Predicting Situational Onset of Aggression in Minimally Verbal Youth with Autism Using Biosensor Data and Machine Learning Algorithms
Abstract
Unpredictable aggressive behavior by youth with autism spectrum disorder (ASD) isolates them from educational, social and family activities. Approximately 2/3 of youth with ASD display aggression, a common reason for treatment referral; yet evidence-based pharmacological and behavioral interventions for aggression in ASD are frequently ineffective. Aggression is particularly impairing in the 30-40 percent of youth with ASD who are minimally verbal (MV-ASD). Aggression may represent a maladaptive attempt to express or modulate physiological arousal arising from distress. We hypothesize that physiological arousal precedes aggressive behavior. We aim to predict aggression in MV-ASD before it occurs using data collected from wrist-worn physiological sensors and behavior observation. Using sophisticated machine learning algorithms linking observable aggression to preceding physiological signals (heart rate, skin conductance), we may identify new opportunities for intervention. Since project launch, we have refined data collection procedures, established processes for behavioral data upload and physiological data transfer to collaborators at NEU, and implemented physiological data quality checks. Staff training has been completed on all procedures including use of biosensors and a smartphone application to code aggression instances, at a high level of inter-rater reliability. 49 youth have been enrolled and data collection has been completed with 25 thus far.
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 01, 2021
- Accession Number
- AD1156799
Entities
People
- Matthew Siegel