The Effects of Response Alternatives on Keeping Track Performance

Abstract

A keeping-track task was used to explore the role of interference in dynamically updating memory. Subjects performed a keeping-track task in which all monitored attributes either shared four response alternatives, or were distributed such that each monitored attribute had its own unique set of four states. The response alternatives were distributed differently on the computer keyboard in these two conditions. Results suggest that the mapping effect seen in Yntema and Meuser's (1960) original experiments is best explained by interference: Accuracy was lowest when all attributes shared the same four states. Spatially distributing the response alternatives benefited accuracy when subjects kept track of multiple attributes with unique response sets, but having to choose from many response keys led to increased response times. These data suggest that keeping-track performance is sensitive to interference from shared attribute sets and to how the response environment is designed.