Characterising the Relationship Between Anxiety, Executive Function, and Restricted and Repetitive Behaviours in Children and Adolescents with Autism Spectrum Disorder

Background: Kanner (1943) first hypothesised that anxiety may underlie restrictive and repetitive behaviours (RRBs) observed in children with Autism Spectrum Disorder (ASD), serving as a maladaptive coping mechanism (Rimland, 1964). RRBs may also be associated with poor executive functions (EF), such as poor impulse control, and inflexibility to shift between different tasks (Turner, 1997). However, little is known about the underlying relationship of anxiety with EF and RRBs in children and adolescents with ASD. Better understanding anxiety and EF's roles in RRBs may inform the development of evidence-based treatments to help reduce RRBs.

Objectives: To characterise the relationship between anxiety, EF (impulse control and flexibility), and RRBs in children and adolescents with ASD, and explore possible mediators of RRBs.

Methods: Participants included 41 high-functioning children and adolescents with ASD between the ages of 5 and 17 years (male = 20, female = 21; IQ Mean = 102.24, SD = 19.03; range 74 to 167). Parents completed all behavioural measures. Anxiety subscale of Child and Adolescent Symptom Inventory – 5 (CASI-5) (Sukhodolsky et al., 2007) measured anxiety. The Inhibition and Shift subscales of Behavior Rating Inventory of Executive Function (BRIEF) measured impulse control and flexibility respectively. Repetitive Behavioral Scales-Revised (RBS-R) measured RRBs. First, Pearson’s correlation evaluated the relationship between anxiety, flexibility, impulse control, and RRBs. Next, hierarchical linear regression further partitioned the variance associated with anxiety and EF in predicting RRBs. Finally, mediation analyses evaluated possible mediating roles of flexibility and impulse control on the relationship between symptom severity of co-occurring anxiety, and RRBs. Age was controlled in all analyses.

Results: We found significant positive correlation between anxiety (CASI-5 raw) and RRBs (RBS-R total raw) (r =.57, p <.001). Greater difficulties in both impulse control (BRIEF-Inhibit raw), and flexibility (BRIEF-Shift raw) were associated with heightened anxiety (r =.46, p <.01; r =.48, p <.01; respectively), and greater RRBs (r =.64, p <.001; r =.78, p<.001; respectively). Hierarchical Linear regression model (Table 1) showed that flexibility accounted for a significant portion of the variance associated with RRBs, R² =.55, F (1,38) = 34.18, p <.001. Adding impulse control did not increase the overall model’s ability to account for variances associated with overall RRBs. Adding anxiety accounted for additional variance associated with RRBs, R² change =.04, p <.05, and significantly improved model’s ability to account for variances associated with RRBs, F change (1,36) = 4.18, p <.05. Mediation analyses revealed that flexibility, and not impulse control, partially mediated the relationship between anxiety and use of RRBs (Figure 1), Sobel’s Z = 2.93, p<.01. Anxiety did not mediate the relationship between flexibility and RRBs.

Conclusions: Results indicate that children and adolescents with ASD who experience greater levels of anxiety display more RRBs, which was partially mediated by poor flexibility and set-shifting. One clinical implication may be that simultaneously targeting symptoms of anxiety and improving flexibility may help reduce RRBs in young people with ASD. Future research can determine whether findings may hold in a lower-functioning sample, and investigate gender differences.