A 2026 meta-analysis of small randomized trials found that virtual reality and related game-based training may improve some thinking-test scores in people with Alzheimer’s dementia. The results did not show that the programs made everyday activities easier or improved quality of life, mood, balance or falls.
The findings offer a limited but potentially useful picture of what these programs may do. The studies tested short courses lasting four to 12 weeks, and none examined whether any cognitive gains continued after training ended.
What the review found
Across eight trials, overall cognition favored VR-based training over control groups. The standardized mean difference was 0.44, with a 95% confidence interval from 0.21 to 0.68. In practical terms, the review found a modest difference in cognitive test performance between the groups, but this does not establish a meaningful change in day-to-day independence.
Short-term memory showed the clearest specific result. Across three trials, the standardized mean difference was 0.62, with a confidence interval from 0.25 to 0.99.
Executive function did not improve clearly overall. Two studies involving at least 24 sessions reported a larger effect, but two studies with fewer sessions did not. Because this pattern came from only two trials on each side, the review treated it as a hypothesis rather than evidence for a specific number of sessions.
Daily life did not change clearly
The review found no statistically clear benefit for activities of daily living. Results across four trials had a standardized mean difference of −0.38, but the confidence interval was wide, ranging from −1.60 to 0.84.
Quality of life also showed no clear improvement across three trials. The same was true for depression, balance and coordination, fall risk and fear of falling. These findings should not be read as proof that VR worsens these outcomes. The available studies were small, and several confidence intervals were broad enough to include benefit, no effect or harm.
The distinction matters for families and care teams. A person may perform somewhat better on a memory or cognition test without finding it easier to dress, prepare food, manage routines or complete other daily tasks. The review did not show that VR slows Alzheimer’s disease or changes its progression.
“VR” covered many different activities
The 11 reports included 361 participants: 182 in VR groups and 179 in control groups. However, two Italian reports followed the same cohort, so the number of unique people may be lower.
Participants had clinically diagnosed Alzheimer’s dementia and ranged in mean age from 73 to 87. Six trials involved mild dementia and five involved moderate dementia. The programs generally ran two to five times per week, with 10 to 40 sessions lasting 20 to 45 minutes.
The interventions were not all the same. The review used a broad definition that included Wii, Kinect, Xbox, video games, computerized cognitive training, screen-based tasks and a small number of fully immersive headset programs. No trial established that a full headset works better than a screen or game console, and no study isolated immersion as the active ingredient.
Why the evidence remains limited
Only two reports were rated at low risk of bias. Three were rated at high risk, while six raised some concerns. The researchers described the overall evidence as weak.
The findings also resemble earlier evidence on cognitive training that did not use VR. A 2019 Cochrane review of 33 trials involving 1,924 people with mild to moderate dementia found a similar global-cognition effect, along with no improvement in daily activities, mood or behavioral symptoms.
No adverse events were reported in the included VR trials. Still, none was large enough to identify uncommon problems, such as falls during exercise-based gaming or motion sickness with headsets.
What this may mean for families
For a person with mild or moderate Alzheimer’s dementia who enjoys games, supervised VR or game-based activity may be a reasonable form of engagement. The likely potential benefit, based on current evidence, is a modest improvement in thinking-test performance during the program.
Families should not assume that the equipment will improve independence, prevent falls or slow the disease. The evidence does not directly apply to mild cognitive impairment, because every participant in the review had diagnosed Alzheimer’s dementia. Longer studies are needed to determine whether any cognitive gains last and whether particular types or amounts of training are more useful.
AI tools were used to assist with the preparation of this article.
Leave a Reply