Investigating Electrodermal Response and Anxiety in Preschool Children with ASD

Background: Estimates suggest that up to 80% of children with autism spectrum disorder (ASD) experience clinically significant levels of anxiety suggesting the need for earlier screening and intervention. However, assessment of anxiety in this population is challenging and, particularly, in younger children. Electrodermal response (EDR), a noninvasive physiological measure of autonomic activity, has been shown to quantify levels of anxiety in typically developing children but evidence shows significant variability in autonomic function in children with ASD. Additionally, the utility of EDR as a measurement for anxiety in preschool-aged children with ASD is relatively unexplored.

Objectives: To investigate EDR among young children with ASD compared to typically developing children during a face-to-face cognitive task and to explore the relationship between EDR and anxiety symptoms.

Methods: Twelve children ages 6-7 years with ASD and no intellectual disability and fifteen age-matched typically developing children participated to date. Groups did not differ in verbal, spatial, or nonverbal intelligence measured with the DAS-2. EDR was measured as participants performed a digit span task, which assesses verbal and working memory. The maximum fluctuation in EDR was recorded for each trial unless the fluctuation did not exceed 0.05 microsiemens. Scores were averaged into 30 second epochs across the first two minutes of the task. Parents responded to the BASC-2 about emotional and behavioral challenges, including anxiety symptoms.

Results: Children with ASD had higher parent reported anxiety, t(25) = -2.73, p=.02. Repeated measures ANOVA was used to examine the EDR between groups and their pattern of response over the first four 30s epochs of the task. A main effect of time, F(3, 72)=3.9, p=.01, indicated that EDR increased throughout the task (i.e., as it became more difficult). EDR also differed by diagnosis, F(1, 24)=3.1, p=.04, due to larger responses for children with ASD. Across diagnostic groups, children with more reported anxiety exhibited greater EDR during the first, r(28)=.48, p=.01, and second, r(28)=.48, p=.009, 30s epochs of the task. When the RM-ANOVA was conducted with anxiety covaried, the main effect of time remained significant, F(3, 69)=3.2, p=.03, and there was a trend for time x anxiety, F(3, 69)=2.6, p=.06, but diagnostic group did not significantly differ, p=0.13.

Conclusions: Children with ASD had significantly higher anxiety levels and EDR during the digit span task compared to typically developing controls. Overall, both groups exhibited increasing EDR as the task became more difficult. However, greater EDR during the first minute of the task was related to higher anxiety levels and, when anxiety was co-varied, the relation between anxiety and EDR over time approached significance. This preliminary data suggests that parents observe significantly more anxiety among children with ASD by early elementary school and that EDR may be a potentially useful index of anxiety. In particular, higher initial reactivity during a task with both social and cognitive demands and differences in EDR patterns over time may capture anxiety among young children. In the coming months, we plan to increase our sample size and further examine the relation between ASD, anxiety, and EDR.