Skip to content

Guide · 6 minute read

Puzzles for older adults: what the research actually says

By Tall Type Press editorial team ·

People who do puzzles often score higher on thinking tests than people who do not. But no study has shown that puzzles prevent dementia or reverse memory loss.

That is the honest summary. The research is interesting, and some of it is encouraging. It is also limited in ways that matter. This article walks through the main studies on puzzles for older adults, and what they can and cannot tell us.

What the big puzzle surveys found

The largest studies on puzzles are surveys. Researchers ask people how often they do puzzles. Then they compare that with scores on thinking tests.

Word puzzles

A team at the University of Exeter studied 19,078 healthy adults aged 50 to 93. They asked how often each person did word puzzles, from “more than once per day” to “never.” Everyone also took online tests of attention, reasoning and memory.

The pattern was clear. On every one of 14 measures, the group that never did word puzzles scored lowest. The authors wrote that word puzzle use “is directly related to cognitive function in adults aged 50 and over.”

They were also careful. They wrote that future work “needs to determine whether engaging in such puzzles can favourably influence cognitive trajectory with age.” In other words, they did not show that puzzles caused the better scores.

Number puzzles

The same team looked at number puzzles in the same 19,078 people. Again, more frequent puzzle use went with better scores on all 14 measures. People who did number puzzles more than once a day did best on 10 core measures. Not all of those differences were statistically significant.

Sudoku

An earlier study from a related group asked over 65,000 participants about their leisure activities. More frequent sudoku “or similar puzzles” was linked with better scores in reasoning, spatial working memory and episodic memory. The authors called for further investigation.

So, does sudoku help with memory? These results are consistent with that idea. They do not prove it.

Why “linked with” is not the same as “causes”

Surveys like these have a built-in problem. They show that two things go together. They cannot show which one drives the other.

There are a few possible explanations:

  • Puzzles help thinking. This is the hopeful reading.
  • Good thinking makes puzzles appealing. People with sharper memory may simply enjoy puzzles more and keep doing them.
  • Something else explains both. Education, health or an active lifestyle could make people both puzzle more and score higher.

The surveys cannot separate these. That is why the authors themselves ask for more research.

Crosswords and memory decline

The Bronx Aging Study followed 488 people who started with no thinking problems. Of these, 101 later developed dementia. In that group, people who did crossword puzzles at the start had a later onset of rapid memory decline, by 2.54 years.

That sounds strong. But note what it measured. It was about when decline sped up, in people who did develop dementia. It does not show that crosswords stopped anyone from getting dementia. The authors said the role of crosswords “should be validated in future clinical trials.”

What trials show: training, games and crosswords

Trials are stronger than surveys. People are assigned to an activity at random, so other differences even out.

The ACTIVE trial

The best known is the ACTIVE trial. It enrolled 2,832 older adults with a mean age of 73.6, living independently. They had ten sessions of training in memory, reasoning or speed of processing. Some had extra booster sessions later.

Ten years on, the reasoning and speed groups still did better on the skills they had trained. Memory training effects had faded. At a mean age of 82, about 60% of trained people were at or above their starting level of daily function, versus 50% of controls. Daily function here meant self-reported “instrumental activities of daily living.”

ACTIVE did not test puzzle books. It tested structured training sessions. The results show that some mental training can have lasting effects. They do not tell us that a word search does the same.

Crosswords versus computer brain games

A smaller trial compared two activities in 107 people with mild cognitive impairment. One group did web-based crosswords. The other did web-based cognitive games. Both trained at home for 12 weeks, followed by booster sessions, and were followed for 78 weeks.

On the main memory and thinking scale, the crossword group improved slightly. The games group worsened slightly. The authors concluded crosswords showed “superior efficacy to games.”

It is one trial, and it was small. It also had no group that did nothing, so it compares two activities rather than activity versus none.

What the National Institute on Aging says

The National Institute on Aging (NIA) gives a balanced view. It says cognitive training “appears to have benefits” for older adults, pointing to ACTIVE.

It also warns: “Beware of claims that playing certain computer and online games can improve your memory and thinking.” NIA says there is not enough evidence that commercial brain-training apps match the ACTIVE training.

On engaging activities in general, NIA is plain. It says evidence for a lasting cognitive effect “is not definitive.”

NIA also lists other steps with research behind them. It recommends that adults get at least 150 minutes of physical activity each week. It notes that high social engagement was associated with better cognitive health in a study of more than 7,000 adults aged 65 and older.

A summary of the evidence

Study Type Who What it found What it cannot show
Brooker, word puzzles (2019) Survey 19,078 adults, 50–93 More puzzling, better test scores Cause and effect
Brooker, number puzzles (2019) Survey 19,078 adults, 50–93 Same pattern for number puzzles Cause and effect
Ferreira, sudoku (2015) Survey Over 65,000 people Sudoku linked with better memory and reasoning scores Cause and effect
Bronx Aging Study (2011) Follow-up 488 people, 101 later with dementia Crosswords linked with a later start of fast memory decline Whether crosswords prevent dementia
ACTIVE (2014) Trial 2,832 adults, mean age 73.6 Reasoning and speed training lasted 10 years Effects of puzzle books
Crosswords vs games (2022) Trial 107 people with MCI Crosswords did better than games Effect versus no activity

So, are puzzles worth doing?

Yes, for reasons that do not depend on a medical claim. Puzzles are enjoyable. They fill quiet time with focus. They can be shared, at a kitchen table or in a care home lounge. The research does not promise more than that, and we will not either.

If you choose brain puzzles for seniors, a few practical points help:

  • Pick what they enjoy. A puzzle that gets done beats one that sits on a shelf.
  • Mix the types. Word searches, sudoku and crosswords each feel different, and variety keeps it fresh.
  • Keep it readable. Small print turns a pleasant puzzle into a strain. See our guide on what font size is large print.
  • Do it together. NIA links social engagement with better cognitive health.
  • Keep moving too. Puzzles are not a substitute for physical activity.

You can try a free daily word search and sudoku on screen, with adjustable letter size. Or print one of our free large print printables. For buying advice, see our guide for caregivers and our article on choosing a large print word search for low vision.

Not medical advice. If you notice changes in memory or thinking, talk to a doctor.

Frequently asked questions

Can puzzles prevent dementia?

No study has shown that. Large surveys link frequent puzzle use with better test scores, but they cannot show cause and effect. The National Institute on Aging says evidence for a lasting benefit from such activities is not definitive.

Are crossword puzzles good for your brain?

They are linked with better scores on thinking tests in large surveys of adults over 50. One small trial in people with mild cognitive impairment found crosswords did better than computer brain games over 78 weeks, but it had no group without training.

Does sudoku help with memory?

A survey of over 65,000 people found that doing sudoku or similar puzzles more often was linked with better scores on memory and reasoning tests. That is an association, not proof that sudoku improves memory.

How often should older adults do puzzles?

There is no proven dose. Choose a rhythm that is enjoyable and easy to keep, and combine puzzles with physical activity and time with other people, which the NIA also recommends.

Sources

  1. National Institute on Aging, Cognitive Health and Older Adults (reviewed 2024)
  2. Brooker H et al., word puzzles and cognitive function, Int J Geriatr Psychiatry 2019 (PubMed)
  3. Brooker H et al., number puzzles and cognitive function, Int J Geriatr Psychiatry 2019 (PubMed)
  4. Ferreira N et al., cognitively stimulating leisure activities, Int J Geriatr Psychiatry 2015 (PubMed)
  5. Pillai JA et al., crossword puzzles and memory decline, J Int Neuropsychol Soc 2011 (PubMed)
  6. Rebok GW et al., ten-year effects of the ACTIVE trial, J Am Geriatr Soc 2014 (PubMed)
  7. Devanand DP et al., computerized games versus crosswords in mild cognitive impairment, NEJM Evidence 2022 (PubMed)

Written by the Tall Type Press editorial team. Facts are taken only from the sources above. This guide is general information, not medical advice. For questions about your eyes, ask an eye doctor.

More guides