267-278.

305 Zhang W, Perry KW, Wong DT, Potts BD, Bao J, Tollefson GD, Bymaster FP (2000) Synergistic effects of olanzapine and other antipsychotic agents in combination with fluoxetine on norepinephrine and dopamine release in rat prefrontal cortex. Neuropsychopharmacology 23:250-262.

Zheng D, LaMantia AS, Purves D (1991) Specialized vascularization of the primate visual cortex. J Neurosci 11:2622-2629.

Zimmerman M, Posternak MA, Chelminski I (2004) Using a self-report depression scale to identify remission in depressed outpatients. Am J Psychiatry 161:1911-1913.

Zipursky RB, Seeman MV, Bury A, Langevin R, Wortzman G, Katz R (1997) Deficits in gray matter volume are present in schizophrenia but not bipolar disorder. Schizophr Res 26:85-92.

Zola-Morgan S, Squire LR, Alvarez-Royo P, Clower RP (1991) Independence of memory functions and emotional behavior: separate contributions of the hippocampal formation and the amygdala. Hippocampus 1:207-220.

306 APPENDICES

Appendix 1 - Affective Picture Rating Scale

307 Appendix 2 - IAPS Photograph Descriptions and Ratings – Male – Run1 Neutral Blocks Valence (SD) Arousal (SD) Description Neutral 1 5.17 (0.94) 2.87 (1.75) Teenager 5.13 (1.04) 3.90 (2.14) EroticMale 5.11 (0.95) 2.87 (2.05) Rug 4.88 (1.38) 2.38 (1.91) Male 4.96 (1.14) 2.68 (1.76) Baskets 5.00 (1.43) 5.15 (2.14) Bees 7.42 (1.47) 5.81 (2.25) Turkey 4.90 (0.94) 3.15 (1.85) LightBulb 4.85 (1.13) 2.68 (1.90) Chair 4.87 (0.94) 2.60 (1.95) Building 4.89 (0.60) 2.09 (1.75) Spoon 5.26 (1.61) 4.45 (2.00) Cowboy Positive Blocks Valence (SD) Arousal (SD) Description Valence (SD) Arousal (SD) Description Neutral 2 5.15 (0.85) 3.45 (1.77) Woman Positive 1 7.70 (1.36) 5.94 (2.28) Mountains 4.87 (1.12) 2.16 (1.63) AbstractArt 8.13 (1.29) 6.41 (2.60) Skier 5.24 (1.54) 3.60 (2.15) Mushroom 7.99 (1.09) 7.19 (1.88) EroticCouple 5.00 (1.08) 3.11 (1.98) Trash 7.67 (1.44) 6.57 (1.94) Sailboat 5.21 (1.34) 5.04 (1.43) MaleFace 7.65 (1.36) 5.35 (2.03) Family 4.88 (1.94) 3.51 (2.17) Boy 8.13 (1.29) 6.41 (2.60) Skier 5.08 (1.27) 3.85 (2.00) Terrorist 8.39 (0.93) 7.02 (2.02) AttractiveFem 5.15 (1.36) 2.73 (1.74) AttractiveMan 7.96 (1.59) 4.76 (2.25) Seal 5.17 (1.32) 4.13 (1.97) HermitCrab 8.02 (1.21) 5.53 (2.07) Puppies 5.17 (1.13) 3.62 (2.13) Outlet 8.02 (1.37) 7.17 (1.76) EroticFemale 4.98 (1.36) 3.48 (2.31) AbstractArt 7.99 (1.09) 7.19 (1.88) EroticCouple 5.13 (1.91) 4.85 (2.45) Gun 7.83 (1.29) 4.21 (2.49) Porpoise

Valence (SD) Arousal (SD) Description Valence (SD) Arousal (SD) Description Neutral 3 4.91 (0.97) 2.99 (1.81) Towel Positive 2 7.99 (1.25) 7.72 (1.45) EroticCouple 4.98 (0.97) 3.06 (2.08) Fabric 7.92 (1.06) 7.25 (1.64) EroticCouple 5.07 (1.27) 2.36 (1.77) Plant 7.67 (1.19) 6.46 (2.22) Rafting 5.17 (1.13) 3.62 (2.13) Outlet 7.73 (1.61) 5.94 (2.30) EroticCouple 5.02 (1.22) 2.15 (1.71) Fan 7.83 (1.29) 4.21 (2.49) Porpoise 4.93 (0.35) 2.73 (1.86) RollingPin 8.21 (1.34) 7.43 (1.97) EroticFemale 5.13 (0.77) 2.68 (1.93) Shadow 8.02 (1.37) 7.17 (1.76) EroticFemale 5.13 (1.56) 2.95 (1.83) Boat 7.69 (1.48) 7.15 (1.81) EroticFemale 4.98 (1.57) 3.70 (2.25) Clock 7.88 (1.10) 7.52 (1.51) EroticFemale 5.17 (1.32) 4.13 (1.97) HermitCrab 7.67 (1.44) 6.57 (1.94) Sailboat 5.24 (1.54) 3.60 (2.15) Mushroom 7.89 (1.26) 4.21 (2.22) Bunnies 5.22 (0.89) 3.21 (1.97) Shopping 7.80 (1.47) 4.20 (2.69) Kitten

Valence (SD) Arousal (SD) Description Valence (SD) Arousal (SD) Description Neutral 4 5.15 (1.36) 2.73 (1.74) AttractiveMan Positive 3 8.39 (0.93) 7.02 (2.02) AttractiveFem 5.11 (1.37) 5.36 (2.14) Snake 7.65 (1.36) 5.35 (2.03) Family 4.96 (1.14) 2.68 (1.76) Baskets 7.77 (1.05) 7.17 (1.93) EroticCouple 4.90 (2.32) 5.56 (2.49) Snake 7.69 (1.59) 4.02 (2.30) Baby 4.87 (1.12) 2.16 (1.63) AbstractArt 7.70 (1.64) 7.43 (1.80) EroticCouple 5.07 (1.27) 2.36 (1.77) Plant 7.80 (1.36) 6.41 (2.18) AttractiveFem 4.94 (0.93) 3.44 (1.93) AbstractArt 7.88 (1.10) 7.52 (1.51) EroticFemale 4.89 (1.33) 4.42 (2.13) Fire 7.96 (1.59) 4.76 (2.25) Seal 5.09 (1.41) 3.93 (2.55) Woman 8.14 (1.24) 6.86 (2.00) Money 4.97 (0.93) 2.92 (1.84) Shoes 7.69 (1.28) 2.77 (2.16) Nature 4.88 (1.11) 3.73 (1.96) Trains 7.67 (1.19) 6.46 (2.22) Rafting 4.94 (1.71) 6.34 (1.94) Bear 8.25 (1.30) 7.80 (1.90) EroticFemale

Valence (SD) Arousal (SD) Description Valence (SD) Arousal (SD) Description Neutral 5 4.95 (1.54) 2.30 (1.90) Book Positive 4 7.99 (1.25) 7.72 (1.45) EroticCouple 4.99 (0.86) 2.79 (1.81) Pole 8.25 (1.30) 7.80 (1.90) EroticFemale 4.90 (0.94) 3.15 (1.85) LightBulb 7.63 (1.30) 6.92 (1.74) EroticCouple 5.22 (0.89) 3.21 (1.97) Shopping 7.89 (1.26) 4.21 (2.22) Bunnies 4.95 (1.54) 2.30 (1.90) Book 7.80 (1.47) 4.20 (2.69) Kitten 4.88 (1.38) 2.38 (1.91) Male 8.21 (1.34) 7.43 (1.97) EroticFemale 4.87 (1.08) 3.02 (1.94) Twins 7.63 (1.30) 6.92 (1.74) EroticCouple 5.24 (1.29) 2.82 (1.93) Runner 7.73 (1.25) 7.12 (1.95) Sailing 4.88 (1.11) 3.73 (1.96) Trains 7.69 (1.59) 4.02 (2.30) Baby 5.09 (1.41) 3.74 (1.79) Musician 7.69 (1.48) 7.15 (1.81) EroticFemale 4.96 (1.05) 2.69 (1.95) Mug 7.80 (1.54) 4.05 (2.30) Baby 4.90 (0.97) 3.62 (1.98) BlowDry 7.70 (1.36) 5.94 (2.28) Mountains

Valence (SD) Arousal (SD) Description Valence (SD) Arousal (SD) Description Neutral 6 5.13 (1.04) 3.90 (2.14) EroticMale Positive 5 7.80 (1.36) 6.41 (2.18) AttractiveFem 5.00 (1.08) 3.11 (1.98) Trash 7.63 (1.60) 3.74 (2.09) NeutBaby 5.26 (1.61) 4.45 (2.00) Cowboy 7.77 (1.05) 7.17 (1.93) EroticCouple 4.95 (0.90) 3.48 (1.82) NeutralGirl 8.02 (1.21) 5.53 (2.07) Puppies 5.21 (1.53) 4.27 (2.02) Turtle 7.73 (1.61) 5.94 (2.30) EroticCouple 4.89 (1.33) 4.42 (2.13) Fire 7.80 (1.54) 4.05 (2.30) Baby 4.90 (0.97) 3.62 (1.98) BlowDry 8.14 (1.24) 6.86 (2.00) Money 4.91 (0.97) 2.99 (1.81) Towel 7.92 (1.06) 7.25 (1.64) EroticCouple 4.87 (0.94) 2.60 (1.95) Building 7.73 (1.25) 7.12 (1.95) Sailing 5.15 (1.18) 3.70 (2.07) Smoking 7.69 (1.28) 2.77 (2.16) Nature 5.15 (1.97) 6.02 (2.25) Aircraft 7.70 (1.64) 7.43 (1.80) EroticCouple 5.24 (1.29) 2.82 (1.93) Runner 7.63 (1.60) 3.74 (2.09) NeutBaby

Averages 5.07 (1.23) 3.45 (1.97) Averages 7.87 (1.33) 6.08 (2.05)

308 Appendix 3 - IAPS Photograph Descriptions and Ratings – Male – Run2 Neutral Blocks Valence (SD) Arousal (SD) Description Neutral 1 5.24 (1.08) 2.96 (1.92) AttractiveMan 4.98 (1.36) 3.48 (2.31) AbstractArt 4.76 (0.86) 2.68 (1.77) EroticMale 4.25 (1.50) 4.98 (2.29) Spider 4.89 (0.60) 2.09 (1.75) Spoon 5.20 (1.52) 2.78 (2.29) Mushroom 4.93 (0.35) 2.73 (1.86) RollingPin 5.15 (0.85) 3.45 (1.77) Woman 5.12 (1.29) 3.08 (2.02) Mushrooms 5.09 (1.41) 3.93 (2.55) Woman 4.85 (1.89) 6.21 (2.18) Shark 5.01 (1.21) 2.51 (1.74) Plate Negative Blocks Valence (SD) Arousal (SD) Description Valence (SD) Arousal (SD) Description Neutral 2 5.11 (0.99) 3.59 (1.98) ElderlyWoman Negative 1 1.96 (1.56) 6.38 (2.26) Soldier 4.94 (0.93) 3.44 (1.93) AbstractArt 1.94 (1.28) 5.55 (2.61) Mutilation 4.85 (1.72) 4.45 (2.06) Truck 2.16 (1.31) 4.69 (2.11) Toddler 4.92 (1.79) 2.31 (1.88) Chess 1.62 (1.39) 5.88 (2.79) BurnVictim 5.08 (1.27) 3.85 (2.00) Terrorist 1.94 (1.39) 6.89 (2.08) Mutilation 4.99 (0.86) 2.79 (1.81) Pole 1.78 (1.26) 5.44 (2.70) Mutilation 5.11 (1.37) 5.36 (2.14) Snake 1.96 (1.44) 5.55 (2.55) OpenGrave 4.98 (1.57) 3.70 (2.25) Clock 2.09 (1.27) 5.31 (2.38) SlicedHand 5.12 (1.29) 3.08 (2.02) Mushrooms 1.98 (1.19) 5.51 (2.70) Tumor 5.13 (0.77) 2.68 (1.93) Shadow 1.80 (1.32) 6.20 (2.55) DeadBody 5.13 (1.56) 2.95 (1.83) Boat 1.90 (1.40) 5.84 (2.41) Dog 4.93 (1.14) 3.85 (1.96) Stilllife 1.50 (1.16) 6.20 (2.71) BurnVictim

Valence (SD) Arousal (SD) Description Valence (SD) Arousal (SD) Description Neutral 3 4.96 (1.05) 2.69 (1.95) Mug Negative 2 2.19 (1.42) 7.12 (1.75) Mutilation 4.95 (1.43) 1.55 (1.36) Basket 1.96 (1.44) 5.55 (2.55) OpenGrave 5.00 (1.48) 3.31 (2.23) Beads 2.21 (1.41) 5.70 (2.60) Mutilation 4.94 (1.71) 6.34 (1.94) Bear 2.06 (1.59) 6.61 (2.13) Mutilation 4.92 (1.79) 2.31 (1.88) Chess 2.07 (1.89) 5.46 (2.60) DuckInOil 4.85 (1.05) 3.69 (1.99) Knives 2.21 (1.86) 6.92 (2.44) Mutilation 4.95 (1.56) 3.90 (2.28) Spider 1.90 (1.40) 5.84 (2.41) Dog 5.15 (1.18) 3.70 (2.07) Smoking 1.90 (1.57) 6.56 (2.11) Mutilation 4.85 (1.89) 6.21 (2.18) Shark 2.10 (1.56) 6.43 (2.26) BurnVictim 5.20 (1.52) 2.78 (2.29) Mushroom 2.09 (1.27) 5.31 (2.38) SlicedHand 5.19 (1.14) 3.18 (1.89) AttractiveMan 2.00 (1.62) 5.65 (2.27) Infant 5.00 (1.43) 5.15 (2.14) Shark 2.10 (1.53) 6.85 (2.13) Attack

Valence (SD) Arousal (SD) Description Valence (SD) Arousal (SD) Description Neutral 4 4.95 (1.43) 1.55 (1.36) Basket Negative 3 2.14 (1.44) 5.45 (2.54) Kids 5.00 (1.10) 3.15 (1.93) Hammer 1.77 (1.31) 6.79 (1.93) BabyTumor 4.95 (0.90) 3.48 (1.82) NeutralGirl 1.88 (1.14) 5.88 (2.34) BurnVictim 4.85 (1.05) 3.69 (1.99) Knives 2.14 (1.44) 5.45 (2.54) Kids 5.15 (1.97) 6.02 (2.25) Aircraft 1.78 (1.26) 5.44 (2.70) Mutilation 5.19 (1.62) 2.44 (1.87) Flowers 1.80 (1.32) 6.20 (2.55) DeadBody 4.88 (1.94) 3.51 (2.17) Boy 1.63 (1.11) 6.84 (2.06) Mutilation 4.85 (1.72) 4.45 (2.06) Truck 1.90 (1.57) 6.56 (2.11) Mutilation 4.97 (0.93) 2.92 (1.84) Shoes 1.84 (1.12) 5.44 (2.78) Mutilation 4.98 (0.97) 3.06 (2.08) Fabric 2.10 (1.56) 6.43 (2.26) BurnVictim 5.11 (0.95) 2.87 (2.05) Rug 1.88 (1.17) 5.10 (2.11) StarvingChild 4.85 (1.13) 2.68 (1.90) Chair 1.69 (1.47) 6.74 (2.37) Mutilation

Valence (SD) Arousal (SD) Description Valence (SD) Arousal (SD) Description Neutral 5 5.17 (0.94) 2.87 (1.75) Teenager Negative 4 1.83 (1.19) 5.54 (2.74) Accident 5.08 (1.02) 3.47 (2.09) Shipyard 1.62 (1.39) 5.88 (2.79) BurnVictim 5.21 (1.21) 3.71 (1.93) EroticMale 1.88 (1.14) 5.88 (2.34) BurnVictim 5.19 (1.62) 2.44 (1.87) Flowers 2.07 (1.89) 5.46 (2.60) DuckInOil 4.90 (2.32) 5.56 (2.49) Snake 1.98 (1.28) 5.85 (2.21) Injury 5.12 (1.60) 3.63 (1.87) Fruit 1.94 (1.28) 5.55 (2.61) Mutilation 4.99 (1.42) 5.61 (1.86) Boy 2.10 (1.66) 6.70 (2.60) Mutilation 5.08 (1.02) 3.47 (2.09) Shipyard 2.10 (1.53) 6.85 (2.13) Attack 5.13 (1.91) 4.85 (2.45) Gun 2.19 (1.42) 7.12 (1.75) Mutilation 5.19 (1.14) 3.18 (1.89) AttractiveMan 2.00 (1.62) 5.65 (2.27) Infant 5.00 (1.48) 3.31 (2.23) Beads 1.84 (1.12) 5.44 (2.78) Mutilation 4.93 (1.14) 3.85 (1.96) Stilllife 1.63 (1.11) 6.84 (2.06) Mutilation

Valence (SD) Arousal (SD) Description Valence (SD) Arousal (SD) Description Neutral 6 5.09 (1.41) 3.74 (1.79) Musician Negative 5 2.16 (1.31) 4.69 (2.11) Toddler 5.12 (1.10) 3.41 (1.80) Man 2.21 (1.41) 5.70 (2.60) Mutilation 5.11 (0.99) 3.59 (1.98) ElderlyWoman 1.98 (1.28) 5.85 (2.21) Injury 5.12 (1.60) 3.63 (1.87) Fruit 2.06 (1.59) 6.61 (2.13) Mutilation 5.00 (1.10) 3.15 (1.93) Hammer 1.83 (1.19) 5.54 (2.74) Accident 5.02 (1.22) 2.15 (1.71) Fan 1.94 (1.39) 6.89 (2.08) Mutilation 5.21 (1.34) 5.04 (1.43) MaleFace 2.10 (1.66) 6.70 (2.60) Mutilation 5.24 (1.08) 2.96 (1.92) AttractiveMan 1.96 (1.56) 6.38 (2.26) Soldier 5.12 (1.10) 3.41 (1.80) Man 1.88 (1.17) 5.10 (2.11) StarvingChild 5.21 (1.53) 4.27 (2.02) Turtle 1.77 (1.31) 6.79 (1.93) BabyTumor 4.87 (1.08) 3.02 (1.94) Twins 1.50 (1.16) 6.20 (2.71) BurnVictim 4.99 (1.42) 5.61 (1.86) Boy 1.98 (1.19) 5.51 (2.70) Tumor

Averages 5.02 (1.30) 3.58 (1.98) Averages 1.94 (1.39) 6.01 (2.38)

309 Appendix 4 - IAPS Photograph Descriptions and Ratings – Female – Run1 Neutral Blocks Valence (SD) Arousal (SD) Description Neutral 1 5.14 (1.22) 4.15 (1.95) Bird 5.14 (1.71) 3.60 (2.17) Farmer 4.92 (1.05) 2.95 (1.95) Factoryworker 5.14 (1.22) 3.47 (1.75) Man 5.19 (1.68) 4.39 (1.97) Cowboy 4.78 (2.10) 5.01 (2.22) Actor 4.97 (1.53) 3.81 (2.02) Smoking 4.71 (1.80) 5.40 (2.16) Coach 4.72 (1.74) 4.28 (2.53) Stilllife 5.00 (1.21) 2.89 (1.87) Outlet 5.03 (0.98) 3.28 (2.16) Towel Positive Blocks 5.18 (1.34) 3.95 (1.99) Agate Valence (SD) Arousal (SD) Description Positive 1 8.43 (1.44) 4.47 (2.82) Seal Valence (SD) Arousal (SD) Description 8.59 (0.99) 5.31 (2.54) Puppies Neutral 2 5.11 (1.19) 3.86 (1.91) NeutMan 8.59 (0.75) 4.02 (2.40) Bunnies 5.15 (1.24) 3.58 (1.72) Girl 8.39 (0.94) 4.73 (2.75) Father 5.18 (1.55) 3.31 (2.07) AttractiveFem 8.50 (1.28) 4.84 (2.97) Baby 4.99 (1.66) 3.58 (2.22) Mushrooms 8.21 (1.29) 5.33 (2.53) Baby 5.06 (1.10) 2.15 (1.70) RollingPin 8.26 (1.17) 4.77 (2.47) Girls 5.09 (0.81) 2.58 (1.74) Bowl 8.16 (1.28) 5.03 (2.25) Father 4.97 (1.16) 3.15 (2.14) Pole 8.29 (1.17) 5.05 (2.67) Father 5.06 (1.22) 2.94 (2.08) Chair 8.51 (1.05) 4.77 (2.68) NeutBaby 5.08 (0.64) 2.72 (1.98) AbstractArt 8.41 (1.07) 3.67 (2.56) Nature 5.15 (1.66) 2.96 (2.05) Plate 8.15 (1.25) 6.07 (2.58) Castle 4.79 (1.09) 2.24 (1.87) Building 5.18 (1.34) 3.95 (1.99) Agate Valence (SD) Arousal (SD) Description Positive 2 8.39 (0.91) 3.33 (2.36) Rabbit Valence (SD) Arousal (SD) Description 8.59 (0.75) 4.02 (2.40) Bunnies Neutral 3 5.14 (1.22) 4.15 (1.95) Bird 8.74 (0.64) 4.97 (2.85) Baby 4.90 (1.31) 2.50 (1.86) Man 8.39 (0.94) 4.73 (2.75) Father 5.03 (1.34) 2.50 (1.76) Girl 8.46 (1.20) 4.94 (2.64) Babies 4.76 (1.66) 3.45 (1.81) ElderlyWoman 8.31 (1.49) 5.29 (2.83) Baby 4.97 (1.53) 3.81 (2.02) Smoking 8.51 (1.05) 4.77 (2.68) NeutBaby 5.02 (1.10) 2.90 (1.83) Twins 8.20 (1.59) 3.67 (2.52) Family 4.99 (1.66) 3.58 (2.22) Mushrooms 8.31 (1.12) 4.48 (2.68) Family 4.89 (0.96) 3.26 (1.96) Mug 8.18 (1.24) 4.76 (2.56) Baby 5.04 (0.87) 2.90 (1.82) HairDryer 8.54 (0.82) 4.88 (2.86) Sunset 4.75 (0.94) 3.20 (1.80) Truck 8.15 (1.25) 6.07 (2.58) Castle 5.06 (1.22) 2.94 (2.08) Chair 4.97 (1.16) 3.15 (2.14) Pole Valence (SD) Arousal (SD) Description Positive 3 8.58 (0.76) 4.42 (2.60) Kitten Valence (SD) Arousal (SD) Description 8.62 (0.85) 5.10 (2.67) Baby Neutral 4 4.90 (1.31) 2.50 (1.86) Man 8.24 (1.07) 5.45 (2.60) Baby 4.95 (1.56) 4.03 (2.22) NeutFace 8.24 (1.07) 5.45 (2.60) Baby 4.77 (1.90) 2.55 (1.76) ElderlyMan 8.16 (1.28) 5.03 (2.25) Father 4.95 (1.09) 2.98 (1.97) Man 8.29 (1.17) 5.05 (2.67) Father 5.02 (1.10) 2.90 (1.83) Twins 8.34 (1.10) 4.53 (2.29) Family 4.81 (1.26) 4.23 (1.76) EroticFemale 8.34 (1.10) 4.53 (2.29) Family 5.10 (1.35) 2.87 (2.09) Mushroom 8.10 (1.15) 3.73 (2.46) Kids 5.07 (1.38) 3.80 (2.07) Mushroom 8.31 (1.12) 4.48 (2.68) Family 4.75 (1.14) 3.70 (2.18) Trains 8.17 (1.23) 4.10 (2.53) Romance 5.03 (1.38) 4.02 (2.11) Checkerboard 8.08 (1.48) 6.16 (2.57) Skier 4.78 (1.07) 3.80 (1.86) AbstractArt 5.17 (1.12) 2.27 (1.77) Tissue Valence (SD) Arousal (SD) Description Positive 4 8.58 (0.76) 4.42 (2.60) Kitten Valence (SD) Arousal (SD) Description 8.59 (0.99) 5.31 (2.54) Puppies Neutral 5 5.14 (1.71) 3.60 (2.17) Farmer 8.50 (1.28) 4.84 (2.97) Baby 4.79 (1.44) 3.36 (1.79) Secretary 8.21 (1.29) 5.33 (2.53) Baby 5.15 (1.24) 3.58 (1.72) Girl 8.46 (1.20) 4.94 (2.64) Babies 4.92 (1.05) 2.95 (1.95) Factoryworker 8.20 (1.59) 3.67 (2.52) Family 5.10 (1.37) 3.54 (2.23) NeutralMale 8.14 (1.53) 5.16 (2.67) Couple 4.84 (0.93) 3.99 (1.89) BlowDry 8.18 (1.24) 4.76 (2.56) Baby 4.94 (0.88) 2.19 (1.72) Fan 8.41 (1.07) 3.67 (2.56) Nature 4.71 (1.10) 3.18 (1.98) Shipyard 8.54 (0.82) 4.88 (2.86) Sunset 4.95 (0.80) 1.87 (1.48) Lamp 8.16 (1.15) 5.80 (2.75) Fireworks 5.00 (0.78) 2.55 (1.65) ClothesRack 8.08 (1.48) 6.16 (2.57) Skier 5.17 (1.12) 2.27 (1.77) Tissue 5.21 (1.58) 5.67 (1.97) VolcanoSkier Valence (SD) Arousal (SD) Description Positive 5 8.43 (1.44) 4.47 (2.82) Seal Valence (SD) Arousal (SD) Description 8.39 (0.91) 3.33 (2.36) Rabbit Neutral 6 4.79 (1.44) 3.36 (1.79) Secretary 8.74 (0.64) 4.97 (2.85) Baby 5.19 (1.68) 4.39 (1.97) Cowboy 8.62 (0.85) 5.10 (2.67) Baby 4.78 (2.10) 5.01 (2.22) Actor 8.26 (1.17) 4.77 (2.47) Girls 5.15 (1.98) 5.71 (1.96) EroticCouple 8.31 (1.49) 5.29 (2.83) Baby 5.07 (1.38) 3.80 (2.07) Mushroom 8.10 (1.15) 3.73 (2.46) Kids 4.91 (0.56) 2.96 (1.97) Hammer 8.25 (1.10) 3.80 (2.17) Couple 5.10 (0.88) 2.67 (1.99) Fork 8.25 (1.10) 3.80 (2.17) Couple 4.75 (0.94) 3.20 (1.80) Truck 8.14 (1.53) 5.16 (2.67) Couple 5.01 (1.13) 2.89 (1.91) Rug 8.17 (1.23) 4.10 (2.53) Romance 5.15 (1.66) 2.96 (2.05) Plate 8.16 (1.15) 5.80 (2.75) Fireworks 5.17 (1.52) 3.52 (1.76) Office 4.78 (1.15) 2.80 (1.94) Cabinet Averages 8.34 (1.14) 4.75 (2.59)

Averages 4.99 (1.29) 3.39 (1.95)

310 Appendix 5 - IAPS Photograph Descriptions and Ratings – Female – Run2 Neutral Blocks Valence (SD) Arousal (SD) Description Neutral 1 4.77 (1.49) 3.63 (2.31) BingeEating 5.18 (1.55) 3.31 (2.07) AttractiveFem 4.81 (1.26) 4.23 (1.76) EroticFemale 4.72 (1.74) 4.28 (2.53) Stilllife 5.03 (0.98) 3.28 (2.16) Towel 5.14 (0.59) 1.94 (1.60) Spoon 4.92 (0.48) 1.97 (1.58) Basket Negative Blocks 4.80 (0.81) 2.36 (1.66) Shoes Valence (SD) Arousal (SD) Description 5.02 (1.11) 2.53 (1.79) Baskets Negative 1 1.48 (0.97) 5.72 (2.43) Toddler 5.20 (1.39) 2.73 (1.72) FireHydrant 1.21 (0.80) 7.77 (1.66) Mutilation 5.03 (1.38) 4.02 (2.11) Checkerboard 1.15 (0.44) 7.30 (2.22) Mutilation 5.17 (1.52) 3.52 (1.76) Office 1.33 (0.74) 7.49 (1.96) DeadBody 1.26 (0.68) 7.39 (1.97) Mutilation Valence (SD) Arousal (SD) Description 1.33 (0.83) 6.27 (2.35) Mutilation Neutral 2 5.11 (1.19) 3.86 (1.91) NeutMan 1.26 (0.56) 7.43 (1.75) Injury 4.88 (1.42) 4.84 (1.82) MaleFace 1.49 (0.81) 5.51 (2.23) InjuredChild 4.97 (1.31) 3.65 (1.89) Women 1.44 (0.95) 7.52 (1.99) Attack 4.86 (1.54) 3.85 (2.31) EroticFemale 1.20 (0.57) 7.55 (1.98) BabyTumor 4.80 (1.63) 4.53 (1.83) BikerCouple 1.50 (0.97) 6.44 (2.00) StarvingChild 5.18 (2.00) 5.82 (2.40) EroticCouple 1.35 (0.71) 6.71 (2.27) DeadMan 5.14 (0.59) 1.94 (1.60) Spoon 4.92 (0.48) 1.97 (1.58) Basket Valence (SD) Arousal (SD) Description 5.15 (0.84) 2.75 (1.86) Mug Negative 2 1.34 (0.71) 6.11 (2.87) Accident 4.71 (1.10) 3.18 (1.98) Shipyard 1.15 (0.44) 7.30 (2.22) Mutilation 5.01 (1.13) 2.89 (1.91) Rug 1.35 (0.96) 7.02 (2.02) BurnVictim 5.00 (0.78) 2.55 (1.65) ClothesRack 1.47 (0.89) 6.98 (2.04) BurnVictim 1.26 (0.68) 7.39 (1.97) Mutilation Valence (SD) Arousal (SD) Description 1.33 (0.83) 6.27 (2.35) Mutilation Neutral 3 4.88 (1.42) 4.84 (1.82) MaleFace 1.44 (0.95) 7.52 (1.99) Attack 4.77 (1.90) 2.55 (1.76) ElderlyMan 1.26 (0.56) 7.43 (1.75) Injury 5.14 (1.22) 3.47 (1.75) Man 1.49 (0.81) 5.51 (2.23) InjuredChild 4.95 (1.09) 2.98 (1.97) Man 1.50 (0.97) 6.44 (2.00) StarvingChild 5.10 (1.35) 2.87 (2.09) Mushroom 1.47 (1.00) 6.45 (2.19) Dog 4.84 (0.93) 3.99 (1.89) BlowDry 1.43 (0.89) 6.76 (2.42) ManOnFire 5.06 (1.10) 2.15 (1.70) RollingPin 4.80 (0.81) 2.36 (1.66) Shoes Valence (SD) Arousal (SD) Description 4.91 (0.56) 2.96 (1.97) Hammer Negative 3 1.41 (0.79) 5.87 (2.13) SadChild 5.02 (1.11) 2.53 (1.79) Baskets 1.21 (0.80) 7.77 (1.66) Mutilation 5.18 (1.53) 3.72 (2.07) Traffic 1.34 (0.71) 6.11 (2.87) Accident 5.13 (1.85) 3.73 (1.92) Boy 1.51 (1.07) 7.13 (1.88) Mutilation 1.18 (0.65) 7.18 (2.12) Mutilation Valence (SD) Arousal (SD) Description 1.18 (0.70) 7.09 (2.49) Mutilation Neutral 4 4.95 (1.56) 4.03 (2.22) NeutFace 1.32 (1.01) 7.33 (2.20) Mutilation 5.03 (1.34) 2.50 (1.76) Girl 1.33 (0.75) 7.61 (1.81) Mutilation 4.76 (1.66) 3.45 (1.81) ElderlyWoman 1.50 (0.97) 6.94 (1.68) DeadBody 4.97 (1.31) 3.65 (1.89) Women 1.35 (0.71) 6.71 (2.27) DeadMan 4.90 (1.23) 2.55 (1.76) Chess 1.38 (1.09) 6.46 (2.34) Cat 5.18 (2.00) 5.82 (2.40) EroticCouple 1.43 (0.89) 6.76 (2.42) ManOnFire 4.96 (1.27) 3.40 (1.85) Mushrooms 5.10 (0.88) 2.67 (1.99) Fork Valence (SD) Arousal (SD) Description 5.20 (1.39) 2.73 (1.72) FireHydrant Negative 4 1.48 (0.97) 5.72 (2.43) Toddler 5.05 (1.19) 3.08 (2.09) Fabric 1.41 (0.79) 5.87 (2.13) SadChild 4.78 (1.07) 3.80 (1.86) AbstractArt 1.35 (0.85) 6.74 (2.41) OpenGrave 4.78 (1.15) 2.80 (1.94) Cabinet 1.29 (0.82) 7.44 (2.21) Mutilation 1.51 (1.07) 7.13 (1.88) Mutilation Valence (SD) Arousal (SD) Description 1.15 (0.73) 7.51 (2.29) BurnVictim Neutral 5 5.10 (1.37) 3.54 (2.23) NeutralMale 1.18 (0.65) 7.18 (2.12) Mutilation 4.77 (1.49) 3.63 (2.31) BingeEating 1.18 (0.70) 7.09 (2.49) Mutilation 4.71 (1.80) 5.40 (2.16) Coach 1.35 (0.96) 7.02 (2.02) BurnVictim 4.86 (1.54) 3.85 (2.31) EroticFemale 1.22 (0.85) 7.15 (2.48) BurnVictim 5.15 (1.98) 5.71 (1.96) EroticCouple 1.33 (0.74) 7.49 (1.96) DeadBody 4.89 (0.96) 3.26 (1.96) Mug 1.20 (0.58) 7.54 (1.78) Soldier 4.79 (1.10) 2.98 (2.11) Stool 5.04 (0.87) 2.90 (1.82) HairDryer Valence (SD) Arousal (SD) Description 4.94 (0.88) 2.19 (1.72) Fan Negative 5 1.35 (0.85) 6.74 (2.41) OpenGrave 5.05 (1.19) 3.08 (2.09) Fabric 1.29 (0.82) 7.44 (2.21) Mutilation 5.08 (0.64) 2.72 (1.98) AbstractArt 1.15 (0.73) 7.51 (2.29) BurnVictim 5.13 (1.85) 3.73 (1.92) Boy 1.32 (1.01) 7.33 (2.20) Mutilation 1.33 (0.75) 7.61 (1.81) Mutilation Valence (SD) Arousal (SD) Description 1.22 (0.85) 7.15 (2.48) BurnVictim Neutral 6 4.90 (1.23) 2.55 (1.76) Chess 1.47 (0.89) 6.98 (2.04) BurnVictim 4.80 (1.63) 4.53 (1.83) BikerCouple 1.50 (0.97) 6.94 (1.68) DeadBody 4.96 (1.27) 3.40 (1.85) Mushrooms 1.20 (0.57) 7.55 (1.98) BabyTumor 5.00 (1.21) 2.89 (1.87) Outlet 1.20 (0.58) 7.54 (1.78) Soldier 5.09 (0.81) 2.58 (1.74) Bowl 1.47 (1.00) 6.45 (2.19) Dog 4.79 (1.10) 2.98 (2.11) Stool 1.38 (1.09) 6.46 (2.34) Cat 5.15 (0.84) 2.75 (1.86) Mug 4.75 (1.14) 3.70 (2.18) Trains Averages 1.34 (0.81) 6.95 (2.14) 4.95 (0.80) 1.87 (1.48) Lamp 4.79 (1.09) 2.24 (1.87) Building 5.18 (1.53) 3.72 (2.07) Traffic 5.21 (1.58) 5.67 (1.97) VolcanoSkier

Averages 4.97 (1.24) 3.34 (1.95)

311