Abstract
Background Increasing efforts toward prevention of stress-related mental disorders have created a need for unobtrusive real-life monitoring of stress-related symptoms. Wearable devices have emerged as a possible solution to aid in this process, but their use in real-life stress detection has not been systematically investigated.
Methods Using ecological momentary assessments (EMA) combined with wearable biosensors for ecological physiological assessments (EPA), we investigated the impact of an ecological stressor (i.e., an exam week) on physiological arousal and affect. With this paradigm we investigated whether we could use wearable devices to detect stress states using machine learning models.
Results During stressful high-stake exam (versus control) weeks, participants reported increased negative affect and decreased positive affect. Intriguingly, physiological arousal was decreased on average during the exam week. Time-resolved analyses revealed peaks in physiological arousal associated with both self-reported stress and self-reported positive affect, while the overall decrease in physiological arousal was mediated by lower positive affect during the stress period. We then used machine learning to show that a combination of EMA and physiology yields optimal identification of stress states.
Conclusions Our findings highlight the potential of wearable biosensors in stress-related mental-health monitoring, but critically show that psychological context is essential for interpreting physiological arousal detected using these devices.
Competing Interest Statement
The authors have declared no competing interest.
Footnotes
Funding Statement: This work was supported by the European Research Council (ERC2015-CoG 682591).
Added an extra analysis after feedback from reviewers, with not changes made to any other analyses run. Elaborated on some points in the discussion. Additionally reduced the overall word count.