Neural Network Associated with Social Cognition and Communication In Adults with ASD: A Structural MRI Connectivity Study

Background:  

Social interaction impairments in adults with autism spectrum disorder (ASD) include difficulties with recognizing emotions, mentalizing, social conversation, and atypical verbal and nonverbal communicative emotional gestures. Previous neuroimaging studies with individuals with ASD have found either abnormal functional or structural changes in the social brain network (orbitofrontal cortices [OFC], fusiform gyri, posterior superior temporal sulci [pSTS], amygdalae, temporo-parietal junction, cingulate gyri and anterior temporal cortices) and the mirror neuron system (inferior frontal gyri, fusiform gyri, inferior occipitotemporal gyri); however, no study has examined their cortical structural anatomical connectivities in relation to impaired social cognition and communication. 

Objectives:  

To determine the relation between brain structures in the social brain and mirror neuron networks with social cognition and communication deficits, by measuring cortical thickness and performing structural connectivity analysis on anatomical MRIs, and ASD symptomatology.

Methods:  

Cortical thickness measurements were made on the anatomical MRIs of 19 adults (age=26.8 ± 5.7 years; 16 males) diagnosed with ASD by a trained neurologist and met criteria specified by the Autism Diagnostic Observation Schedule-Generic (ADOS-G) and the Autism Diagnostic Interview Revised (ADI-R). We examined the association between cortical morphology and autism symptom severity in social and communication domains as determined by the ADOS. Data were analyzed using a general linear model controlling for the effects of age and sex, and regional differences in cortical thickness were examined using a cluster analysis, correcting for multiple comparisons according to random field theory with a p<0.05. 

Results:  

A directed search in the left OFC revealed a significant inverse correlation (p=0.01) with thickness in this region and scores for the social and communication domains of the ADOS. We then explored the structural connectivity associated with the left OFC to identify a network of brain areas related with social cognition and communication in ASD. Based on a whole brain search, significant positive correlations were seen in the left parahippocampal gyrus (p=0.02) and the right mid-cingulate gyrus (p=0.001).  Non-significant negative structural correlations were seen in the right posterior superior temporal gyrus and the lateral temporo-occipital gyrus.

Conclusions:  

Our results indicate that varying levels of social cognition and communication impairments in adults with ASD rely on differing cortical networks. Less severe deficits in social cognition and communication were associated with greater thickness in limbic and paralimbic brain regions while more severe impairments showed an inconsistent pattern of regional connectivity more associated with temporal lobe structures.