The Cognitive Eye: Indexing Oculomotor Functions for Mental Workload Assessment in Cognition-Aware Systems

Cognitive load assessment has become an integral part of human-centered design as it enables cognition-aware systems to adapt to users' needs. Among the range of physiological metrics, task-evoked pupillary responses have been implemented successfully to determine mental workload. Only recently, smooth pursuit eye-movements as another oculomotor indicator was suggested to index cognitive processing demands. The current study provides a comparison of both parameters with regard to the capacity to differentiate task load in an auditory n-back task. Results replicate earlier findings, suggesting pupil diameter to display the level of mental workload. However, task loads during high levels of difficulty are not distinguishable. Task difficulty also influences gaze behaviour during smooth pursuit eye-movements, although far less sensitive. A clear distinction is achieved only between low and high workloads. In contrast, a combination of control measures (reaction times, accuracy rates) is suited to distinguish between all n-back task stages.