Working memory (WM), the core component of executive functions (EFs), is crucial to early academic success and even the whole life of achievement of individual. Numerous studies have found that targeted training can improve WM... [ view full abstract ]
Working memory (WM), the core component of executive functions (EFs), is crucial to early academic success and even the whole life of achievement of individual. Numerous studies have found that targeted training can improve WM and that children at low socioeconomic status (SES) benefited from targeted cognitive training. The present study investigated whether computerized WM training can yield near and far transfer on WM and fluid intelligence (Gf) in typically developing 7-year-olds with a pretest-training-posttest design.
Twenty-six low SES first-graders participated the WM training group, and another 25 first-graders with relatively high-SES was assigned as control group. During the experiment, participants in the control group received the school routines normally. For the WM training group, children performed a 15-min daily training session for 20 days (5 days a week). The training program (Loosli, et al., 2012) was designed to improve the visual working memory capacity. All the children received Backward Digit Span Task (BDS) for WM capacity, Raven Standard Progressive Matrices (RSPM) for Gf within a week before and after the training.
Children’s data was analyzed using mixed 2*2 (group * time) ANOVA. For the BDS task, results demonstrated that both the main effect of time (F (1, 49) = 5. 66, p < 0.05，partial η square = 0.10) and group (F (1, 49) = 11. 84, p < 0.01，partial η square = 0.22) were significant. The interaction effect between group and time was also significant (F (1, 49) = 13.93, p < 0.01， partial η square= 0.20) and further analysis showed that this was principally driven by changes across the intervention periods for the training group (p < 0.01). There was no significant group difference at pretest. For the RSPM, the main effect of time was significant (F (1, 49) = 3.85, p = 0.05，partial η square= 0.07). Further analysis showed that the training group improved significantly from the pretest to post-test (p < 0.05) while the control group didn’t. There was no significant group difference at pretest, either.
In conclusion, the targeted WM training can improve WM ability in typically developing low-SES children, transfer to Gf could also be elicited.
