Key Takeaways

  • A 2019 study of 19,000+ adults aged 50 and over found that regular puzzle-solvers outperformed non-solvers on every cognitive measure tested
  • Frequent solvers scored the equivalent of 8 to 10 years younger on tests of attention, reasoning, and working memory
  • Number puzzles like sudoku showed the strongest link to spatial working memory, more so than word puzzles alone
  • The evidence is correlational, not causal, but it is large-scale and consistent
  • Regularity beats intensity: four short sessions per week showed stronger results than occasional longer ones

Most of us first picked up a sudoku puzzle for the simple pleasure of it: that satisfying moment when the last digit clicks into place and the grid resolves into order. But beyond the enjoyment, a quieter question has been running through neuroscience and geriatric research for years. Does this kind of structured mental exercise actually do anything lasting for the aging brain? A large study out of the University of Exeter offers one of the most substantial answers yet.

The Study Behind the Headlines

In 2019, researchers at the University of Exeter and King's College London published findings from the PROTECT study, one of the largest online investigations of cognitive aging ever conducted in the UK. Over several years, the team tracked more than 19,000 adults aged 50 and over, regularly assessing memory, attention, and reasoning alongside detailed surveys about daily habits. These included how often participants engaged with number and word puzzles.

The headline result was striking: adults who regularly solved number and word puzzles performed significantly better on every cognitive assessment than those who did not. On tests of attention, reasoning, and working memory, frequent puzzle-solvers scored as if they were roughly eight to ten years younger than non-puzzlers of the same chronological age.

The results appeared in the International Journal of Geriatric Psychiatry. Lead researcher Dr. Anne Corbett was careful in her language. The study design was observational, meaning it established association rather than direct cause. But she noted the associations were robust across multiple cognitive domains and held up after adjusting for age, education level, and general health status.

19,000+ adults aged 50 and over tracked
8 to 10 years younger on cognitive tests
4x per week for strongest results

What the Tests Actually Measured

The cognitive battery used in PROTECT was not a casual quiz. Participants completed structured tasks designed to measure three distinct capabilities:

  • Short-term and working memory — holding information in mind and manipulating it in real time, such as following multi-step instructions or tracking a conversation
  • Grammatical reasoning — applying logical rules quickly under mild time pressure
  • Spatial working memory — tracking the position of items across time and updating that mental map as conditions change

These are practical, real-world skills. They determine whether you can remember where you put your keys, work through a problem without losing the thread, or navigate a new route without GPS.

Puzzle-solvers outperformed non-solvers on all three. The relationship appeared dose-dependent: more frequent engagement consistently produced stronger scores. Within the puzzle category, number puzzles showed a particularly strong link to spatial working memory, even stronger than word puzzles alone.

"Those who reported doing puzzles more frequently had brain function equivalent to being around a decade younger on some tests." Dr. Anne Corbett, University of Exeter

Why Sudoku, Specifically?

The spatial working memory result is not surprising once you consider what sudoku actually demands. Unlike a crossword (primarily a memory retrieval exercise), sudoku requires you to simultaneously track candidates across nine rows, nine columns, and nine three-by-three boxes. You are not recalling facts. You are constructing and updating a spatial model in your head in real time.

Every cell elimination forces you to revise that model. Every constraint propagates across the grid. It is a sustained spatial reasoning exercise disguised as a number puzzle, and that distinction likely explains why number puzzles showed a disproportionately strong effect on spatial working memory tests in the PROTECT data.

The Science: What Could Explain This?

Cognitive Reserve

The most widely cited explanation is cognitive reserve: the idea that mentally stimulating activities build a kind of structural resilience in the brain over time. Think less "sharpening a pencil" and more "building a wider road." The brain's underlying architecture changes with age regardless, but people with higher cognitive reserve compensate for those changes more effectively and maintain function for longer.

Neuroplasticity

There is also growing evidence that neuroplasticity does not stop at adulthood. The brain's ability to strengthen and reorganize neural connections in response to repeated use continues throughout life. Puzzles that require sustained attention, logical inference, and working memory may reinforce specific cognitive pathways over time. Whether this represents genuine neurological remodeling or highly practiced familiarity with certain types of tasks is still being debated.

General Engagement

A third explanation worth considering: people who do puzzles regularly tend to be more cognitively engaged with life in general. They may read more, stay socially active, and maintain a variety of stimulating habits. Separating the specific effect of puzzle-solving from this broader pattern is methodologically difficult, and most researchers acknowledge this openly.

The Caveats Worth Knowing

What "observational" means in practice

An observational study records what people already do and looks for patterns. It cannot randomly assign someone to "do sudoku for five years" and compare outcomes against a control group. That means it can identify associations but cannot confirm that puzzles directly cause better cognition. This is not a flaw in the research; it is the honest limit of the study design.

Selection bias is a legitimate concern. People who regularly solve puzzles may already be cognitively sharper at baseline, or more educated, more socially active, or more health-conscious in ways that independently protect the brain. The PROTECT team applied statistical controls for many of these variables, but no observational study can account for everything.

Randomized controlled trials (where participants are assigned to either puzzle or not and then followed over years) would provide considerably stronger evidence. A handful of small trials have been conducted, with mixed but broadly encouraging results. The honest scientific position remains: the evidence is consistent and suggestive, but not yet clinically conclusive.

The Enjoyment Factor

One PROTECT finding that rarely gets attention: cognitive benefits were most consistent among people who reported genuinely enjoying their puzzle habit. This matters more than it might initially seem.

Mental activity pursued under chronic stress (as a chore, an obligation, or anxious self-optimization) activates neurological pathways quite different from those engaged during genuinely pleasurable activity. Elevated cortisol from chronic stress actually impairs some of the same cognitive systems that puzzles are thought to support. A tense, rushed puzzle session may not deliver the same benefit as a calm, unhurried one.

Sudoku done slowly and without stakes is genuinely relaxing for many people. The focused attention it requires functions as a mild form of mindfulness, narrowing awareness to a single structured task and quieting background noise. Whether or not future research confirms the cognitive benefits conclusively, that quality has value on its own terms.

Five Practical Takeaways

  1. Aim for regularity, not marathon sessions. The PROTECT data suggests four or more sessions per week outperforms one long weekly session. Ten minutes daily is more useful than ninety minutes on Sunday.
  2. Prioritize enjoyment over performance. Do not race the clock until you want to. A solved puzzle that left you frustrated may not be offering the same neural benefit as one that left you satisfied.
  3. Start easier than you think you need to. The cognitive workout comes from sustained attention and candidate-tracking, not from struggling. Easy puzzles done regularly outperform hard puzzles done occasionally.
  4. Combine with other habits. The strongest protective effects in aging research tend to come from people who maintain multiple forms of stimulation: puzzles, reading, social engagement, physical activity.
  5. Do not treat it as medicine. There is no proven dosage. The best reason to solve sudoku regularly is still that it is an enjoyable, low-cost activity that challenges real cognitive skills. The potential long-term benefit is a reasonable bonus.

Sources & Further Reading

  1. Corbett, A. et al. (2019). Cognitive activity and onset of depression and anxiety in older adults, findings from the PROTECT study. International Journal of Geriatric Psychiatry. Journal homepage
  2. The PROTECT Study, University of Exeter and King's College London. protect-study.org
  3. Wikipedia: Cognitive Reserve
  4. Wikipedia: Neuroplasticity
  5. Wikipedia: Working Memory