Autism as a Predictor of Deviant Peer Association: Testing for Dual Systems Model Mediation Effects.
Autistic teens pick fewer deviant friends, and their naturally lower thrill-seeking explains why.
01Research in Context
What this study did
Wojciechowski (2026) asked whether autistic youth hang out with rule-breaking peers. The team tested two brain systems: sensation-seeking and impulse control. They measured how each system might link autism to deviant peer contact.
Participants were autistic and neurotypical youth. Self-report surveys tracked peer choices and personality traits. The study used mediation stats to map the pathway.
What they found
Autistic youth reported fewer deviant peer contacts. Lower sensation-seeking, not better impulse control, explained the drop. The result supports a protective effect built into autism traits.
The finding flips the usual risk story. Less drive for new thrills steers autistic teens away from trouble-making friends.
How this fits with other research
Humphrey et al. (2011) saw more solitary and aggressive peer acts in autistic teens within mainstream classes. The two papers seem to clash, but the gap is about setting. Thomas sampled broadly; Neil watched real-time playground behavior. Different lenses, different stories.
Adams et al. (2024) extend the peer picture. They show that when autistic teens feel their social life falls short, loneliness rises and depression follows. Thomas adds the front-end: choosing fewer deviant peers may shrink one source of social stress.
Werner et al. (2025) refine the idea further. They found loneliness hinges on emotional quality, not minutes spent with peers. Taken together, autistic youth may avoid risky crowds, yet still feel lonely if interactions feel negative.
Why it matters
You can stop assuming every autistic client is prey to ‘bad-influence’ friends. Screen sensation-seeking first. Low seekers may self-insulate, so teach safe, pro-social ways to meet their lower thrill needs. For high seekers, stay alert to peer risk. Tailor social skills groups around shared interests rather than pushing typical teen hangouts that may overload or bore them.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Add a sensation-seeking checklist to your intake packet and use scores to decide whether peer-risk education or safe social exposure fits the client.
02At a glance
03Original abstract
PURPOSE: Deviant peer association is a robust risk factor for offending, that is, any behavior that violates a criminal law. This can include behaviors like drug use, violence, and theft; among others. Autism is a neurodivergent condition characterized by both atypical cognitive development and challenges with social interaction and communication. Despite these impairments, there is a dearth of research focused on the relationship between autism and deviant peer association. Relatedly, there exists no research examining the role that dual systems model constructs (sensation-seeking and impulse control) as mediators of this relationship. This study sought to address these gaps in the literature by examining autism as a predictor of deviant peer association in childhood and whether either or both impulse control or sensation-seeking significantly mediate this relationship. MATERIALS: Waves 1, 3, and 4 of the Adolescent Brain Cognitive Development study were analyzed. Structural equation modeling was used to examine these direct and indirect relationships of interest. RESULTS: Results indicated that autism was associated with lower deviant peer association scores. Lower sensation-seeking scores among youth with autism significantly mediated this relationship, but impulse control did not. CONCLUSION: Autism may be a protective factor against deviant peer association due to diminished sensation-seeking relative to neurotypical peers. Future research should take a developmental perspective on these relationships to examine whether they are consistent at later stages of the life-course.
Journal of autism and developmental disorders, 2026 · doi:10.1080/14789940600589464