A dynamic ocular index for assessing mental workload across visual tasks

Mental workload changes dynamically as people interact with visually demanding tasks, yet existing ocular measures typically focus on individual features and may overlook coordinated changes across the ocular system. We hypothesized that workload-related changes are reflected in the deviation patterns of multiple ocular signals relative to baseline state. Here, we introduce the Ocular Dynamics Deviation Index (ODDI), which integrates deviations in pupil and gaze behavior to characterize dynamic ocular responses during visual tasks. We evaluated the validity of ODDI in two experiments with 32 participants, including a controlled visual letter discrimination task and three simulated visual display terminal (VDT) tasks. ODDI reliably differentiated mental workload in the controlled experiment, showed greater sensitivity than conventional ocular measures, identified specific time windows for workload discrimination, and was significantly associated with task performance. These findings generalized across diverse VDT scenarios, including monitoring, tracking, and resource management tasks. Our results suggest that ODDI advances mental workload assessment beyond isolated ocular features by enabling individualized characterization of dynamic ocular changes, providing a foundation for real-time workload monitoring in complex visual environments.