The growing interest in preserving cognition through video games in the ageing population is evidenced by its wealth of literature. However, there has been a lack of consensus on video games efficacy and even limited understanding on the factors moderating its effectiveness. The present review sought to evaluate video game effects, relative to no-game control condition, on cognitive functions in older adults. The functions included processing speed, attention, delayed memory, reasoning, and executive functions (inhibition, shifting and updating memory). Personal and methodological moderators were explored to explain the variability in cognitive effects using meta-regression. Through a systematic literature search of online databases, 27 intervention studies were eligible, with a total of 1126 participants, for analysis. Using random-effect models, small but significant training effects were found on updating memory but none were detected in other cognitive functions. The heterogeneity across studies was in general high for all cognitive functions and was partially accounted for, in all cognitive functions except for inhibition and updating, by different moderators such as the proportion of females, outcome measure parameters, training intensity, and game type. The findings highlight limited effects of video games on cognitive functions that were largely influenced by factors related to study design.