Massive study finds no support for generalizability of benefit from brain training software

In a powerhouse of a publication, Adrian Owen and his colleagues have taken a broadside at the field of brain fitness software.  Together with the BBC popular science program Bang Goes the Test, they recruited 11,430 people who completed an average of twenty-four 10 minute training sessions over the six weeks that the experiment was run.  All of these individuals underwent cognitive testing before and after brain training, and the results were pretty clear: brain training provided no benefit other than improvement on the task that people trained with.  In their words:

In our view these results provide no evidence to support the widely held belief that the regular use of computerized brain trainers improves general cognitive functioning in healthy participants beyond those tasks that are actually being trained.

There will be much debate about this study in the days and weeks to come, much of it focused on the details, which, as we all know, is where the devil habitually resides.   In particular, it will be very important for experts to examine the specific tests being employed and compare them to what has been previously published, especially for those brain training regimes that have reported improvements in fluid intelligence.  One thing is clear: the numbers of participants make it difficult to refute Owen et al.’s findings, and certain make replication a challenge.

Personally, I see Owen et al.’s study as a healthy development, indicating that the field is moving from adolescence into early adulthood.  Indeed, with pharmaceuticals of all stripes we are quite accustomed to optimistic early results only to find in later, larger trials (especially after the drugs are released into use by the general population) that a subtle effect is seen which was not observed in earlier clinical trials.  Moreover, it is likely to spur further investigation, which again can only be healthy. The results, and our interpretation of those results, are likely to change over time.

I will admit to being among those who had hoped that brain training would produce impressive effects.  But in the end, I am a slave to data.   Anything less would be wishful thinking.

For more about the study, see the following video from the Nature Video.

Image source: Nature Publishing Group and Rex Features

Hat tip to Sofia Lombera for alerting me to the video


3 thoughts on “Massive study finds no support for generalizability of benefit from brain training software

  1. As the CEO of Posit Science, I was surprised at the authors’ sweeping conclusion as the study’s headline should have been, “BBC’s Brain Training Did Not Work.” It overreaches in generalizing that since their methods did not work, all methods would not work. Their conclusion would be like saying, “I cannot run a mile in under 4 minutes and therefore it is impossible to do.” Many people can’t, but we know from independent, randomized, controlled studies that you can improve brain performance. There will be many more detailed questions about methodology, e.g., length of time training, specific training completed, assessments used, but those are secondary.

    It is a good result in that it shows how hard it is to build products that create a meaningful difference in people’s lives. I also hope the results will move people to ask themselves if they are playing brain games for fun or do they expect to see a benefit from their effort.

    I feel passionately about this topic becuase Posit Science has built products that have been rigorously tested by folks like the Mayo Clinic, Johns Hopkins, and USC’s Center for Gerontology and shown real-world improvements including lower crash risk, improved memory and better health. You can read more at

  2. There is a big difference between serious brain training programs based on peer reviewed research and casual brain games that have no scientific validation. Frankly, the market hasn’t done a very good job of explaining what a real brain training program is, including the need for scheduled blocks of training time. Here is a site that helps to separate validated brain training programs from casual brain games :

  3. Pingback: Racing to restoring cognitive function | Neuroethics at the Core

Comments are closed.