Do Puzzles Keep Your Brain Sharp? What the Research Actually Says

"Do puzzles keep your brain sharp?" gets answers both for and against. Here is a careful look at a few of the studies people usually point to, and at how far each claim goes.

A note on where we stand: this blog is written alongside three puzzle apps, so we have a reason to like puzzles, but the evidence is murky, so we are not making any health claims at this point. Everything below is about puzzles in general, and the studies mostly cover crosswords and Sudoku-style number puzzles, not Hitori, Nurikabe or Norinori specifically.

Finding 1: puzzle fans score better, in a big sample

Two large online studies drawn from the PROTECT cohort in the UK asked adults aged 50 and over how often they did word puzzles or number puzzles, then gave them a battery of computerized cognitive tests.

• In the number-puzzle study, 19,078 people aged 50 to 93 took part. How often people did number puzzles was significantly related to performance on all 14 cognitive measures, including attention, reasoning and memory.
• The word-puzzle study found the same direction: more frequent word-puzzle use went with better scores on 14 measures, and people who never did them scored lowest.

The catch is built into the design. These studies measured people once. They cannot tell whether puzzles improve thinking, or whether people who already think quickly are more likely to enjoy puzzles. The authors say as much: the results are not evidence that puzzles caused the better scores.

Finding 2: a crossword study with a hint of delayed decline

A 2011 study followed 488 older adults in the Bronx who started out with no cognitive problems. Of those, 101 went on to develop dementia. Among that group, people who reported doing crosswords at the start showed a delay of about 2.5 years before their memory began to decline quickly.

This is a much narrower claim than "crosswords prevent dementia." It is an observational result in a group that all developed dementia, it says nothing about people who did not, and the authors themselves call for clinical trials to test the idea.

Finding 3: solving puzzles lights up a recognizable network

Brain-imaging research does not usually look at puzzles specifically, but it has looked at problem solving. A 2018 meta-analysis combined 280 imaging experiments from 131 papers and found that problem solving in general, whether mathematical, verbal or visual, draws on a shared network spread across the frontal, parietal and cingulate regions of the brain.

That tells you which parts of the brain are busy while you solve. It does not show that regular puzzling strengthens them.

What about "brain-training" games?

The strongest review we found is a 2016 look at the evidence behind commercial brain-training products by Daniel Simons and colleagues. Their conclusion was that people reliably get better at the tasks they practice; there is much less evidence that the gain carries over to closely related tasks, and little evidence that it carries over to everyday thinking. Their advice: if you want to get better at a specific task, training may help; otherwise, stay skeptical of anything further.

What we could not find

We looked for studies showing that puzzle-solving makes people live longer and did not find any. Claims that puzzles raise IQ or prevent dementia are not supported by the studies we saw.

A fair way to read all this

• Puzzle fans tend to score higher on thinking tests. The research does not tell us that the results are specificially caused by puzzles, and education and lifestyle likely play a role.
• Puzzles are a pleasant, low-risk way to spend an evening, and a reasonable part of a lifestyle that also includes exercise, sleep and company. That is a judgment call, not a research result.

The best reason to solve a puzzle is that it is satisfying to finally see the whole grid click into place. If your brain gets a little healthier along the way, consider that a bonus.

Sources

• Brooker H. et al. The relationship between the frequency of number-puzzle use and baseline cognitive function in a large online sample of adults aged 50 and over. Int J Geriatr Psychiatry 2019;34(7):932-940.
• Brooker H. et al. An online investigation of the relationship between the frequency of word puzzle use and cognitive function in a large sample of older adults. Int J Geriatr Psychiatry, doi:10.1002/gps.5033.
• Pillai J.A., Hall C.B., Dickson D.W., Buschke H., Lipton R.B., Verghese J. Association of crossword puzzle participation with memory decline in persons who develop dementia. J Int Neuropsychol Soc 2011;17(6):1006-1013.
• Bartley J.E. et al. Meta-analytic evidence for a core problem solving network across multiple representational domains. Neurosci Biobehav Rev 2018;92:318-337.
• Simons D.J. et al. Do "brain-training" programs work? Psychol Sci Public Interest 2016;17(3):103-186.