Social Reward and Alpha Asymmetry in ASD

Background: The social motivation hypothesis posits that individuals with autism spectrum disorder (ASD) are less motivated to socially engage because they find interactions less rewarding than their typically developing (TD) peers. Our previous work corroborated this by measuring neural activity during reward anticipation (Stavropoulos & Carver, 2013; 2014). Our studies focused on event-related potentials (ERPs), which provided information about the timing of neural activity. Electrophysiology data also provides information about oscillatory patterns. Of particular relevance is activity in the alpha band (8-12Hz). Reduction in alpha is thought to reflect increased engagement, and left-hemisphere dominant alpha asymmetry (more left activation than right) is hypothesized to reflect approach motivation and reward.

 

Objectives: The current investigation measured alpha asymmetry during reward anticipation. We performed novel EEG analyses on brain activity in children with and without ASD (Stavropoulos & Carver, 2013). In this study, children were presented with reward indicators accompanied by face or non-face stimuli. Non-face stimuli were comprised of scrambled faces in the shape of arrows.

 

Methods: Participants were 6-8 year-old children with (n = 20) and without (n = 20) ASD. EEG data in the alpha band (8-12Hz) was extracted using EEGlab. Data were analyzed from four sets of left/right hemisphere (L/RH) electrodes in frontal, central, parietal, and temporal locations. To measure alpha asymmetry, log values from LH electrodes were subtracted from comparable RH electrodes for each location (frontal, central, parietal, temporal). ANOVAs were conducted on each location separately with Condition as a within-subjects variable, and Group as a between-subjects variable.

Results: A significant Condition x Group interaction was observed for temporal electrode locations (p = .053). Follow-up tests revealed (1) marginal effects of condition in the ASD group (p = .058); ASD children demonstrated greater LH activation for the arrow versus face condition, and (2) a marginally significant effect of the face condition between groups (p = .066); TD children evidenced greater LH activation when anticipating faces compared to children with ASD. Finally, we explored correlations between severity of ASD symptoms and alpha asymmetry in temporal electrodes. A significant correlation was observed between alpha asymmetry in the arrow condition and ADOS severity score (p= .01), Figure 1. Children who evidenced greater LH activation for the arrow condition had more severe symptoms of ASD.

Conclusions: We provide novel evidence that alpha activity during reward anticipation is different in children with and without ASD. Children with ASD evidenced more left-dominant activation when anticipating rewards accompanied by arrows compared to rewards accompanied by faces. Given that actual rewards were the same in both conditions, we suggest that children with ASD experience more approach-motivation prior to non-social versus social stimuli. Interestingly, this difference is correlated with symptom severity. Finally, TD children evidenced more left-dominant activation when anticipating social stimuli compared to children with ASD. Taken together, the findings support the social motivation hypothesis, and extend it by suggesting that children with ASD may have too much non-social motivation, which comes at the expense of social motivation.