ABSTRACT
This chapter describes a method we have developed over the past 5 or 6 years for constructing neurobiologically plausible computational models of behavior. The method is quite general and can be applied to a wide variety of behavioral phenomena. The components in the models that are constructed with this method are units that correspond to groups of similar cells in specific brain regions (e.g., cor-
tical columns or hypercolumns). The resulting models are also quite flexible with respect to the types of data they can be tested against. For example, we describe detailed methods for fitting the models to single-cell recording data, fMRI data, and human behavioral data.
