Autistic youth have impaired executive functioning (EF) and these challenges increase throughout adolescence. Deficits in EF have been associated with poor adult outcomes, decreased availability for learning, and linked to academic outcomes. Specifically, flexible problem solving is greatly reduced in autistic youth. We aim to investigate how flexibility contributes to learning in autistic youth and their typically developing peers.

Participants included 44 teens with (n=22) and without (n=22) ASD. All teens were 14-18 years old (ASD M = 15.77, SD = 1.05; TD M = 15.73, SD = .96) with FSIQ 3 70 (ASD M = 105.92, SD= 16.17; TD M = 107.93, SD = 10.14). Teens with ASD met DSM-5 criteria for autism supported by the SCQ and/or ADOS-2. All participants completed the California Verbal Learning Task (Child or Third Edition) over zoom. Parents reported on their child’s flexibility skills on the Flexibility Questionnaire (FQ) which encompasses 5 subscales: routines/rituals, transitions, special interests, social flexibility, and generativity. Independent samples t-tests examined group differences in performance on the CVLT, measured by the long delay recall and learning slope. Bivariate correlations examined the relationship between learning and flexibility in the autistic group. Linear regression was used to determine how flexibility contributes to learning above and beyond age, gender, and diagnosis.

Autistic youth had significantly lower scores on the CVLT long-delay recall (t = 2.311, p = .026) and the learning slope (t = 1.186, p = .038) than their typically developing peers. Special interests on the FQ were related to both performance on the first trial of the CVLT (r = -.482 p = .023) and the short delay cued recall (r = -.469 p = .028) in the autistic group. Fewer transition related problems were predictive of higher scores on the CVLT above and beyond age, gender, and diagnosis (R2 = .366, B = - .088, p = .030).

Learning was significantly lower in autistic teens compared to their typically developing peers. Additionally, autistic youth had a less steep learning slope than their typically developing peers. Autistic youth may focus on the first group of words only and don’t flexibly update their list to add new words as the trials go on. Autistic youth with more restricted interests have difficulty with the first trial due to the shifting required when starting a new unfamiliar task. Additionally, youth with a greater focus on their own interests had more difficulty on cued recall indicating that these youth may have had trouble shifting when prompted to use semantic categories. Fewer challenges with transitions were a significant predictor of learning above and beyond age, gender, and diagnosis. Previous research has shown large discrepancies between parent-report and laboratory-based tasks in autistic youth. This project highlights two unique measures of different modalities that show similarities in their ratings emphasizing their potential as good representations of overall skills. Future research should utilize a larger sample size to continue to examine the role of flexible problem solving in working memory and learning in autistic youth.