Modularity

Abstract One approach to the perception/cognition distinction is based on the cognitive architecture of the mind, the relatively fixed structures within which perception and cognition operate. One type of architectural theory is the modularity view. Another is the dimension restriction hypothesis. But even if the mind is not modular, there are significant partial truths in modularity theses. For example, perception has significant dimensions of informational encapsulation. Jerry Fodor (1983) characterized modules in terms of a list of nine diagnostic properties that are supposed to apply to input systems but not to central cognition. Those properties are: domain specificity, mandatory operation, limited central accessibility, fast processing, informational encapsulation, “shallow” outputs, fixed neural architecture, characteristic and specific breakdown pattern, and characteristic ontogenetic pace and sequencing. Each module was supposed to have its own “database” and its own algorithms which were available to the computations of that module but not to other modules. This chapter discusses Fodor’s notion of modularity and goes through his criteria one by one, concluding that there is some truth in the modularity thesis but that it is substantially wrong.