Blog

Can puzzles really help prevent dementia?

You've probably seen the headlines: "Do crosswords every day and never get Alzheimer's!" The reality is messier — but more interesting. Here's what the research actually shows, minus the hype.

▶ Play today Back to blog Try MemoryGlyph

The honest answer: it's promising, not proven

No one can say "puzzles prevent dementia" with certainty. But there's real evidence that regular mental activity — including puzzles — is associated with slower cognitive decline and better brain health. The key word is associated. Correlation isn't causation, and researchers are still untangling what's really happening.

That said, the findings are genuinely encouraging. Here's what stands out.


What the research actually found

9–11% lower dementia risk

A 10-year study of 10,000+ adults (published in JAMA Network Open) found that people who regularly did puzzles, games, or chess were 9–11% less likely to develop dementia than those who didn't.

2.5 years of delayed decline

Research from the Einstein Aging Study found that crossword puzzlers showed memory decline starting about 2.5 years later than non-puzzlers — even among those who eventually developed dementia.

Better scores across cognitive domains

A study of 19,000 adults (University of Exeter / King's College London) found that frequent puzzle solvers performed better on tests of memory, attention, and reasoning — with sharper scores the more often they played.


The Duke study: crosswords vs. brain games

One of the most interesting recent trials came from Duke and Columbia Universities, published in the New England Journal of Medicine. Researchers compared crossword puzzles against a commercial "brain training" app (Lumosity) in people with mild cognitive impairment.

The crossword group showed: — Better cognitive improvement — Better daily functioning — Less brain shrinkage on MRI The researchers called this "hitting the trifecta" — something no Alzheimer's drug has achieved.

The takeaway isn't "crosswords are magic." It's that simple, engaging mental activity may be as powerful (or more) than expensive, gamified alternatives — and way more accessible.


Why does this work? (Probably)

The leading theory is cognitive reserve. Think of it like a buffer: the more you exercise your brain, the more capacity you build to handle damage before symptoms appear. Puzzles seem to strengthen neural pathways and possibly even promote new connections.

Working memory workout

Holding rules, patterns, and constraints in mind while solving — that's a real mental rep. Games like NumberGlyph and MemoryGlyph train exactly this.

Attention practice

Short, focused tasks help you practice staying on one thing. In a world of constant distraction, that's valuable.

Pattern recognition

Spotting patterns, eliminating possibilities, updating your mental model — these are executive functions that matter for everyday life.


What puzzles won't do

Time to manage expectations. No puzzle will:

The National Academy of Medicine's 162-page report on cognitive aging emphasises that physical activity and blood pressure control have the strongest evidence. Mental activity is "encouraging but inconclusive" — which is honest, not dismissive.


The real win: a habit you'll actually keep

Here's what I keep coming back to: the best brain exercise is the one you'll actually do. Consistently. For years.

That's why I built NumberGlyph and MemoryGlyph to be small and calm. One puzzle. One to three minutes. No streaks to stress about, no gamified pressure. Just a quick daily rep.

The science says: consistent engagement matters more than which specific puzzle you do. So pick something you enjoy — and show up daily.

Try it yourself

Start with one game. See if you like it. Come back tomorrow.

▶ NumberGlyph 🧠 MemoryGlyph 🎯 FocusGlyph

Sources

Texas A&M (2024) — Study on games, puzzles, and cognitive function in adults with mild cognitive impairment. Read →

Duke University / NEJM Evidence — Crossword puzzles vs. computerised brain games in mild cognitive impairment. Read →

Alzheimer's Drug Discovery Foundation — Overview of puzzle research and cognitive reserve. Read →

Alzheimer's Society (UK) — Study of 19,000 adults on word and number puzzles. Read →

PMC / Einstein Aging Study — Crossword puzzles and memory decline in persons who develop dementia. Read →

Note: This is general information, not medical advice. If you're concerned about memory or cognitive changes, please talk to a doctor. The evidence for brain training is encouraging but still evolving.

Menu