Neither crosswords nor brain training apps are magical for cognitive health, and the honest research favors variety. Brain training apps with adaptive, personalized training (particularly speed-of-processing exercises) have somewhat stronger evidence for measurable cognitive improvement — the ACTIVE Trial being the strongest single study. Crosswords are genuinely good cognitive stimulation but are one activity in a healthy mental diet, not a targeted intervention. The best answer is often both.
The debate is older than smartphones: are the daily crossword puzzles your grandmother swears by actually good for your brain, or are modern brain training apps a genuine upgrade? Here is what the research honestly says and how to pick what fits you.
First: what both are actually trying to do
Crosswords are word-based logic puzzles that engage vocabulary, general knowledge, pattern recognition, and problem-solving. They have been recommended for cognitive health for a century — often without much specific evidence backing the recommendation.
Brain training apps (like Zenelia, BrainHQ, Elevate, or Lumosity) are targeted cognitive exercises designed to train specific mental skills — attention, memory, processing speed, or practical abilities like mental math. They typically use adaptive difficulty, personalized programs, and metrics-based progress tracking.
Both engage the brain. But they engage it differently, and the research on their effects differs meaningfully.
What the research says about crosswords
Doing crosswords is probably genuinely good for cognitive stimulation — but not for the specific reason most people think.
What we actually know:
- Crosswords engage vocabulary, general knowledge, and problem-solving. These are cognitively demanding activities.
- Regular crossword solving is associated with delayed onset of memory decline in some observational studies of older adults. However, this is correlation, not proof of causation — people who do crosswords may already have healthier cognitive habits generally.
- A 2022 randomized trial in the NEJM Evidence journal compared crosswords to computerized cognitive training in adults with mild cognitive impairment. The result was interesting: crosswords produced slightly better cognitive outcomes than the cognitive training program used in that specific study. But this finding is more nuanced than headlines suggested (more on this below).
What we do not have strong evidence for:
- Crosswords “preventing” dementia through a specific mechanism. The research shows association, not causation, and the size of any specific crossword effect is modest.
- Crosswords being uniquely beneficial compared to other complex cognitive activities. Reading, learning a language, playing bridge, or doing puzzles likely produce similar general effects.
The honest bottom line on crosswords: they are a good form of cognitive stimulation, particularly for vocabulary and problem-solving. They probably contribute to cognitive health as part of a broader engaged mental life. They are not a specific targeted intervention.
What the research says about brain training apps
Brain training has a more nuanced research picture, both stronger and weaker than the marketing suggests.
What we actually know:
- Brain training apps reliably improve you at the specific tasks you practice. Near transfer (improvement in closely related skills) is well established.
- The ACTIVE Trial — a large NIH-funded study following older adults for years — found that structured cognitive training, particularly speed-of-processing training, produced measurable, lasting benefits in older adults. This is the strongest single piece of evidence for targeted brain training.
- Apps built specifically on ACTIVE Trial research, particularly BrainHQ and Zenelia, have this evidence backing their approach.
- Adaptive, personalized training (where the program adjusts to your weak areas) has better evidence than fixed one-size-fits-all programs.
What we do not have strong evidence for:
- Broad far transfer — the claim that training on abstract cognitive games makes you significantly “smarter” in unrelated real-world tasks. This is where most marketing overreach happens.
- Brain training preventing dementia as a specific claim. Some evidence, particularly for speed-of-processing training in older adults, but far from definitive.
- Most abstract brain training games producing meaningful transfer to daily life. Games that look like real-life skills (Elevate’s writing and math practice, for example) sidestep this problem by training the skills directly.
The honest bottom line on brain training apps: they reliably improve specific skills. Some approaches (speed-of-processing training in older adults especially) have real evidence for lasting benefit. Broad claims about “making you smarter” are usually overstated.
For the fuller picture, see our guide on whether brain training apps actually work.
About that 2022 study — the honest nuance
Some readers may have seen headlines about the NEJM Evidence study finding crosswords beat brain training. The honest picture is more complicated than the headlines:
The study specifically compared crosswords against a computerized cognitive training program in adults with mild cognitive impairment. The training program used was a general cognitive training approach, not specifically the speed-of-processing training that has the strongest ACTIVE Trial evidence.
Both interventions produced cognitive benefits. Crosswords did slightly better on the specific outcome measured in that study.
The study does not mean crosswords are universally better than all brain training. It means one specific brain training program performed slightly worse than crosswords in one specific population with a specific condition.
What it does suggest: engaged cognitive activity — of many kinds — is likely more important than the specific format. A daily commitment to cognitive engagement, whatever the specific activity, is probably more valuable than choosing the “right” cognitive activity and doing it inconsistently.
The honest comparison
Crosswords:
- Strengths: cheap or free (newspaper subscriptions, apps like NYT Games), engaging, socially discussable, no learning curve for the concept, adaptable to your interest and time.
- Weaknesses: primarily verbal — limited variety in cognitive domains. Can plateau (if you always do the same-difficulty puzzles).
- Best for: users who enjoy word puzzles, want low-cost cognitive stimulation, and are already generally cognitively engaged.
Brain training apps:
- Strengths: targeted training across multiple cognitive domains, adaptive difficulty, metrics tracking, some approaches (speed-of-processing) with genuine research backing.
- Weaknesses: cost (typically $35-$120/year), some marketing hype, most far transfer claims overstated.
- Best for: users who want targeted training in specific cognitive areas, appreciate metrics and progress tracking, or specifically want research-backed cognitive health interventions.
The genuinely honest answer: both, and other things
The best approach for cognitive health is variety of cognitive engagement, not choosing one activity as uniquely superior. Research consistently supports:
Multiple types of cognitive stimulation — reading, puzzles, brain training, learning new skills, engaging conversations, complex hobbies.
Combined with the lifestyle fundamentals — regular exercise, quality sleep, social engagement, cardiovascular health, managing depression. See our how to prevent cognitive decline guide for the full picture.
Consistency over intensity — 20 minutes of cognitive engagement daily beats 2 hours once a week.
If you enjoy crosswords, do them. If you enjoy brain training, do that. Ideally both, alongside reading, learning, socializing, and moving your body. The person who does only crosswords daily and nothing else is likely getting less cognitive benefit than the person doing crosswords, a brain training session, walks with friends, and reading.
When to choose which
Choose crosswords if:
- You genuinely enjoy word puzzles (enjoyment matters for consistency)
- You want low or zero cost
- You want an activity that is socially discussable
- You prefer paper-based or non-app engagement
- You are already generally engaged cognitively
Choose brain training if:
- You want targeted training across multiple cognitive domains
- You value adaptive difficulty and metrics
- You specifically want research-backed cognitive health training (particularly Zenelia or BrainHQ, both built on ACTIVE Trial research)
- You are an older adult specifically focused on cognitive protection (speed-of-processing training has the strongest evidence for this)
Do both if:
- You have the time and interest
- You want variety in cognitive engagement (which is what the research actually supports)
The best brain training apps to try
If you decide to try brain training, here are the honest picks by fit:
Zenelia — best overall for adaptive AI training built on ACTIVE Trial research. $44.99/year. See our Zenelia review.
BrainHQ — deepest research record, particularly for speed-of-processing training in older adults. $96/year. See our BrainHQ review.
Elevate — practical skill training that sidesteps the transfer problem. $39.99/year. See our Elevate review.
For the full comparison, see our best brain training apps roundup.
Common questions
Do crosswords prevent dementia? They probably contribute modestly to cognitive health as one form of cognitive stimulation. They are not a proven specific intervention. See our how to prevent cognitive decline guide for what the evidence actually supports.
Are Sudoku or other puzzles similar to crosswords? They engage different cognitive skills (logic vs vocabulary) but likely have similar general benefits as cognitive stimulation.
How often should I do crosswords or brain training? Daily is ideal — most research on cognitive stimulation shows dose-response effects with consistency.
What if I do not enjoy either? Find a cognitively engaging activity you do enjoy. Learning a language, reading challenging material, playing bridge, learning an instrument — any of these engages your brain productively. Enjoyment matters more than which specific activity you choose because it determines whether you actually do it consistently.
The bottom line
Both crosswords and brain training apps are genuinely useful forms of cognitive engagement. Brain training apps with strong research foundations — particularly Zenelia and BrainHQ, both built on ACTIVE Trial research — have somewhat stronger evidence for targeted cognitive benefit, especially speed-of-processing training in older adults. Crosswords are excellent general cognitive stimulation and pair well with other activities.
The honest answer for cognitive health is not “which one” — it is variety of cognitive engagement combined with the lifestyle fundamentals (exercise, sleep, social engagement, cardiovascular health). Pick what you will actually do consistently, layer multiple types of engagement, and give it years rather than weeks.
For deeper context, see our do brain training apps actually work guide, how to prevent cognitive decline guide, or best brain training apps roundup.