Extending an integrated computational model of the time-based resource-sharing theory of working memory.

The time-based resource-sharing (TBRS) model envisions working memory as a rapidly switching, serial, attentionalrefreshing mechanism. Executive attention trades its time between rebuilding decaying memory traces and processingextraneous activity. To thoroughly investigate the implications of the TBRS theory, we integrated TBRS within the ACT-Rcognitive architecture. This allowed us to test the TBRS model against both participant accuracy and RT data in a dual taskenvironment and in particular, determine the patterns in these data directly attributable to working memory limitations. Inthe current work, we extend the model to include articulatory rehearsal, which allows us to examine suppression effects.Additionally, we use the model to predict performance under a larger range of cognitive load. These predictions enable astronger test of the TBRS model that would not be possible without our complete computational account of TBRS and thegeneral assumptions of the ACT-R framework.