tion, showing how theoretical models, influenced by the newly developing area of computer science, could provide tools for investigating attention, perception, and memory, problems that had been regarded as totally intractable by earlier behaviorist approaches.By this time, Baddeley's rat experiment had appeared, but Hullian psychology had disappeared. The rapidity with which theoretical fashions changed impressed him and encouraged adopting a style of theorizing whereby the theory kept quite close to the empirical data, in the hope that it would prove more durable, if less exciting, than more ambitious flights of theoretical speculation.Conrad was a brilliant applied psychologist who at this time was carrying out extensive work on the design of codes that is still highly relevant, though rarely heeded. Broadbent's work had suggested the need to consider separate short-term memory (STM) and long-term memory (LTM) systems, and Conrad's observation of acoustic intrusion errors in the recall of letter codes suggested that STM relies on an acoustic code.At this time, Baddeley's own work was principally concerned with LTM, but while Conrad was away on sabbatical, he began to play with Conrad's new phenomenon. Like the sorcerer's apprentice, he was captivated by it and has been working on acoustic similarity ever since. In particular, work with a colleague and friend, Harold Dale, seemed to indicate that whereas STM relied principally on an acoustic code, LTM was more dependent on semantic coding. Baddeley also developed an interest in applied research, including the underwater study of divers, which subsidized a hobby and subsequently led to work with Duncan Godden showing very clear context-dependent memory effects.After nine years in Cambridge, the opportunity arose to join the new University of Sussex. The experimental psychology department there was led by Stewart Sutherland, like Donald Broadbent, a charismatic figure, though of a rather more rumbustious temperament. The department proved to be an extremely open and intellectually lively one. The need to teach also broadened Baddeley's horizons, as did the amval of his first doctoral students, Arnold Wilkins and John Richardson, both of whom have gone on to distinguished research careers. His first postdoctoral fellow, Graham Hitch, joined Baddeley in a project investigating the relation between STM and LTM, a link that was becoming increasingly problematic for existing theories. Their results suggested the need to abandon the idea of a unitary STM in favor of a multicomponent system, to which they applied the already existing term working memory. The model was presented in an invited chapter in 1974 (Baddeley & Hitch, 1974) and has continued to drive the work of both Baddeley and Hitch since that time.The opportunity to work with Elizabeth Warrington at the National Hospital, Queen's Square, London, convinced ' Articulatory was changed to phonological to emphasize the fact that this subsystem is not limited to the articulatory component. The term sketchpad was ...