Working Memory Training for Healthy Older Adults: The Role of Individual Characteristics in Explaining Short- and Long-Term Gains

Abstract

Objective: The aim of the present study was to explore whether individual characteristics such as age, education, vocabulary, and baseline performance in a working memory (WM) task-similar to the one used in the training (criterion task)-predict the short- and long-term specific gains and transfer effects of a verbal WM training for older adults. Method: Four studies that adopted the Borella et al. (2010) verbal WM training procedure were found eligible for our analysis as they included: healthy older adults who attended either the training sessions (WM training group), or alternative activities (active control group); the same measures for assessing specific gains (on the criterion WM task), and transfer effects (nearest on a visuo-spatial WM task, near on short-term memory tasks and far on a measure of fluid intelligence, a measure of processing speed and two inhibitory measures); and a follow-up session. Results: Linear mixed models confirmed the overall efficacy of the training, in the short-term at least, and some maintenance effects. In the trained group, the individual characteristics considered were found to contribute (albeit only modestly in some cases) to explaining the effects of the training. Conclusions: Overall, our findings suggest the importance of taking individual characteristics and individual differences into account when examining WM training gains in older adults.

Keywords: age; general cognitive ability; individual characteristics; individual differences; older adults; training benefits; working memory baseline performance; working memory training.

Figure 1

Group (trained vs. control) X Session (pre-test vs. post-test vs. follow-up) interaction from the best model for each measure of interest. Categorization Working Memory Span task (A), Dot Matrix task (B), Forward (C) and Backward (D) digit span tasks, Cattell test (E), Pattern Comparison task (F), Stroop Color task (G) and intrusion errors in the Categorization Working Memory Span task (H). CWMS, Categorization Working Memory Span Task; CWMS intrusions, intrusion errors in the Categorization Working Memory Span Task; RTs, Response Times. Segments represent the 95% credibility intervals.

Figure 2

Fitted values from the best model for the trained group alone for each variable of interest. Categorization Working Memory Span task (A), Dot Matrix task (B), Forward (C) and Backward (D) digit span tasks, Cattell test (E), Pattern Comparison task (F), Stroop Color task (G) and intrusion errors in the Categorization Working Memory Span task (H). CWMS, Categorization Working Memory Span Task; CWMS intrusions, intrusion errors in the Categorization Working Memory Span Task; RTs, Response Times; cwmspre, CWMS baseline performance level.