Skin conductance beats heart rate for detecting stress during exercise

A new study by Esther Bosch tackles a long-standing challenge in automated stress recognition: separating the biometric signatures of acute psychological stress from those of physical exertion. The research, published on arXiv, tested 19 participants in a 2x3 within-subjects design. Cognitive stress was induced via an n-back arithmetic task combined with social pressure and financial reward, across three activity conditions: idle sitting, walking, and stationary cycling. Five physiological signals were measured: tonic electrodermal activity (EDA), trapezius electromyography (EMG), heart rate, heart rate variability (RMSSD), and respiration rate.

The results reveal a clear sensor hierarchy for real-world use. Tonic EDA showed a robust, additive response to both cognitive stress (r=0.48) and physical exertion (r=0.67), with no significant interaction – meaning it can detect stress regardless of activity level. Heart rate and trapezius EMG were driven almost exclusively by physical exertion, with no reliable sensitivity to the stress task. RMSSD was strongly suppressed by physical activity and showed only marginal sensitivity to cognitive load. Respiration rate was dominated by physical activity, with no reliable stress effect in the primary analysis. These findings highlight tonic EDA as the most informative single channel when cognitive stress must be identified in physically active populations, offering immediate implications for wearable stress monitors and adaptive systems in transportation, occupational settings, and fitness tracking.