The brain training industry generates billions in annual revenue selling the idea that playing games on your phone for 10 minutes a day will make you smarter, sharper, and protected against cognitive decline. The science is considerably less enthusiastic.
This is the honest version.
The 2014 Stanford Letter
In 2014, a group of 75 neuroscientists and cognitive psychologists published an open letter in the Stanford Center on Longevity newsletter. Their message was blunt: the marketing claims made by brain training companies were not supported by the available evidence. They wrote that while training on specific tasks did improve performance on those tasks, "the companies often make claims about broader improvements in cognitive performance that go well beyond what the science shows."
The letter attracted immediate attention because of who signed it: researchers who had spent careers studying cognition, not lobbyists for competing industries. Their concerns were methodological: studies funded by brain training companies often showed results that independent replications could not reproduce.
The FTC Action on Lumosity
In January 2016, the US Federal Trade Commission fined Lumosity two million dollars for deceptive advertising. The FTC charged that Lumosity's claims, which included preventing memory loss, protecting against Alzheimer's disease, and improving performance in school and work, had no credible scientific backing.
Lumosity settled without admitting guilt but agreed to stop making health claims and to notify subscribers of the action. The case marked a turning point: it established regulatory scrutiny over cognitive training claims and made the industry more cautious in its marketing language.
What the Research Actually Shows
The picture is nuanced rather than simply negative. Brain training research consistently shows:
- Task-specific improvement is real. Practice on a memory training task improves performance on that specific task and closely related variants. This is genuine learning, not noise.
- Transfer to other skills is limited. The same practice rarely improves untrained cognitive tasks at a level that is clinically or practically meaningful. You get better at the game, not at cognition generally.
- Far transfer is largely absent. Improving your working memory span in a digit-span game does not consistently improve your performance at work, in school, or in daily life tasks. This is the claim that matters most, and the evidence is weak.
What DOES Transfer to Better Cognition
The evidence is much stronger for several non-game interventions:
Aerobic exercise
This has the strongest evidence base of anything in cognitive enhancement research. Regular cardio exercise (30 minutes, most days) consistently improves memory, attention, processing speed, and executive function across age groups. The mechanisms include increased neuroplasticity, improved blood flow to the brain, and BDNF (brain-derived neurotrophic factor) release. The effect sizes are meaningful, not marginal.
Sleep
Sleep is when memory consolidation happens. Consistently poor sleep degrades nearly every cognitive measure. People chronically short on sleep perform worse on memory, attention, reaction time, and decision-making tasks, often without realising how impaired they are. Fixing sleep produces cognitive benefits that no app can match.
Learning genuinely new complex skills
Studies on musicians, chess players, and language learners show broader cognitive improvements than brain training games produce. The key factors appear to be genuine novelty, meaningful challenge, and social or real-world engagement. Learning a musical instrument in your 50s shows better cognitive benefits than playing working memory games for the same amount of time.
Social engagement
Cognitively stimulating social interaction, conversation, argument, collaboration, appears to have real benefits. Isolation is consistently associated with cognitive decline in aging adults.
Honest Assessment of Popular Apps
| App | Evidence Quality | What It Is Genuinely Good For |
|---|---|---|
| Lumosity | Weak (FTC action) | Entertainment, specific task performance |
| Elevate | Moderate | Vocabulary, verbal skills, writing quality |
| Duolingo | Strong | Actual language learning with genuine transfer |
| Dual n-back | Mixed | Working memory span (limited transfer) |
Where Cognitive Tests Genuinely Help
Free cognitive tests, including the ones on this site, are genuinely useful for specific purposes that do not require overselling:
- Establishing a baseline. If you do not know your current reaction time, memory span, or processing speed, you have no reference point for tracking changes.
- Tracking over time. Testing weekly over months reveals genuine trends. Are you slower after nights of bad sleep? Faster after you started running? These are real, actionable insights.
- Spotting patterns. Many people discover they perform measurably worse on certain days, after certain foods, or at certain times of day. This kind of self-knowledge is practically useful.
- Fun and friendly competition. Comparing scores with friends or colleagues is genuinely enjoyable and provides motivation to maintain habits that support good performance.
Reaction time, memory, aim, sequence, maths and Stroop. Track weekly. The trends are the insight.
What to Actually Do
If you want to support your cognitive performance based on what actually works:
- Walk 30 minutes daily. Aerobic exercise is the highest-evidence cognitive intervention available.
- Sleep seven to eight hours consistently. No app comes close to the cognitive benefit of fixing your sleep.
- Learn one genuinely hard new skill. An instrument, a language, a craft, something with real challenge and meaningful output.
- Use cognitive tests to track, not to train. They tell you where you are, but they do not reliably take you somewhere better on their own.
Run the reaction time test once a week for a month. See whether your sleep and exercise habits show up in the numbers.