Disentangling the association between cognitive flexibility and anxiety in autistic youth: real-world flexibility versus performance-based task switching.

Introduction: Anxiety symptoms are highly prevalent among autistic youth yet remain under-recognized and undertreated, in part, due to a limited conceptual understanding of autistic cognition. Emerging evidence suggests that autistic differences in cognitive flexibility (CF) may be associated with a greater likelihood of developing and maintaining anxiety symptoms, relative to non-autistic youth. However, further work is needed to elucidate mechanisms of anxiety vulnerability that could inform potential targets for anxiety treatment in autistic youth. The current study aims to examine the associations between CF and anxiety in this population. Given the complexity of CF as a multifaceted construct, we used a multi-method approach to measure CF in order to tease apart its relationship with anxiety symptoms. Specifically, we hypothesized that real-world flexibility difficulties (assessed by survey measures), and poorer task switching performance (assessed by neurocognitive lab measures), would be associated with higher anxiety. Methods: Participants included forty 8-17-year-old autistic children (11 female, 29 male), along with their parents. Anxiety symptoms were measured by parent- and self-report using the Anxiety Scale for Children with Autism Spectrum Disorder. Task switching performance was measured using the Trail Making, Verbal Fluency, and Color-Word Interference tests from the Delis-Kaplan Executive Function System, administered to children in a controlled lab setting. Flexibility challenges in real-world settings were assessed by parent- and self-report using the Flexibility Scale and Shift subscale from the Behavior Rating Inventory of Executive Functioning (BRIEF-2; BRIEF-SR). Autistic traits were measured by parent-report using the Social Responsiveness Scale (SRS-2). Results: Multiple linear regression analyses revealed that reduced real-world flexibility was a significant predictor of anxiety, according to both parent- and self-report, even after controlling for autistic traits. Poorer task switching performance, however, was not associated with increased anxiety. Discussion: Our findings are consistent with previous literature suggesting that real-world CF challenges are linked to anxiety in autistic youth. The current study also offers preliminary evidence for the distinction between real-world CF and performance-based CF regarding their relationship to anxiety among autistic youth. Findings may help inform targeted assessment and treatment approaches for anxiety in this vulnerable population.