Age-Related Differences in the Association Between Executive Function and Social Responsiveness in Autism Spectrum Disorder: A Multi-Method Study

Abstract

Background: Executive function (EF) deficits are a core cognitive feature of autism spectrum disorder (ASD) and are closely associated with social responsiveness. Previous research has primarily focused on children with ASD, whereas how specific executive components relate to social functioning in adults remains less clear. This study examined whether patterns of association between EF and social responsiveness differ between children and adults with and without ASD. Methods: Data were obtained from the Autism Brain Imaging Data Exchange II (ABIDE II), including 423 participants aged 8-23 years (ASD = 184; controls = 239). EF was evaluated using the Behavior Rating Inventory of Executive Function (BRIEF/BRIEF-A), and social responsiveness was assessed with the Social Responsiveness Scale (SRS). Covariates of age, sex, and full-scale IQ (FIQ) were controlled using entropy balancing in children and multiple regression in adults. Hierarchical regression, moderated mediation analysis, and latent profile analysis (LPA) were conducted to examine the moderation, mediation, and heterogeneity effects, respectively. Results: Across both child and adult samples, individuals with ASD exhibited significantly higher T-scores than controls on nearly all BRIEF and SRS subdomains after covariate adjustment (all adjusted p < 0.01), indicating widespread EF and social responsiveness impairments. Moderation analyses revealed no significant age group × EF interaction, indicating that the association between EF and social responsiveness was consistent across development. Mediation analysis revealed age-specific pathways, with EF broadly mediating social responsiveness in adults but showing more selective mediation in children. LPA identified four distinct subtypes, which were independent of age, sex, and FIQ. Conclusions: EF–social responsiveness associations were evident across development, but the functional contribution of specific executive components became more differentiated with age. Working memory showed greater relative prominence in adulthood. Latent profile analysis revealed heterogeneity in how executive difficulties align with social challenges, supporting developmentally informed assessment and clinical interpretation rather than direct treatment recommendations.