Is Sudoku Good for Your Brain? What the Research Shows

In 2019, a research group from the University of Exeter and King’s College London published a paper in the International Journal of Geriatric Psychiatry on how often people do number puzzles and how well they do on cognitive tests (Brooker et al., doi:10.1002/gps.5085). It looked at about 19,000 adults aged 50 and over. The more often people said they did number puzzles such as sudoku, the better they tended to score on tests of memory, reasoning and attention.

That study is often quoted as evidence that sudoku is good for the brain. But it does not show that doing sudoku makes you sharper. In this post I look at what sudoku actually asks of you, and then at what the research has and has not shown, with sources.

I should say up front that I make a sudoku app called Sudoku365, and on the Japanese App Store its name includes the word for brain training. So this post keeps what can be said separate from what cannot.

What sudoku asks of you

The rule of sudoku is only this: 1 to 9, once each, in every row, column and 3 × 3 box. While solving, you mostly use three things.

Reasoning. “This row already has a 5. So does this box. So the 5 can only go here.” You draw a conclusion from the rules. Sudoku moves forward by reasons, not by guessing.

Working memory. When you solve without pencil marks, you hold candidates in your head, “this cell is 3 or 8”, while looking somewhere else. It is the ability to keep a few things in mind for a short time while you think.

Attention. You scan the grid row by row, column by column and box by box, looking for what you have missed.

The important point is that using an ability in sudoku does not mean that ability gets stronger outside sudoku. That is where the research comes in.

What the large survey found

The 2019 paper used data from PROTECT, an online study in the UK. About 19,000 people aged 50 to 93 reported how often they did number puzzles and took online tests of attention, reasoning, memory and similar abilities. Those who reported doing puzzles more often scored better on many of the measures. The same group published a matching paper on word puzzles such as crosswords, with similar results (doi:10.1002/gps.5033).

The researchers were clear about the limits. In the university’s announcement, Dr Anne Corbett said: “We can’t say that playing these puzzles necessarily reduces the risk of dementia in later life.” (University announcement via ScienceDaily)

Correlation is not cause

This was an observational study. It compared how often people did puzzles with how they scored, at one point in time. It shows that the two go together. It does not show which one causes the other.

There are at least three possible explanations.

1. Doing puzzles helps keep cognitive abilities up.
2. People who already have stronger cognitive abilities are the ones who enjoy puzzles.
3. Something else, such as education, health or free time, affects both.

An observational study cannot fully separate these. If the second is true, people do not score well because they do sudoku. They do sudoku because they score well. “People who do sudoku are sharp” and “doing sudoku makes you sharp” are different claims.

What trials have shown

To test cause, you need an intervention study, where people are assigned at random and only one group does the puzzles.

A search turns up no large randomized trial of sudoku on its own. A close one is a study by Devanand and colleagues published in NEJM Evidence in 2022 (doi:10.1056/EVIDoa2200121). It assigned 107 people aged 55 to 95 with mild cognitive impairment to either online crossword puzzles at home or computerized brain-training games, and followed them for 78 weeks. The crossword group did better on a cognitive test and on a measure of daily functioning.

But there was no group that did nothing. The research team itself said the findings need to be confirmed in a larger trial with an inactive control group (Columbia University Department of Psychiatry). And the puzzles were crosswords, not sudoku.

Research on brain training in general keeps finding the same pattern: people get better at what they practise, while spread to other abilities and to daily life is unclear. I wrote about that in Does Brain Training Work? What Research Shows and What It Doesn’t. Sudoku can be seen inside that same frame.

What can and cannot be said

To sum up.

What can be said: sudoku uses reasoning from rules, working memory for candidates and attention to scan the grid. If you keep doing sudoku, you get better at sudoku. Middle-aged and older adults who often do number puzzles tend to score better on cognitive tests.

What cannot be said: that sudoku makes you sharper outside sudoku, that it reduces forgetfulness, or that it prevents dementia. Current research does not support saying any of these with confidence.

Reasons to keep playing anyway

Even if the effects cannot be claimed, there are other reasons to solve.

First, sudoku is an enjoyable puzzle in its own right. You find a reason for one cell, and the next cell follows from it. That feeling is worth having without asking for anything more.

Second, there is value in a daily habit. Sitting down at a set time for ten minutes or so and giving your attention to one thing can help shift your mood and give the day a rhythm.

Third, you can play with other people. Solving the same puzzle with family or friends and talking about it is a different pleasure from solving alone.

Sudoku is not a substitute for exercise, sleep or seeing people, though. If forgetfulness is getting in the way of daily life, please talk to a doctor rather than an app.

About Sudoku365

Sudoku365 is a sudoku app I make. It runs on iPhone, iPad and Android. There are four difficulties, Easy, Medium, Hard and Expert, with 300 puzzles each, 1,200 in total, and all 1,200 have been checked to have exactly one solution.

The app is built to help you keep playing every day. There is a Daily Challenge, one puzzle a day shared worldwide, a streak that counts the days you clear a puzzle, and Play reminders you can set for each day of the week. Play together connects two phones over Bluetooth: in Race you and the person next to you solve the same puzzle, and in Solve together you fill in one grid as a pair. No internet or account is needed.

The app is free with ads, and a one-time Remove Ads purchase turns them off. For the rules, see Sudoku Rules: What First-Timers Need to Know. For the first techniques to learn, see How to Solve Sudoku: The First Techniques Beginners Should Learn.

Sudoku is not a medicine for the brain. It is a well-made puzzle.