Websites such as make some bold promises about the effectiveness of computer-based brain-training programs. The site claims:

“Harness your brain’s neuroplasticity and train your way to a brighter life”
“Your brain’s abilities are unique. That’s why your Personalized Training Program adapts to fit your brain and your life goals.”
“Just 10 hours of Lumosity training can create drastic improvements. Track your own amazing progress with our sophisticated tools.”

Wow – in just 10 hours I can become smarter by playing fun video games personalized to my brain. I’m a huge fan of video games, and I would love to justify this hobby by saying that I’m training my brain while I play, but what does the scientific evidence have to say about such claims?

Not surprisingly, the published evidence is complex and mixed.

Before I summarize that evidence, let me describe the variables with which brain-training research must contend. First there are various target populations who likely will not respond in the same way to brain-training interventions. These include: healthy children, healthy young adults, healthy older adults, children with some form of cognitive impairment or developmental delay, adults with traumatic brain injury, older adults with mild cognitive impairment, and older adults with Alzheimer’s disease or other forms of dementia.

Most studies do indeed pick a target population or two on which to focus. Each of these populations need to be considered separately when reviewing the literature.

The second important variable is the brain function that is being evaluated. There is no single measure of brain function or intelligence. Studies typically identify the following distinct functions:

Memory is the ability to encode, store, and recall information. Memory can be further divided into recognition, recall, verbal, visual, episodic, and working memory. Each type of memory has specific tasks associated with that memory function.

Attention is the ability to focus one’s perception on target visual or auditory stimuli and filter out unwanted distractions.

Executive function is ability to strategically plan one’s actions, abstraction, and cognitive flexibility – the ability to change strategy as needed. A classic test for executive function is trail-making, drawing a line from A-1-B-2, etc., which requires quickly switching from numbers to letters and back again.

Reaction time and processing speed are related functions that deal with how quickly someone can react to stimuli and process information, respectively.

Another very important variable in brain-training studies is generalizability – to what extend does training in one specific task increase performance on other tasks, and how far from the trained task does the effect extend? For example, does training in a visual memory task improve verbal memory, and does any memory training improve executive function?

Intervention types generally break down to three categories – classic training tasks, neuropsychological training (which involves multiple tasks at once), and video games.

Finally, studies need to account for the duration of any training effect. If there is an effect, how long does it last after the period of training ends?

The above variables must be considered in addition to all the generic factors that influence the rigor of any clinical study – number of subjects, randomization, effect size, statistical significance, proper blinding, adequate control group, accounting for multiple comparisons, drop-out rate if any, dose-response (in this case, duration and intensity of training) and replicability.

With all of these variables to account for it will take a great deal of research to understand the true effects of computer-based brain training of each type for various outcomes and on various populations. Not surprisingly, existing research is just scratching the surface of addressing all the potential questions regarding brain-training.

A 2012 systematic review by Kueider et. al. identified 151 computerized training studies published between 1984 and 2011 involving healthy older adults. That is not many studies, resulting in only a few studies for each intervention and target population. Of the 151 studies identified, only 38 met the review’s inclusion criteria.

For the full results of this review, I suggest you read the original article, which is available open-access at the link above. It’s not really possible to summarize the full results in less space than the review itself, so there is no reason to duplicate it here. To give an overview, however, in each category there were only a few studies, and most studies were relatively small. My overall impression, therefore, is that much more research needs to be done.

Studies generally found positive effects from brain-training (not surprising for small preliminary studies), but in most cases results were mixed with some positive and some negative studies. Brain-training was generally found to be as effective as traditional book and pencil training, but less labor intensive.

Effects were strongest for the task that was trained, with highly variable outcomes in terms of generalizability. Overall tasks generalized either not at all or only to closely related tasks, but not across the board or to very different tasks. For example, there seemed to be no cross-over effect between visual spacial cognitive function and verbal cognitive function.

In this review classic training tasks had the biggest effect on working memory, processing speed, and executive function. Neuropsychological tasks had the most improvement on memory and visuospacial ability. Video games had a positive impact on reaction time and processing speed.

A more recent 2013 review and meta-analysis of studies involving healthy children and adults concluded:

…that memory training programs appear to produce short-term, specific training effects that do not generalize. Possible limitations of the review (including age differences in the samples and the variety of different clinical conditions included) are noted. However, current findings cast doubt on both the clinical relevance of working memory training programs and their utility as methods of enhancing cognitive functioning in typically developing children and healthy adults.

A 2013 study of brain-training in older adults with mild cognitive impairment or dementia found no statistically significant difference in the treatment and control groups, but a tendency toward better performance in the treatment group, only for the more mildly affected subjects.


Computer-based brain-training is a promising intervention for maintaining and improving cognitive function in healthy and perhaps mildly impaired individuals, primarily because it is convenient, less labor intensive than traditional methods, and cost effective.

Existing research, however, is inadequate to rigorously address all of the variables of brain-training interventions. There do appear to be a few patterns in the existing research, however.

  • Brain-training is effective, whether designed as classic cognitive tasks, combined tasks, or video games.
  • Effects are mostly restricted to the specific tasks being trained and do not significantly generalize to other tasks or cognitive functions. Effects tend to be short lived, although evidence here is very mixed.
  • Computer-based brain-training does not appear to be significantly different in outcome from traditional pencil and paper-based training, but is less labor intensive.
  • I could find no published evidence to support any claims for individualized programs.

In short, brain-training does not seem to make you smarter, but will make you better at whatever task you perform. This can be simply a training effect – you will get better at anything you do repetitively. This is no more an effect of brain plasticity than any generic learning. Suggestions that such brain-training makes your brain function better in any way other than simply learning the task that is being practiced is not evidence-based.

Another way to look at all this is that the very concept of “brain-training” is probably flawed. It is useful as a marketing slogan, but does not seem to be based in reality. “Brain-training” is just a fancy term for good old-fashioned learning, but is meant to invoke an image of cutting edge neuroscience and brain plasticity which is not supported by evidence. It’s just learning.

The bottom-line recommendations I would make from existing data are this:

  • Engaging in various types of cognitively demanding tasks is probably a good thing.
  • Try to engage in novel and various different types of tasks. These do not have to be computer-based.
  • Find games that you genuinely find fun – don’t make it a chore, and don’t overdo it.
  • Don’t spend lots of money on fancy brain-training programs with dramatic claims.
  • Don’t believe the hype.

Finally, there is a clear need for further research. We need many large rigorous studies that control for multiple variables.

Posted by Steven Novella

Founder and currently Executive Editor of Science-Based Medicine Steven Novella, MD is an academic clinical neurologist at the Yale University School of Medicine. He is also the president and co-founder of the New England Skeptical Society, the host and producer of the popular weekly science podcast, The Skeptics’ Guide to the Universe, and the author of the NeuroLogicaBlog, a daily blog that covers news and issues in neuroscience, but also general science, scientific skepticism, philosophy of science, critical thinking, and the intersection of science with the media and society. Dr. Novella also contributes every Sunday to The Rogues Gallery, the official blog of the SGU.