Working Memory for Water-Escape Loci in Rodents
Larry W. Means East Carolina University
During the past 20 years, behavioral neuroscientists have shown increasing interest in studying working memory, memories for information that is situation or trial dependent. Accordingly, several variations of delayed matching-to-sample (DMTS) ( Blough, 1959; Dunnett, Evenden, & Iversen, 1988; Mishkin & Delacour, 1975; Parkinson & Elsmore, 1989; Roitblat, 1980) and other trial-dependent tasks, including the radial arm maze tasks ( Beatty & Shavalia, 1980; Olton & Samuelson, 1976), the working memory version of the Morris water maze task ( Morris, 1984), the platform with holes tasks ( Barnes, 1979; Oades & Isaacson, 1978; van der Staay, Kretch ing, Blokland, & Raaijmakers, 1990), novel object exploration ( Ennaceur & Delacour, 1988), and others have been developed. Although all of these tasks require the use of trial-dependent information, they differ with respect to what the subject must remember, the motor demands that are made on the subject, the source of motivation and reinforcement, and how readily they can be acquired.
While studying several related working memory problems over the past five years, we have tested rodents on several versions of water-escape DMTS (also called win-stay, lose-shift), and delayed nonmatching-to-sample (DNMTS) (also called win-shift, lose-stay). This work has resulted in the development of a DMTS task that we feel is an excellent assay of working memory. The task that we currently use ( Means, 1993) has several features that make it valuable for behavioral neuroscience research: