How Is Working Memory Training Likely to Influence Academic Performance? Current Evidence and Methodological Considerations.

Working memory (WM) is one of our core cognitive functions, allowing us to keep information in mind for shorter periods of time and then work with this information. It is the gateway that information has to pass in order to be processed consciously. A well-functioning WM is therefore crucial for a number of everyday activities including learning and academic performance (Gathercole et al., 2003; Bull et al., 2008), which is the focus of this review. Specifically, we will review the research investigating whether improving WM capacity using Cogmed WM training can lead to improvements on academic performance. Emphasis is given to reviewing the theoretical principles upon which such investigations rely, in particular the complex relation between WM and mathematical and reading abilities during development and how these are likely to be influenced by training. We suggest two possible routes in which training can influence academic performance, one through an effect on learning capacity which would thus be evident with time and education, and one through an immediate effect on performance on reading and mathematical tasks. Based on the theoretical complexity described we highlight some methodological issues that are important to take into consideration when designing and interpreting research on WM training and academic performance, but that are nonetheless often overlooked in the current research literature. Finally, we will provide some suggestions for future research for advancing the understanding of WM training and its potential role in supporting academic attainment.

Working Memory Training is Associated with Long Term Attainments in Math and Reading.

Training working memory (WM) using computerized programs has been shown to improve functions directly linked to WM such as following instructions and attention. These functions influence academic performance, which leads to the question of whether WM training can transfer to improved academic performance. We followed the academic performance of two age-matched groups during 2 years. As part of the curriculum in grade 4 (age 9-10), all students in one classroom (n = 20) completed Cogmed Working Memory Training (CWMT) whereas children in the other classroom (n = 22) received education as usual. Performance on nationally standardized tests in math and reading was used as outcome measures at baseline and two years later. At baseline both classes were normal/high performing according to national standards. At grade 6, reading had improved to a significantly greater extent for the training group compared to the control group (medium effect size, Cohen's d = 0.66, p = 0.045). For math performance the same pattern was observed with a medium effect size (Cohen's d = 0.58) reaching statistical trend levels (p = 0.091). Moreover, the academic attainments were found to correlate with the degree of improvements during training (p < 0.053). This is the first study of long-term (>1 year) effects of WM training on academic performance. We found performance on both reading and math to be positively impacted after completion of CWMT. Since there were no baseline differences between the groups, the results may reflect an influence on learning capacity, with improved WM leading to a boost in students' capacity to learn. This study is also the first to investigate the effects of CWMT on academic performance in typical or high achieving students. The results suggest that WM training can help optimize the academic potential of high performers.

Cogmed WM training: reviewing the reviews.

Does Cogmed working-memory training (CWMT) work? Independent groups of reviewers have come to what appears to be starkly different conclusions to this question, causing somewhat of a debate within scientific and popular media. Here, various studies, meta-analyses, and reviews of the Cogmed research literature will be considered to provide an overview of our present understanding regarding the effects of CWMT. These will particularly be considered in light of two recent critical reviews published by Melby-Lervåg and Hulme ( 2013 ) and Shipstead, Hicks, and Engle ( 2012 ) and their arguments and conclusions assessed against current available evidence. Importantly we describe how the conclusions drawn by Melby-Lervåg and Hulme appear to contradict their own findings. In fact, the results from their meta-analysis show highly significant effects of working-memory (WM) training on improving visuospatial WM and verbal WM (both ps < .001). In addition, analyses of long-term follow-ups show that effects on visuospatial WM remain significant over time (again at p < .001). Thus, the analyses show that WM is indeed improved using WM training, and the highest effect sizes are achieved using CWMT (compared with other training programs). We also conclude that there is current evidence from several studies using different types of outcome measures that shows attention can be improved following CWMT. In a little more than a decade, there is evidence that suggests that Cogmed has a significant impact upon visual-spatial and verbal WM, and these effects generalize to improved sustained attention up to 6 months. We discuss the evidence for improvements in academic abilities and conclude that although some promising studies are pointing to benefits gained from CWMT, more controlled studies are needed before we can make strong and specific claims on this topic. In conclusion, we find that there is a consensus in showing that WM capacity and attention is improved following CWMT. Due to the importance of WM and attention in everyday functioning, this is, on its own, of great potential value.

Polymorphisms in the dopamine receptor 2 gene region influence improvements during working memory training in children and adolescents.

Studying the effects of cognitive training can lead to finding better treatments, but it can also be a tool for investigating factors important for brain plasticity and acquisition of cognitive skills. In this study, we investigated how single-nucleotide polymorphisms (SNPs) and ratings of intrinsic motivation were associated to interindividual differences in improvement during working memory training. The study included 256 children aged 7-19 years who were genotyped for 13 SNPs within or near eight candidate genes previously implicated in learning: COMT, SLC6A3 (DAT1), DRD4, DRD2, PPP1R1B (DARPP32), MAOA, LMX1A, and BDNF. Ratings on the intrinsic motivation inventory were also available for 156 of these children. All participants performed at least 20 sessions of working memory training, and performance during the training was logged and used as the outcome variable. We found that two SNPs, rs1800497 and rs2283265, located near and within the dopamine receptor 2 (DRD2) gene, respectively, were significantly associated with improvements during training (p < .003 and p < .0004, respectively). Scores from a questionnaire regarding intrinsic motivation did not correlate with training outcome. However, we observed both the main effect of genotype at those two loci as well as the interaction between genotypes and ratings of intrinsic motivation (perceived competence). Both SNPs have previously been shown to affect DRD2 receptor density primarily in the BG. Our results suggest that genetic variation is accounting for some interindividual differences in how children acquire cognitive skills and that part of this effect is also seen on intrinsic motivation. Moreover, they suggest that dopamine D2 transmission in the BG is a key factor for cognitive plasticity.

Computerized training of non-verbal reasoning and working memory in children with intellectual disability.

Children with intellectual disabilities show deficits in both reasoning ability and working memory (WM) that impact everyday functioning and academic achievement. In this study we investigated the feasibility of cognitive training for improving WM and non-verbal reasoning (NVR) ability in children with intellectual disability. Participants were randomized to a 5-week adaptive training program (intervention group) or non-adaptive version of the program (active control group). Cognitive assessments were conducted prior to and directly after training and 1 year later to examine effects of the training. Improvements during training varied largely and amount of progress during training predicted transfer to WM and comprehension of instructions, with higher training progress being associated with greater transfer improvements. The strongest predictors for training progress were found to be gender, co-morbidity, and baseline capacity on verbal WM. In particular, females without an additional diagnosis and with higher baseline performance showed greater progress. No significant effects of training were observed at the 1-year follow-up, suggesting that training should be more intense or repeated in order for effects to persist in children with intellectual disabilities. A major finding of this study is that cognitive training is feasible in this clinical sample and can help improve their cognitive performance. However, a minimum cognitive capacity or training ability seems necessary for the training to be beneficial, with some individuals showing little improvement in performance. Future studies of cognitive training should take into consideration how inter-individual differences in training progress influence transfer effects and further investigate how baseline capacities predict training outcome.

Dopamine, working memory, and training induced plasticity: implications for developmental research.

Cognitive deficits and particularly deficits in working memory (WM) capacity are common features in neuropsychiatric disorders. Understanding the underlying mechanisms through which WM capacity can be improved is therefore of great importance. Several lines of research indicate that dopamine plays an important role not only in WM function but also for improving WM capacity. For example, pharmacological interventions acting on the dopaminergic system, such as methylphenidate, improve WM performance. In addition, behavioral interventions for improving WM performance in the form of intensive computerized training have recently been associated with changes in dopamine receptor density. These two different means of improving WM performance--pharmacological and behavioral--are thus associated with similar biological mechanisms in the brain involving dopaminergic systems. This article reviews some of the evidence for the role of dopamine in WM functioning, in particular concerning the link to WM development and cognitive plasticity. Novel data are presented showing that variation in the dopamine transporter gene (DAT1) influences improvements in WM and fluid intelligence in preschool-age children following cognitive training. Our results emphasize the importance of the role of dopamine in determining cognitive plasticity.

Gains in fluid intelligence after training non-verbal reasoning in 4-year-old children: a controlled, randomized study.

Fluid intelligence (Gf) predicts performance on a wide range of cognitive activities, and children with impaired Gf often experience academic difficulties. Previous attempts to improve Gf have been hampered by poor control conditions and single outcome measures. It is thus still an open question whether Gf can be improved by training. This study included 4-year-old children (N = 101) who performed computerized training (15 min/day for 25 days) of either non-verbal reasoning, working memory, a combination of both, or a placebo version of the combined training. Compared to the placebo group, the non-verbal reasoning training group improved significantly on Gf when analysed as a latent variable of several reasoning tasks. Smaller gains on problem solving tests were seen in the combination training group. The group training working memory improved on measures of working memory, but not on problem solving tests. This study shows that it is possible to improve Gf with training, which could have implications for early interventions in children.

The SNAP25 gene is linked to working memory capacity and maturation of the posterior cingulate cortex during childhood.

BACKGROUND: Working memory (WM) is the ability to retain task relevant information. This ability is important for a wide range of cognitive tasks, and WM deficits are a central cognitive impairment in neurodevelopment disorders such as attention-deficit/hyperactivity disorder (ADHD). Although WM capacity is known to be highly heritable, most genes involved remain unidentified. METHODS: Single nucleotide polymorphisms in genes previously associated with cognitive functions or ADHD were selected for genotyping. Associations of these with WM tasks were investigated in a community sample of 330 children and young adults. One single nucleotide polymorphisms was also investigated in an independent sample of 88 4-year-old children. Furthermore, association between brain structure and activity, as measured by magnetic resonance imaging techniques, and single nucleotide polymorphisms alleles were estimated in 88 participants. RESULTS: Genotype at rs363039, located in the gene coding for synaptosomal-associated protein, 25 kDa (SNAP25) was associated to WM capacity in both samples. Associations in the community sample were also found with measures of other cognitive functions. In addition, this polymorphism affected the gray matter and brain activity in the posterior cingulate cortex, an area included in the so-called default mode network previously correlated to regulation of attention and hypothesized to be implicated in ADHD. CONCLUSIONS: A novel gene-brain-behavior network was identified in which a genotype located in SNAP25 affects WM and has age-dependent effects on both brain structure and brain activity. Identifying such networks could be a key to better understanding cognitive development as well as some of its disorders.

Measuring working memory capacity with greater precision in the lower capacity ranges.

Working memory capacity is usually measured as the number of stimuli correctly remembered. However, these measures lack precision when assessing individuals with low capacity. This study aimed to create a more precise measure of visuospatial working memory capacity, using intra-level differences in difficulty between items. In two experiments, children aged 4-6 years (N = 97) were tested on a large number of items. Data showed a large variability of difficulty within each level and the factors contributing to this variability were identified. This variability can be used to provide a precise measure of working memory capacity in the lower ranges.