Does brain training actually work?

Two questions that keep getting muddled together

Almost every argument about brain training is really two different arguments happening at once. Separating them clears up most of the confusion.

Near transfer: getting better at the thing you practised

If you play 2048 for a week, you will get better at 2048. You will spot merges faster, keep the big tile in a corner without thinking about it, and stop making the move that traps your board. This is not disputed by anyone. Practising a task improves performance on that task, and usually on very similar tasks too.

Far transfer: getting better at everything else

The commercial claim is bolder. It says that practising a puzzle will sharpen some general capacity — attention, reasoning, "mental sharpness" — which then pays off in places you never practised: schoolwork, your job, driving, remembering where you put your keys.

This is the claim that researchers have spent fifteen years arguing about, and it is where the evidence gets thin.

The gym analogy that actually holds up

Going to the gym makes you measurably better at lifting weights. It does not automatically make you a better swimmer, even though both use the same muscles. The strength is real; the transfer to an untrained activity is limited and has to be earned separately.

Cognitive training works the same way. Getting better at remembering sequences of flashing squares makes you better at remembering sequences of flashing squares. Whether it makes you better at following a lecture is a separate question with a much less encouraging answer.

What the large reviews concluded

When researchers pool the results of many training studies — rather than cherry-picking the ones that worked — a consistent pattern shows up. People improve substantially on the trained tasks. Improvements on untrained tasks are small, inconsistent, and often disappear once you account for practice effects and the fact that participants who expect to improve tend to try harder on the tests.

This is not a fringe position. In 2014, more than seventy cognitive scientists and neuroscientists signed an open letter stating that the evidence for broad, lasting transfer from brain-training games was weak, and that the strongest marketing claims were running well ahead of the data.

Regulators have said much the same thing. The US Federal Trade Commission reached a multi-million dollar settlement with the maker of one of the best-known brain-training products over claims that its games would improve everyday cognitive performance and protect against cognitive decline. Separately, the US FDA has cleared a prescribed video game as an adjunct treatment for attention difficulties in children — but that is a narrowly defined medical use, evaluated in specific patients, and not evidence that casual puzzle play makes healthy adults sharper.

What does seem to support long-term thinking ability

The list here is unglamorous, which is probably why it gets less marketing budget than puzzle apps:

• Regular aerobic exercise. Among the best-supported findings in the whole field.
• Sleep. Both the amount and the consistency. Memory consolidation happens while you sleep, so a bad night costs you more than the hours lost.
• Cardiovascular health. What is good for your heart tends to be good for your brain, over decades.
• Learning genuinely complex new skills. A new language, a musical instrument, a demanding hobby — things with a long learning curve and constant feedback.
• Education and social contact. Both consistently associated with better cognitive outcomes later in life.

Being honest about the strength of this evidence matters too: most of it comes from observational studies and correlations, not from randomised trials that run for forty years, because nobody can run those. "Associated with" is not the same as "causes."

So why play puzzles at all?

Because they are a genuinely good way to spend five minutes. That is not a consolation prize.

A short puzzle is a clean break: it has a beginning and an end, it fully occupies your attention so the thing that was stressing you gets displaced for a few minutes, and it gives you a tiny sense of completion that a scrolling session never does. Plenty of people use one as the punctuation mark between two blocks of work.

What puzzles are not is a shortcut to general intelligence, and you should be suspicious of anyone selling one.

A quick test for spotting overblown claims

• Does the claim name a specific benefit, or just "brain health" and "sharpness"?
• Is the benefit measured on the game itself, or on something genuinely different from the game?
• Does the study have a control group that did something else equally engaging?
• Does the improvement last, or is it measured the same day as the training ended?
• If a subscription costs money, what exactly is being promised for it?

Try it yourself

Five minutes, one puzzle, then back to whatever you were doing. Keep a streak if it helps you come back tomorrow — that habit is worth more than any transfer effect.

Play Flow Free  →

Editorial note: This article is for general educational purposes. Cognitive effects vary between individuals, and claims about specific benefits should be based on evidence from well-designed studies.