steady

Memory and learning · Keeping the brain plastic

Delete the brain-training app

Evidence: solid

Do this

If the goal is a sharper head, swap the app time for a real skill. The app makes you better at the app.

How to do it

  1. Take the app minutes and put them into one real skill. Learning digital photography improved memory in older adults; easy tasks didn't.
  2. Pick something you're bad at. Hard and new is what showed an effect, not familiar and pleasant.
  3. Walk, and sleep. Both have better evidence for memory than any game.
  4. Enjoy the app if you like it, as a game. Just don't count it as training.

Where the gain stopsTap a word for what it means

  1. Practicetunes that one task

    Repeat a game and you get better at that game. In the biggest trial, every trained task improved.

  2. Working memorygains, then fades

    Training it lifted working-memory scores in the short term. The verbal gains didn't last to follow-up.

  3. Transfermostly stops there

    The gains rarely spread to other tasks, and there's little evidence they reach daily life.

  4. Youbetter at the app

    A real skill, exercise and sleep have better evidence for a sharper head. None of them is a sure thing either.

Why it works

Brain-training games make you better at brain-training games. In the biggest trial, over eleven thousand people trained for six weeks and improved on the tasks they practised and nothing else. A review of more than a hundred studies found the same: little or no transfer to memory, attention or daily life. Evidence: solid

What has shown some transfer: learning a demanding real skill, aerobic exercise, sleep. Time with people is good for other things, but in the best trial it didn't move memory. Evidence: decent All of them are harder than an app. That's not a coincidence. Evidence: worth trying

The long readEvery step of the chain, each study with its numbers, and what's still unproven. 3 min read.

What happens in your brain

Practice tunes the task you practise. Repeat a game and you get better at it. In the biggest trial, every trained task improved. Evidence: solid

Working memory gets a short bump. Working memory is what holds a few things in mind while you use them. Across 23 studies, training it improved working-memory scores in the short term. For verbal working memory the gain didn't last to follow-up; for visual-spatial there was limited evidence it might. Evidence: solid Those studies mixed children and adults, some with ADHD or other conditions, so not all of it is about a healthy adult with an app. Evidence: solid

Then the gain mostly stops. The same review found no convincing sign that working-memory training spread to reasoning, attention, reading or arithmetic. Evidence: solid The large review of brain-training studies found plenty of evidence for the trained tasks, less for closely related ones, and little for distant tasks or everyday thinking. Evidence: solid

What the brain seems to need is harder. In a 14-week trial, older adults spent about sixteen hours a week learning digital photography, quilting or both. Their memory improved more than that of people who did social activities or easy familiar tasks: clearly for photography, not for quilting on its own. Evidence: decent That's the case for a real skill over an app.

The proof

  • 11,430 adults aged 18 to 60 trained online for six weeks, at least ten minutes a day, three times a week, on tasks for reasoning, memory, planning and attention. They improved on every task they trained. On untrained tests, even closely related ones, they did no better than a control group who answered general-knowledge questions online. Evidence: decent More than 52,000 signed up; the results come from those who finished, which is one limit. Evidence: decent
  • A 2016 review looked at the 132 papers brain-training supporters listed as evidence, plus papers cited by the main companies. It found extensive evidence for gains on trained tasks, less for related tasks, and little for distant tasks or everyday life. None of the studies met all the design standards the reviewers set. Evidence: solid
  • A review of 23 studies of working-memory training found short-term gains in working memory itself and no convincing spread to other skills. Evidence: solid
  • In a trial of 2,802 healthy older adults followed for ten years, speed-of-processing training went with a lower rate of dementia than no training. Memory and reasoning training didn't. The speed result only just cleared the usual bar for chance. Evidence: worth trying

How to do it

  • Take the app minutes and put them into one real skill. Evidence: worth trying In the trial that showed it, the dose was about sixteen hours a week for fourteen weeks. Evidence: decent
  • Pick something you're bad at. Hard and new is what improved memory; social alone and familiar alone didn't. Evidence: decent
  • Walk, and sleep. Walking grew the hippocampus of older adults in a year-long trial. Evidence: decent
  • Enjoy the app if you like it, as a game. Just don't count it as training.

Watch out

The app does make you better at the app. That part is real. Evidence: solid

You'll read that one kind of training cuts dementia risk. That comes from the ten-year trial above: only speed training showed it, not memory or reasoning training, the result was borderline, and it needs repeating before anyone should rely on it. Evidence: worth trying

In 2016 Lumosity paid 2 million dollars to settle US Federal Trade Commission charges that its ads claimed, without the science to back them, that its games improved everyday performance and delayed age-related decline. Evidence: solid

That brain games build cognitive reserve hasn't been shown. Evidence: decent

Arrows to move, Enter to open, Esc to close.