Remi and Rouse: Quantitative Models for Long-Term and Short-Term Priming in Perceptual Identification

This chapter presents two models of priming. The primary task under consideration is the identification of words presented visually at threshold. The first model, REMI (Retrieving Effectively from Memory, the 'I' stands for implicit), is a model for long-term priming in implicit memory. It explains repetition priming effects by assuming that during study of a word some contextual information is added to the corresponding lexical trace. This contextual information stored during the study task will tend to match the contextual information present during the test task, leading subjects to prefer studied words over non-studied words. The second model, ROUSE, is a model of short-term priming. ROUSE stands for Responding Optimally with Unknown Sources of Evidence, and it is able to explain an intricate pattern of results.