Prediction of Self-Perceived Stress and Arousal Based on Electrodermal Activity – a paper by Tomppa Pakarinen, Julia Pietilä and Hannu Nieminen has been presented at the European Biomedical Conference EMBC in Berlin in July.
The researchers were keen to understand how individuals perceive stress and how devices can capture it. The need for this kind of study has been there for a long time. Today, prolonged stress is a common cause of work-related health problems and has major negative impact on employee wellbeing and productiveness. Being able to measure exposure to stress long term would provide a valuable tool for improving workplace and personal wellbeing and potentially reduce health-related problems.
In contrast to some earlier studies, the researchers in this study attempted to simulate actual work-related stress rather than induce extreme reactions.
Electrodermal activity (EDA) in evaluation of mental state
Questionnaires are commonly used to measure the individual’s subjective perception of stress. Physiological measurements are used for assessing the physiological responses related to stress and arousal. The most common measures are heart rate variability (HRV), electrodermal activity (EDA), heart rate, electroencephalography, respiration, and skin temperature. In previous research, HRV has been the most commonly used method.
Electrodermal activity (EDA) reflects the functions of the autonomic nervous system and was chosen to be applied in this study. It is often used for the evaluation of different mental states such as short and long-term stress. In the study, test subjects were exposed to a 3-phase test (relaxation, arousal, stress) during which EDA was recorded, and the self-perceived stress and arousal were assessed.
The results are promising for the use of EDA as a long-term measurement of work-related stress
In this study, the research team was able to reliably classify relaxation, arousal and stress-inducing phases of simulated work with high accuracy (94.1% with BIOPAC, 82.8% with Moodmetric smart ring), using a number of EDA features.
When comparing EDA to subjective questionnaires, the self-perceived stress and arousal were classified with much lower accuracy of 60.5–72.2%. Based on the results, it is possible that individuals are less able to recognize and interpret the level of stress they are experiencing in a particular situation than the measuring devices reading their EDA.
Overall the results are promising for the use of EDA as a long-term measurement of stress at work.
For Moodmetric the research is an important, continued validation of accuracy. The easy-to-use Moodmetric smart ring can provide information on the stressfulness of work-related situations almost as accurately as respective laboratory equipment designed to measure EDA.
Access to the paper will be available later at the EMBC site
Niina Venho, CEO, +358 40 710 0487, [email protected]