It has long been a puzzle why the benefits of intensive cognitive training are so limited. In a new study, Professor Susan Gathercole and colleagues from the MRC Cognition and Brain Sciences Unit put forward a new theory that training in unfamiliar working memory tasks involves acquiring new cognitive routines rather than improving existing working memory capacities. These routines can only be successfully applied to other similar tasks, limiting transfer. The process of constructing new routines draws on general cognitive resources rather than working memory itself. For tasks such as digit span that are already supported in verbal short-term memory, new routines are not necessary and this reduces still further transfer to other similar tasks. The theory is supported by a meta-analysis of training studies and by re-analysis of two large training datasets. It suggests there is little prospect that training working memory with highly artificial tasks can have a meaningful impact on a broad range of everyday cognitive abilities that rely on an extensive array of routines acquired over many years. An alternative approach of designing training to enhance performance in specific activities may be more useful.

The full article can be read here: Working memory training involves learning new skills

Please contact Professor Susan Gathercole for more information.