Abstract
Autism spectrum disorder (ASD) is associated with negative social outcomes and with impaired social competence in general. There is a wealth of research on cognitive factors that underpin these social challenges. Social cognition is perhaps the most studied contributor to social challenges; however, the lack of ecologically valid assessment tools raises questions about the translation of study results to everyday social functioning. Research has also primarily focussed on categorical approaches to social cognition (i.e. clinical versus control groups). However, there is growing evidence for the utility of a dimensional approach to research, where ASD symptomology is investigated across all children on a continuum, regardless of diagnosis. We adopted a dimensional framework and novel assessment tools from the Test for the Evaluation of Emotions and Socialisation (TEEMS) platform to investigate the relationship between social cognition, everyday social competence and autism symptomology in a sample of 668 children aged 4–18 years. Structural equation modelling showed that social cognition test errors predicted poor social competence. This relationship diminished when intelligence (IQ) and autism symptomology were included as predictors, suggesting a potential moderating effect of IQ and autism symptomology on the association between social cognition and social competence. Furthermore, social cognition errors predicted autism symptomology, which in turn predicted social competence. These findings support the use of a dimensional approach – focussing on symptomology across all children, regardless of diagnosis, rather than clinical samples alone. The findings demonstrate further evidence for the utility of TEEMS as a measure of social cognition.