Knowledge of results and decision making performance in visual monitoring

Abstract

Two experiments were conducted to investigate the role of knowledge of results (KR) on decision making performance in a simple visual monitoring task in which subjects were required to detect aperiodic, long-duration brightness changes and ignore short duration changes. Both experiments demonstrated the characteristic decrease in percent of signals detected over the watch period when KR was not given. In addition, the results of each study were analyzed in terms of the Theory of Signal Detectability. Experiment I resulted in significant changes in log β and d' when cumulative total events KR and cumulative detect response KR were provided. Experiment II factorially compared the effects of cumulative total events KR, correct detection KR, and false alarm KR. Correct detection KR was effective in increasing d' over the monitoring session. The implications of the results of each study were discussed in terms of a decision making point-of-view. It was concluded that the role of KR on decision performance is complex and depends upon the specific type of KR presented.