Examining the Links between Restricted Interests/Repetitive Behaviors, Respiratory Sinus Arrhythmia, and Anxiety: Do Repetitive Behaviors Mediate the Relation between Rsa and Anxiety?

Background: Recent research points to a relation between restricted interests/repetitive behaviors (RRBs) and respiratory sinus arrhythmia (RSA) in ASD (Condy, Scarpa, & Friedman, 2017; Matsushima et al., 2016). Specifically, findings demonstrate a negative link between RSA and ASD symptomology, that is, greater physiological dysregulation (lower RSA) is associated with greater autism severity. Co-occurring anxiety has also been related to both poorer vagal control (Hollocks, Howlin, Papadopoulos, Khondoker, & Simonoff, 2014) and more RRBs (Spiker, Lin, Van Dyke, & Wood, 2012) among samples of youth with ASD.

Objectives: The purpose of this study was to examine the relations between RRBs, RSA, and anxiety and test the mediating role of RRBs in the association between RSA and anxiety among a sample of adolescents with ASD. It was hypothesized that RSA would be negatively associated with RRBs, both RRBs and RSA would be positively related to anxiety, and that RRBs would mediate the association between RSA and anxiety.

Methods: One-hundred fifteen adolescents with ASD participated in this study. Participants presented for a RCT of a social skills intervention; data presented here were from pretest. Presence of ASD was confirmed using the ASDOS-G and IQ above 70 was confirmed with the KBIT-2. Parents or caregivers completed the Social Responsiveness Scale (only the Autistic Mannerisms subscale was used here as a measure of RRBs; SRS-AM) and the Social Anxiety Scale for Adolescents (SAS). Adolescents completed a three-minute eyes-open resting state paradigm during which heart rate data was collected.

Results: A Process mediation model was used to test the mediating role of RRBs (SRS-AM) in the relation between RSA and anxiety. A negative relation was uncovered between RSA and RRBs F(1, 113) = 3.96, p < 0.049, r² = 0.03, indicating that more RRBs were associated with greater physiological dysregulation. Analyses also revealed that both RRBs (b = 0.78, p < .001) and RSA (b = 2.24, p = .02) were significant predictors of anxiety F(2, 112) = 18.08, p < .001, r² = 0.24. The indirect effect of RSA on anxiety through RRBs, however, was nonsignificant (b = -1.06 [CI: -2.54-0.10]).

Conclusions: The hypotheses were partially supported. Results provide further evidence for the notion that greater physiological dysregulation is associated with more RRBs. In constrast to hypotheses, however, results suggest that more anxiety is associated with better physiological regulation, controlling for RRBs. Although there were links between RSA and both RRBs and anxiety, RRBs did not explain the association between RSA and anxiety. Rather, RRBs explained additional variance in anxiety over and above RSA alone. Therefore, physiological dysregulation seems to affect anxiety, but does not do so via increased RRBs.