This study examined the effects of transitions in task demand on vigilance performance and perceived mental workload. Task demand was manipulated through variations in background event rate–the rate of cascade of neutral events which must be monitored in order to detect critical signals. As is typical in vigilance research, overall performance varied inversely with event rate in all phases of the study. The post-transition performance of observers shifted from a fast-to-slow event rate (high-to-low task demand) remained below that of their continuous slow event rate controls, and was thus unaffected by the shift. In contrast, the post-transition performance of monitors shifted in the opposite direction, slow-to-fast event rate, was affected by the shift. In this case, the performance of the shifted observers fell below that of their continuous fast event rate controls. These results challenge prior findings indicating that psychophysical contrast is the representative outcome of shifts in information-processing demand in vigilance tasks (Krulewitz, Warm, & Wohl, 1975). Consistent with previous findings, workload scores, as indexed by the NASA-TLX, fell at the mid-to-upper level of the scale. Shifted observers who experienced both high and low levels of task demand during the vigil showed differences in composite ratings on the Mental Demand subscale. These results serve to caution that workload measurements obtained through the NASA-TLX at the end of an experimental session containing variations in task demand do not simply reflect an averaging of the observer's demand experiences.
Deterioration of signal detectability during a vigilance task as a function of background event rate
