Type "chess brain" into a search engine. You'll hit the same sentence over and over: chess rewires your brain. Sometimes with a technical word for backup, neuroplasticity, and two studies as collateral: London taxi drivers and jugglers.

Both studies exist. They're excellent. They were published in top-tier journals.

And they don't say what they're made to say. In both cases, the most interesting result is the one that never gets reported.

What the Word Means, and What It Doesn't Say

Neuroplasticity refers to the nervous system's capacity to change in response to experience. It's an established, uncontested fact, and it's a general property of nervous tissue.

It's also a nearly empty observation when it gets used as an argument.

Saying "chess changes your brain" amounts to saying that something happened when you learned something. Learning a phone number changes your brain. Watching a TV series changes your brain. If that weren't the case, you wouldn't remember anything. The term only becomes informative once you specify which change, how long it lasts, and above all what it's good for.

Those are exactly the three points where popular science stops.

The London Taxi Drivers, and the Half We Forget

The most cited study in the field.

In 2000, Eleanor Maguire and colleagues published in the Proceedings of the National Academy of Sciences a structural MRI comparison between London taxi drivers and matched controls. London cabbies are a textbook case: to get their license, they have to pass "the Knowledge," a formidable exam requiring memorization of thousands of streets and routes.

The famous result: the drivers' posterior hippocampus is significantly larger than that of controls. And the volume correlates with years on the job.

That's what everyone remembers. Now, the rest of the sentence, in the same paper.

The anterior hippocampus, meanwhile, was negatively correlated with seniority. The more years a driver had on the job, the more developed his posterior region and the smaller his anterior region.

This isn't a technical detail. It changes the nature of the result. This isn't a hippocampus getting bigger: it's a redistribution inside a structure. A specialization, with everything that word implies about compensation elsewhere.

The popular image, a brain you build up like a bicep, doesn't match what Maguire observed. The correct image is a budget being reallocated.

The Jugglers, and the Third Scan

The second canonical study, even more instructive, and here it's the protocol itself that contains the answer.

In 2004, Bogdan Draganski and colleagues published in Nature a short and elegant experiment. Twenty-four volunteers, average age 22, none of whom could juggle. They were split into two matched groups. First scan for everyone: no grey matter difference between the two groups.

The experimental group then got three months to learn the three-ball cascade. Second scan once they managed it, with a precise criterion: hold it for sixty seconds.

The result: a significant, bilateral expansion of grey matter in the mid-temporal area hMT/V5, specialized in processing complex visual motion, as well as in the left posterior intraparietal sulcus. Exactly the regions you'd expect for a task that consists of tracking balls in the air.

That's usually where the citation stops. But the experiment includes a third scan.

Three months later. During that period, strict instructions: none of the participants trained or tried to improve. The authors note that by that stage, most were no longer able to juggle properly.

And the grey matter expansion had decreased.

The word Draganski and colleagues use in their own results is transient. It appears in the paper's abstract, and the word order leaves no ambiguity: a "transient and selective" structural change.

The study most often cited to demonstrate that learning durably transforms the brain actually demonstrates that it transforms it as long as you keep practicing.

The Logical Leap

Grant everything above, and add the studies conducted directly on chess players, which did reveal structural differences between experts and novices. Our article on chess and the brain goes through the detail.

One problem remains, and it's the real subject.

A structural change proves that an activity has been practiced. It proves nothing about a benefit, and above all nothing about a benefit elsewhere.

The myth's implicit reasoning goes like this: chess changes the brain, therefore chess improves the brain, therefore chess improves everything the brain does. Two slippages in one sentence. "Change" is not "improve." And a local improvement, in the circuits the task recruits, is not a general improvement.

Fortunately, this question doesn't need to be settled by reasoning. It's directly testable, and it has been tested.

The Decisive Test: Transfer

The question is called transfer. Does a skill acquired in one domain carry over to another? Near transfer, between neighboring tasks, is well documented. Far transfer, to unrelated domains, is the point of dispute.

In 2016, Giovanni Sala and Fernand Gobet published in Educational Research Review a meta-analysis of the question in chess: 24 studies, 5,221 participants, 40 effect sizes. The overall result is encouraging. An average effect of 0.34, moderate but real. In mathematics, 0.38. In reading, 0.25. And 0.43 for programs exceeding twenty-five hours of instruction.

If the matter stopped there, the myth would be true.

But the authors themselves flag the problem, and it's a massive one: almost none of these studies included an active control group. In other words, children doing chess were compared to children doing nothing in particular. And in that comparison, everything counts: an adult's attention, novelty, the teacher's enthusiasm, the fact of learning something difficult. None of that is specific to chess.

So Sala and Gobet redid the work properly. In 2017, two experimental studies published in Learning & Behavior compared chess instruction to a control activity that was equally demanding and equally supervised, on mathematical problem solving.

The effect specific to chess becomes indistinguishable from zero.

That same year, in Current Directions in Psychological Science, they extended the finding to chess, music and working memory training. Their most disturbing observation isn't the absence of an effect, it's its structure: the size of the measured effect is inversely related to the methodological quality of the study. The more rigorous the protocol, the smaller the effect gets.

That's the typical signature of an effect that doesn't exist. When a phenomenon is real, improving the measurement makes it sharper. When it vanishes as you look more carefully, you were looking badly.

Our article on does chess make you smarter develops this case on the school performance side. Here, the point is narrower and sharper: the neuroplasticity argument doesn't survive the behavioral test. It doesn't matter what the imaging shows if the announced benefit can't be measured in practice.

What Remains True

The conclusion isn't that chess does nothing. It does a lot, just not what it gets credited with.

You genuinely get better at chess, and that's a considerable cognitive transformation. An experienced player doesn't see the same board as a beginner. He perceives structures where the other sees pieces, what researchers call chunking. That skill is real, measurable and deep. It's just specific.

Some less spectacular effects hold up better. The link between sustained cognitive activity and cognitive reserve in old age is documented, even if it raises causality problems, and our article on chess and Alzheimer's discusses its limits.

And pleasure doesn't need neurological justification. That may be the real stake of this whole article.

Why the Myth Holds Up So Well

Three reasons, and none of them has anything to do with the strength of the evidence.

It makes a passion respectable. Saying "I play chess because I like it" exposes you to having to defend the time spent. Saying "I'm developing my brain" ends the conversation. The neuroplasticity argument is a social justification before it's a scientific claim.

It sells. School programs, apps, online courses: the promise of a general cognitive benefit is worth infinitely more than the promise of playing good chess. The audience for "your child will improve at math" is a hundred times bigger than the audience for "your child will play better."

Imaging is convincing. A colored brain image wins people over far beyond what it demonstrates. That's a documented effect: the mere presence of a brain illustration increases the perceived credibility of an explanation, even when that explanation is a bad one.

What to Do With All This

Don't choose chess for your brain. Choose it because the game is beautiful, difficult and endless. That's a sufficient reason, and it's the only one that holds up over ten years.

Be suspicious of "rewire." The verb has no verifiable content. Faced with any claim of cognitive benefit, ask the three questions: which precise change, measured how, and compared to which control group. The third is almost always the one that's missing.

Remember the third scan. The jugglers' grey matter expansion decreased when they stopped. That's probably the most useful lesson in this whole article, and it's encouraging in its own way: what gets built by practice is maintained by practice. No capital is ever acquired for good.

Separate the real question from the fake one. The fake question is whether chess improves the brain in general. The good one is how to improve at this specific game, which is difficult, measurable, and far more interesting.

Key Takeaways

The two studies invoked to prove that chess durably transforms the brain each contain a result that contradicts that use. In London taxi drivers, the posterior hippocampus grows but the anterior one shrinks: that's a redistribution, not a gain. In jugglers, the grey matter expansion regressed as soon as practice stopped, and the authors themselves call it transient.

As for the announced benefit, it has been tested directly. With a control group doing something else equally demanding, the effect specific to chess on academic and cognitive skills becomes indistinguishable from zero. And the effect size shrinks as protocol quality goes up.

Chess makes you good at chess. That's already a lot, and it was always the only honest promise.

After reading: next time you come across "chess rewires the brain," look for the control group in the study being cited. If it isn't mentioned, or if it wasn't doing anything comparable, you have your answer.

This article is part of a series on psychology applied to chess. See also the Dunning-Kruger effect in chess and survivorship bias in chess.

Sources

  • Maguire, E. A., Gadian, D. G., Johnsrude, I. S., Good, C. D., Ashburner, J., Frackowiak, R. S. J., & Frith, C. D. (2000). Navigation-related structural change in the hippocampi of taxi drivers. Proceedings of the National Academy of Sciences, 97(8), 4398-4403.
  • Draganski, B., Gaser, C., Busch, V., Schuierer, G., Bogdahn, U., & May, A. (2004). Neuroplasticity: Changes in grey matter induced by training. Nature, 427, 311-312.
  • Sala, G., & Gobet, F. (2016). Do the benefits of chess instruction transfer to academic and cognitive skills? A meta-analysis. Educational Research Review, 18, 46-57.
  • Sala, G., & Gobet, F. (2017). Does chess instruction improve mathematical problem-solving ability? Two experimental studies with an active control group. Learning & Behavior, 45, 414-421.
  • Sala, G., & Gobet, F. (2017). Does far transfer exist? Negative evidence from chess, music, and working memory training. Current Directions in Psychological Science, 26(6), 515-520.