Introduction

Over the past several chapters, we have developed our accuracy–effort framework based upon a top-down view of strategy selection. That is, we have assumed that a decision maker has a repertoire of strategies, evaluates the costs and benefits of those strategies given his or her a priori perceptions of the decision task, and then selects the particular strategy that represents the best accuracy–effort tradeoff for the task at hand. As we have seen in the previous chapter, there are data supporting such a goal-directed process of strategy selection. However, assuming both the use of single prototypical strategies and the notion of a top-down process of strategy selection is certainly too restrictive. Problem solvers not only use information extracted from the original problem definition in deciding what strategy to use but also use information they have already explored to identify promising paths for further search (Langley et al., 1987). That is, as people learn more about the problem structure during the course of making a decision, they can change their processing to exploit that structure. Such a view implies that individuals construct choice strategies on the spot during the course of making a decision (Bettman, 1979), which is a more bottom-up or data-driven view of strategy usage. Processing, in other words, can be opportunistic (Hayes-Roth & Hayes-Roth 1979).

Several examples may clarify this concept. For instance, an individual may intend to evaluate a set of gambles using a weighted adding (expected value) strategy.