Abstract

To date, previous meta-analyses reported that computerized cognitive training (CCT) was a clinically beneficial intervention for cognitive function in older adults with mild cognitive impairment (MCI). However, little is known about the efficacy of narrowly defined cognition-specific CCT excluding commercial video games on cognitive function in patients with MCI. Randomized controlled trials (RCTs) of CCT in older adults with MCI were searched through CINAHL, Embase, Medline, PubMed, and PsychINFO. The overall cognitive domains, global cognitive function, attention, memory, working memory, and executive function were pooled separately for MCI. The overall effect on cognitive function in MCI across 8 trials was moderate (Hedges' g = 0.48, 95% confidence interval (CI) = 0.003-0.974). There was no significant publication bias. Moderate to large effects were found for global cognitive function, memory, and working memory, with the exception of attention, and executive function. Cognition-specific CCT was beneficial in improving global cognitive function, memory, and working memory in older adults with MCI. Therefore, this intervention warrants long-term trials with a larger number of subjects to investigate the effect on conversion from MCI to dementia.