Comparison of Bayesian and regression approaches to the study of information processing in judgment

In recent years there have been several hundred studies within the rather narrowly-defined topic of information utilization in judgment and decision making. Much of this work has been accomplished within two basic schools of research, which we have labeled the “regression” and the “Bayesian” approaches. Each has its characteristic tasks and characteristic information that must be processed to accomplish these tasks. For the most part, researchers have tended to work strictly within a single approach and there has been minimal communication between the resultant subgroups of workers. Our objective here is to present a review and comparative analysis of these two approaches. Within each, we examine (a) the models that have been developed for describing and prescribing the use of information in decision making; (b) the major experimental paradigms, including the types of judgment, prediction, and decision tasks and the kinds of information that have been available to the decision maker in these tasks; (c) the key independent variables that have been manipulated in experimental studies; and (d) the major empirical results and conclusions. In comparing these approaches, we seek the answers to two basic questions. First, do the specific models and methods characteristic of different paradigms direct the researcher's attention to certain problems and cause him to neglect others that may be equally important? Second, can a researcher studying a particular substantive problem increase his understanding by employing diverse models and diverse experimental methods?