Detecting momentary reward and affect with real-time passive digital sensor data

This study explores the capability of passive digital sensor data from smartphones and smartwatches to predict self-reported ecological momentary assessments (EMA) of affect, motivation, interest, and pleasure in activities in an unseen test sample. Using data from 245 depressed participants with high-to-low anhedonia (195 train, 50 test) generating 23,812 EMA sessions, we evaluated whether behaviors and physiological factors could detect subjective states. For 11 of 15 EMA questions asked, machine learning models exceeded random chance in the fully-held-out test sample, suggesting detectable signals between passive measures and subjective states. Dependent on the sensor type, the optimal aggregation periods ranged from 15 minutes to 3 hours, with generally at least two hours of data being required. Subgroup analyses revealed variations in model performance by demographics, depression severity, and anhedonia severity. These findings demonstrate the potential for passive digital sensing to help monitor aspects of mental health on a large scale.