Background Previous findings have been mixed regarding the relationship between maternal depressive symptoms and child cognitive development. The objective of this study was to systematically review relevant literature and to perform a meta-analysis. Method Three electronic databases (PubMed, EMBASE, PsycINFO) were searched. Initial screening was conducted independently by two reviewers. Studies selected for detailed review were read in full and included based on a set of criteria. Data from selected studies were abstracted onto a standardized form. Meta-analysis using the inverse variance approach and random-effects models was conducted. Results The univariate analysis of 14 studies revealed that maternal depressive symptoms are related to lower cognitive scores among children aged ⩽56 months (Cohen's d = −0.25, 95% CI −0.39 to −0.12). The synthesis of studies controlling for confounding variables showed that the mean cognitive score for children 6–8 weeks post-partum whose mothers had high depressive symptoms during the first few weeks postpartum was approximately 4.2 units lower on the Mental Developmental Index (MDI) of the Bayley Scales of Infant and Toddler Development (BSID) compared with children with non-symptomatic mothers (B̂ = −4.17, 95% CI −8.01 to −0.32). Conclusions The results indicated that maternal depressive symptoms are related to lower cognitive scores in early infancy, after adjusting for confounding factors. An integrated approach for supporting child cognitive development may include program efforts that promote maternal mental health in addition to family economic wellbeing, responsive caregiving, and child nutrition.