Is Intelligence Hereditary?

Is intelligence hereditary? The short answer

Partly, and more than many people expect. For about a century researchers have compared the test scores of identical twins, fraternal twins, siblings, parents and adopted children. The pattern is consistent: the more genes two people share, the more alike their IQs tend to be. Across modern studies, genetic differences account for something like half of the spread in scores within the populations studied, and more than that by adulthood.

So "IQ is genetic" is half right, and often misunderstood. A heritability of 50% does not mean half of your intelligence came from your parents and half from your school. It does not mean your score was set at conception. And it says nothing about why one country or group might score differently from another.

What heritability means (and what it doesn't)

Heritability is a statistic about a group. It estimates how much of the variation in a trait among people in a particular population, at a particular time, lines up with genetic differences between them. A review of the concept in Nature Reviews Genetics calls it a "dimensionless population parameter" (Visscher, Hill and Wray, 2008). Four consequences follow.

• It is not a share of one person. Genes and experience build every ability together, so no part of your IQ is "the genetic half". A review of 10 robustly replicated findings in behavioural genetics stresses that heritability "does not refer to a single individual but rather to individual differences in a particular population" (Plomin and others, 2016).
• High heritability does not mean fixed. The same review states plainly that "heritability does not imply immutability". Height is the classic example: about 80% heritable, yet average height has risen markedly in many populations, most likely because of better nutrition and medical care (Visscher and colleagues). A trait can owe a lot to genes and still respond to the right help, in the way glasses correct poor eyesight.
• It can change with circumstances. Heritability describes "what is" in one population, not "what could be". Because it depends on how varied both genes and surroundings are in that group, the estimate can differ between places and eras. Our page on heritability and family income shows how the estimate has varied with family circumstances in some countries but not others.
• It says nothing about differences between groups. Visscher and colleagues list this as a common misconception: heritability "should not be used to make predictions about mean changes in the population over time or about differences between groups". Their example is height again. In the mid-19th century, white American men were about 9 cm taller than Dutch men; by the end of the 20th century, Dutch men were about 5 cm taller, a reversal they describe as likely environmental rather than genetic. By the same logic, average IQ scores could climb for decades in the Flynn effect even though IQ is substantially heritable.

How researchers separate genes from environment

Families share both genes and homes, so researchers look for comparisons where one varies more than the other.

• Twin studies. Identical twins share all their inherited DNA; fraternal twins share about half of the variants that differ between people, like any brothers and sisters. If identical pairs are more alike in IQ than fraternal pairs raised in the same homes, the extra similarity points to genes. A meta-analysis of almost every twin study published over 50 years, covering 17,804 traits and 14,558,903 partly overlapping twin pairs, found an average heritability of 49% across all traits, from height to personality to illness (Polderman and others, 2015).
• Adoption studies. Adopted children share genes with one family and a home with another. These studies show genes matter, and that a better-resourced home can lift scores, as our guide to increasing IQ sets out.
• Twins raised apart. Identical twins separated early share genes but not homes. The best-known project, from Minnesota, is covered on our page about twins reared apart, which linked roughly 70% of IQ variation to genetic differences.

Each design has weak points: identical twins may be treated more alike than fraternal twins, for example, and parents often choose partners of similar ability. The DNA methods below rest on different assumptions, so where they agree, the conclusion is sturdier.

How alike are relatives' IQs?

The most quoted summary of family resemblance comes from a 1981 review in Science, which pooled 111 studies, 526 correlations and 113,942 pairs of relatives (Bouchard and McGue, 1981). The chart shows its weighted average correlations. A correlation of 1 would mean pairs always matched; 0 would mean no resemblance at all.

Two patterns stand out. First, resemblance climbs with genetic closeness: cousins barely resemble each other, siblings and parent and child are moderately alike, identical twins most of all. The authors judged the overall profile "consistent with a polygenic mode of inheritance", meaning many genes each adding a little. Second, the environment shows up too. Identical twins raised apart (0.72) are less alike than those raised together (0.86), siblings raised apart less alike than those raised together, and adopted children who share nothing but a home still correlate 0.34.

Read these numbers with care. They are averages over studies that varied a lot, the review describes "a marked degree of heterogeneity", and some rows rest on small samples: only 65 pairs of identical twins raised apart and 203 pairs of separated siblings. The data are also more than 40 years old, and the effect of a shared home on IQ fades during adolescence, as the next section explains. As a rough illustration, doubling the gap between identical (0.86) and fraternal (0.60) twins gives about 0.52, close to the "about half" that later studies report.

Why heritability rises with age

Surprisingly, genetic influence on IQ grows as people get older. Our page on how heritability changes from childhood to adulthood walks through a study of 11,000 twin pairs in which the estimate climbed from 41% at age 9 to 66% at 17, and our guide to what IQ is traces the wider rise from about 20% in infancy to perhaps 80% later in adult life.

At the same time, the influence of the shared family home on IQ is found "up until adolescence and then diminishes as adolescents begin to make their own way in the world" (Plomin and others, 2016).

The leading explanation is not that new genes switch on. A review of the genetics of intelligence argues that "small genetic differences are magnified as children select, modify and create environments correlated with their genetic propensities" (Plomin and Deary, 2015). A child who enjoys reading asks for more books; over years, such choices add up. Rank order on IQ tests also stays fairly steady across life (see IQ stability from 11 to 90).

Finding the genes: DNA studies

Twin studies never look at DNA. Since the early 2010s, researchers have been able to measure hundreds of thousands of genetic markers directly and ask how much of the variation in IQ they capture.

SNP heritability: the DNA-based estimate

An early study of 3,511 unrelated adults from England and Scotland, with data on 549,692 common genetic markers (SNPs), estimated that these markers captured 40% of the variation in crystallized-type intelligence and 51% in fluid-type intelligence (Davies and others, 2011). The authors treated these as lower bounds, since the method misses rarer variants, and saw the results as consistent with "many genes of small effects" (see fluid vs crystallized intelligence for the two types).

DNA-based estimates usually come in below twin estimates. In one sample analysed both ways, the figure for intelligence was 0.35 from DNA against 0.47 from twins (Plomin and Deary, 2015). The difference is called missing heritability. Part of it is probably rarer variants that standard genetic chips miss; Plomin and Deary expected whole-genome sequencing to fill some of the gap.

No single "IQ gene"

Genome-wide studies of hundreds of thousands of people have found hundreds of regions of the genome linked to intelligence, but each one shifts scores by a minute amount, and a combined "polygenic score" still predicts only a few per cent of the differences between individuals. Our page on polygenic scores and IQ gives the figures from the largest studies. The finding fits one of the best-replicated results in behavioural genetics: heritability "is caused by many genes of small effect" (Plomin and others, 2016).

Comparing brothers and sisters

Population-wide genetic links can be inflated by things that travel with genes, such as family wealth or parents' own behaviour. Siblings share parents and a home but inherit different variants, so comparing them strips much of this out. A study of 178,086 siblings from 19 cohorts found that genetic associations with cognitive ability shrank by about 22% within families, and the DNA-based heritability estimate fell from 0.24 to 0.14 (Howe and others, 2022). Genes still mattered within families, but part of the apparent genetic signal was really about the family environment that goes with those genes.

Is intelligence inherited from your mother?

A viral claim says children, or at least sons, get their intelligence from their mother because "intelligence genes" sit on the X chromosome. The evidence does not support it.

Where it came from. According to a Snopes fact check, the story spread in September 2016 through an article on a site called Second Nexus, which drew on a March 2016 blog post from Psychology Spot and a Cosmopolitan/Good Housekeeping piece citing that same post. Snopes found no new research behind it (only 1 of the blog's 14 references was from the 2010s) and rated the claim "Unproven" (Kasprak, 2016). A KQED science column reached the same verdict. A co-author of the study the viral post seemed to be citing told KQED that it had no data on fathers' IQ at all; it was a study of breastfeeding and IQ, the topic of our breastfeeding page (Starr, 2016).

What the science says.

• Both parents pass on X chromosomes. A mother always gives her child an X; a father gives a son his Y but a daughter his X. So even if key genes were on the X, daughters would inherit them from both parents.
• The X has a real but limited role. Genes on the X chromosome appear over-represented among those known to cause intellectual disability, and some researchers have suggested this could add to the slightly wider spread of male scores (Johnson, Carothers and Deary, 2009). Commentators pointed out that estimates of how many intelligence genes lie on the X rest on "doubtful premises" (Craig, Haworth and Plomin, 2009). (See sex differences in IQ spread.)
• Family data show no maternal bias. The 1981 review of 111 family studies found no consistent pattern when it compared male, female and mixed-sex pairs of relatives, and said the absence of any sex effect fits a polygenic model "that does not posit the existence of sex linkage" (Bouchard and McGue, 1981).

A mother's role in the early home is a separate question from genetics, and fathers supply half of a child's genes either way.

Is reaction time genetic?

Yes, to a moderate degree. Reaction time is how fast you respond to a simple signal; our processing speed guide explains how it is measured and how it relates to IQ.

• In Dutch 16-year-old twins, heritability was 64% for simple reaction time and 62% for choice reaction time. The modest links between reaction time and IQ in that sample were explained by genetic influences (Rijsdijk, Vernon and Boomsma, 1998). Estimates were lower when the twins were retested at 18, which the authors put down mainly to changes in how the tasks were given.
• In 166 identical and 190 fraternal twin pairs in Australia, the genetic correlation between choice reaction time and IQ was −0.45 to −0.70, meaning much of the genetic influence on speed overlapped with genetic influence on IQ (Luciano and others, 2001). The study also found that 43% of the genetic variance in IQ was unrelated to speed or working memory.
• Perceptual speed, measured through inspection time (how briefly a shape must be shown for you to judge it), was 46% heritable in a Dutch sample of 688 adults from 271 families, and its small links with IQ were entirely due to shared genetic factors (Posthuma, de Geus and Boomsma, 2001).

On any one day, sleep, attention, practice and equipment move reaction time too, so a single slow result says little about your genes.

Genes and environment work together

The nature versus nurture debate assumes the two can be pulled apart. For intelligence, research shows they are tangled together.

• Parents pass on genes and a home. So a link between, say, books at home and a child's IQ can reflect either. In one adoption study, a measure of the home environment correlated 0.44 with 2-year-olds' mental development in ordinary families but 0.29 in adoptive families, where parents and children share no genes (Plomin and others, 2016).
• Children shape their surroundings. Parenting partly reflects children's own characteristics, and environments are chosen as well as given. Studies have found genetic influence on measures of family, neighbourhood, school and work environments, and a review of 55 studies found an average heritability of 0.27 across 35 such measures (Plomin and others, 2016).
• Like marries like. Spouses correlate about 0.40 for intelligence, far more than for personality (about 0.10) or height and weight (about 0.20), and partly because people pair up by education (Plomin and Deary, 2015). This "assortative mating" increases genetic variation in the next generation, one reason heritability for IQ is high.

True or false?

Is each statement true or false? Make your call, then check the evidence.

What this means for you

Are you born with your IQ? You are born with genetic tendencies that make some kinds of learning easier or harder, but not with a number. The environments you meet, and later choose, help decide how those tendencies show up. Twin research makes the same point from the other direction: no behavioural trait is anywhere near 100% heritable, which the 2016 review calls "the strongest available evidence for the importance of environmental influence".

• Your parents' scores are a loose guide, not a ceiling.
• Environment still counts. Schooling, a stimulating home and good health have measurable effects, especially in childhood; our guide to increasing IQ sets out which ones have evidence behind them.
• DNA tests can't tell you your IQ. Current polygenic scores predict only a small fraction of the differences between individuals.
• Measure the ability, not the ancestry. To see where you stand now, take a well-built test when rested. You can try our free IQ test, which tests fluid, logical, spatial and numerical reasoning; our average IQ guide helps you read the result.

Frequently asked questions

Sources

• Visscher, P. M., Hill, W. G. and Wray, N. R. (2008). Heritability in the genomics era: concepts and misconceptions. Nature Reviews Genetics, 9(4), 255–266.
• Plomin, R., DeFries, J. C., Knopik, V. S. and Neiderhiser, J. M. (2016). Top 10 replicated findings from behavioral genetics. Perspectives on Psychological Science, 11(1), 3–23.
• Plomin, R. and Deary, I. J. (2015). Genetics and intelligence differences: five special findings. Molecular Psychiatry, 20(1), 98–108.
• Bouchard, T. J. and McGue, M. (1981). Familial studies of intelligence: a review. Science, 212(4498), 1055–1059.
• Polderman, T. J. C., Benyamin, B., de Leeuw, C. A., Sullivan, P. F., van Bochoven, A., Visscher, P. M. and Posthuma, D. (2015). Meta-analysis of the heritability of human traits based on fifty years of twin studies. Nature Genetics, 47(7), 702–709.
• Davies, G. and others (2011). Genome-wide association studies establish that human intelligence is highly heritable and polygenic. Molecular Psychiatry, 16(10), 996–1005.
• Howe, L. J. and others (2022). Within-sibship genome-wide association analyses decrease bias in estimates of direct genetic effects. Nature Genetics, 54(5), 581–592.
• Johnson, W., Carothers, A. and Deary, I. J. (2009). A role for the X chromosome in sex differences in variability in general intelligence? Perspectives on Psychological Science, 4(6), 598–611.
• Craig, I. W., Haworth, C. M. A. and Plomin, R. (2009). Commentary on "A role for the X chromosome in sex differences in variability in general intelligence?" (Johnson et al., 2009). Perspectives on Psychological Science, 4(6), 615–621.
• Rijsdijk, F. V., Vernon, P. A. and Boomsma, D. I. (1998). The genetic basis of the relation between speed-of-information-processing and IQ. Behavioural Brain Research, 95(1), 77–84.
• Luciano, M., Wright, M., Smith, G. A., Geffen, G. M., Geffen, L. B. and Martin, N. G. (2001). Genetic covariance among measures of information processing speed, working memory, and IQ. Behavior Genetics, 31(6), 581–592.
• Posthuma, D., de Geus, E. J. C. and Boomsma, D. I. (2001). Perceptual speed and IQ are associated through common genetic factors. Behavior Genetics, 31(6), 593–602.
• Kasprak, A. (2016). Intelligence is inherited only from your mother? Snopes, 11 October 2016.
• Starr, B. (2016). Debunking the idea that a kid's intelligence comes mostly from mom. KQED Future of You, 29 September 2016.