Deriving and Testing Production System Models of Sequential Decision‐Making Behavior

The present paper has two major goals.The first goal is to provide a detailed description of a new approach to constructing models of individual decision-making (DM) behavior that satisfy three general requirements.These requirements are that models of individuals:1. Possess the ability to predict the DM behavior of a given individual in a relatively complex decision environment.In particular, it is important that the models have an ability to represent individual variations in DM behavior. Possess the property of indicating relationships between an individual'sbehavior and an individual's representatioh of the world.It is difficult both to explain variations in the DM behaviors of different decision makers and to understand changes in DM behavior over time without some explicit reference to an individual's internal representation of the world.3.Not place unreasonable demands on computational resources during construction or use.Recent advances in our knowledge of both computation and human cognitive processing suggest procedures for constructing new classes of models that permit us to satisfy such requirements (see, e.g., Smith 1983;Smith, Pellegrino, and Golledge 1982).The second goal is to demonstrate the ability of the approach to satisfy these criteria by applying such procedures to the results of a study of DM behavior in a housing market search context.