Utilizing Resting-State fMRI to Explore Insula Functional Connectivity in Individuals with and without Autism Spectrum Disorders

Background: Under the theory of embodied simulation put forth by Gallese and colleagues (2005, 2006), impairments in brain regions that integrate internal and external information may lead to deficits in social information processing (SIP). Recent neuroimaging studies report that the insula plays a key role in behavioral aspects that mediate SIP. The insula consists of three cytoarchitectectonically distinct regions: the anterior insula (AI), which participates in emotional salience detection and attentional control; the posterior insula (PI), which is involved in sensorimotor integration, body orientation, environmental monitoring, and response selection; and a transitional dysgranular zone. Autism spectrum disorders (ASD) are commonly characterized by deficits in social interaction and communication. Given these observed social behavioral impairments in ASD and the role played by the insula in SIP, investigating the functionality of this area in ASD is important.

Objectives: To further understand the role of the insula in ASD, we investigated differences in functional connectivity (FC) during the acquisition of resting-state functional magnetic resonance imaging (rs-fMRI) data in male children and adolescents with and without ASD.

Methods: Resting-state FC data was collected in 30 male participants (Age: 10-17 years). Thirteen participants had ASD diagnoses (Age: M=14.6; range 10.2-18) and 17 individuals were typically-developing controls (Age: M=14.4; range: 10-17.5). With seed-based FC measures, we compared resting FC in AI and PI networks between groups. We utilized the Social Communication Questionnaire (SCQ) to assess social functioning.

Results: We found lower FC in the ASD group between AI and superior frontal gyrus (SFG) and thalamus when compared to controls. We also noted hypoconnectivity between PI and inferior parietal lobule (IPL), fusiform gyrus (FG), and lentiform nucleus/putamen in comparison to controls. Conducting a regression analysis using SCQ scores as a continuous variable (without having a distinction between groups), we observed lower SCQ scores were associated with higher FC between: AI and SFG; AI and thalamus; PI and FG; and PI and IPL.

Conclusions: It has been hypothesized that the heterogeneity observed in ASD may be associated with variation in connectivity within and between neural networks. In our investigation, we found hypoconnectivity between the insula and several brain areas (e.g., IPL, FG) in participants with ASD relative to controls. When using SCQ scores from the ASD and control groups to create a continuum, we observed lower scores were associated with higher FC between the insula and some of the same regions. These findings suggest further investigation of the insula’s role in SIP across a spectrum of social abilities is needed.