Wearable Monitoring of Cognitive Load: A Comparative Sensor Study

Wearable sensing technologies are becoming increasingly relevant for tracking autonomic responses associated with cognitive workload across laboratory and ecological settings. Nevertheless, the extent to which different devices provide reliable and task-sensitive physiological markers is still not fully established. While several studies have investigated electrodermal activity (EDA) and cardiac-derived variability using established wearable systems, there is currently limited literature evaluating recently introduced multimodal devices such as the Empatica EmbracePlus, and no publicly available datasets providing simultaneous comparative recordings across multiple wearable platforms including this device. In this study, electrodermal activity, electrocardiography and photoplethysmography were simultaneously recorded with Polar H10, Shimmer3 GSR+, and Empatica EmbracePlus, while participants performed two cognitively demanding tasks. Features were extracted from the recorded biosignals in each experimental condition. Statistical analysis revealed task-related autonomic modulations across signals. Pearson correlation analysis revealed feature- and device-dependent cross-device agreement, with lower correlations observed between the newer wearable devices and selected reference devices. Overall, these findings emphasize the critical role of device characteristics and sensor placement in wearable-based cognitive stress assessment.