Electrodermal Activity, Behavior, and Context in Autistic People with High Support Needs: A Retrospective Ecological Study

Background/Objectives: Wearable biometric sensors may help characterize autonomic arousal in everyday settings among autistic people with high support needs, a population underrepresented in naturalistic research. This study aimed to describe patterns of autonomic arousal measured using electrodermal activity (EDA) and to explore their relationship with behavioral and contextual records compatible with distress or maladaptive stress. Methods: We retrospectively analyzed routine-care data from 57 autistic people with high support needs attending two specialized services. EDA was recorded for up to 10 days and integrated with behavioral and contextual records collected through a mobile application and stored in ABmonitor. Phase 1 retrospectively analyzed these data using principles derived from Ecological Momentary Assessment. Phase 2 descriptively compared pre- and post-adjustment EDAmax values in five cases with documented contextual adjustments, drawing on principles of Ecological Momentary Intervention, without causal inference. Results: In Phase 1, 19.3% of participants remained within the same arousal category across days, whereas 80.7% showed variable levels. Overall, 22 of 57 participants (38.6%) showed high autonomic arousal on at least one day; among them, 12 had behavioral records compatible with distress, and 5 had a specific contextual factor documented. In Phase 2, descriptive reductions in EDAmax ranging from 4 to 23 µS were observed after contextual adjustments. Conclusions: EDA provides relevant information about autonomic arousal but should not be interpreted as a direct marker of stress. Multimodal ecological monitoring may help generate person-centered clinical hypotheses and inform individualized support in real-world settings.