ABSTRACT
This chapter explores the value of formal (mathematical and computer) models in psychology. Research on factors that have been shown to bias and limit unaided human reasoning is briefly reviewed, and it is noted that psychologists are susceptible to these errors, just as their subjects are. Characteristics of formal models are discussed in relation to such errors, in an effort to identify the ways in which models can and cannot aid scientific thought. Some limitations of the modeling approach are also discussed. It is argued that because psychological models greatly oversimplify the domains to which they are applied, model evaluation is a complex matter. The measure of a model's value lies not in its ability to fit data, but in how much we can learn from it.
