A comparison of two working memory tasks in aphasia

Background: Overall, there is growing consensus that working memory (WM) should be routinely assessed in individuals with aphasia as it can contribute significantly to their level of language impairment and be an important factor in treatment planning. However, there is still no consensus in the field as to which tasks should be used to assess WM in aphasia. The two main alternatives are adapted complex span tasks and N-back tasks. Both have been used interchangeably in previous studies of WM in aphasia, even though the correspondence between the two tasks has not been properly established. Aims: The current study investigates the relationship between two WM tasks—complex span and N-back tasks—in a large sample of individuals with aphasia. The relationships of these tasks to measures of language comprehension are also explored, as well as differences in performance patterns between individuals with nonfluent and fluent aphasia. Methods & Resources: Forty-four participants with aphasia (nonfluent: n = 27; fluent: n = 13; mixed: n = 4) were examined with a modified listening span task (Ivanova & Hallowell, 2014), an auditory verbal 2-back task, and a standardised Russian language comprehension test. Outcomes & Results: Results revealed a moderate relationship between the two WM measures, but demonstrated a divergence in terms of their relationship to language comprehension. Performance on the modified listening span task was related to language comprehension abilities, but performance on the 2-back task was not, suggesting that the two tasks primarily index different underlying cognitive mechanisms. Furthermore, the relationship between the modified listening span task and language comprehension was significant for individuals with non-fluent aphasia, but not for those with fluent aphasia. Conclusions: Overall, the data demonstrate that while performance of individuals with aphasia was related on the two tasks, the two tasks cannot be substituted for one another without further inquiries into their underlying differences. ARTICLE HISTORY Received 6 November 2015 Accepted 24 March 2016