Are behavioral measures useful for detecting cognitive workload during human-computer interaction?

Abstract

Commonly used techniques for measuring cognitive workload during human-computer interactions can be cumbersome or intrusive to task performance. In the current work, we examine the utility of heuristic behavior analysis, including keystroke dynamics, mouse tracking, and body positioning for measuring cognitive workload during direct interactions between humans and computers. We present a method for modeling behavioral measures as well as physiological and neurophysiological data using probabilistic, statistical, and machine learning algorithms for real-time estimation of human states. We believe this discussion will inform the capability to provide estimates of cognitive workload in real-world scenarios.