The prediction-oriented middle ground between behaviorist and cognitivist consumer research

Abstract

Classic behaviorist and cognitivist approaches to consumer research are based on very different assumptions concerning the nature of the cognitive processes that drive consumption. Classic behaviorist research assumes that the underlying cognitive processes are extremely complex, and the effort to model them is not likely to lead to useful insights. In contrast, classic cognitivist research assumes that the underlying cognitive processes can be precisely captured with simple cognitive models. The current paper highlights the value of a middle ground. This middle ground favors the assumption that while the underlying processes are highly complex, their impact can be approximated with prediction-oriented cognitive models. The potential of this approach is illustrated with a review of previous research demonstrating how a simple prediction-oriented model can shed light on four interesting consumption puzzles: Buying and not using, the backfiring of certain safety devices, too much and insufficient checking, and underconsumption and overconsumption.