The impact of time-of-day and time-on-task on cognitive processing by telephone operators

Mental workload is a topic that has received considerable attention from the human factors research community. Research efforts to understand the relationship between mental workload and cognitive processing performance have resulted in development of cognitive models such as Multiple Resource Theory. However, this particular model has two shortcomings: (1) it oversimplifies cognitive processes and tasks, rendering them inapplicable to a real workplace, and (2) it fails to address the effects of time-of-day on cognitive processing performance. Evidence suggests that predictable cognitive performance rhythms exist, which vary over time-of-day in parallel with 24 hour circadian cycles of physiological arousal. In order to investigate time-of-day effects on cognitive processing performance, the current study used “real-world” performance data, continuously collected for several days from 60 telephone operators. Performance on two complex high-processing load and two “automatic” low-processing load tasks were analyzed over two shifts. Hypothesis 1 proposed that performance on the four tasks are significantly impacted by time-of-day and/or time-on-task, displaying performance trends as predicted by prior research. Hypothesis 2 proposed that significantly different performance trends are apparent in comparisons between tasks resulting from differences in load, displaying inverse performance relationships, as predicted by prior research. Trend analyses using ANOVA were conducted with between-subject time-of-day and task load factors, and a within-subject time-on-task factor. Performance of three out of the four tasks displayed significant time-of-day effects. Task load comparisons yielded a significant difference in task performance depending on task load for one set of tasks. Performance trend diagrams were compared in terms of shape and direction to trends for comparable processes found in the literature. Certain performance trends in the current study displayed agreement with performance trends in the literature, while others did not. Despite the uncontrolled nature of field data, the results suggest that for certain tasks, cognitive performance varies as a function of time-of-day in concurrence with similar relationships found in controlled experiments. The findings, which provide partial support for both hypotheses, are discussed in the context of possible intervening performance factors. The potential for applying the results of the current study to a “real-world” workplace is discussed.