Divergent Patterns of Speaker-Intention Discrimination and Emotional Valence Ratings in Self-Reported Autistic and Non-Autistic Adults.

Autistic adults often interpret verbal and non-verbal cues underlying non-literal language differently compared to non-autistic people, reflecting differences in communicative style and shared context. Research on non-literal language in autism in controlled settings has typically depended on static stimuli that lack ecological validity and primarily focused on children, leaving non-literal language processing in autistic adults relatively understudied. The present study examined how self-reported autistic and non-autistic adults infer speaker intentions and emotional valence in dynamic, naturalistic videos. Participants (N = 66) viewed 50 brief videos from the Relational Inference in Social Communication database, depicting literal, sarcastic, teasing, or prosocial lie scenarios. After each video, they identified the speaker's communicative intent and rated the emotional response of the addressee. Participants also completed two measures of autistic characteristics (Autism Spectrum Quotient; Comprehensive Autistic Trait Inventory). Self-reported autistic participants showed lower speaker-intention discrimination than non-autistic participants, whereas valence ratings were similar across intentions. Autistic traits predicted lower speaker-intention discrimination in non-autistic adults but were unrelated (Comprehensive Autistic Trait Inventory) or positively associated (Autism Spectrum Quotient) in self-reported autistic adults. These findings highlight shared and differing aspects of social communication and support combining categorical and trait-based approaches to understand individual differences.Lay AbstractAutistic adults often interpret non-literal language such as sarcasm differently, which can affect everyday communication. However, prior research studies have often used written or static materials that do not reflect real-life conversations. In this study, we examined how self-reported autistic and non-autistic adults understand sarcasm, teasing, and prosocial lies using short, realistic video clips. We used videos showing actors having a short conversation. After each video, we asked participants to identify the intention that was used (literal, sarcastic, teasing, or prosocial lie) and to evaluate how happy the person receiving the statement felt. Participants also reported whether they had an autism diagnosis or self-identified as autistic and completed two questionnaires measuring autistic traits (Autism Spectrum Quotient; Comprehensive Autistic Trait Inventory). We found that autistic adults were less accurate than non-autistic adults in identifying speaker intentions, but both groups judged the emotional impact of the statements in similar ways. Among non-autistic adults, higher levels of autistic traits were linked to more difficulty identifying speaker intentions. In contrast, autistic traits were not linked, or were even linked to better performance, in autistic adults. These findings suggest that autistic and non-autistic people may use both shared and different strategies when interpreting social communication and highlight the importance of considering both diagnosis and individual traits.