Controlled and Automatic Processing during Tasks Requiring Sustained Attention. A New Approach to Vigilance

摘要

This investigation attempts to predict when a vigilance decrement will occur by analyzing performance within the automatic and controlled processing perspective (Schneider and Shiffrin, 1977). The normal observed vigilance curve is assumed to be composed of a controlled process vigilance decrement, a simple task practice effect, and in some instances automatic process learning. This analysis explains some conflicts in the vigilance literature. An experiment which examined vigilance decrements found to significant decrement in automatic processing and significant decrements in control processing. A second experiment replicated the first and showed the control processing performance vigilance decrement reduced when memory load was increased and number of channels was decreased. A third experiment showed that when subjects were required to continually and consistently allocate controlled processing resources to display locations the vigilance decrement was maximal. The results disconfirm the habituation of neural responding vigilance hypothesis. The results indicate vigilance decrements are largest when subjects must consistently apply control processing resources. Suggestions are made for structuring tasks in order to obtain maximal sustained performance when observers are faced with tasks requiring vigilance over extended periods of time. Methods for developing automatic processing to minimize vigilance decrements are proposed. (Author)