Passive sensing in mental health: Challenges and opportunities

The articles in this special issue are a good representation of the current state of passive sensing research, illustrating both its potential and significant challenges. These articles span multiple areas of sensing, including behaviors in the real world (e.g., global positioning system [GPS], actigraphy, and spoken language), online behaviors (e.g., app use and texting), behaviors that straddle the real world and online (e.g., phone use, text messages, and keystrokes), and psychophysiological states. As noted in the editors' introduction, the consistency of findings in both Filip's metaanalysis (Filip et al., 2025) and other systematic analyses (De Angel et al., 2022) indicates something is "there." But effect sizes remain stubbornly in the small to at best medium range. Small effect sizes are perhaps unsurprising. Passive sensing provides objective behavioral and physiological data, while clinical targets typically rely on subjective symptom ratings. Objective and subject measurements of the "same" constructs use inherently different vantage points and tend to show only modest correlations (e.g., the modest correlations between objective neuropsychological tests and self-reported cognitive symptoms). In this commentary, the author discusses a few reasons for these limited effect sizes and potential uses for passive sensing. (PsycInfo Database Record (c) 2025 APA, all rights reserved).