Real-Time Prediction of Fluctuations in Cognitive Workload
Human operators often experience large fluctuations in cognitive workload that can lead to sub-optimal performance, ranging from overload to neglect. Help from automated support systems could potentially address this issue, but to do so the system would ideally need to be aware of real-time changes in operators’ cognitive workload, so it can provide help in times of peak demand and take advantage of troughs to elicit operator engagement. We used the ISO standard Detection Response Task (DRT) to measure cognitive workload approximately every 4 seconds in a demanding task requiring monitoring and refuelling of a fleet of unmanned aerial vehicles (UAVs). We showed that the DRT provided a valid measure that can detect changes in workload due to changes in the number of UAVs. We used a cross-validation analysis to assess whether measures related to task performance immediately preceding the DRT could be used to predict detection performance as a proxy for cognitive workload. Although the simple occurrence of task events had weak predictive ability, composite measures that tapped operators’ situational awareness with respect to fuel levels were much more effective. We conclude that real-time prediction of operators’ cognitive workload shows promise as an avenue for improved human-automation teaming.