Cusp catastrophe models for cognitive workload and fatigue: a comparison of seven task types

The study introduces a nonlinear paradigm that addresses several unresolved problems concerning cognitive workload and fatigue: (a) how to separate the effects of workload versus fatigue, (b) whether the upper boundaries of cognitive channel capacity are fixed or variable, and how multitasking produces a bottleneck phenomenon, (c) that prolonged time on task can produce performance decrements but also produce improvements in task performance associated with practice and automaticity, and that (d) task switching can alleviate fatigue but could be mentally costly. This study describes two cusp catastrophe models that have become useful for separating the workload and fatigue performance phenomena and explores the role of task switching and multitasking in both performance phenomena. In the experiment, 105 undergraduates completed seven computer-based tasks seven times under one of four experimental conditions: tasks fully alternated, tasks aggregated with the multitask module performed first, aggregated with the multitask module performed last, and where the participants chose the task order themselves. Results supported both the cusp models such that fatigue effects were stronger for tasks with higher memory or attentional demand, and were often counteracted by practice effects; spelling ability acted as a compensation variable in most cases, and the intervening amount of work done acted as the bifurcation variable. For cognitive workload, catastrophic shifts in performance were noted between the single tasks and the multitask, with relative difficulty of the single task acting as the load (asymmetry) variable and the flexible task ordering condition as the bifurcation variable.