The Memory Machine

The foundations of an adequate cognitive science that binds the cognitive activities of human beings into a coherent conceptual system with the neurological basis of these activities has been slow to develop. The problem is partly due to the complexity of the relationships that must be set up between a naturalistic analysis of the discursive practices such as remembering, and the brain mechanisms by which they are accomplished. There are plenty of intermediate models of cognitive ‘mechanisms’ but they have been developed for the most part with little attention to the ontological constraints on model building that are central to the physical sciences. The transition from human activity to neurological hypothesis can be accomplished by a second step of modeling in which standard cognitive models are revised by the use of connectionist architectures, to provide a foundation for plausible neurological hypotheses. The argument is set out in the context of the psychology of remembering