Cognitive workload classification using cardiovascular measures and dynamic features

Monitoring cognitive workload has the potential to improve performance and fidelity in human decision making through a real-time monitoring model. Multiple studies have shown a successful binary classification of high and low workload using various methods and often focused on multiple physiological signals. A more detailed detection of cognitive workload is needed for a meaningful and reliable workload monitoring tool. This study focuses on trinary workload classification of parameters extracted from the cardiovascular system. The experiment was validated with the use of a database containing 96 participants performing tasks designed to induce slight variations in cognitive workload. Two distinct supervised learning classifying methods were used and their likelihood score used for the classification schemes of (1) each heartbeat and (2) each task screen. The results show that the support vector classifier outperforms the random forest with the average misclassification rate of 20.44% using the whole screen classification scheme instead of individual heartbeat classification.