Effects of Workload and Workload Transitions on Attention Allocation in a Dual-Task Environment: Evidence From Eye Tracking Metrics

Abstract

High mental workload, in addition to changes in workload, can negatively affect operators, but it is not clear how sudden versus gradual workload transitions influence performance and visual attention allocation. This knowledge is important as sudden shifts in workload are common in multitasking domains. The objective of this study was to investigate, using performance and eye tracking metrics, how constant versus variable levels of workload affect operators in the context of a dual-task paradigm. An unmanned aerial vehicle command and control simulation varied task load between low, high, gradually transitioning from low to high, and suddenly transitioning from low to high. Performance on a primary and secondary task and several eye tracking measures were calculated. There was no significant difference between sudden and gradual workload transitions in terms of performance or attention allocation overall; however, both sudden and gradual workload transitions changed participants’ strategy in dealing with the primary and secondary task as compared to low/high workload. Also, eye tracking metrics that are not frequently used, such as transition rate and stationary entropy, provided more insight into performance differences. These metrics can potentially be used to better understand operators’ strategies and could form the basis of an adaptive display.

Affective Workload Allocation for Multi-human Multi-robot Teams

“…Moacdieh et. al [60] found that the preference-based workload allocation method has benefits on workload allocation, while Wigboldus et. al [61] proposed that capability-based workload has greater productivity and effectiveness.…”

Design and validation of a computer-based task for the induction of a mental workload spectrum

“…B. Moacdieh et al. 2020 ). Die unterschiedlichen Rahmenbedingungen der einzelnen Studien erschweren den direkten Vergleich der angewandten Messgeräte und -verfahren untereinander.…”

Leveraging Initial Cognitive Load to Predict User Response to Complex Visual Tasks

“…Whether the studies use touch screen, fNIRS, eye-tracking data to measure performance and subjective cognitive load, a further work is needed to better understand the relationship between cognitive load and predicting task performance. A recent study has investigated the difference between sudden and gradual workload transitions in terms of performance during a task but not at the initial of the task [41]. Being able to determine the cognitive load of a task at the initial of the task can be used to improve those interfaces that are sensitive to time.…”