Limitations/Future Directions
The current study has several limitations. Our pilot sample was recruited from an ongoing longitudinal study, so feasibility may be influenced by the familiar research setting or increased adherence to habituation protocols at home. Still an unfamiliar team member administered all procedures to ensure the social challenges were comparable. The convenience sample also prevented concurrent measurement of developmental level. Additionally, MRP task demands excluded children using single or no words. Taken together, experimental design modifications and larger sample sizes may elucidate how RSA relates to externalizing behavior in a more representative sample of autistic children.
Future studies may also benefit from collecting additional physiological measures. While there is no “gold-standard” method for computing RSA, methods are differentially impacted by artifact (Larsen et al., 2010; Lewis et al., 2012) and by measuring RSA using obtained respiration versus normative respiration levels (Graziano & Derefinko, 2013). Studies collecting both ECG and respiration could directly test the impact of different RSA computation methods. Furthermore, given RSA and behavioral outcomes may be dynamically impacted by co-occurring sympathetic activity (Fenning et al., 2019), obtaining measures of PEP or galvanic skin response could further elucidate relations between physiological response and externalizing behavior.