Skip to main content

Does ‘Brain Training’ Actually Work?

An online citizen science project is recruiting 30,000 volunteers to assess whether exercises to improve memory and attention are valid—and if so, for whom

If there were an app on your phone that could improve your memory, would you try it? Who wouldn’t want a better memory? After all, our recollections are fragile and can be impaired by diseases, injuries, mental health conditions and, most acutely for all of us, aging.

A multibillion-dollar industry for brain training already capitalizes on this perceived need by providing an abundance of apps for phones and tablets that provide mental challenges that are easily accessible and relatively inexpensive.

The three of us, and many others, have provided evidence that carefully formulated exercises can improve basic cognitive skills and even lead to better scores on standard IQ tests. At the same time, brain training has become a profoundly controversial endeavor. Some researchers, including one of us, have expressed deep reservations about both its reliability and its validity. There was even a consensus statement issued calling brain training into question, which, in turn, resulted in a counterresponse from researchers who defended it.


On supporting science journalism

If you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.


In the case of memory training, for instance, study results have been inconsistent, and even meta-analytic approaches that combine data across studies come to differing conclusions. Undoubtedly, there is an enormous amount of hyperbole surrounding the field with many companies exaggerating the potential benefits of using their apps.

The main controversies center around the extent to which the practice of these skills results in actual benefits that are consequential for your daily life. Does recalling an increasing number of digits help you remember to take your medication, do better on a school exam, remember the name of the person whom you met yesterday or even make better life choices?

Some scientists question whether this is even possible. Others argue that we should consider the brain like our muscles, which can be exercised and toned. In this analogy, daily challenges, even demanding ones such as reading a detailed newspaper article or solving an algebra problem, might not be sufficiently challenging to furnish an adequate work-out for the brain.

Just as athletes engage in strength and conditioning by repeatedly exercising certain muscle groups and their respiratory and cardiovascular systems, targeted repetition of memory exercises may be the key to strengthening and conditioning our memory processes. Memory training apps require tracking a large number of objects while one is distracted with a secondary task (such as making mental calculations or navigating the landscape of a game). That degree of difficulty and repetition, however, may be rare in daily life, which is the gap that memory apps aim to fill.

If brain training works, the field holds enormous promise to help people with cognitive impairments, and to aid individuals who are recovering from cancer or perhaps even COVID-19. Some affirmation of the potential for cognitive training could be seen in the FDA’s recent approval of a brain training game to treat ADHD.

Critics, however argue that while the concept is appealing, the overall evidence does not suffice to demonstrate that core brain processes can be truly improved. Despite what many apps and brain training companies will tell their customers, scientists have not uncovered the key ingredients that make an intervention effective, nor the recipes that would best address the diverse needs of those seeking help. Furthermore, most apps available to consumers have not undergone scientific validation at all.

So how do we reconcile the mixed evidence in the field and at the same time cut through the hype? We suggest that some of the confusion may result from the fact that little consideration has been given to who would benefit most from the brain training apps that are supported by research studies. Will it only be those who have some form of memory impairment, or can it also help those eager for self-improvement even though they are already functioning relatively well?

Although the jury is still out, there is evidence that short-term working memory training can provide benefits to relatively high functioning individuals, such as college students. For vision training, there are suggestions that even elite athletes can benefit. Still, whether one has a memory impairment or not, it is likely that, similar to diet or exercise, brain training does not benefit everyone in the same way.

We suggest that much of the debate, and the lack of consensus, revolves around the wrong scientific questions being asked. Specifically, the current paradigm is dominated by population health and research methods devoted towards group averages, when what most of us want to know is whether something is right for us.

To illustrate the problem, consider the hypothetical situation in which one person in 10 gains a profound benefit from a particular memory training app. In the population health model, outcomes are averaged across all individuals who received the intervention, and thus the profound benefit experienced by the few will be washed out by the lack of an effect in the many.

Repeat this experiment across a number of different memory training apps, each potentially providing positive effects for different subgroups, and the collection of benefits experienced by some individuals will be hidden by the inappropriate research methods applied. The population health model, while highly appropriate when its assumptions are upheld, simply does not apply well in a diverse population where different people may interact with the training app in distinct ways and therefore show a range of benefits.

To overcome these limitations, our team is currently leveraging the power of citizen science. Similar to a large-scale study in the United Kingdom (Brain Test Britain, promoted by Cambridge University and the BBC), we are seeking to recruit thousands of participants to help us uncover the potential merits of memory training. But unlike Brain Test Britain’s simple question of whether brain training works, we are looking to engage the U.S. population in a new challenge to test why and for whom brain training works, and under which conditions.

To accomplish our goal, we have launched a new study funded by the National Institutes of Health that aims to recruit 30,000 volunteers to participate in a memory training study that compares multiple approaches to train working memory. The study will use a common set of assessment measures to evaluate potential training gains, and it will focus on individual differences. Anyone older than 18 can join our study and help generate the data required to change the debate and move forward with a new paradigm of precision brain training. If you might be interested in joining our trial, go to the registration site at the University of California, Riverside.

Only by including a very large number of participants and evaluating how different training approaches and their outcomes relate to particular individuals can we address these controversies once and for all. It may be the case that most of the benefits will be found in those who have a condition that impairs their cognitive abilities, or we may find that high-functioning individuals may benefit from training.

We will attempt to solve this riddle by applying statistical models to examine how baseline responses on questionnaires and assessments predict what gains each participant may receive from the different training types. If successful, the study will help us determine what factors may be most revealing of a given individual’s likelihood of gaining from memory training, as well as which form of training may be best for that person.

Our goal is to avoid a one-size-fits-all approach. Instead, we want to advance a new model based upon the premise that people are diverse in their cognitive strengths and needs, and therefore require the type of interventions that would serve them best.