Neural Mechanisms of Emotion Regulation with Circumscribed Interests in Adults with ASD

Background: Restrictive and repetitive behaviors (RRB) are a core feature of autism spectrum disorder (ASD), and circumscribed interests (CI) are one of the core characteristics of RRB. Individuals with ASD also evidence increased affective responses (i.e. hyperarousal and positive valence) toward CI (Sasson, Dichter, & Bodfish, 2012). Intensity of CI are known to impact difficulties in executive function and social responsiveness (Anthony et al., 2013). The ability to modulate one’s affective responses is known as emotion regulation (ER) and deficits in ER around CI may therefore be related to impairment with these day-to-day difficulties arising from CI. Although there are noted atypical neural mechanisms of ER with social information in ASD (Richey et al., 2015), mechanisms of ER have not been studied in the context of CI. We hypothesize that the lateral prefrontal cortex (PFC), a core area in modulating behavior, may underlie altered ER for CI in ASD. 

Objectives: To examine activation of the PFC in the context of CI under varying states of ER in ASD and control groups. 

Methods: A total of 27 adults (ASD=13; Control=14) participated in this study. Groups were matched on age (ASD M=26.1; Control M=27.4) and IQ (ASD M=113.3; Control M=116.3). ASD diagnosis was confirmed with the Autism Diagnostic Observation Schedule. Functional magnetic resonance imaging (fMRI) data were collected during a cognitive appraisal task on a 3T General Electric Signa Excite HD scanner. Each individual was asked to bring in ten photographs of their CI, which were used as stimuli for the task. After receiving standardized training in cognitive emotion regulation techniques, participants viewed each image for 4 seconds (free viewing period) and were then asked to “Think Positive,” or “Think Negative” about CI images while undergoing fMRI. Whole group activation for the pre-instruction [<] post-instruction periods contrast showed activation of the medial PFC (mPFC). A region of interest (ROI) was identified using a 40-voxel sphere in the mPFC from which parameter estimates were extracted.

Results: Contrary to predictions, significant clusters of activation were observed in the medial, rather than lateral prefrontal cortex. ROI analyses of mPFC clusters revealed no group differences between ASD and control subjects during the baseline (pre-instruction) or think-negative conditions (p>0.25). However, for the think-positive condition, the control group showed significantly increased activation in the mPFC, compared to the ASD group, t(17.78)=2.12, p<0.05, d=0.85. 

Conclusions: Individuals with ASD showed diminished function in mPFC during think-positive instructions for CI. These findings may indicate impairment in constraining positive affect toward CI, and thus provide evidence for targeting emotion regulation in the context of treatment.