Independent public reference library

Ageing biology, biomarkers, interventions, and research literacy.

Cognitive Training and Healthy Ageing Evidence

Key Takeaways

Cognitive training uses repeated, structured practice to improve a defined cognitive ability, such as memory strategy, reasoning, working memory, attention, or processing speed. It differs from general cognitive stimulation, which includes broader activities such as discussion, reading, games, or learning without a standardized task and progression rule. [1] [2]

Who This Is Useful For

This page is for readers evaluating claims that cognitive exercises can preserve cognition or prevent dementia during otherwise healthy ageing. It separates improvement on a trained task from transfer to other cognitive abilities, everyday independence, and clinically diagnosed dementia. [1] [7]

How Cognitive Training Could Produce Change

Practice can make task-relevant operations more efficient and can teach explicit strategies. Memory training may teach mnemonic organization, reasoning training may teach pattern detection, and adaptive speed training may progressively shorten the time available to detect and locate visual information. These are plausible learning effects, but a gain in a practised operation does not by itself demonstrate slower biological brain ageing. [3] [7]

The central research question is transfer. “Near transfer” means improvement on an untrained measure that closely resembles the training; “far transfer” means improvement in a different ability or in daily life. Reviews find extensive evidence for learning the trained tasks, less evidence for near transfer, and limited evidence for far transfer or everyday cognitive benefit. [7] [8]

Evidence at a Glance

Outcome Level Main Finding Interpretation Main Limitation
Trained tasks Performance usually improves with structured practice. [7] Older adults retain capacity for task-specific learning. Practice effects can be mistaken for a general cognitive change.
Cognitive tests Meta-analyses report small average effects, varying by domain and study design. [1] Some near transfer is supported. Effects are heterogeneous and are weaker with demanding control conditions.
Everyday function ACTIVE found modest long-term differences in self-reported difficulty with instrumental activities for some groups. [4] Functional transfer is possible but not broad or immediate. Self-report, selective attrition, and multiple outcomes complicate interpretation.
Dementia diagnosis ACTIVE follow-ups found lower incidence after speed training, especially with booster training, but not after memory or reasoning training. [5] [6] One specific training method has a long-term preventive signal. Diagnosis was assessed secondarily and later through Medicare claims rather than as the original primary endpoint.

What the ACTIVE Trial Tested

ACTIVE randomly assigned 2,832 community-dwelling adults aged 65 or older to memory training, reasoning training, visual speed-of-processing training, or a no-contact control. Each programme began with ten group sessions; eligible participants in the training arms were later randomized to receive booster sessions. The interventions targeted separate abilities rather than a single measure of “brain health.” [3]

At ten years, reasoning and speed training retained effects on their targeted abilities, whereas the memory-training difference was no longer statistically significant. Participants in all three training groups reported less difficulty with instrumental activities of daily living than controls, although the absolute differences were modest and objective functional measures showed a less uniform pattern. [4]

The Dementia Findings

A secondary ACTIVE analysis over approximately ten years found a lower risk of incident dementia in the speed-training group, with a larger association among participants assigned additional booster sessions. Memory and reasoning training were not associated with significantly lower dementia risk. Dementia was not the trial's original primary outcome, and cases were identified through an algorithm combining study assessments and reported clinical information. [5]

A 2026 study linked surviving ACTIVE participants to Medicare claims for up to 20 years. Assignment to speed training with boosters was associated with a 25% lower hazard of a claims-based Alzheimer's disease or related dementia diagnosis than control; the memory and reasoning arms again showed no significant reduction. Because claims capture recorded diagnoses rather than standardized research adjudication, and because the linked sample excludes participants without suitable Medicare data, this result strengthens but does not settle the causal interpretation. [6]

The specificity is important. ACTIVE tested a particular adaptive visual-processing protocol. Its result cannot be generalized automatically to crossword puzzles, educational activity, working-memory games, or commercial applications with different tasks and training schedules. [6] [7]

What Systematic Reviews Add

A meta-analysis of 52 randomized computerized-training studies involving 4,885 cognitively healthy older adults found a small average improvement in cognitive test performance. Effects differed across domains; processing speed, visuospatial skills, and some memory measures improved, while attention and executive function did not show significant pooled effects. The review did not establish durability or dementia prevention. [1]

A Cochrane review focused on programmes lasting at least 12 weeks found little evidence that computerized training maintains cognition in cognitively healthy people aged 65 and older. Certainty was limited by small trials, inconsistent outcomes, and risk of bias. The contrast with shorter-training meta-analyses illustrates how eligibility criteria and comparison groups can change a pooled conclusion. [2]

Why Control Groups Matter

A no-contact control cannot separate the specific content of training from expectations, social contact, computer exposure, and repeated testing. Active controls better match these experiences, but may themselves train cognitive skills and reduce the apparent contrast. Reviews of brain-training claims identify weak controls, selective outcome reporting, small samples, and inadequate blinding as recurring reasons for overestimating general benefit. [2] [7]

Evidence Quality and Interpretation

Confidence is high that people can improve on practised cognitive tasks and moderate that some programmes produce small improvements on closely related cognitive measures. Confidence is lower for broad transfer to everyday cognition because results vary with the trained domain, outcome, comparator, and follow-up period. [1] [7] [8]

The long follow-up and randomized origins of ACTIVE make its speed-training result unusually informative. However, the dementia analyses were not the original primary test, later follow-up relied on subsets of the randomized cohort, and the finding has not established a class effect for cognitive training. Independent trials designed around clinically adjudicated dementia are needed to test the result directly. [5] [6]

What This Does Not Mean

Practical Interpretation Examples

Related Reading

References

  1. Lampit, A., Hallock, H., & Valenzuela, M. (2014). Computerized cognitive training in cognitively healthy older adults: a systematic review and meta-analysis of effect modifiers. PLOS Medicine. https://doi.org/10.1371/journal.pmed.1001756
  2. Gates, N. J., et al. (2019). Computerised cognitive training for maintaining cognitive function in cognitively healthy people in late life. Cochrane Database of Systematic Reviews. https://pubmed.ncbi.nlm.nih.gov/30864187/
  3. Ball, K., et al. (2002). Effects of cognitive training interventions with older adults: a randomized controlled trial. JAMA. https://doi.org/10.1001/jama.288.18.2271
  4. Rebok, G. W., et al. (2014). Ten-year effects of the Advanced Cognitive Training for Independent and Vital Elderly cognitive training trial on cognition and everyday functioning in older adults. Journal of the American Geriatrics Society. https://pmc.ncbi.nlm.nih.gov/articles/PMC4055506/
  5. Edwards, J. D., et al. (2017). Speed of processing training results in lower risk of dementia. Alzheimer's & Dementia: Translational Research & Clinical Interventions. https://pubmed.ncbi.nlm.nih.gov/29201994/
  6. Coe, N. B., et al. (2026). Impact of cognitive training on claims-based diagnosed dementia over 20 years: evidence from the ACTIVE study. Alzheimer's & Dementia: Translational Research & Clinical Interventions. https://doi.org/10.1002/trc2.70197
  7. Simons, D. J., et al. (2016). Do “brain-training” programs work? Psychological Science in the Public Interest. https://pubmed.ncbi.nlm.nih.gov/27697851/
  8. Kelly, M. E., et al. (2014). The impact of cognitive training and mental stimulation on cognitive and everyday functioning of healthy older adults: a systematic review and meta-analysis. Ageing Research Reviews. https://pubmed.ncbi.nlm.nih.gov/24607830/
Educational Disclaimer

This page summarizes research on structured cognitive training in populations and does not provide an individualized prevention or treatment recommendation. New or progressive cognitive symptoms can have multiple causes and warrant appropriate clinical assessment rather than interpretation through training scores alone.