ASD Symptoms Are Associated with Reduced LPP As Youth Learn about Inconsistent Peers during Social Interactions.

Background: Social skills deficits are a primary autism spectrum disorder (ASD) symptom. Failure to learn about others in new or unpredictable social situations may contribute to this deficit. Prediction and reevaluation of others during iterative social interactions is essential for learning (Behrens, Hunt, Woolrich, & Rushworth, 2008) to appropriately engage in social interactions, but is less essential once learning has occurred (Davis & Levine, 2013). Quicker learning to predict another’s behavior may improve social skills in individuals with ASD symptoms. The late positive potential (LPP), an event-related potential associated with evaluating and attending to emotional stimuli (Hajcak et al., 2009), is enhanced to unpleasant social stimuli and reduced after reevaluation (Foti & Hajcak, 2008). Individuals with ASD exhibit a blunted LPP to social stimuli (Benning et al., 2016). Yet, little is known about how ASD symptomology relates to the changes in the LPP as individuals learn about their peers during social interactions.

Objectives: Examine if learning moderates relations between ASD symptomology and changes in the LPP over iterative social interactions.

Methods: Electrophysiological responses were measured via EEG as neurotypical youth (N = 54, M_age=12.11, SD_age=2.91; 22 male) were evaluated by purported peers who provided consistent (100% positive or negative) or inconsistent (50% positive and negative) evaluation in a Virtual-School context. Participants randomized to the control condition were informed of each peer’s reputation for being nice, mean, or unpredictable prior to entering the Virtual-School (N=29). Those in the learning condition learned each peer’s reputation through their evaluation alone (N=25). In this group, learning was confirmed based on ratings participants provided about each peer after every two interactions (nice-vs-unsure-vs-mean). Change scores for the LPP after social evaluation (400-1000ms) were calculated by subtracting early (first 48) from late (last 48) interactions. Moderation analyses determined the extent to which changes in LPP (X) differentially related to ASD symptoms (Y; subscales of the SRS-2, Constantino & Gruber, 2005) in the learning and the control groups (W).

Results: The learning group had greater changes in peer rating between early and late interactions for all peers, demonstrating learning occurred (p’s<0.05). Group moderated the relation between change in LPP to inconsistent peer evaluation and ASD symptoms of social awareness, communication, and cognition (B’s>0.49, p’s<0.05). Specifically, in the learning group, reduced LPP to inconsistent social evaluation in late-vs-early interactions was associated with more severe ASD symptoms. This relation was not observed in the control group. No effects were observed for consistent social interactions (B’s<0.34, p’s>0.16).

Conclusions: More severe ASD symptoms are associated with reduced LPP responses to inconsistent social evaluation when individuals learn about their peers through first-hand experience. Reduced LPP responses may indicate social evaluation from inconsistent peers becomes less salient throughout learning for those with ASD symptoms. This may increase reevaluation and difficulties predicting inconsistent peers which could exacerbate problems generalizing social skills to novel social situations (Bellini, Peters, Benner, & Hopf, 2007). Thus, targeting learning in novel or inconsistent social interactions may enhance the generalization of social skills in those with ASD symptoms.