How long does it take for brain games to make a difference?

Studies that show benefits typically observe people who engage 4+ days per week over months or years. The 2003 NEJM study tracked participants for 5+ years. The PROTECT study tracked people who had been doing puzzles regularly. There is no evidence that 30 days of brain training produces measurable cognitive change.

Are free brain games as good as paid ones?

There is no evidence that paid brain-training apps produce better outcomes than free traditional puzzles like crosswords, Sudoku, and word games. The mechanisms that appear to drive cognitive benefit (sustained engagement, novel mental challenge, varied task types) work the same way regardless of price. Lumosity charged $11.99/month and had to be sued by the FTC for its marketing claims; meanwhile, free crossword puzzles have decades of supportive observational research.

Do brain games actually work? What the science says (2026)

Q: Do brain games actually work?

Yes for cognitive aging, no for general intelligence. Multiple long-term studies (NEJM 2003, PROTECT 2019, ACTIVE 2017, Wilson 2013) show that frequent puzzle-solving over years is associated with reduced dementia risk and slower cognitive decline. They do NOT show that brain games make you smarter at unrelated tasks. The 2014 consensus statement signed by 70+ cognitive scientists explicitly warned against this kind of claim, and in 2016 the FTC fined Lumosity $2 million for overstating it.

Q: What was the Lumosity FTC ruling?

In January 2016 the Federal Trade Commission fined Lumos Labs $2 million for deceptive advertising. The FTC ruled that Lumosity had no scientific basis for its claim that brain training reduced or delayed cognitive impairment from age, dementia, Alzheimer's, stroke, traumatic brain injury, PTSD, ADHD, chemo-fog, and the side effects of Turner syndrome. The settlement required Lumos Labs to notify subscribers of the ruling.

An honest look at what the published science says — the four studies that show real benefit, the FTC ruling that exposed the marketing, and what the evidence actually supports.

The marketing claim vs. the science

If you've seen a brain-training ad in the past decade, you've probably been told that 10 minutes a day will make you measurably smarter, sharper, and protected against cognitive decline. That marketing has a problem: most of it isn't supported by the published research, and one major company — Lumosity — was fined $2 million by the Federal Trade Commission in 2016 for making it.

But the opposite reaction — "brain games are a scam" — is also wrong. There's a real and meaningful body of evidence showing benefits, just not the benefits the apps were selling. Here's what the actual literature says.

The four studies that matter

These are the most-cited papers in the field. They're not perfect — observational studies have inherent limits — but they're the strongest evidence we have, and they all point in the same direction.

47% lower dementia risk

Older adults who did crossword puzzles 4+ days per week vs. those who did them once per week.

Verghese et al., Leisure Activities and the Risk of Dementia in the Elderly. New England Journal of Medicine, 2003. 469 adults aged 75+, followed 5.1 years on average.

8–10years younger brain age

Regular word-puzzle users performed at this brain-age equivalent on reasoning and short-term memory tests.

Corbett et al., An Online Investigation of the Relationship Between the Frequency of Word Puzzle Use and Cognitive Function in Older Adults. International Journal of Geriatric Psychiatry, 2019. 19,078 healthy adults 50+, the PROTECT study.

29% reduced dementia risk over 10 years

Participants who received speed-of-processing cognitive training. This is the only one of these four that's a randomized controlled trial — the gold standard.

Edwards et al., Speed of Processing Training Results in Lower Risk of Dementia. Alzheimer's & Dementia: Translational Research, 2017. Follow-up from the ACTIVE trial — 2,802 adults 65+, 10-year follow-up.

32% slower cognitive decline

Older adults who frequently engaged in cognitively stimulating activities (reading, puzzles, games) compared to those who rarely did.

Wilson et al., Life-span cognitive activity, neuropathologic burden, and cognitive aging. Neurology, 2013. Rush Memory and Aging Project. 294 participants tracked until death with detailed brain autopsy.

So that's the case for. Now the case against the marketing:

The Lumosity FTC ruling

In January 2016, the Federal Trade Commission fined Lumos Labs (the company behind Lumosity) $2 million for deceptive advertising. The FTC's complaint stated that the company "preyed on consumers' fears about age-related cognitive decline, suggesting their games could stave off memory loss, dementia, and even Alzheimer's disease" without scientific support for those claims.

Lumos Labs had specifically claimed brain training reduced or delayed cognitive impairment from age, dementia, Alzheimer's, stroke, traumatic brain injury, PTSD, ADHD, chemo-brain, and even the side effects of Turner syndrome. The FTC found there was no scientific basis for any of it. As part of the settlement, Lumos Labs was required to notify subscribers and offer cancellations.

The 2014 consensus statement

Two years before the FTC ruling, more than 70 cognitive scientists signed an open letter through the Stanford Center on Longevity and the Berlin Max Planck Institute. The relevant line:

From the consensus statement: “We object to the claim that brain games offer consumers a scientifically grounded avenue to reduce or reverse cognitive decline when there is no compelling scientific evidence to date that they do.”

The signatories weren't anti-puzzle. They were specifically anti-overclaiming. The position the field arrived at: regular cognitive engagement is associated with healthier aging, but no app that promises specific transferable IQ gains can back that up.

So what actually works?

Synthesizing across the studies, three things appear to matter:

Consistency over years, not weeks. Every study showing meaningful benefit involved sustained engagement — the NEJM crossword study tracked 5+ years, ACTIVE followed 10 years. There is no evidence that a 30-day brain-training challenge produces lasting change.
Variety beats specialization. The Wilson 2013 study specifically found that varied cognitive engagement (reading, puzzles, games, learning) correlates with slower decline. Doing only Sudoku, every day, forever, isn't supported by the same evidence as switching between word, logic, and memory tasks.
Frequency matters more than duration. The cited studies generally measured "4+ days per week" or "regular engagement," not "two hours per day." Ten focused minutes daily appears to clear the bar in most of the data.

What this means for everyday use

If you're playing puzzles for fun, the science is mostly irrelevant — do it because you enjoy it. If you're playing them for cognitive benefit, the honest read of the evidence is:

You're probably doing something good for long-term brain health.
You're not going to get measurably "smarter" at unrelated tasks like work performance, driving, or remembering names.
Free puzzles work the same as paid apps. The mechanism (sustained novel mental challenge) is what matters — not the brand or the subscription price.
Variety beats hyper-optimization. A daily mix of word, logic, and memory puzzles has stronger evidence behind it than mastering one genre.

How Puzzle Cottage applies this

Honestly, the design of Puzzle Cottage — 10 different daily puzzles across word, logic, math, and memory categories — is built around what the research actually supports. Not because we want to claim cognitive benefits we can't prove, but because the format that the science endorses (variety, daily cadence, no specialization) happens to also be the format that makes a puzzle site fun to come back to.

The site is free. No account, no subscription, no upgrade path. We don't need you to subscribe for the cognitive benefits to apply — if anything, the research suggests the format you use matters more than the platform.

Frequently asked questions

Do brain games actually work?

Yes for cognitive aging, no for general intelligence. Long-term studies consistently associate frequent puzzle-solving with reduced dementia risk and slower cognitive decline in older adults. They do not support the claim that brain games make you smarter at unrelated tasks — that claim got Lumosity fined by the FTC in 2016.

How long do I need to play to see benefits?

The studies that show benefit involve people who engaged 4+ days per week over years, not weeks. There's no evidence that a 30-day brain-training challenge produces lasting change. Daily 10–15 minutes over months and years appears to be the threshold across multiple studies.

Are paid brain-training apps better than free puzzles?

There's no evidence that paid apps outperform traditional puzzles like crosswords, Sudoku, and word games. The mechanism that appears to drive benefit (consistent novel mental challenge) works the same regardless of price. The largest study with positive results — the 2003 NEJM crossword paper — was about a free, 100-year-old puzzle format.

Does Sudoku prevent dementia?

Sudoku is associated with reduced dementia risk in observational studies, but no randomized trial has proven causation. It's a member of the broader category of cognitively demanding leisure activities that the research suggests is protective. It is not a guaranteed prevention strategy.

Why did Lumosity get sued?

In January 2016 the FTC fined Lumos Labs $2 million for advertising claims that brain training could stave off cognitive impairment from age, dementia, Alzheimer's, stroke, traumatic brain injury, and other conditions. The FTC found there was no scientific basis for those claims. Subscribers had to be notified and given the option to cancel.

What's the most evidence-based puzzle to start with?

There is no single "best" puzzle — the consistent finding is that variety matters. A daily mix of word puzzles (like Connections or Anagram Hunt), logic puzzles (Sudoku), and memory tasks (Memory Match) covers the categories that the research supports.