Early Childhood Movement- and Arts-Based Experiences and Executive Functions: A Systematic Review and Meta-Analysis

Abstract

Highlights What are the main findings? Across 15 Study IDs and 29 model-ready effects, average benefits were observed for inhibition, working memory and cognitive flexibility, but all prediction intervals crossed the null, and GRADE certainty was low. The quantitative core comprised 12 primarily structured programmes and three guided-play/structured-hybrid programmes; no free or unstructured-play intervention entered the meta-analysis. What are the implication of the main findings? Benefits should not be generalised to movement, play or arts participation per se; pedagogical structure and deliberately embedded executive demands may influence outcomes. Future trials should report task progression, complexity, teacher guidance, child agency, implementation fidelity and cluster-adjusted analyses sufficiently to identify active programme components. Abstract Background/Objectives: Movement- and arts-based experiences are frequently promoted as developmentally appropriate contexts for supporting cognition in early childhood, but effects may differ across executive-function domains, pedagogical formats and study contexts. Methods: This PRISMA 2020 systematic review and meta-analysis was registered in PROSPERO (CRD420261363924). PubMed/MEDLINE, Web of Science, Scopus, ERIC, PSICODOC, CINAHL and SPORTDiscus were searched from inception through April 2026. One immediate-post effect per Study ID and core domain was retained. Random-effects models used restricted maximum likelihood with Hartung–Knapp adjustment. Study-level risk of bias was assessed with RoB 2 or ROBINS-I, missing-evidence risk with ROB-ME and certainty with GRADE. Intervention content, pedagogical format and embedded cognitive demands were also coded descriptively. Results: The review included 130 reports representing 125 unique studies. Fifteen Study IDs contributed 29 core effects: 10 for inhibition (n = 675), 10 for working memory (n = 697) and nine for cognitive flexibility (n = 642). Twelve core programmes were primarily structured, and three used guided-play/structured-hybrid formats; no free/unstructured-play intervention entered the meta-analysis. Mean effects favoured the intervention for inhibition (Hedges’ g = 0.52, 95% CI 0.15 to 0.90), working memory (g = 0.42, 95% CI 0.09 to 0.75) and cognitive flexibility (g = 0.60, 95% CI 0.22 to 0.98). Heterogeneity was substantial, and all 95% prediction intervals crossed the null. Estimates remained favourable after excluding Serious risk-of-bias judgments. Age-exclusion and ICC-sensitivity analyses retained positive effects; a worst-case exclusion of unadjusted class- or cluster-assigned studies reduced precision for inhibition and working memory. SMD-specific modified-precision regressions did not indicate small-study associations for inhibition (p = 0.529) or working memory (p = 0.619); cognitive flexibility was not formally tested because k = 9. ROB-ME indicated some concerns, and GRADE certainty was low for all three domains. Conclusions: Favourable average effects should not be attributed to movement or arts participation per se. Pedagogical structure and embedded cognitive demands may influence outcomes, but substantial inconsistency, unit-of-analysis uncertainty and low certainty preclude identification of a superior model. Funding: There was no external funding.