BACKGROUND: There is growing literature examining the effects of balance training on cognitive function in individuals with multiple sclerosis (MS), yet the findings remain inconsistent. This study aimed to investigate methodological characteristics of balance training studies and examine the effects of this modality on cognitive function. METHODS: This study performed literature search using MEDLINE, EMBASE, PsycINFO, SPORTSDiscus, and CINAHL databases from inception to April 2024. Randomized and non-randomized controlled trials of balance training studies that assessed cognitive function were included. The Tool for the Assessment of Study Quality and Reporting in Exercise was used to evaluate methodological quality of included studies. We calculated effect sizes as standardized mean differences (SMD) using Hedge's g and employed multi-level random-effects model for meta-analysis. RESULTS: Fourteen articles with 1,020 participants were included in this review, and a subset of seven studies with 26 effect sizes were further used for meta-analysis. Balance training indicated medium effects on global cognition (SMD [95 % CI] = 0.39 [0.13, 0.64]), processing speed (SMD [95 % CI] = 0.36 [0.00, 0.73]), and visuospatial memory (SMD [95 % CI] = 0.53 [0.12, 0.93]) and large effect on verbal memory (SMD [95 % CI] = 0.66 [0.25, 1.06]). We found that intervention characteristics and exercise doses significantly moderated the effects on cognition. CONCLUSIONS: Balance training may effectively improve overall and specific domains of cognitive function in persons with MS. Exercise doses of at least 120 min per week and a total duration of 720 min or more were significant moderators for greater cognitive improvement.