Cognitive workload and affective state: A computational study using Bayesian networks

This paper uses Bayesian networks to investigate the impact of three different kind of inputs, namely, physiological, cognitive and affect features, on workload estimation, from a computational point of view. The ability of the proposed models to infer the workload variation of subjects involved in successive tasks demanding different levels of cognitive resources is discussed, in term of two criteria to be jointly optimized: the diversity, i.e. the ability of the model to perform on different subjects, and the accuracy, i.e., how close from the (subjectively estimated) workload level the model prediction is.