Detection of Microbehavior Intervals for Predicting Mental Health: Clinically Relevant and Advanced Multimodal Temporal Analysis
Sapir Gershov, Charlotte E Hilberdink, Yiwen Zhao, Sarah B Birnbaum, Victoria Mueller, Stephen P Wall, Katharina Schultebraucks · 2026
Read Detection of Microbehavior Intervals for Predicting Mental Health: Clinically Relevant and Advanced Multimodal Temporal Analysis by Sapir Gershov, Charlotte E Hilberdink, Yiwen Zhao, Sarah B Birnbaum, Victoria Mueller, Stephen P Wall, Katharina Schultebraucks (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13254507.txt.
Description
Sapir Gershov and colleagues detect distress in health care workers from fleeting changes in behavior during interviews in a 2026 study. Burnout and trauma stress are common in this workforce, yet stigma and weak self-report tools hide many cases. The team focused on microbehavior intervals, brief unplanned shifts across face, gaze, posture, and gesture that surface when questions touch emotional ground.
Workers recorded a video interview answering five charged questions about their jobs, then completed validated burnout and trauma surveys. Computer vision converted the recordings into time series for each channel, and an unsupervised anomaly model flagged those intervals without human labels. About twenty intervals appeared per interview on average. Features from those moments trained a classifier that sorted workers into four profiles: moderate to severe burnout, subthreshold trauma symptoms, both together, or resilient. An explanation analysis then identified the features driving each prediction.
Catalog record
| CREATOR | Sapir Gershov, Charlotte E Hilberdink, Yiwen Zhao, Sarah B Birnbaum, Victoria Mueller, Stephen P Wall, Katharina Schultebraucks |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 43 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |