ABSTRACT
This chapter aims to present a body of new experimental evidence which provides a firm basis for the working memory hypothesis. It describes a series of experiments on the role of memory in reasoning, language comprehension, and learning. An attempt is made to apply comparable techniques in all three cases in the hope that this will allow a common pattern to emerge, if the same working memory system is operative in all three instances. The chapter suggests a dichotomy between working memory and the recency effect, in contrast to the more usual view that both recency and the memory span reflect a single limited capacity short-term buffer store (STS). The most devastating evidence against the hypothesis that STS serves as a crucially important working memory comes from the neuropsychological work of Shallice and Warrington. The chapter discusses the effect of factors which might be supposed to influence a working memory system, should it exist, across a range of cognitive tasks.
