MIT CogNet, The Brain Sciences ConnectionFrom the MIT Press, Link to Online Catalog
SPARC Communities
Subscriber : Stanford University Libraries » LOG IN

space

Powered By Google 
Advanced Search

 

Benefits of Computational Modeling for Cognitive Neuroscience Studies of Verbal Working Memory.

 David E. Meyer, David E. Kieras, Shane Mueller and Travis L. Seymour
  
 

Abstract:
Verbal working memory (VWM) has become a major topic of research in cognitive neuroscience over the past decade. Stimulated by theoretical proposals of Baddeley et al. (1974, 1975, 1986), cognitive neuroscientists have demonstrated the existence and identified the brain loci of several VWM components, including phonological storage, subvocal rehearsal, and the "central executive" (e.g., Paulesu et al., 1993; Petrides et al., 1993; Cohen et al., 1994; Awh et al., 1996; Schumacher et al., 1996). Nevertheless, much remains to be learned about the algorithmic operations of these components and the neural mechanisms that implement them. In order for this crucial further progress to take place, detailed formal modeling rather than loose qualitative theorizing is needed. Consequently, using the Executive-Process Interactive Control (EPIC) architecture developed by Meyer and Kieras (1997), we have begun formulating precise computational models of VWM. Our models provide explicit thorough specifications of the symbolic codes, memory stores, maintenance activities, retrieval operations, and executive processes associated with human performance in prototypical paradigms such as the serial memory-span task. With these specifications, accurate quantitative fits to representative empirical data sets can be obtained. As a result, new insights from our research may significantly stimulate future cognitive neuroscience studies of VWM.

 
 


© 2010 The MIT Press
MIT Logo