Does cognitive training prevent age-related mental decline?

Does cognitive training work on its own?

The short answer is: not very well, and not consistently. When cognitive training is given as a standalone intervention, the results are mixed and often underwhelming. For example, a 2025 study that combined strength or aerobic training with cognitive training found no significant cognitive improvement on the Montreal Cognitive Assessment (MoCA) after 12 weeks, even though physical fitness improved [4]. Similarly, a 2022 trial of mindfulness training and exercise found no significant effect on episodic memory or executive function at 6 or 18 months in older adults with subjective cognitive concerns [5]. These results suggest that cognitive training by itself is not a reliable shield against age-related decline.

On the other hand, some studies do show benefits. A 2022 study of multidomain cognitive training in healthy older adults found significant improvements in story recall and reasoning after 12 weeks of twice-weekly sessions [7]. A 2025 pilot study of a home-based hybrid analog-digital cognitive training program also found improvements in executive function after 12 weeks [9]. The key difference may be that these programs were more intensive or targeted multiple cognitive domains, but even then, the effects were limited to specific areas and did not prevent overall decline.

What actually works: combining cognitive training with other lifestyle changes

The strongest evidence points to multidomain interventions—combining cognitive training with exercise, diet, social engagement, and vascular risk management—as the most effective approach. The landmark FINGER trial, a large randomized controlled trial of 1,260 at-risk older adults, found that a two-year program of diet, exercise, cognitive training, and vascular monitoring led to a small but statistically significant improvement in cognitive function compared to general health advice [10]. The intervention group showed a mean change in neuropsychological test battery Z-score of 0.20 versus 0.16 in controls, a difference that, while modest, was meaningful at the population level.

Other studies reinforce this. A 2024 trial combining resistance exercise with cognitive training found that moderate-intensity exercise produced the largest cognitive gains, with MoCA scores improving significantly more than with cognitive training alone [1]. A 2024 multicomponent intervention that included cognitive training plus lifestyle guidance (diet, sleep, exercise) for older adults with mild cognitive impairment or subjective cognitive decline also showed significant improvements in memory, executive function, and attention, with effect sizes ranging from 0.27 to 0.71 [3]. Adding social engagement to cognitive training further boosted cognitive and emotional outcomes in a 2025 study [11]. The pattern is clear: cognitive training works best as part of a broader lifestyle package.

Can diet, exercise, or supplements replace cognitive training?

Not really—at least not based on the strongest trials. The MIND diet, a hybrid of the Mediterranean and DASH diets, was tested in a large 3-year randomized trial of 604 older adults with a family history of dementia. Despite high hopes, the diet did not significantly improve global cognition or brain MRI measures compared to a control diet [8]. Similarly, spermidine supplementation, which showed promise in animal studies, failed to improve memory in a 12-month trial of older adults with subjective cognitive decline [6].

Physical activity alone has a weak but real association with better late-life cognition. A meta-analysis of 104 studies involving over 340,000 participants found that physical activity was associated with a reduced risk of cognitive impairment, but the effect was small—a pooled risk ratio of 0.97—and disappeared in follow-ups longer than 10 years [2]. Exercise improves physical fitness and may slow decline, but it is not a standalone solution. The best strategy appears to be combining multiple approaches, as the FINGER trial and others demonstrate [10][3].

About These Sources

This answer is built on 11 peer-reviewed studies — published from 2015 to 2025, 6 from 2024 or later, 6 in Q1 journals, collectively cited 617 times — selected as the most relevant from 15 studies that passed quality screening, drawn from 88 papers retrieved from a database of over 500 million.

Sources used in this answer

Related research questions