How to choose a brain-training app

Start with what you actually want

The right app depends on what you expect from it, so settle that first. If you want a broad promise — a sharper mind, a younger brain, protection from decline — no app can honestly give it to you, and the evidence guide explains why. If you want a short daily challenge, a repeatable measure of a few specific skills, and a habit with some structure, several apps can, and the rest of this guide is about telling them apart.

A useful rule: the more an app promises, the less it has probably thought about what its numbers mean.

Does it say where each task comes from?

Nearly every brain-training game descends from a laboratory task — Stroop, n-back, Corsi block-tapping, go/no-go, the flanker task. An app that names the origin is telling you two things: that it knows what the task was designed to measure, and that it is not pretending to have invented cognition. An app that dresses the same task in a cartoon and gives it a proprietary name is not necessarily worse as a game, but it has chosen to hide the one piece of information that lets you judge what you are training.

Look for a brief before each game that says what the task is, what it descends from, and what skill it exercises in a sentence you could repeat to someone.

A fixed benchmark or a moving one?

This is the question most apps hope you will not ask. Almost all of them adapt: the difficulty rises as you improve, to keep you working near the edge of what you can do. That is exactly right for training. It is useless for measurement, because a score from an adaptive task is a score at whatever difficulty the engine had chosen that day. If your number is flat while the levels climb, you have improved; if the number climbs, you might have improved or the engine might have eased off.

Ask how the benchmark is separated from training, how often it runs, and whether the result feeds back into the difficulty. The good answer is: separate, regular, and no.

Does it invent a brain age, percentile or ranking?

A percentile needs a representative population. A brain age needs a validated mapping from scores to age that holds for people who are not the app’s own users. A leaderboard needs everyone to be playing under comparable conditions. None of these usually exist, and an app that shows them anyway has decided that a motivating number matters more than a true one.

The honest alternative is to compare you only with yourself, and to say plainly that this is what is being done. That is less exciting on the first day and much more useful in the sixth month. Be especially wary of anything phrased to sound like a medical assessment; a game is not a cognitive screen, and if you are worried about your memory or thinking the right next step is a doctor.

Does it work offline, and does it need an account?

These two questions are related, because an app that needs a server usually wants an account to go with it. Ask:

• Can I play with no connection at all? An app that trains attention should not fail on a plane or in a basement.
• Is an account required to start? An email address is a small thing to give up, but it is the beginning of a relationship you did not ask for, and it usually means your play history lives somewhere other than your phone.
• Where is my data, and can I get it out? A local database you can export is the strongest answer. A cloud profile you cannot download is the weakest.
• Are there ads or trackers? Check the privacy label on the store listing, not the marketing page.

Accessibility

Brain-training games lean heavily on colour, speed and animation, which makes them unusually hostile to some players unless the developer has thought about it.

• Colour. Any game that asks you to tell two colours apart should offer a palette chosen to stay distinct for the common forms of colour-vision deficiency, and switching to it should not reset your progress.
• Motion. The app should honour the system Reduce Motion setting, and ideally offer a mode without timers for people for whom the clock is the problem rather than the task.
• Language. If you are not training in your first language, check that the whole app is translated — game rules and achievements, not just the menus — and that changing language keeps your history.
• Input. Look for games that use a range of interactions rather than tapping alone; it matters for coordination and for people with limited fine motor control.

How is the free tier limited?

Almost every app is free to install. The question is what free actually gets you, and the honest answers are easy to recognise.

• By content, not by time. A fixed set of games that stays free forever is a real free tier. A seven-day “trial” that then locks everything is a paid app with a delay.
• Enough to judge the app. If the free games are all from one category, you cannot tell whether the rest is any good. One free game in each skill area is a fair test.
• The same engine, the same measurements. A free tier that shows a different, cruder progress view is a demo, not a tier.

Subscription honesty

If you do pay, the price is the least of it. Check whether the subscription is sold through the app store, so it can be cancelled from your phone’s settings rather than through a support form; whether local pricing is shown before you commit; whether there is a free trial that converts automatically, and if so whether the app tells you when; and whether cancelling deletes your history or simply returns you to the free tier. An app that keeps your data and your free games after you cancel is behaving well. One that holds them hostage is not.

The checklist

1. Does every game say which task it descends from and what it measures?
2. Is there a fixed-difficulty benchmark, separate from the adaptive training, that you can re-run and compare?
3. Does it refuse to show a brain age, percentile or ranking it cannot justify?
4. Does it work fully offline?
5. Can you use it without an account, and export your data?
6. Are there no ads and no trackers, according to the store privacy label?
7. Is there a colour-safe palette, Reduce Motion support and full translation?
8. Is the free tier limited by content, permanently, with something free in every skill?
9. Is the subscription sold through the store and cancellable without losing your history?

Braintide, whose site hosts these guides, was built to answer yes to each of these, and you can check every one against its landing page before installing. Whatever you choose, how you use it matters more than which one it is.

Frequently asked questions

What is the best brain training app?

The one that is honest about what it measures. Look for named task origins, a fixed-difficulty benchmark separate from adaptive training, no invented brain age or percentile, offline use, no required account, real accessibility options and a free tier limited by content rather than time. Any app that meets those is a reasonable choice; the differences after that are taste.

Is there a brain training app without a subscription?

Most apps use subscriptions, but the fairer ones offer a permanent free tier with a real selection of games rather than a time-limited trial. Check whether the free games cover every skill area and whether they stay free forever before deciding whether the paid tier is worth it to you.

Are free brain training apps any good?

A free tier can be perfectly good if it is limited by content rather than by time and gives you a game in each skill. The thing to avoid is a free app funded by advertising and tracking, which is a bad fit for something meant to train attention.

Do I need an account for a brain training app?

You should not. Everything a brain-training app does can run on the device, and the best ones keep your history in a local database you can export. An account is a sign that your data is going somewhere else.

Not a medical tool: Braintide is a game, not a cognitive assessment or a treatment. These guides summarise published research for general interest and are not medical advice. If you are worried about your memory or thinking, talk to a doctor.