Signs of High Intelligence: What the Research Supports, and What It Does Not

The Scorecard

Most lists of intelligence signs mix three kinds of claim: findings from large meta-analyses, findings from single small studies, and claims with no study at all. The table sorts the most common ones by the strength of the evidence behind them.

A correlation of .20 means the trait explains about 4 percent of the differences in test scores. Plenty of curious people score average, and plenty of incurious people score high. That is the honest frame for every sign below. For what high means on the scale, see what is a high IQ and what an IQ of 130 means.

Supported: Speed of Thought

Psychologists have timed people on simple choices for more than a century. Sheppard and Vernon reviewed 50 years of that research in 2008 and found a mean correlation of r = −.24 between reaction times and intelligence across 172 studies. The sign is negative because shorter times go with higher scores.

Speed here means milliseconds on elementary tasks, such as deciding whether a number appeared in a short list. It does not mean talking fast, answering first in meetings or finishing exams early. Those habits reflect confidence and temperament as much as cognition.

Supported: Holding More in Mind

Working memory is the mental workspace that holds information while you use it: a phone number while you dial, a clause while you parse a long sentence. Ackerman, Beier and Boyle pooled the research in Psychological Bulletin in 2005 and estimated the true-score correlation between working memory and general ability at ρ = .479.

That is a substantial link, and the authors' central point was that it falls well short of 1.0. Working memory and intelligence overlap; they are not the same thing. In daily life the signal looks like following a complex argument without notes or doing multistep arithmetic in your head. Techniques such as the memory palace improve recall of specific material, which is a different capacity.

Supported, Weakly: Curiosity and Openness

The largest study of personality and intelligence to date, a 2022 meta-analysis in Psychological Bulletin by Jeromy Anglim and five colleagues, pooled 272 samples and 162,636 people. Openness to experience was the strongest Big Five correlate of intelligence, at ρ = .20. Its intellectual side ran a little higher: .25 for the NEO's Ideas facet and .26 for the Intellect aspect of a second inventory.

The other traits barely registered. Extraversion, conscientiousness and agreeableness each correlated with intelligence at roughly zero; neuroticism at ρ = −.09. Hard work, sociability and kindness say almost nothing about a person's test score. A taste for abstract ideas, reading and debate says a little more.

Vocabulary deserves a note. Kanazawa and Perina, citing a 1957 review by Miner, reported a median correlation of .83 between vocabulary and general intelligence, which is why vocabulary subtests anchor most IQ batteries.

Viral: The Messy Desk

The claim that geniuses keep messy desks traces largely to a 2013 paper in Psychological Science by Vohs, Redden and Rahinel of the University of Minnesota. Participants worked in either a tidy or a cluttered room. Those in the tidy room donated more to charity and chose healthier snacks. Those in the messy room generated new uses for ping-pong balls that judges rated as more creative, and they were likelier to choose a novel product.

Nobody's intelligence was tested. The study manipulated the room, not the person, so it says something about how surroundings shift thinking for a few minutes. It says nothing about whether people who keep messy desks score higher on IQ tests.

Viral: Swearing

The swearing claim comes from Kristin Jay and Timothy Jay, writing in Language Sciences in 2015. College students listed as many words as they could in one or two minutes: words beginning with F, A and S, animal names, and curse words. People who produced more of one kind produced more of the others. In a third study of 126 students, taboo fluency correlated positively with openness and neuroticism.

The finding rebuts an old prejudice, that people swear because they lack words. It does not show that swearing signals intelligence. The study measured verbal fluency, not IQ, used only college students, and the authors noted that knowing taboo words is different from using them.

Viral: Staying Up Late

Satoshi Kanazawa and Kaja Perina published "Why night owls are more intelligent" in Personality and Individual Differences in 2009. They used the National Longitudinal Study of Adolescent Health, which measured verbal IQ in adolescence with the Peabody Picture Vocabulary Test and asked about sleep habits years later, when respondents were 18 to 28.

The association held: the highest-scoring group went to bed about 48 minutes later on weeknights than the lowest-scoring group. But the statistical models explained only about 3 to 8 percent of the variation in sleep timing, sleep was self-reported, and the test measured vocabulary rather than general ability. The authors' evolutionary explanation remains a hypothesis. Staying up late will not raise anyone's score.

How to Actually Find Out

Every sign above describes an average across a population. None works as a diagnosis for an individual, and none should be used to label a child. Signs in young children are a separate subject, covered in signs of genius in children; there, as here, a licensed psychologist interprets the scores.

The reliable route is a full-scale test such as the WAIS-IV, given and interpreted by a licensed psychologist; see WAIS-IV explained. A free practice IQ test is a reasonable screener, not a measurement; are online IQ tests accurate explains why. Intelligence also comes in more shapes than one score captures: read emotional intelligence vs IQ and browse the full set of IQ guides.

Editorial standard: IQ figures for historical figures are retrospective estimates and are labeled as such; scores for living people appear only when the person disclosed them or a reputable outlet reported them. Corrections: contact the editors.