Associations Between Cortical Thickness and Social Cognition in Autism Spectrum Disorders

Background: Cortical thickness atypicalities in individuals with autism spectrum disorders (ASD) have been reported in several brain regions. Normative studies have shown that cortical thickness is associated with socio-emotional capacities such as theory of mind.  Recent studies with individuals with ASD have explored the relationship between cortical thickness and socio-communicative abilities using scores from diagnostic measures such as the Autism Diagnostic Interview–Revised.  The use of these assessments however precludes a comparison with typically developing (TD) peers. 

Objectives: The present study examined emotion recognition in individuals with and without ASD using the Reading the Mind in the Eyes task (RMET), and explored whether overall performance, specific valances, and the level of difficulty were associated with differences in cortical thickness. 

Methods: 51 youth with ASD and 15 TD controls (7-17 years of age) were included in these analyses.   Participants were part of a multi-site imaging study (NeuroDevNet) at two Canadian sites, the Hospital for Sick Children, and the Montreal Neurological Institute. Participants were imaged using a Siemens Allegra 3T MRI system. A T1-weighted high-resolution automatic scan was obtained at the following parameters: TR=2300 ms, TE=2.96 ms, flip angle=90, FoV=256 x 192 x 240, slice thickness=1mm. Total imaging time was approximately 5 minutes. The CIVET automated analysis pipeline was using for preprocessing analysis. SurfStat and SAS were used for statistical analysis. We explored (1) baseline differences in age and IQ, (2) performance differences in RMET total scores, valence scores (positive, negative and neutral) and difficulty scores (easy and difficult), (3) the relation between RMET total, valence and difficulty scores, and cortical thickness between groups and (4) the effect of site.  

Results: Groups were similar on age (ASD = 13.03±2.78, TD = 12.91±2.60, p=0.58), but significantly different on IQ (ASD = 95.55±18.88, TD = 116.06±7.97, p<0.001). RMET total scores were significantly poorer in individuals with ASD compared to TD youth (ASD = 16.41±0.61, TD = 20.29±1.16, p=0.004).  After controlling for IQ, scores were no longer significantly different (ASD = 16.96±0.54, TD = 18.29±1.09, p=0.29). IQ was not controlled for in subsequent analyses. A significant interaction was observed between accuracy and cortical thickness between groups in the inferior frontal gyrus (Brodmann area; BA 45, p=0.05) and the cingulate gyrus (BA 32, p=0.003) for overall performance; the superior frontal gyrus (BA 11, p<0.001) and the medial prefrontal gyrus (BA 10, p<0.001) for negative emotions; and the middle frontal gyrus (BA 11, P=0.003), the inferior frontal gyrus (BA 44, p=0.008), and the cingulate gyrus (BA 24, p=0.016) for difficult emotions, after controlling for age and correcting for multiple comparisons.  The results did not change significantly when controlling for site. In all cases, TD individuals showed increased cortical thickness related to improved accuracy, whereas no such relation was observed in individuals with ASD.  No significant interactions were found between performance and cortical thickness for positive or neutral emotions, and easy items (all p>0.05).  

Conclusions: These findings suggest that cortical thickness across the brain may relate to social perception in different ways in individuals with and without ASD.