Does Wordle Make You Smarter? What the Research Shows for IQ

There is no evidence that Wordle makes you smarter: no study has measured IQ before and after people took up the game. In the only large dataset on repeated play, covering more than 22 million games, people barely got better at Wordle itself (a correlation of r = -0.04 between how many puzzles someone had played and how many guesses they needed), so there is little room for the game to be improving anything broader.

What the research does show is that Wordle draws on reasoning. In the first psychometric study of the game, good players were mainly better at figural reasoning, the kind of pattern-finding measured by matrix puzzles, and only weakly better on vocabulary. This page covers what the game asks of your brain, what that study found, why daily practice does not spread to general intelligence, and why the people who play it may have been sharp before their first green square.

Wordle gives you six tries to find a hidden five-letter word. After each guess, every letter is marked green (right letter, right place), yellow (in the word, wrong place) or grey (not in the word). Josh Wardle released it in 2021, and The New York Times Company bought it in 2022.

The game looks like a vocabulary test, but the vocabulary demands are modest. The answers are common words that almost every adult reader knows. The hard part is using the colour feedback to rule out candidates and choosing the next guess so that it tells you as much as possible. Information theory gives a precise way to think about this: a guess that cuts the remaining possibilities in half provides one "bit" of information. Bertsimas and Paskov solved the game exactly by treating it as a sequence of decisions, and found that a perfect strategy needs 3.42 guesses on average, starting with SALET.

That makes Wordle closer to a deduction puzzle than to a crossword. Crossword skill depends heavily on general knowledge: in four studies of about 200 adults each, Hambrick, Salthouse and Meinz found knowledge was the strongest predictor of crossword proficiency, while abstract reasoning had no direct relation to it.

In 2026 a team at the University of Kassel published the first systematic study of Wordle as a cognitive test. It was a "registered report", meaning the methods and hypotheses were peer reviewed and fixed before the data were collected, and the final version has been accepted by the journal Collabra: Psychology. The researchers gave 370 German-speaking adults 10 puzzles each from a German version of the game, along with tests of matrix reasoning, verbal fluency (naming as many words as possible that fit a rule), vocabulary and general knowledge.

• Reasoning came first: Wordle performance correlated r = 0.50 with figural reasoning, a measure of "fluid intelligence", the ability to solve new problems.

• Word retrieval came second: The correlation with verbal fluency was r = 0.31.

• Vocabulary mattered little: The correlation with vocabulary was r = 0.22, and with general knowledge it was r = 0.06, essentially zero.

• Age and education: Younger adults did slightly better (r = -0.20), and more educated adults did slightly better (r = 0.15). Daily reading time was unrelated to performance.

The authors had expected the opposite pattern, with vocabulary leading. They concluded that Wordle is a problem-solving task that combines quick word retrieval with reasoning about the feedback. A few caveats apply. These correlations are estimated at the level of the underlying abilities, which corrects for measurement error and makes them larger than raw score correlations would be. Ten puzzles gave a reliability of only 0.68, so the authors describe Wordle as a coarse indicator of ability and not a precise measure. Most participants rarely played Wordle, and none of these numbers are IQ scores. Our page on what vocabulary tests measure explains why vocabulary is usually such a strong marker of verbal ability, which makes its weak link to Wordle notable.

A correlation of 0.50 means that a player one standard deviation above average in reasoning would be expected to sit about half a standard deviation above average in Wordle skill. The direction runs from ability to game. Nothing in this design shows that the game builds reasoning.

The best answer comes from a second paper by the same group, which analysed 872 daily puzzles from a German Wordle site, played by 76,095 people over two years. Across more than 22 million player-puzzle combinations, the correlation between how far someone was into their playing history and how many guesses they needed was r = -0.043. The authors called this a negligible association: on average, playing more did not make people noticeably better. In the registered report, people who said they played Wordle often did a little better (r = 0.20), but that may reflect who keeps playing.

Psychologists separate "near transfer", getting better at the trained task and close relatives of it, from "far transfer", gains in broad abilities such as general intelligence. Sala and Gobet's meta-analyses of chess, music and working-memory training found that far-transfer effects shrink toward zero as study quality improves. A major review of commercial brain-training programmes by Simons and colleagues reached a similar conclusion: people reliably improve on the trained tasks, and there is little evidence that the gains reach distant tasks or everyday thinking. Wordle has not even been tested for far transfer, and its own practice curve is nearly flat.

The nearest training evidence comes from other word puzzles. Murphy and colleagues randomly assigned 53 university students to four weeks of word puzzles built around letter structure, puzzles built around meaning, or a diary-writing control. The structure group improved more on a letter-based verbal fluency task, which is a narrow skill and a near-transfer result in a small sample. The large crossword studies found that years of puzzle experience did not reduce age-related declines in fluid ability.

What about keeping the brain young? In the PROTECT study of 19,078 adults aged 50 and over, people who did word puzzles more often scored better on cognitive tests, a cross-sectional link covered in full on our page about puzzles and IQ. Salthouse's review of the "use it or lose it" idea found many studies showing that mentally active people score higher, but very few showing that they decline more slowly, which is what the claim actually requires.

Part of the appeal is that Wordle really does tap thinking skills. A 2024 meta-analysis of 52 samples found that scores from computer games correlated about r = 0.30 with traditional cognitive tests (0.45 after correcting for measurement error), whether or not the games were designed as tests. Doing well at a game that correlates with ability feels like exercising that ability.

Selection explains the rest. The players who are good at Wordle and stick with it are more likely to be people who already reason well, enjoy word play and have the education and leisure to make a daily puzzle part of their routine. In the registered report, good players also used deliberate strategies, such as opening with common letters and avoiding repeated letters. Both self-reported strategy use and an information-based measure of how much the first two guesses narrowed the field correlated r = 0.36 with performance. A better opening word improves a Wordle score, and no study has found that it improves anything else.

Knowledge of how words are spelled helps too. In the large puzzle dataset, the letter patterns of the answer, such as how common its letters are in each position and whether a letter repeats, predicted much of how hard each puzzle was. How spelling itself relates to intelligence is covered on our page about spelling and IQ.

No study has shown that Wordle raises IQ. Data from more than 22 million games show that people barely improve at Wordle itself with practice, and there is no evidence of gains on broader tests.

Probably somewhat, on average. In the only study of the question, Wordle skill correlated 0.50 with a reasoning test, though with 10 puzzles per person the game was too unreliable to judge any single player.

Less than it looks. Wordle performance correlated only 0.22 with vocabulary and close to zero with general knowledge, because the answers are common words and the challenge lies in using the feedback.

There is no Wordle-specific evidence. People who do more word puzzles score better on cognitive tests in later life, but these studies cannot show that puzzles slow decline, and anyone worried about memory changes should talk to a doctor.

An exact mathematical solution found that SALET gives the lowest average number of guesses, 3.42 with perfect play. Strategy improves your Wordle score, which is a different thing from raising your IQ.

Wordle is a short, enjoyable reasoning puzzle, and the people who are good at it tend to be good at reasoning. There is no evidence that playing it makes anyone smarter, and the largest dataset suggests it barely makes people better at Wordle. If you want an accurate measure of your reasoning ability, take a full-length online IQ test.

1. Achaa-Amankwaa, P., Walter, J., & Schroeders, U. (2026). Wordle: A game-based assessment of verbal ability? [Stage 2 registered report, preprint of a manuscript accepted in Collabra: Psychology]. PsyArXiv. [doi.org/10.31234/osf.io/pa89m_v5](. doi.org

2. Achaa-Amankwaa, P., Walter, J., Hübner, V., Hübner, P., & Schroeders, U. (2026). Modeling item difficulty in large-scale game-based cognitive ability assessment: An illustration using Wordle [Preprint]. PsyArXiv. [doi.org/10.31234/osf.io/dxybj_v2](. doi.org

3. Bertsimas, D., & Paskov, A. (2025). An exact solution to Wordle. Operations Research, 73(3), 1384-1394. [doi.org/10.1287/opre.2022.0434](. doi.org

4. Hambrick, D. Z., Salthouse, T. A., & Meinz, E. J. (1999). Predictors of crossword puzzle proficiency and moderators of age-cognition relations. Journal of Experimental Psychology: General, 128(2), 131-164. [doi.org/10.1037/0096-3445.128.2.131](. doi.org

5. Sala, G., & Gobet, F. (2017). Does far transfer exist? Negative evidence from chess, music, and working memory training. Current Directions in Psychological Science, 26(6), 515-520. [doi.org/10.1177/0963721417712760](. doi.org

6. Simons, D. J., Boot, W. R., Charness, N., Gathercole, S. E., Chabris, C. F., Hambrick, D. Z., & Stine-Morrow, E. A. L. (2016). Do "brain-training" programs work? Psychological Science in the Public Interest, 17(3), 103-186. [doi.org/10.1177/1529100616661983](. doi.org

7. Murphy, M., Spillane, K., Cully, J., Navarro-Pardo, E., & Moret-Tatay, C. (2016). Can word puzzles be tailored to improve different dimensions of verbal fluency? A report of an intervention study. The Journal of Psychology, 150(6), 743-754. [doi.org/10.1080/00223980.2016.1182887](. doi.org

8. Brooker, H., Wesnes, K. A., Ballard, C., Hampshire, A., Aarsland, D., Khan, Z., Stenton, R., McCambridge, L., & Corbett, A. (2019). An online investigation of the relationship between the frequency of word puzzle use and cognitive function in a large sample of older adults. International Journal of Geriatric Psychiatry, 34(7), 921-931. [doi.org/10.1002/gps.5033](. doi.org

9. Salthouse, T. A. (2006). Mental exercise and mental aging: Evaluating the validity of the "use it or lose it" hypothesis. Perspectives on Psychological Science, 1(1), 68-87. [doi.org/10.1111/j.1745-6916.2006.00005.x](. doi.org

10. Bipp, T., Wee, S., Walczok, M., & Hansal, L. (2024). The relationship between game-related assessment and traditional measures of cognitive ability: A meta-analysis. Journal of Intelligence, 12(12), 129. [doi.org/10.3390/jintelligence12120129](. doi.org

Hero image: Wooden Block Letters Vision, by WOKANDAPIX, CC0. Via Wikimedia Commons.