Converging sources of evidence and theory integration in working memory: A commentary on Morey, Rhodes, and Cowan (2019)

Morey et al. (2019) offer a critique of the dominance of the multiple component framework of working memory in the interpretation of patterns of impairment and sparing in individuals with focal brain damage associated with specific impairments of immediate, serial-ordered verbal recall. They argue that the lack of pure cases of verbal short-term memory impairments, that recognition performance is higher than recall in such patients, that digits are remembered better than other verbal material, and that problems with replicability in patient studies undermine this traditional theoretical interpretation of the impairments from which these patients suffer. They further speculate that an alternative theoretical framework for working memory, incorporating embedded processes and perception-action links offers a more plausible account of the data from these patients. This commentary points to a range of errors and misconceptions in the arguments presented, notably that such patients are not as rare as suggested, that their recognition is actually no better than their recall, that digits offer substantial advantages for memory, and that results have been shown to be replicable between and within individuals. It is proposed that attempts to integrate more recent theoretical developments in working memory with those shown previously to be highly successful in accounting for impairments in these patients, and for generating hypotheses and accounts across a wide range of contexts may be a more fruitful approach to advancing understanding of cognition in the healthy and damaged brain.