Assessing affective and physiological responses in a psychiatric inpatient sample of individuals with autism spectrum disorder

Current prevalence estimates suggest that Autism Spectrum Disorder (ASD) affects 1 in 59 children in the United States. Up to 90% of individuals with ASD engage in challenging behaviors, including disruptive or dangerous aggression towards others and self-injurious behavior. Challenging behaviors can negatively impact social, emotional, and educational development in youth with ASD and are one of the most common reasons for referral to behavioral health care services. Recent theoretical work suggests that impairments regulating emotion may account for the emergence and maintenance of challenging behaviors in ASD. However, the inclusion of individuals with ASD who engage in high rates of challenging behaviors is rare in existing research due to recruitment and data collection challenges, including the inability of many on the autism spectrum to provide reliable self-reports about their emotional experience. Advances in wearable biosensors are making it increasingly feasible to collect peripheral physiological data unobtrusively during task demands and rest in this population. The current thesis systematically explores the relationship between perceived affect ratings and ambulatory electrodermal activity (a peripheral physiological measure of the sympathetic nervous system) in 62 individuals with ASD recruited from two psychiatric inpatient units who participated in a standardized series of affect-inducing tasks taken from the Laboratory Temperament Assessment Battery (Lab-TAB). This work aims to bridge existing gaps by including in research a marginalized segment of the ASD population, and by adding to our current understanding of affective and physiological profiles of youth with ASD who are more behaviorally challenged. Overall, results suggest that psychiatrically hospitalized youth with ASD display less neutral and more negative and positive affect, as well as higher physiological reactivity (as measured by the coefficient of variation [CV]) in response to affect-inducing tasks compared to preceding and subsequent non-demand rest periods. Effects of demographics (age, gender) and clinical measures (verbal ability and ASD severity) on affective and physiological responses are also discussed. Although a direct relationship between physiological responses and rated affect was not clear based on Aim 3 analyses, a similar response pattern in rated affect and EDA variability (CV) to affect-inducing tasks (Aims 1&2) was observed. Limitations, future directions, and clinical utility of the current work are also discussed.--Author's abstract