Use of wearable physiological sensors to predict cognitive workload in a visuospatial learning task

Area of Science:

• Human-computer interaction
• Cognitive science
• Biomedical engineering

Background:

• Cognitive workload, or mental strain, is linked to decreased performance.
• Understanding and predicting cognitive workload is crucial for optimizing human performance and safety.

Purpose of the Study:

• To investigate the efficacy of physiological monitoring in predicting cognitive workload and performance.
• To establish correlations between physiological signals and self-reported workload and task performance.

Main Methods:

• Twenty-one participants underwent physiological monitoring (heart rate, heart rate variability, electrodermal activity, skin temperature, electromyographic activity) during rest and cognitive tasks.
• Tasks included a visuospatial learning simulation (crane operation) and the Stroop test.
• Physiological data were analyzed in relation to self-reported frustration, workload, and task performance.

Main Results:

• Task performance, specifically learning in the crane operation simulation, correlated inversely with self-reported frustration.
• Physiological metrics like heart rate, electrodermal activity, and trapezius muscle activity increased significantly during the simulation compared to rest.
• Muscle activity and heart rate variability (high frequency power) were significantly associated with self-reported workload and simulation performance.

Conclusions:

• Physiological monitoring provides a viable method for assessing cognitive workload and predicting performance.
• These findings support the real-time application of physiological monitoring for decision-making in areas such as injury risk assessment and personalized training programs.