Are crosswords good
for your brain?

Crosswords are the most studied puzzle of all, and the studies are better than the headlines they generate. Three pieces of evidence matter: a long-running cohort in the Bronx, a randomised trial in people with early memory problems, and a very large online survey. Together they support a modest claim and rule out a grand one.

The Bronx Aging Study: a delay, not a shield

Researchers followed 488 people aged 75 to 85 who were cognitively healthy at enrolment, testing them every 12 to 18 months. Among the 101 who eventually developed dementia, those who had been doing crosswords at baseline reached the point where memory decline accelerates about 2.54 years later than those who had not.^[1] Once decline began, it was steeper in the crossword group, a pattern consistent with cognitive reserve: the brain compensates for longer, then the compensation runs out.^[4]

Two cautions. The study is observational, so crossword habits may be a marker of other protective traits rather than a cause. And the finding is about people who developed dementia; it says nothing about whether crosswords stopped anyone developing it.

The randomised trial: crosswords beat games

In 2022 Devanand and colleagues published the first randomised trial pitting crosswords against computerised cognitive games.^[2] 107 adults with mild cognitive impairment, average age 71, trained four times a week for 12 weeks and then had booster sessions out to 78 weeks. The crossword group declined less on the ADAS-Cog scale, did better on a measure of daily functioning, and lost less brain volume on MRI. Those are clinically meaningful outcomes.

What the trial did not include was a no-training control, so it shows crosswords were better than that particular games package, not that crosswords beat doing nothing. Its participants already had memory impairment, so it does not speak directly to healthy adults. It remains the best experimental evidence for any single puzzle type.

The large survey: a dose-response pattern

PROTECT asked 19,078 adults over 50 how often they did word puzzles and tested them on nine cognitive measures.^[3] Scores rose with frequency: the more often people did word puzzles, the better their attention, reasoning and memory. As a cross-sectional study it cannot separate cause from selection, but the graded relationship is what you would expect if the habit contributed something.

What to take from it

• Regular crosswords are associated with later onset of memory decline and better current thinking skills in large observational samples.
• In one randomised trial, crosswords outperformed computer games for people with early impairment on cognition, function and brain volume.
• No evidence shows crosswords prevent dementia, and the observational findings may partly reflect who chooses to do them.

If you enjoy crosswords, that is a good reason to keep doing them. If you want the same word-and-reasoning workout in a different shape, Brisk’s daily anagram and meaning questions cover similar ground in two minutes. The techniques in how to solve anagrams transfer directly to crossword anagram clues.