ABSTRACT
Put in the most general terms, the question I wish to address is how to account for the responses that occur in short-term recall. A simplistic answer is that they are just reports of the items remembered from the input list. That answer does not suffice because we observe errors, that is, responses that do not correspond to items in the input list. A somewhat more sophisticated answer arising from a large body of research is embodied in what I will term the trial-unit model, which constitutes a generalization of the familiar “modal moder of short-term memory (Baddeley, 1976; Murdock, 1961) and seems to be assumed implicitly by most investigators of short-term recall. This model, more aptly termed a framework for specific models, is illustrated in Figure 9.1.
