Working memory training format in older adults: individual versus group sessions.

Working memory (WM) training has been shown to increase the performance of participants in WM tasks and in other cognitive abilities, but there has been no study comparing directly the impact of training format (individual vs. group) using the same protocol. Therefore, the aim of this study was to compare the efficacy of the Borella et al. three session verbal WM training offered in two different formats on target and transfer tasks. This study was conducted in two waves. In the first wave, participants were randomized into individual training (n = 11) and individual control conditions (n = 15). In the second wave, participants were randomized into group training (n = 16) and group control conditions (n = 17). Training consisted of three sessions of WM exercises and participants in the active control condition responded to questionnaires during the same time. There was significant improvement for both training conditions at post-test and maintenance at follow-up for the target task, other WM tasks, processing speed, and executive functions tasks. The ANOVA results showed that the training gains did not depend on the WM training format. However, the effect size analyses suggested that this intervention can be more effective, at short term and follow-up, when provided individually. To conclude, this study showed that providing this training collectively or individually does not change the training benefits, which increases the possibilities of its use in different contexts.

Verbal working memory training in older adults: an investigation of dose response.

The WM training protocol proposed by Borella et al. found specific and transfer effects among seniors, however, the studies were carried out in the same socio-cultural context and variations in the procedure were never tested. The present study aimed at analyzing the efficacy of Borella et al.'s training, in terms of short and long-term benefits, in a different socio-cultural context (Study 1), and the effect of change in the training's length (duplicating the number of sessions (Study 2). Participants were randomly assigned to a trained group (N = 18 for Study 1, and N = 23 for Study 2) and active control group (N = 28 for Study 1, and N = 27 for Study 2), and evaluated at pre, post-test and six-month follow-up for verbal WM task (criterion task), and for visuospatial and verbal WM, inhibition, processing speed, executive function, and fluid intelligence measures (transfer tasks). The trained groups had higher performance in all tasks when compared with active control groups after training and at 6 month follow-up. The longer training (Study 2) generated similar gains as the original protocol, with some advantage in far transfer tasks at post-test and follow-up. Study limitations include the small sample sizes. In conclusion, this training was effective in a different socio-cultural context and adding three sessions to the protocol did not significantly change training impact.

Categorization Working Memory Span Task: Validation study of two Brazilian alternate versions.

BACKGROUND: The Categorization Working Memory Span Task (CWMS task) is a complex working memory (WM) span test that has been used previously to assess age and individual differences in WM as well as the relationship between WM and complex aspects of cognition such as listening comprehension. Two alternate versions of the task have been developed to be used as outcome variables in WM training protocol, which mirrors the task involved in the CWMS task. OBJECTIVES: (1) To translate and adapt the CWMS task for Brazilian Portuguese; (2) to test the equivalence of its 2 alternate versions; (3) to examine temporal stability; (4) to examine the influence of age and schooling on CWMS task; (5) to establish its relationship with other tests of WM. METHODS: Eighty-one older adults completed version A, and 86 completed version B of the CWMS task. After 6 months, a subsample (n = 85) completed the same version of the task. RESULTS: Versions A and B of the task generated comparable scores. Both versions had adequate temporal stability, which was higher for the CWMS total recall, which is the classical variable in this task to represent WM performance, when compared to the other variables generated by the test. The CWMS task variables were moderately correlated with schooling and other cognitive tests (Mini Mental State Examination, Letter-Number Sequencing, Spatial Span Backward, Digit Span Forward). CONCLUSIONS: The 2 versions of the CWMS task were equivalent and stable temporally. The task was influenced by schooling and global cognition.