The claim and where it comes from
In 2014, two open letters about brain games reached opposite conclusions. The first, a consensus statement from more than 70 scientists, said that brain games do not offer a scientifically grounded way to improve cognitive functioning or to stave off cognitive decline [1]. A few months later, 133 scientists and practitioners replied that the literature contains many demonstrations of benefits for a wide range of cognitive and everyday activities [1].
That dispute led to a comprehensive review by seven researchers, published in 2016 [1]. It set out best practices for training studies and used them to evaluate every peer-reviewed intervention study cited by leading brain-training companies; the authors suggested that the two camps had partly disagreed because they used different standards of evidence [1].
This page summarises that review and the meta-analyses around it. For a beginner's overview, see What is brain training?. For the narrower question of intelligence, see our review of brain games and IQ.
What does "work" mean?
A programme can work at one level and fail at the next, so it helps to ask five separate questions.
| Level | The question | Example |
|---|---|---|
| Trained task | Do you get better at the exercise itself? | A higher n-back level after weeks of n-back |
| Near transfer | Do you improve on similar tasks you did not practise? | A different working-memory test |
| Far transfer | Do you improve on different abilities? | Reasoning, reading, arithmetic |
| Daily life | Does anything change outside the tests? | Fewer forgotten appointments, better focus at work |
| Durability | Does the gain last after training stops? | Still there months or years later |
Marketing usually promises the last three; the strongest evidence is about the first.
What the reviews and meta-analyses found
The 2016 review. Simons and colleagues found extensive evidence that brain-training interventions improve performance on the trained tasks, less evidence that they improve closely related tasks, and little evidence that they improve distantly related tasks or everyday cognitive performance [1]. Many studies had major shortcomings in design or analysis, and none met all of the best practices the authors set out [1].
Working-memory training. A meta-analysis of 23 studies (30 comparisons) in children and adults, including clinical samples, found reliable short-term improvements in working-memory skills [2]. For verbal working memory, these gains were not sustained at follow-up; for visuospatial working memory, limited evidence suggested they might last [2]. There was no convincing evidence that the training generalised to verbal or nonverbal ability, inhibition, word decoding or arithmetic [2].
Across kinds of training. Drawing on meta-analyses of working-memory training, video games, music and chess, several of them their own, Sala and Gobet concluded that the effect on domain-general cognitive skills is minimal, and that the variation between studies is accounted for by design quality and statistical artefacts [3]. A second-order meta-analysis by the same group (a meta-analysis of meta-analyses) found that working-memory training does produce near transfer to memory measures, while far-transfer effects were small or null [4]. When placebo effects and publication bias were controlled for, the far-transfer effect was zero, whatever the population or type of training [4].
Older adults. A meta-analysis of 52 randomised trials (4,885 participants) of computerised training in cognitively healthy older adults found a small overall effect (g = 0.22) [5]. Effects were small to moderate for nonverbal memory, working memory, processing speed and visuospatial skills, very small for verbal memory (g = 0.08), and not significant for executive functions or attention [5]. Home-based training was ineffective compared with group-based training, more than three sessions a week was ineffective compared with three or fewer, and there was no evidence that working-memory training by itself was effective [5]. The authors note that the results are limited to healthy older adults and do not address how long the effects last [5].
The ACTIVE trial
ACTIVE is one of the largest randomised trials of cognitive training. It enrolled 2,832 independent-living volunteers aged 65 to 94 in six US cities and assigned them to ten group sessions of memory, reasoning or speed-of-processing training, or to a no-contact control group [6]. Memory training taught mnemonic strategies for word lists and stories, reasoning training taught strategies for spotting patterns, and speed training used increasingly complex tasks on a computer [6].
After two years, each kind of training had improved the ability it targeted, and booster sessions added to the reasoning and speed gains [6]. No training effects on everyday functioning were detected; the authors noted that daily function had barely declined in any group, so a longer follow-up was needed to test it [6].
After ten years, reasoning and speed training still showed effects on the abilities they had targeted (effect sizes 0.23 and 0.66), but memory training no longer did [7]. Each trained group reported less difficulty with instrumental activities of daily living, such as housework, finances and shopping, than the control group; about 60% of trained participants, against 50% of controls, were at or above their starting level of self-reported function [7].
The ten-year result is often quoted, so its limits matter:
- It is self-reported. Participants rated their own difficulty with daily tasks [7]. The trial's performance-based measures of everyday problem solving and everyday speed showed no effect of training [7].
- It is less decline, not improvement. Self-rated function declined in all groups as people aged; the trained groups declined less [7].
- The control group did nothing. A no-contact group is weaker than an active control, and cannot rule out differences in expectations between groups [8] (see below).
- Many people dropped out. Only 44% of participants were still in the study at ten years, although dropout was similar across groups [7].
- The setting was specific. Participants were independent-living US volunteers, recruited in 1998–1999 and trained in group sessions, not people using an app at home [6].
ACTIVE shows that targeted training can produce lasting gains in some trained abilities. Whether it changed daily life remains uncertain.
Why results conflict
If reviews broadly agree, why do individual studies keep reporting wider benefits? Four reasons recur.
Control groups. Retaking a cognitive test tends to raise the score even without training [9]. In a randomised study of 102 healthy young adults, the Stroop-training group, an active control group and a no-training group all improved substantially from pre-test to post-test on every transfer task, and the Stroop training added no transfer of its own [10]. Without the control groups, that general improvement could have been mistaken for a training effect. For n-back training and for music instruction, Sala and Gobet report that much of the variation between studies was explained by whether the control group was active or passive, with effects close to zero against active controls [3].
Expectations. If trained participants expect to improve and the control group does not, any difference between them could come from those expectations rather than from the training. An active control group helps, but only when its members expect as much improvement as the trained group, and almost no psychological intervention studies check that they do [8].
Publication bias. When studies with non-significant results go unpublished, the average effect in a literature is inflated [3]. Corrections for this bias matter: in the second-order meta-analysis, far transfer disappeared once publication bias and placebo effects were accounted for [4]. Not every analysis finds the problem, though; the older-adult meta-analysis found no systematic evidence of publication bias [5].
Study quality. The 2016 review found major shortcomings in design or analysis in many training studies [1]. Summarising that review, Sala and Gobet point to passive control groups, small samples and selective reporting as sources of inflated effects, and note that better-designed studies find smaller ones [3].
What is still uncertain
- Daily life. ACTIVE's self-reported benefit was not matched by its performance-based everyday measures [7], and almost no intervention studies match expectations between groups [8].
- Delivery and dose. The findings about group versus home training and session frequency in older adults come from moderator analyses that compare different studies [5], not from trials that tested these choices directly.
- Durability. ACTIVE suggests that some trained-ability gains can last years, while in the working-memory meta-analysis verbal gains were not sustained at follow-up.
- Specific groups. This page is about cognitive training in general, mostly in healthy people. Whether training helps people with particular medical conditions is a separate clinical question.
- Specific products. Results for one programme do not carry over automatically to another, so a study of one product says little about a different app.
What the research says
What this means if you are choosing an app
- Expect to get better at the app's games. That is the effect the evidence supports.
- Read claims level by level. A claim about game scores is not a claim about your memory at work. Be sceptical of promises about everyday life, intelligence or protection against ageing.
- Ask whether the app itself was tested, and against what: an active control group with similar expectations is much stronger evidence than a group that did nothing.
- Practise the real skill when you can. If you want to remember what you study or focus at work, practise that directly.
- Track your own outcome. Our guide on how to measure progress in brain training shows how to separate practice gains from real change.
What this means at LaserMind
We grade computerised cognitive training E2 (conditional evidence) for "some domains, mostly older adults; app-specific", and we apply the same limits to our own exercises: they are based on standard research tasks but have not been tested in trials of their own. Your results are compared only with your own history. A check-in with untrained tasks every four weeks, and a report that keeps trained tasks, untrained tasks and real life apart, are there so that you can see for yourself which level, if any, has changed. How we review evidence explains our grades.
Try it on LaserMind
▶ N-BackSee the trained-task effect first-hand: track your n-back level over a few weeks.
Try it on LaserMind
▶ Number SpanA classic span task. Practice raises your span here; whether it reaches daily life is the open question.
Frequently asked questions
So is brain training useless? No. It reliably makes you better at the tasks you practise, which can be satisfying and useful in itself. The evidence simply does not show that those gains spread to everyday memory, attention or general ability.
Why do some apps cite studies showing benefits? The 2016 review examined exactly those studies, the peer-reviewed intervention studies cited on the websites of leading brain-training companies, and found that many had major shortcomings in design or analysis [1].
What the research says
Last evidence check 2026-10-11In healthy older adults, computerized training had small effects on memory, processing speed and visuospatial skills, but not on executive functions or attention; supervised group sessions worked, home-based training showed no effect. Trained & similar tasks [5]
In the ACTIVE trial, memory, reasoning and speed training each improved the ability that was trained, but no training effects on everyday functioning were detected after two years. Trained & similar tasks [6]
A major review found strong evidence that training improves the trained tasks, less for closely related tasks, and little evidence that it improves everyday cognitive performance. Daily life [1]
References
- Simons, D. J., Boot, W. R., Charness, N., Gathercole, S. E., Chabris, C. F., Hambrick, D. Z., & Stine-Morrow, E. A. L. (2016). Do "brain-training" programs work? Psychological Science in the Public Interest, 17(3), 103–186. https://doi.org/10.1177/1529100616661983
- Melby-Lervåg, M., & Hulme, C. (2013). Is working memory training effective? A meta-analytic review. Developmental Psychology, 49(2), 270–291. https://doi.org/10.1037/a0028228
- Sala, G., & Gobet, F. (2019). Cognitive training does not enhance general cognition. Trends in Cognitive Sciences, 23(1), 9–20. https://doi.org/10.1016/j.tics.2018.10.004
- Sala, G., Aksayli, N. D., Tatlidil, K. S., Tatsumi, T., Gondo, Y., & Gobet, F. (2019). Near and Far Transfer in Cognitive Training: A Second-Order Meta-Analysis. Collabra: Psychology, 5(1), 18. https://doi.org/10.1525/collabra.203
- 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, 11(11), e1001756. https://doi.org/10.1371/journal.pmed.1001756
- Ball, K., Berch, D. B., Helmers, K. F., et al. (2002). Effects of cognitive training interventions with older adults: A randomized controlled trial. JAMA, 288(18), 2271–2281. https://doi.org/10.1001/jama.288.18.2271
- Rebok, G. W., Ball, K., Guey, L. T., 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, 62(1), 16–24. https://doi.org/10.1111/jgs.12607
- Boot, W. R., Simons, D. J., Stothart, C., & Stutts, C. (2013). The Pervasive Problem With Placebos in Psychology: Why Active Control Groups Are Not Sufficient to Rule Out Placebo Effects. Perspectives on Psychological Science, 8(4), 445–454. https://doi.org/10.1177/1745691613491271
- Hausknecht, J. P., Halpert, J. A., Di Paolo, N. T., & Moriarty Gerrard, M. O. (2007). Retesting in selection: A meta-analysis of coaching and practice effects for tests of cognitive ability. Journal of Applied Psychology, 92(2), 373–385. https://doi.org/10.1037/0021-9010.92.2.373
- Talanow, T., & Ettinger, U. (2018). Effects of task repetition but no transfer of inhibitory control training in healthy adults. Acta Psychologica, 187, 37–53. https://doi.org/10.1016/j.actpsy.2018.04.016