A Formative Study of Brief Affective Text as a Complement to Wearable Sensing for Longitudinal Student Health Monitoring

2026-05-14

AI Summary

Wearable devices often fail to capture the individual psychological context influencing health, limiting the utility of passive sensing in digital health. Addressing this gap, this study leverages a year-long cohort of university students, integrating physiological data from Oura rings with biweekly ultra-brief open-ended worry entries (median: three words) to demonstrate, for the first time, that such low-burden textual inputs significantly enhance the psychological interpretability of physiological signals. Combining dictionary-based methods, general-purpose pretrained and domain-adapted NLP models, intra-individual mixed-effects modeling, and zero-shot classification, the analysis reveals that emotional valence—not topical content—serves as the critical signal: academic worries correlate with reduced physical activity, while language indicative of emotional exhaustion is strongly associated with poorer sleep quality and lower heart rate variability. Notably, general-purpose pretrained embeddings outperform domain-adapted models across most health outcomes.

Abstract

Wearable devices capture physiological and behavioral data with increasing fidelity, but the psychological context shaping these outcomes is difficult to recover from sensor data alone, limiting passive sensing utility for digital health. We examined whether ultra-brief naturalistic concern text could serve as a scalable complement to passive sensing. In a year-long study of 458 university students (3,610 person-waves) tracked with Oura rings, participants responded bimonthly to an open-ended prompt about what concerned them most; responses had a median length of three words. We compared dictionary-based, general pretrained, and domain-adapted NLP approaches using within-person mixed-effects models across nine sleep and physical activity outcomes. Weeks dominated by academic concern framing were associated with lower physical activity; weeks characterized by emotional exhaustion language were associated with poorer sleep quality and lower heart rate variability. General pretrained embeddings outperformed domain-adapted models for most outcomes, with domain adaptation showing relative advantage for autonomic outcomes. Zero-shot classification of concern topics produced no significant associations, while affective dimensions across all three methods were consistently associated with outcomes, indicating emotional register rather than topical content carries the signal. These findings offer design guidance: ultra-brief affective prompts enrich the psychological interpretability of passive physiological data at minimal burden.

Problem

• passive sensing
• psychological context
• wearable devices
• digital health
• student health monitoring

Innovation

• ultra-brief affective text
• passive sensing
• digital health
• natural language processing
• within-person modeling