Does brain training work? What the research shows and what we promise

September 26, 2026 · 9 min read · Tyamka team

What studies from 2008 to 2017 found about brain games, why getting better at a game is not getting better at everyday tasks, and what Tyamka does not claim.

You get better at the games you practise, and the evidence that this spreads to other tasks or to everyday life is weak. That is the conclusion of major reviews of the research, and it is the reason Tyamka makes no promises about memory or attention. This article goes through what the studies measured, where researchers disagree, and how to read claims made by any brain training app, ours included. If you want to see a practice effect for yourself, play a few rounds of the Stroop test and watch your points change from one round to the next.

What "works" means

A brain game can "work" in three different ways, and most arguments about brain training come from mixing them up.

• You get better at the game itself. This almost always happens. You learn the rules, find a strategy, get used to the screen and stop wasting time.
• You get better at similar tasks. Researchers call this near transfer. For example, after weeks of one memory task you might also do better on another memory task you never practised.
• You get better at different tasks and at daily life. This is far transfer. Reading faster, forgetting fewer things at work, doing better at school.

The first one is not in doubt. The debate is about the second and especially the third, because that is what most advertising for brain training has promised.

A large online study, 2010

In 2010 Adrian Owen and colleagues published a study in Nature called "Putting brain training to the test". 11,430 people trained online several times a week for six weeks on tasks meant to exercise reasoning, memory, planning, visuospatial skills and attention. Participants improved on every task they trained. The researchers found no evidence of transfer to untrained tasks, even to tasks closely related to the trained ones.

The study has limits. People trained on their own at home, and not everyone trained the same amount. Still, it was large, and it asked the right question. The question was not whether people improve at what they practise, but whether that improvement shows up anywhere else.

Two open letters, 2014

In 2014 two groups of scientists published open letters with opposite conclusions. The first, a consensus statement from an international group of more than 70 scientists, said brain games do not offer a scientifically grounded way to improve cognition or to stave off decline. Several months later a group of 133 scientists and practitioners replied that the research contains many demonstrations of benefits. Both letters are described in the review below, which was written partly to settle the disagreement.

The big review, 2016

In 2016 Daniel Simons, Walter Boot, Neil Charness, Susan Gathercole, Christopher Chabris, David Hambrick and Elizabeth Stine-Morrow published "Do 'Brain-Training' Programs Work?" in Psychological Science in the Public Interest. It runs to more than 80 pages. The authors took every published intervention study that brain training companies cited as support for their products and checked each one against a set of best practices, such as an active control group, several measures of each ability and clear reporting.

Their findings, in their own terms:

• extensive evidence that training improves performance on the trained tasks;
• less evidence that it improves performance on closely related tasks;
• little evidence that it improves performance on distantly related tasks or everyday cognitive performance.

They also found that many of the studies had major problems in design or analysis, and that none of the cited studies followed all of the best practices they considered essential. That does not prove brain training can never transfer. It means the studies companies relied on did not show it.

Working memory training

Working memory training got its own wave of attention. In 2008 Susanne Jaeggi and colleagues reported in PNAS that young adults who trained on a demanding n-back task, where you say whether an item matches the one shown a set number of steps earlier, also improved on tests of fluid intelligence. The paper was titled "Improving fluid intelligence with training on working memory", and it suggested that more training brought bigger gains.

Later studies tried to repeat this with stricter designs. In 2013 Thomas Redick and colleagues gave young adults 20 sessions of adaptive dual n-back and compared them with a group that trained on visual search (an active placebo) and a group that did nothing. Both training groups got better at their own task. Neither showed transfer to any of the tests of intelligence, multitasking, working memory capacity or perceptual speed.

Meta-analyses, which pool many studies, point the same way. Monica Melby-Lervåg and Charles Hulme (2013) combined 23 studies and found reliable short-term gains in working memory itself, but no convincing evidence that they generalised to other skills such as verbal and nonverbal ability, attention or arithmetic. Giovanni Sala and Fernand Gobet (2017) summarised three meta-analyses of chess, music and working memory training and found that the effects got smaller as the study designs got better, especially when there was an active control group. They concluded that far transfer rarely occurs.

For background on the working memory model these studies refer to, see what working memory is.

Research by the companies

Brain training companies publish studies of their own, and it would be unfair to skip them. The best known is a 2015 trial in PLoS ONE by Joseph Hardy and colleagues. 4,715 people who completed the study trained on a set of Lumosity exercises about 15 minutes a day, at least five days a week, for 10 weeks, while an active control group did crossword puzzles. The training group improved more on an aggregate of cognitive assessments, with a small effect (Cohen's d of about 0.26).

The paper states that most of the authors worked for Lumos Labs, the company behind Lumosity, and held stock options in it. Lumosity's own website, as of 26.09.2026, describes the study, says that more research is needed to connect better assessment scores with everyday tasks, and notes that both groups lost about half of their participants before the end. That is a fair summary. A small gain on tests that resemble the training is a different claim from a gain in daily life.

The FTC case against Lumosity, 2016

In January 2016 the US Federal Trade Commission announced a settlement with Lumos Labs. The FTC alleged that the company had advertised, without scientific support, that its games would help people perform better at work, at school and in sport, and would delay age-related decline and other health conditions. Lumos Labs agreed to pay $2 million. A $50 million judgment was suspended because of the company's financial condition. The order required competent and reliable scientific evidence before any future claims about real-world performance or health, and it required the company to let subscribers who had signed up for auto-renewal between 2009 and 2014 cancel easily.

This settlement did not rule on whether any single game has any effect. It ruled on advertising. The lesson for anyone making brain games is simple. You do not say the game improves your life unless you can prove it.

Why being better at the game is not the same

When your time in a game drops, several things are going on, and most of them belong to the game.

• You learn the task. You stop reading the rules and start acting on them. In the Schulte table you learn where to look first. In the Stroop test practice makes the delay between word and ink colour smaller, though in Stroop's own 1935 experiments it did not go away (more in the Stroop effect explained).
• You find a strategy. Grouping digits into chunks lets you hold more of them. That is a real skill, but it is a skill for that kind of task.
• You get used to the device. Where the buttons are, how the screen responds, how long the pause before a signal tends to be.
• You are in a different state. Sleep, time of day, noise and mood all change a single result.

Each of these can make your score go up without anything changing outside the game. That is why a rising best score is good news about the game and says little about anything else.

What Tyamka promises

Our games page and FAQ say the same thing in plain words. Points and your best show how you did in a game, not what you are capable of. We do not claim that the games improve memory or attention outside them.

We make the games because they are short and fun, and because trying to beat your own best is a good reason to come back. Every level compares you only with your previous result, not with other players and not with a norm. If you play for ten minutes on the bus instead of scrolling, that is a fine reason on its own, and it needs no promise from us.

Our article on the Schulte table and what it trains applies the same view to one game.

How to read claims from any brain training app

Whatever you use, a few questions help sort careful claims from marketing.

1. What exactly improved? Better at the game, at a similar test or at daily life. Each needs different evidence.
2. Was there an active control group? A group that did something else engaging, like crosswords or visual search, not a group that did nothing.
3. Who ran the study? Company research can be sound, but a conflict of interest should be stated, as it was in the Hardy paper.
4. How many people dropped out? If half the participants leave, the ones who stay may differ from the ones who leave.
5. Was daily life measured, or only tests? Self-reports and lab tests are not the same as doing better at work.

If an app promises better memory, sharper focus at work or protection from decline, ask where it shows that. If it only says the games are fun and track your results, it is making a claim it can keep.

Sources

• Owen A. M. et al. (2010). Putting brain training to the test. Nature, 465, 775–778.
• Jaeggi S. M. et al. (2008). Improving fluid intelligence with training on working memory. PNAS, 105(19), 6829–6833.
• Redick T. S. et al. (2013). No evidence of intelligence improvement after working memory training. Journal of Experimental Psychology: General, 142(2), 359–379.
• Melby-Lervåg M., Hulme C. (2013). Is working memory training effective? A meta-analytic review. Developmental Psychology, 49(2), 270–291.
• Hardy J. L. et al. (2015). Enhancing cognitive abilities with comprehensive training. PLoS ONE, 10(9), e0134467.
• Simons D. J. et al. (2016). Do "brain-training" programs work? Psychological Science in the Public Interest, 17(3), 103–186.
• Sala G., Gobet F. (2017). Does far transfer exist? Current Directions in Psychological Science, 26(6), 515–520.
• US Federal Trade Commission (2016). Lumosity to pay $2 million to settle FTC deceptive advertising charges.

If you want to see how other free brain game sites compare on price and ads, read brain games without a subscription. All our games are on the all games page.