Violations of the Speed-Accuracy Tradeoff Relation

Abstract

Beginning with the seminal work by Simon (1955), decision making has been conceptualized in terms of information-processing systems that take anticipated future consequences as input, and after processing this information for some period of time, produce an action as output. Decision researches have considered two primary ways to measure the “performance” of a decision process. One way is in terms of the “accuracy” of the process—the probability that the process selects the “best” action, where best is defined by an accepted normative theory (e.g., Expected Utility Theory). The other way is in terms of effort—the average amount of time or number of mental operations required to reach a decision. The ideal decision process would be the one that maximizes accuracy and minimizes effort. Unfortunately, these two performance measures tend to be positively related—a decrease in effort usually produces a decrease in accuracy. Thus the decision maker has to tradeoff effort for accuracy. How this is done is the primary question addressed in this chapter.