Neural Mechanisms of Social and Non-Social Reward and Their Relation to Autistic Traits

Background: Social and non-social rewards are regarded as important modulators of behaviour. Recent models of autism have highlighted attenuated behavioural responsiveness to social incentives (such as pictures of smiling faces), with social motivation deficit theories suggesting that those with autism do not properly anticipate and appreciate the pleasure of social stimuli. Functional neuroimaging has also highlighted reduced neural responsiveness to social reward in ASD in regions associated with reward processing, which taken together may partially explain the reduced social motivation commonly seen in autism. However, only one study has examined the temporal correlates of social and non-social reward processing in ASD. In an ERP study, Kohls and colleagues (2011) found that compared to typically developing individuals, those with ASD displayed reduced P3 amplitude to reward versus non-reward irrespective of social or non-social reward type, suggesting a general reward processing deficit in autism. Because the P3 component is thought to reflect motivated attention to reward signals, a lower P3 amplitude to reward cues in autism may therefore reflect reduced motivated attention.

Objectives: The current study followed-up this work by investigating the differential effects of social and non-social reward on goal-directed behaviour in young adults using Event-Related Potentials (ERPs), examining modulation of reward sensitivity by level of autistic traits. A specific goal of the study was to improve upon previous operationalisations of social and non-social reward by incorporating more ecologically-valid reward stimuli.

Methods: 40 typically developing young adults were pre-screened with the Autism Spectrum Quotient and Broad Autism Phenotype Questionnaire for high (N=20) or low (N=20) levels of autistic traits. Event-related potentials were recorded with a 128 channel HydroCel Geodesic Sensor Net while subjects performed a cued incentive go/no-go task during simulated observation by a peer. Three identical blocks were administered (with performance indicated by on-screen dots) across three reward conditions: 1) Social reward, positive feedback on performance by the observer (simulated with pre-recorded video); 2) Non-social reward, receipt of candy (also displayed on video); 3) No reward, a video of mosaic pictures. ERPs were time locked to in-task performance cues, and P3 amplitude and latency were extracted across centroparietal leads.

Results: Analyses in progress contrast P3 amplitude and latency using mixed measures ANOVAs with within-subjects factors of reward type (social/non-social/no-reward) and hemisphere (left/right), and a between-subjects factor of autistic traits (high/low). We predict a better performance and larger P3 amplitude for reward trials than no-reward trials, and that compared to those with low autistic traits, those with high levels will display reduced P3 amplitude to reward vs. no-reward, with the largest effect seen for social reward.

Conclusions: Results will shed light on the temporal dynamics of social and non-social reward processing and their relation to autistic traits. The experimental paradigm represents a significant innovation in the study of social and non-social reward, offering a valuable tool currently being applied to individuals with clinical levels of autistic traits. Insight into vulnerabilities in reward processing is critical for understanding social function in ASD as well as interpreting co-morbidities with psychiatric disorders.