Wearable Research Reveals Gaps in Student Stress Detection
A systematic review published 30 March 2026 synthesizes 134 studies (Jan 2020–Dec 2025) on wearable-based stress detection in college-aged students, finding electrodermal activity used in 57.5% (n=77) and support vector machines as the most common best-performing model in 33.3% (n=45). The authors report heavy reliance on preexisting datasets — 62.8% (n=84), with ~80% (n=67) using the 15-participant WESAD — and call for more diverse data and temporal modeling.
Key Points
- 1Reports electrodermal activity as the most used physiological signal in 57.5% (n=77) of studies.
- 2Highlights reliance on small preexisting datasets, notably WESAD used in ~80% (n=67) of reused datasets.
- 3Recommends broader, diverse datasets and temporal models to improve generalizability and real-world deployment.
Scoring Rationale
Solid, peer-reviewed systematic review with credible methods; scores high on credibility and topic relevance but offers incremental novelty focused on a student subpopulation. The score reflects clear findings about dataset overuse, methodological gaps, and same-day publication timeliness.
Sources
Public references used for this report.
Practice interview problems based on real data
1,625 SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems
