Search

Highlights • Non-computerised training can improve working memory and near-transfer skills. • Training other skills (physical activity, play, inhibition) is beneficial indirectly. • Tapping into attentional resources (executive control) is key to task effectiveness. • Further studies need clear theoretical underpinnings and rigorous methodology. • Outcome measurement and dosage need greater consideration., It has been suggested that diverse interventions applied within children’s everyday contexts have the potential to improve working memory (WM) and produce transfer to real-world skills but little is known about the effectiveness of these approaches. This review aims to examine systematically the effectiveness of non-computerised interventions with 4–11 year olds to identify: (i) their effects on WM; (ii) whether benefits extend to near- and far-transfer measures; (iii) if gains are sustained over time; (iv) the active ingredients; and (v) the optimum dosage. Searches were conducted across 12 electronic databases using consistent keywords. Papers were screened by title and abstract (n = 6212) and judged against pre-defined eligibility criteria (n = 63). Eighteen papers were included in the review. They used a range of non-computerised WM intervention approaches that included: (i) adapting the environment to reduce WM loads; (ii) direct WM training with and without strategy instruction; and (iii) training skills which may indirectly impact on WM (physical activity, phonological awareness, fantastical play and inhibition). Both direct training on WM tasks and practicing certain skills that may impact indirectly on WM (physical activity, fantastical play and inhibition) produced improvements on WM tasks, with some benefits for near-transfer activities. The common ingredient across effective interventions was the executive-loaded nature of the trained task i.e., training on a task that taps into attentional and processing resources under executive control and not just the storage of information. Few studies reported dosage effects, measured far-transfer effects (n = 4), or tested the durability of gains over time (n = 4). The lack of a clear theoretical framework in many of the included studies resulted in ambiguous predictions about training and transfer effects, and inadequate use of outcome measures. Methodological issues also constrain the strength of the evidence, including: small samples sizes; an absence of blinding of participant and outcome assessors; and lack of active control groups. Further well-designed and controlled studies with clear theoretical underpinnings are required to expand and enhance the evidence base. The heterogeneity of the interventions and of the study designs (randomised and non-randomised) in the included papers limited the synthesis of evidence across studies. However, this diversity enabled the identification of key ingredients, notably the training of executive-loaded WM tasks, which can help inform novel approaches to WM intervention in everyday contexts.

Two developments have contributed to the convergence of views about the benefits of real-life and inquiry-based pedagogies in mathematics learning. First, the mathematics teaching community is increasingly focused on the learning of mathematics that involves the transfer of prior knowledge to novel problem-solving situations, a key element in recent characterizations of the notion of Numeracy. Second, research about human cognition in domain-specific learning suggests that problem-solving activity provides better contexts in which to observe the construction of creative connections of disparate information. The question is how can we examine the above cognitive processes, as these are played out in Numeracy contexts? (Tan, Educ Res Policy Pract 6:101–114, 2007) identified dialogue and inquiry as important themes of the psychology that girds problem-based learning. In this report, we take up Tan’s suggestion that research needs to make learner’s cognition more visible by immersing a cohort of Malaysian and Australian students in arithmetic word problems. Students were required to explain and justify their solutions. The frameworks of dialogue representation and schema guided our analyses that focused on students’ readings, explanations, representations and reflections about a given set of problem contexts. Results showed that both the groups experienced difficulty in representing far-transfer numeracy problems. However, Australian children tended to develop longer and more varied explanations in comparison to their Malaysian counterparts. Potential implications for classroom practices, policy-making and future research directions are explored.