Does brain training work?

Does brain training improve your thinking?

It improves the thing you trained. A 2016 meta-analytic review covering 87 publications and 145 experimental comparisons found large gains on tasks similar or identical to the training, and no convincing evidence of improvement on intelligence, verbal ability, word decoding, reading comprehension or arithmetic.

Melby-Lervåg, Redick and Hulme published this in Perspectives on Psychological Science, and it remains the clearest statement of the position. The distinction they draw is between near transfer, improvement on tasks resembling what you practised, and far transfer, improvement on the general capacities the products advertise.

Near transfer is solid. Far transfer is where the field has repeatedly failed to deliver, and the failure holds specifically when training is compared against an active control group doing something else engaging, rather than against people doing nothing.

That control group detail is the crux. Comparing training against an untreated group measures the effect of doing an interesting task and being paid attention to. Several of the field's more encouraging results shrink or vanish once the comparison is fair.

Why do the studies disagree?

The authors name the reasons directly: small samples inflating effect estimates, publication bias favouring positive results, untreated control groups overstating benefits, and an absence of analyses linking the trained gains to any broader outcome.

A field with these features produces a steady supply of encouraging individual studies and no accumulating picture. That is what has happened here.

Worth adding that the size of a person's improvement on the trained task did not predict how much they improved elsewhere. If training were building a general capacity, you would expect those to move together.

So is brain training worthless?

No, and the honest framing matters. If you enjoy it, it is a pleasant cognitive activity and you will genuinely get better at it. What the evidence does not support is buying it expecting a sharper mind in the rest of your life.

Near transfer has real uses. Training that resembles the actual task you care about can help with that task. The problem is the advertised promise of general improvement.

If the goal is your thinking across the whole of life, the evidence points elsewhere: sleep, physical activity, social connection, and continuing to learn genuinely difficult things. Learning an instrument or a language is not brain training, and it is a real cognitive demand with the ordinary rewards of being able to do something new.