ORIGINAL ARTICLE The Role of the Fusiform-Amygdala System in the Pathophysiology of Autism Isabel Dziobek, PhD; Markus Bahnemann, MD; Antonio Convit, MD; Hauke R. Heekeren, MD, PhD Context: Autism is a condition of unknown origin with cal findings by analyzing correlations with emotional face– well-documented impairments in social perception and processing performance. cognition. Main Outcome Measures: Fusiform gyrus cortical Objective: To assess the relevance of the fusiform- thickness, amygdala volume, emotional face processing. amygdala system to the pathophysiology of autism spec- Results: We found a specific local increase in cortical trum conditions. thickness of the fusiform gyrus and associated impair- ments in face processing in individuals with autism. Ana- Design: Cross-sectional case-control study. tomical covariance between amygdala volume and the in- crease in fusiform gyrus local thickness was significantly Setting: University hospital. smaller in the group with autism spectrum conditions. Participants: A total of 27 adults with autism spec- Conclusions: Our data provide the first anatomical evi- trum conditions and 29 age-, sex-, and intelligence quo- dence of an abnormal amygdala-fusiform system and its tient–matched typically developed healthy controls. Pa- behavioral relevance to face-processing deficits in au- tients were assessed according to DSM-IV criteria using tism spectrum conditions. In light of recent evidence of the Autism Diagnostic Interview–Revised. the involvement of the fusiform gyrus and amygdala in social perception as well as the areas of social cognition Interventions: We applied an automated measure- and emotional awareness, all of which are relevant to au- ment to estimate fusiform gyrus cortical thickness and a tism, our findings might represent a core pathophysi- manual tracing method to obtain amygdala volumes. We ological mechanism of autism. analyzed volumetric covariance among these brain re- gions and assessed the functional relevance of anatomi- Arch Gen Psychiatry. 2010;67(4):397-405 UTISM SPECTRUM CONDI- meaning from faces. The amygdala is an im- Author Affiliations: tions (ASC) involve severe portant part of this system and plays a cen- Max Planck Institute for abnormalities in social tral role in processing the social relevance Human Development, Berlin, interaction, communica- of information gleaned from faces.6,8 Ani- Germany (Drs Dziobek, tion, and repetitive behav- mal studies have identified strong recipro- Bahnemann, and Heekeren); ior.1 Deficits in face processing are among cal projections between the amygdala and Max Planck Institute for A 2-4 9,10 the core features of ASC. Although these fusiform gyrus, and recent functional Human Cognitive and Brain impairments represent only a small frac- magnetic resonance imaging studies have Sciences, Leipzig, and Cluster tion of the reported deficits in social per- provided evidence of fusiform activity of Excellence, Languages of 5 11-13 Emotion, Freie Universität ception and social cognition, their assess- modulation by the amygdala. In fact, Berlin, Berlin, Germany ment is of particular importance because viewing emotional faces increases the cou- (Drs Dziobek and Heekeren); faces represent a crucial primary source of pling between the amygdala and fusiform Berlin NeuroImaging Center, social information, and their decoding is a gyrus.14 Notably, patients with amygdala le- Department of Neurology, precursor for more complex social inference. sions do not show increased activation in Charite´ University Medicine In typically developed individuals, face the fusiform cortex when shown fearful Berlin, Berlin, Germany perception involves a distributed set of brain faces.11 Together, these findings empha- (Drs Bahnemann and regions comprising a core and an ex- size that the concerted action of the amyg- Heekeren); and Center for Brain tended system.6 The core system mediates dala and fusiform gyrus is crucial for emo- Health, New York University the visual analysis of faces and comprises, tional face processing. School of Medicine, New York, 15 and Nathan Kline Institute for among other parts, the lateral fusiform gy- Schultz recently presented a patho- 7 Psychiatric Research, rus (fusiform face area [FFA] ). The ex- physiological model of autism based on Orangeburg, New York tended system consists of regions acting in concerted functioning of the amygdala and (Dr Convit). concert with the core system to extract FFA. This model postulates that a poten- (REPRINTED) ARCH GEN PSYCHIATRY/ VOL 67 (NO. 4), APR 2010 WWW.ARCHGENPSYCHIATRY.COM 397 Downloaded from www.archgenpsychiatry.com at Mt Sinai School Of Medicine, on April 8, 2010 ©2010 American Medical Association. All rights reserved. nicians to participate in the study. Each subject underwent a Table. Demographic Variables of Participants With ASC videotaped diagnostic interview based on which diagnoses of and Controls ASC were made according to DSM-IV criteria1 by consensus of 1 psychiatrist (A.C.) and 1 psychologist (I.D.). Diagnostic dis- Mean (SD) crepancies were resolved based on the video material by in- cluding a third person trained in diagnosing autism spectrum ASC Control a conditions in adults. In addition, diagnoses were confirmed with Variable (n=27) (n=29) P Value the Autism Spectrum Quotient25 and the Autism Diagnostic In- Sex terview-Revised (ADI-R)26 in the 18 participants with avail- Male 20 22 .88 able parental informants. A group of 29 healthy controls (22 Female 7 7 men and 7 women), chosen to match the ASC group as closely Age, y 42.0 (11.3) 44.9 (14.6) .39 as possible with respect to age, education, and intelligence quo- Education, y 16.5 (1.6) 16.4 (1.5) .81 tient (IQ), also participated in the study (Table). Estimated WAIS-R IQ 111 (8) 113 (7) .17 b Ͻ All study participants had medical (including electrocar- Face processing 21.7 (3.4) 25.4 (1.9) .001 diogram, blood pressure, and routine blood tests), neurologi- Mental rotationc 17.2 (8.6) 20.2 (9.7) .23 cal, neuropsychological, psychiatric, and neuroradiologic (mag- Spatial visualizationc 25.7 (8.2) 26.9 (7.1) .57 netic resonance imaging) examinations. Any present or prior evidence of significant neurological, psychiatric, or medical dis- Abbreviations: ASC, autism spectrum condition; WAIS-R IQ, Wechsler Adult Intelligence Scale–Revised Intelligence Quotient. ease led to exclusion from the study. All participants gave in- a P values reflect level of significance from independent samples t test and formed written consent, and the research protocol was ap- ␹2 as appropriate. proved by the local ethics committee. b Maximum score, 28. c Maximum score, 40. MEASURES AND STATISTICAL ANALYSIS tial congenital abnormality of the amygdala in ASC would lead to a lack of orienting responses to socially salient Neuropsychology stimuli such as faces, which would then prevent the de- Intellectual Functioning. To assess intellectual functioning, we velopment of an expertise for faces mediated by the FFA. used the Shipley Institute of Living Scale,27 comprising a vo- In support of that model, structural magnetic resonance cabulary and an abstract thinking test. Based on a sum of the 16,17 imaging studies have provided evidence of amygdala raw scores of the tests, we estimated Wechsler Adult Intelli- abnormalities in individuals with ASC. In addition, there gence Scale–Revised IQs.28 is recent evidence from voxel-based morphometry stud- ies of increased gray-matter volume in the fusiform gy- Facial Emotion Recognition. Participants were given 28 pic- rus in individuals with ASC.18,19 Several functional neu- tures of facial expressions of happiness, sadness, anger, fear, sur- roimaging studies have also shown reduced activation of prise, disgust, and neutral expressions.29 A word list of these ba- the FFA20,21 and altered amygdala function21,22 in indi- sic emotional states was displayed simultaneously, and the subject viduals with autism during face processing. was required to correctly identify the displayed emotion. Evidence of pathophysiological involvement of the General Visual Processing. To ascertain the specificity of vi- amygdala-fusiform system in ASC can only be convinc- sual processing impairments to faces, we gave a mental rotation ing, however, if structural changes in this system can be test and a spatial visualization test.30 Specifically, in the mental related to functional deficits in social perception and/or rotation task, participants were required to mentally rotate let- social cognition. In the present study, we used a direct ters 0°, 60°, 120°, or 180° to indicate if they were displayed as measurement of cortical thickness of the fusiform gy- mirror images for a period of 120 seconds. In the spatial visual- rus, given that it might be more sensitive to subtle changes ization task, participants were asked to count surfaces of increas- than voxel-based morphometry, which involves con- ingly complex 3-dimensional objects for a period of 240 seconds. founding factors introduced by normalization,23 in 27 Group differences were assessed with independent samples ␣ Ͻ adults with ASC and 29 controls. To obtain amygdala vol- t tests. Analyses were 2-tailed and the level was set at P .05. umes, we used a well-validated and highly reliable manual All statistical procedures were performed using the Statistical tracing method.24 Furthermore, we ascertained anatomi- Package for the Social Sciences version 15.0 (SPSS Inc, Chi- cago, Illinois). cal relationships between the fusiform gyrus and the amyg- dala by