Does Brain Training Work? What Research Shows and What It Doesn't

Search for “does brain training work” and you will find roughly as many articles saying yes as saying no. I make a brain-training app called BrainDojo, so I can’t really sidestep the question. In this post I sort out what the main studies have found and what is still unknown, with links to the sources.

I’ll say this up front: I don’t think the question has a yes-or-no answer. Once you separate what improves from what doesn’t, the research is actually fairly clear.

The real question is transfer

By brain training I mean short tasks in arithmetic, memory, attention and so on, repeated in the hope of improving cognitive function. It comes in many forms: paper workbooks, console games, phone apps.

Researchers are not mainly asking whether you get better at the task you practise. You usually do. The question is whether that improvement spreads to tasks you didn’t practise, or to everyday life, like work or running a household. This is called transfer. Spreading to similar tasks is sometimes called near transfer, and spreading to quite different abilities or to daily life is called far transfer. Most of the argument about brain training is about whether far transfer happens.

In 2014, scientists split into two camps

In 2014, the Stanford Center on Longevity and colleagues published a consensus statement on the brain-training industry, signed by more than 70 scientists. It said brain games do not offer a scientifically grounded way to improve cognitive function or stave off decline. A few months later, 133 scientists and practitioners replied with an open letter arguing that the literature is full of demonstrations of benefit.

How can two groups read the same literature and reach opposite conclusions? The next paper was written to answer exactly that.

What a 2016 review found

In 2016, psychologist Daniel Simons and colleagues published a long review in Psychological Science in the Public Interest titled “Do ‘Brain-Training’ Programs Work?” They took the peer-reviewed studies that brain-training companies cite as evidence and checked each against a set of best practices.

Their summary: there is extensive evidence that training improves performance on the trained tasks, less evidence that it improves closely related tasks, and little evidence that it improves distantly related tasks or everyday cognitive performance. They also found that many studies had major flaws in design or analysis, and that none of the cited studies met all the conditions needed to draw clear conclusions about benefits for everyday life.

Some large studies

An online experiment with more than 10,000 people. In “Putting brain training to the test”, published in Nature in 2010, Adrian Owen and colleagues had 11,430 participants train online several times a week for six weeks on tasks covering reasoning, memory, planning, attention and more. Performance improved on every trained task, but there was no evidence of transfer to untrained tasks, even closely related ones.

The ACTIVE trial. This is a large randomised controlled trial with older adults in the United States. In the first report in JAMA in 2002, 2,832 people aged 65 to 94 received ten sessions of training in memory, reasoning or speed of processing. Each kind of training improved the ability it targeted, and the gains lasted two years. At two years, however, no effect on everyday functioning was detected.

The ten-year follow-up found that the reasoning and speed training effects on the trained abilities were still there after ten years. The memory training effect was not. Each training group also reported less difficulty with everyday activities such as shopping or managing money than the control group did. Two cautions, though. That result is self-reported, and ACTIVE used instructor-led group sessions, which look quite different from a modern app.

The FINGER trial. This two-year trial in Finland enrolled 1,260 people aged 60 to 77 at elevated risk of dementia. One group received a combined programme of diet, exercise, cognitive training and vascular risk monitoring. The other received general health advice. The combined group did slightly better on cognitive tests. But cognitive training was only one part of the package, so the effect of brain training alone can’t be separated out.

In 2017, the US National Academies report Preventing Cognitive Decline and Dementia: A Way Forward described the evidence for cognitive training as encouraging but inconclusive, and not strong enough to justify a public health campaign promoting it.

When marketing went too far

In January 2016, the US Federal Trade Commission announced that the company behind the Lumosity app would pay $2 million to settle charges of deceptive advertising. The FTC said claims that the games improve performance at work and school and slow age-related decline were not backed by science. When I see strong claims about brain training, I try to remember this case.

What we know and what we don’t

Putting it together, here is what we know. You get better at the tasks you practise. Gains on similar tasks are limited. There is no clear evidence of gains in distant abilities or in everyday life.

Plenty is still unknown. What kind of training, how much, and for whom, would make a meaningful difference in daily life? How much weight should we give small signals like the self-reported results in ACTIVE? And how far can a few minutes a day in an app be compared with programmes like ACTIVE or FINGER? Honestly, there are still not many high-quality long-term studies of short app-based training.

If you keep going anyway

I think the most honest way to see brain training is as practice at the task itself. Keep doing arithmetic drills and you will get faster at arithmetic. Current research does not let us say that this will make you better at planning your workday.

The rising scores inside an app are also mostly practice effects. You get used to the format, and the numbers go up. That is fine in itself, but I would not read a rising score as proof that you have become smarter.

Even so, I don’t think it is pointless. Sitting down for a few minutes at a set time and giving one task your full attention can be a useful daily routine, a way to reset your mood or give the day some rhythm. It is not a substitute for exercise, sleep or talking with people, though. The fact that FINGER was a combined programme says a lot about that.

If forgetfulness or trouble concentrating is getting in the way of your life, please talk to a doctor rather than an app. Brain training is not a tool for that.

About BrainDojo

BrainDojo is an iPhone and iPad app I build with all of the above in mind. The app itself states “(This app is not a medical device.)” I don’t claim it treats or prevents anything.

The training consists of short mini-games grouped into lanes such as Calculation Boost, Memory Boost, Attention Boost and Handwriting Training, along with reading aloud and copying text. You answer by handwriting with your finger, speaking, or tapping. The free plan gives you two “Try” games a day, the Brain Dojo Check once a day, and a calendar where you stamp the days you trained. BrainDojo Plus adds Today’s Training and every game. Neither plan has ads.

The Brain Dojo Check is a short test that picks one game from each of three areas: processing speed, inhibitory control and working memory. The games are simplified app versions of cognitive tasks used in research, such as the flanker task and go/no-go, and the result is not a medical diagnosis. Your score will probably rise as you keep taking it, but as I wrote above, I believe much of that rise comes from getting used to the tasks. I think it works best as a rough read on how you are doing that day.

Getting Started with BrainDojo walks through the first session, BrainDojo vs Other Ways to Keep a Daily Brain-Training Habit compares it with other options, and BrainDojo Free vs Plus explains the two plans.