Abstract
While the search for a theory of rehabilitation by Caramazza and Hillis is welcomed, the model they propose is criticised on the grounds that it offers little more than a recommendation that we should wait until rehabilitation has occurred, and then attempt to understand it. The theme of this comment is to propose that it is necessary not only to understand the systems being treated, but also to have a model of how they can be changed by experience. Examples are given of ways in which recent developments in the cognitive neuropsychology of learning and memory may contribute to improved rehabilitation methods. It is suggested that an adequate theory of rehabilitation must combine a knowledge of the systems undergoing treatment with an understanding of how the principles of learning may be applied to their modification.