The microstructure of dual-task interaction. 4. Sleep deprivation and the control of attention.

The experiment was designed to investigate attention-switching behaviour in sleep-deprived subjects and involved the same dual-task paradigm reported in earlier studies. The task specified by instructions to be the main task, was the five-choice serial-response task; the secondary task was an auditory-verbal digit task which occurred at random time intervals. A surprising result was that, in terms of overall response rates, the sleep-deprived group, although slower, preserved the dual-task priority structure better than the control group, who slowed on both tasks in the combined condition. The control result was different in this respect from control data from previous research, but it is important that within the sleep-deprived group there was evidence that ability to preserve the priority structure was unaffected. Analysis of the intervals in which the two tasks concurred gave a different result. Sleep-deprived subjects seemed less able to control the fine structure of the interval and tended to persist in producing serial responses at the expense of the secondary-task response. The evidence supported the view that both sleep-deprived and control subjects were not able to process information in parallel and, thus, the difference in interval patterning was due to the difference in attention-switching processes. The possibility that sleep deprivation might result in loss of attentional control is considered briefly.