Verbal and non-verbal working memory in aphasia: What three n-back tasks reveal

Background: Researchers have found that many individuals with aphasia (IWA) present with cognitive deficits that may impact their communication, and perhaps underlie their language-processing deficits (e.g., Erickson et al., 1996 Erickson, E., Goldinger and LaPointe, L. 1996. Auditory vigilance in aphasic individuals: Detecting non-linguistic stimuli with full or divided attention. Brain and Cognition, 30: 244–253. Crossref], PubMed], Web of Science ®] , Google Scholar]; Murray et al., 1997 Murray, L.L., Holland, A.L. and Beeson, P.M. 1997. Accuracy monitoring and task demand evaluation in aphasia. Aphasiology, 11: 401–414. Taylor & Francis Online], Web of Science ®] , Google Scholar]; Wright et al., 2003 Wright, H. H., Newhoff, M., Downey, R. and Austerman, S. 2003. Additional data on working memory in aphasia. Journal of International Neuropsychological Society, 9: 302 Google Scholar]). However, many investigations of cognitive ability in aphasia have included measures that may be considered “language heavy”; they require overt lexical, semantic, and/or phonological processing to follow the task instructions and/or formulate a response. Few have considered the amount of linguistic processing required to perform the task. Subsequently, it is not clear if poorer performance by IWA on cognitive tasks compared to neurologically intact (NI) participants is due to a deficit in the respective cognitive domain or due to the inability of IWA to perform the task because of their language difficulties.

Aims: The purpose of the current study was to explore the effect of varying linguistic processing demands in the context of a dynamic working memory task—an n-back task for participants with and without aphasia.

Method & Procedures: This study compared differences on three different n-back tasks within and across groups for individuals with aphasia and NI matched peers. Participants completed three different n-back tasks; stimuli for the tasks varied in “linguistic load”. For each n-back task participants completed two levels of difficulty: 1-back and 2-back.

Outcomes & Results: The aphasia group performed significantly worse than the NI participants across the n-back tasks. All participants performed significantly better with the stimuli that carried a higher linguistic load (i.e., the fruit), than with the fribbles (semi-linguistic) and blocks (non-linguistic). All participants performed significantly better on the 1-back than the 2-back working memory task. Unlike the NI participants, IWA performed equally poorly with the fribbles and the blocks in the 2-back task.

Conclusions: Overall, the performance of individuals with aphasia on working memory tasks that varied in their linguistic load was similar to the control group but reduced. However, unlike the NI participants, IWA were less skilled at rapidly utilising linguistic knowledge to increase performance on the fribbles, demonstrating the further decrement in working memory that results from a decreased ability to utilise a linguistic strategy to increase performance on verbal working memory tasks. The results of this study indicate that language ability has a significant influence on performance on working memory tasks and should be considered when discussing cognitive deficits in aphasia.