Performance in a complex multiple-task environment during a laboratory-based simulation of occasional night work.

A study was carried out to examine the impact of occasional night work on simulated process control using a complex task environment. The 21 student participants were tested during 2 6-hr simulated shifts (daytime and night). In addition to the primary system management task, the simulation allowed measurement of fault diagnosis behavior, monitoring and control actions, and two secondary tasks--alarm reaction time and system status checks (prospective memory)--as well as subjective state. Consistent with predictions from compensatory control theory, night work did not impair system performance, although monitoring and control were reduced (supported by subjective reports of increased use of risky "corner-cutting" strategies). Secondary tasks showed an increase in alarm reaction time during night work, but there was no effect on prospective memory and no clear pattern of change in subjective state. Actual or potential applications of this research include the design of complex systems for nighttime operation.