Examining the Relations Between Children's Vagal Flexibility Across Social Stressor Tasks and Parent- and Clinician-Rated Anxiety Using Baseline Data from an Early Intervention for Inhibited Preschoolers

Abstract

Early behavioral inhibition (BI) is a known risk factor for later anxiety disorder. Variability in children's parasympathetic nervous system (PNS) functioning may provide insight into the substantial heterogeneity in anxiety outcomes for children high in BI. However, gaps persist due to an over-reliance on static measures of functioning, which limits our ability to leverage PNS functioning to identify risk for anxiety. We address these gaps using baseline data from an early intervention study of inhibited preschoolers by characterizing vagal flexibility (VF), an index of non-linear change in PNS functioning, across social stressor tasks and by examining the associations between VF and anxiety. One hundred and fifty-one parents and their 3.5- to 5-year-old children were selected on the basis of BI to participate in an early intervention program (ClinicalTrials.gov registration: NCT02308826). A structural equation modeling framework was used to model children's VF across tasks designed to mimic exposure to novel social interactions and to test the predictive links between VF and anxiety. Children who showed less VF, characterized by less suppression and flatter recovery, were rated by both parents and clinicians as more anxious. Moreover, a multiple group model showed that children meeting diagnostic criteria for social anxiety disorder demonstrated significantly less VF across social stressor tasks. Among inhibited youth, reduced VF is a risk factor for anxiety and may reflect an individual's reduced capacity to actively cope with external demands. Study results contribute to our understanding of the regulatory processes underlying risk for anxiety in early childhood.

Keywords: Anxiety; Biomarkers; Developmental Psychopathology; Psychophysiology.

Figure 1

Model-implied RSA levels across social stressor tasks.

Figure 2

Model-implied RSA levels across social stressor tasks as a function of meeting diagnostic criteria for clinically-significant impairment in social anxiety. Group intercepts did not significantly vary, Wald’s X²(1) = 0.87, p = 0.35, but group slopes did significantly vary, Wald’s X²(1) = 7.07, p = 0.0078.