The post-1965 literature on the use of secondary tasks in the assessment of operator workload was surveyed. Twelve classes of tasks were identified; the most frequently used were choice reaction time, memory, monitoring, and tracking. The literature review did not suggest a single best task or class of tasks for the measurement of workload. Limitations in using secondary tasks are discussed, and directions for future research are presented.

The tasks used in 41 studies of the effects of alcohol on human performance were classified in terms of the abilities required for task performance. For each category of task, median performance computed across all studies within the category was plotted as a function of both dosage and the time between alcohol administration and task performance. It was found that the curves relating performance to dosage differed as a function of the ability requirements of the task. The effects on performance of time from the start of drinking to the initiation of task performance also depended upon the particular abilities required by the task. Despite many differences among the specific tasks involved, their categorization according to ability requirements enabled an integration of results and the establishment of tentative functional relationships between performance, dosage, and time from ingestion.

No abstract

No abstract

The effects on performance of the value of detecting a signal, the cost of a miss or false detection, and the size of the set from which the signals were drawn were studied in an auditory vigilance task. Seventy-two subjects were randomly assigned to each cell of a factorial arrangement of the cost and load variables and required to detect and identify each of several 49 db SPL pure tones differing only in frequency. Analyses of the number of correct detections, correct identifications, false detections and detection response time indicated a significant performance decrement with time for all measures and suggested that increasing costs for misses and false detections led to poorer detection performance while value had no effect. Load effected only identification performance, as higher loads led to a decrease in the percentage of signals correctly identified. The ď and β statistics of signal detection theory, indicated sensitivity to be invariant with manipulations of costs and with time. These findings imply that the performance decrement during a vigil is due to an increased strictness in the criterion the subject sets for deciding whether or not a signal was present. The cost factors were effective in manipulating performance by causing changes in the subjects' decision criteria.