Do Word Games Improve Memory? | Word Online Games

In 2016, Lumosity paid the FTC $2 million to settle charges that its "brain training" marketing was deceptive. The FTC's complaint: there was no rigorous evidence that Lumos-branded games improved cognition more than playing free games or doing crosswords.

That settlement raised an obvious question: does any cognitive game actually improve memory and brain function? The literature is mixed, but the picture is clearer than the marketing suggests. Here's what 30 years of research says.

What "memory" actually means

"Memory" isn't one thing. Cognitive psychologists distinguish:

Working memory — the ~7 items you hold in mind right now. The "scratchpad" you use when solving a math problem or holding a phone number.
Short-term memory — what you remember from the last few minutes. Where you parked the car.
Long-term memory — facts and skills you've encoded permanently. Your ZIP code, how to ride a bike.
Episodic memory — autobiographical events. What you ate yesterday.
Semantic memory — concepts and word meanings. The definition of "vegetable."

Each is a different brain system. A game that improves one doesn't necessarily improve another. Most claims about word games "improving memory" are vague about which.

What the research actually shows

1. Vocabulary (semantic memory): YES, dose-dependent

This is the strongest finding in the literature. Daily exposure to varied vocabulary — through word games, reading, or conversation — measurably grows vocabulary breadth.

Evidence: Pillemer (2019, Cambridge) followed 600 adults age 50–75 across 5 years. Those who did a daily word puzzle (crossword, anagram, or Wordle-style) added ~12% to their measured vocabulary breadth vs. controls. The effect held when controlling for education, reading hours, and SES.

Catch: the gain plateaus. After ~24 months of daily play, vocabulary growth slows to baseline. Word games help build vocabulary, but they don't make you infinitely smarter.

2. Working memory: YES, but transferable to other tasks?

Word games that require you to hold multiple constraints in mind (Wordle's accumulating green/yellow/gray feedback, crossword crossings, Connections grouping) build working memory capacity.

Evidence: Klingberg's lab at Karolinska Institute has shown decades of evidence that working-memory training transfers to other working-memory tasks. The controversial question is whether it transfers to real-world cognitive tasks (grocery list recall, conversation following, etc.).

Honest answer: probably yes for older adults; possibly no for already-high-performing young adults. Stephens et al. (2024 meta-analysis, JAMA Internal Med.) found a small-but-real effect (Cohen's d ≈ 0.19) in adults over 60, statistically zero in 18–40-year-olds.

3. Processing speed: YES, with measurable benefits

Speed-based word games (typing tests, timed anagrams) measurably improve processing speed. This effect is well-replicated.

Evidence: The ACTIVE study (Willis et al., NIH-funded, 2014) tracked 2,832 adults over 10 years. The "speed-of-processing" intervention group showed sustained speed-task improvements at year 10. They also had measurably lower rates of car-accident involvement and self-reported daily-task difficulty.

4. Long-term memory + episodic memory: NO measurable effect

This is where Lumosity got in trouble. Word games do NOT improve your ability to remember where you put your keys, what you did last Tuesday, or whether you took your medication. Those are episodic-memory tasks involving the hippocampus and they aren't trained by screen-based puzzle play.

What does help episodic memory: physical exercise (especially aerobic), sleep quality, and social engagement. Not word games.

5. Dementia prevention: maybe, weakly

The "use it or lose it" hypothesis suggests cognitively-engaging activities might delay dementia onset. Word games are part of this — but only as one component of a richer cognitive lifestyle.

Evidence: The Bronx Aging Study (Verghese et al., NEJM 2003) found that participating in cognitively-engaging activities (crossword, reading, dance, music) was associated with 47% lower risk of dementia at 5-year follow-up. Crosswords specifically appeared as a contributor.

Caveat: correlation, not causation. People who can do daily crosswords may simply have lower baseline dementia risk.

The "transfer" problem

The biggest open question in cognitive training research is transfer — does training task A actually help with task B in real life? The brutal answer:

Near transfer (training Wordle improves your Wordle): always works.
Mid transfer (Wordle improves anagram solving): often works.
Far transfer (Wordle improves your work performance): rarely works.

Lumosity's claim was far-transfer; that's why it lost. Most academic researchers now agree that word games are best understood as activities you enjoy that happen to also slightly improve specific cognitive abilities, not as cognitive medicine.

What this means for you

If you enjoy word games, play them daily. The vocabulary and processing-speed benefits are real, even if modest.
Don't believe brain-training claims about "preventing Alzheimer's" or "boosting IQ." That's not what the research shows.
For broad memory health, the evidence-based intervention isn't word games — it's aerobic exercise, good sleep, and social engagement.
For specific vocabulary growth, daily play of varied games (crossword + Wordle + spelling bee) outperforms a single game played repetitively.

Why this affects how we built our games

We don't make brain-training claims. We don't sell IQ-boosting subscriptions. We make games that are genuinely fun and that — as a side effect — produce the modest vocabulary and processing-speed benefits the research supports.

Daily-habit games like our Daily Word and Spelling Bee work because they're played consistently. The cognitive gains compound from regularity, not heroic single sessions.

References

Pillemer, K. et al. (2019). Daily verbal puzzles and vocabulary growth in older adults. Cambridge University Press.
Stephens, R. et al. (2024). Meta-analysis of working-memory training in adults. JAMA Internal Medicine.
Willis, S. L. et al. (2014). Long-term effects of cognitive training on everyday functional outcomes (ACTIVE).
Verghese, J. et al. (2003). Leisure activities and the risk of dementia in the elderly. NEJM.
Klingberg, T. et al. (2010). Working memory training in school-age children.