Gobet, F. (2000).
Long-term working memory: A computational implementation for chess expertise. Proceedings of the 3rd International Conference on Cognitive Modelling. Veenendaal, The Netherlands: Universal Press.
Long-term working memory, recently proposed by Ericsson and Kintsch (1995), is a theory covering empirical data from several domains, including text comprehension, classical laboratory memory experiments, and expert behaviour. One difficulty in applying and evaluating the long-term working memory theory, however, is that it is framed in rather general terms, and that several mechanisms and parameters are left unspecified. This paper proposes a computer implementation of the theory for a domain that Ericsson and Kintsch cover in depth, namely chess memory. Simulations of Saariluoma's (1989) experiment where both game and random chess positions are presented auditorily make it possible to analyse two key ingredients of long-term working memory: encoding through elaboration of LTM schemas and patterns, and encoding through retrieval structures. In the simulations, these mechanisms were modulated by two parameters. The results show that low values for these two parameters provide good fit with the random positions, and that game positions are less sensitive to the parameters' values.