Prediction of Challenging Behaviors Associated with Profound Autism in a Classroom Setting Using Wearable Sensors

2026-05-17

AI Summary

This study addresses the high-risk challenging behaviors—such as self-injury, aggression, and elopement—exhibited by students with severe autism in classroom settings, which pose significant safety threats and disrupt instruction. For the first time in authentic classroom environments, the research leverages wearable devices to collect multimodal physiological time-series data, including accelerometry, electrodermal activity, and skin temperature, to fine-tune state-of-the-art foundation models for prospective prediction of these behaviors. The resulting system achieves a predictive lead time of up to 10 minutes prior to behavioral onset, with an AUC-ROC of 0.78, thereby transcending the constraints of prior laboratory-bound paradigms and establishing a technical foundation for proactive classroom intervention strategies.

Abstract

Autism Spectrum Disorder (ASD) is characterized by challenges with social interaction and communication and by restricted or repetitive patterns of thought and behavior, with significant variability in presentation. Approximately a quarter of children with ASD are classified as having profound autism, who often exhibit challenging behaviors, such as self-injurious behavior, aggression, elopement, or pica, that pose serious safety risks and disrupt learning in educational settings. Prior work has applied wearable sensors and machine learning to detect challenging behaviors, but has been largely confined to controlled laboratory environments. This work demonstrates that predicting challenging behavior episodes is feasible in a real-world special education classroom. We collected approximately 110.7 hours of labeled multimodal wearable data comprising accelerometry, electrodermal activity (EDA), and skin temperature from 9 children and young adults aged 10 to 21 years across standard classroom sessions. We fine-tuned state-of-the-art foundation models for multimodal wearable time-series analysis and show that challenging behavior episodes can be predicted up to 10 minutes in advance with an AUC-ROC of 0.78. These results establish a concrete foundation for developing proactive in-class intervention systems that enable teachers to minimize the safety risks of challenging behaviors in special education classrooms

Problem

• profound autism
• challenging behaviors
• classroom setting
• wearable sensors
• behavior prediction

Innovation

• wearable sensors
• multimodal time-series
• foundation models
• challenging behavior prediction
• profound autism