Effects of Noise and of Signal Rate upon Vigilance Analysed by Means of Decision Theory

Abstract

A vigilance task was performed in which regular flashes of light were monitored for an occasional flash of greater brightness. Following every flash a decision of signal present, signal absent, or doubtful had to be recorded. Two separate groups received high and low signal rates with a variety of the task in which the flashes occurred only on one lamp; another group received a high signal rate divided between three simultaneously flashing lights. An analysis in terms of decision theory showed that detrimental changes during the watch period were entirely attributable to movement of the subject's criterion for reporting a signal: his sensitivity to the signals if anything improved during the work period. In addition, criterion changes in the presence of intense noise occurred at high signal frequencies, even when only one source of information was involved: thus showing that division of attention between different sources is not essential for harmful effects of noise. In addition, criteria under quiet conditions were different at different signal rates.