Abstract
Performance in a task of visual vigilance in which both ‘wanted’ and ‘ unwanted ’ signals occurred was observed in an experiment in which signal frequency and wanted signal probability (ratio of wanted to unwanted signals) were varied independently. It was found that: 1. A sixfold increase in the probability that any signal occurring would be a wanted one resulted in a considerable improvement in the efficiency with which those wanted signals were detected, whereas a similar increase in the frequency with which signals were presented produced no significant alteration in efficiency. 2. The improvement in performance appeared during the first ten minutes of the watch and remained in evidence throughout the session. 3. The improvement was most marked in the case of signals appearing near the periphery of the display. 4. Individual differences in detection efficiency were related to the frequency with which false reports were made. Implications for the theory of vigilance behaviour, and the relevance of the results to the organization of monitoring work, are considered in the discussion.

This publication has 7 references indexed in Scilit: