Self-Report and Parent-Report Reveal Similar Patterns of Executive Function Problems in Autistic Adolescents

Background: Parents of autistic individuals^a consistently report global executive functioning (EF) problems in their children, with the greatest problems occurring in flexibility and meta-cognition (Granader et al., 2014; Wallace et al., 2016; White et al., 2017). Unlike parent report of EF, self-report allows autistic individuals to offer insight into their own experiences. As suggested by prior research, self-report provides the opportunity to obtain a more accurate understanding of an individual’s behavior (Verhulst & van der Ende, 1992; Robins, Fraley & Krueger, 2009). To date, however, research has only investigated self-report of EF in older autistic adults (Davids et al., 2016).

Objectives: To investigate whether autistic youth self report EF problems and if their self-reported profile yields a similar pattern of EF problems to that reported about them by their parents.

Methods: Archival data on the Behavior Rating Inventory of Executive Function (BRIEF) self-report from 134 autistic youth (101 male, ages 11.0–18.75 years, FSIQ≥71) and BRIEF informant report from parents (n=134) were evaluated. One-sample T-tests were utilized to determine if the self-ratings represented statistically significant elevations in EF problems compared to the expected T-score mean of 50. The parent- and self-report versions of the BRIEF have the same three domains in the Behavior Regulation Index (BRI), but only share four of the same domains in the Metacognition Index (MCI); see Figure 2. Therefore we ran two general linear model (GLM) repeated measures ANOVAs to a) examine the self-report BRIEF profile in autistic youth and b) compare subscale score patterns from the BRIEF self- and parent-report in the same autistic youth.

Results: Autistic youth reported significantly elevated EF problems compared to the expected mean for all BRIEF domains (p’s<.005) except for the Inhibit domain (ns); see Figure 1. A GLM repeated measures ANOVA showed a main effect of BRIEF domain score for autistic youth, F(5.39, 242.26)=2.98, p<.01). A second GLM repeated measures ANOVA showed a main effect of BRIEF domain score, F(4.84, 509.22)=5.35, p<.001, and of rater, F(4.84, 879.57)=9.24, p<.001, as well as an interaction between rater and domain score, F (1,265)=80.03, p<.001, indicating that parents report higher levels of problems than their children report; see Figure 2. Post-hoc paired-sample T-tests show peak problems in the BRI Shift subdomain for both parent, (t=71.22, p<.005), and self-report, (t=57.84, p<.05), across all domains except for self-report of MCI Task Completion (ns) and parent-report of MCI Working Memory (ns).

Conclusions: These findings suggest that autistic youth self report a profile of EF problems to a greater degree than the normative BRIEF sample. Although they report less severe problems than their parents report about them, autistic youth identify a pattern of difficulties that emphasizes flexibility problems and metacognitive deficits, as do their parents. These results suggest that autistic youth recognize and can identify their own EF challenges. Therefore, assessing the lived experiences of autistic youth is critically important to clinical and research endeavors involving autistic individuals.

^a Identity-first language rather than person-first language is used frequently in this abstract, consistent with practice among autistic self-advocates.