How Heritable Is Intelligence? What the Numbers Really Mean

You may have heard that intelligence is “50% genetic,” “80% genetic,” or something in between. All of those numbers can sound authoritative—and none is a universal percentage for a person.

Intelligence is substantially heritable, but heritability is a population statistic rather than a personal genetic score. Depending on age, sample, test, and research method, estimates vary widely. Twin studies often find that genetic differences account for a larger share of variation in intelligence by late adolescence and adulthood than they do in childhood.

This article is about the number itself: what heritability means, why it changes, and why different methods produce different estimates. For the broader nature-and-nurture question, see Is IQ Genetic or Learned?

What Does “Heritability” Actually Mean?

Heritability describes how much of the variation between people in a particular population is statistically associated with genetic differences in that population.

Suppose a study reports an IQ heritability estimate of 60%. That does not mean 60% of your intelligence came from your genes. It means that, under the conditions of that study, genetic differences accounted for about 60% of the observed differences in the measured trait across the people studied.

The distinction matters because heritability depends on the range of environments and genetic variation present in the sample. Change the population or the conditions, and the estimate can change too.

So What Percentage of Intelligence Is Heritable?

There is no single percentage that works across the lifespan. One of the clearest demonstrations came from a large international twin study using about 11,000 twin pairs from four countries.

Claire Haworth and colleagues estimated the heritability of general cognitive ability at 41% at age 9, 55% at age 12, and 66% at age 17. The increase was approximately linear across those ages.

A later meta-analysis of longitudinal twin and adoption studies reached the same broad conclusion: genetic influences on cognitive ability tend to become more prominent across childhood and adolescence. The analysis covered 11 unique samples and roughly 11,500 twin or sibling pairs.

Why Would Heritability Increase With Age?

This seems backward at first. Children have had fewer years of experience, while adults have accumulated decades of schooling, work, hobbies, and personal choices. You might expect environmental differences to become more important over time.

One explanation researchers study is gene–environment correlation. As people grow, they gain more freedom to choose activities and settings that fit their tendencies. Someone who enjoys verbal challenges may read more, choose language-heavy subjects, and spend time with people who reinforce those interests. A person drawn to spatial problems may seek very different experiences.

Over time, those self-selected environments can amplify earlier differences. The 2013 longitudinal meta-analysis found evidence that genetic influences present earlier in childhood become increasingly amplified after about age 8.

Why Do DNA Studies Give Different Numbers?

Twin and adoption studies infer genetic influence from patterns of resemblance between relatives. Modern molecular-genetic studies can also compare unrelated people directly using measured DNA.

Those methods do not capture exactly the same thing. A study of 6,815 unrelated adults in Generation Scotland estimated that common measured SNPs accounted for about 29% of the variation in general cognitive ability. Earlier genome-wide work using a different sample and method estimated that common genetic variants tagged by SNP arrays accounted for substantial portions of variation in fluid and crystallized abilities as well.

Why can these DNA-based numbers be lower than some twin-study estimates? Common-SNP methods capture only part of the genetic variation that family designs can reflect. Rare variants, imperfectly tagged variants, non-additive effects, and other sources of genetic resemblance can sit outside a particular SNP-based estimate.

This gap is one reason “the heritability of intelligence” is better understood as a family of estimates produced by different methods.

Does High Heritability Mean Intelligence Is Fixed?

No. Heritability tells you about variation in a group; it does not tell you how changeable a trait is.

A trait can be highly heritable and still respond to environmental conditions. Education provides a clear example: a meta-analysis of 42 data sets involving more than 600,000 participants found consistent evidence that additional schooling improves cognitive-test performance. Heritability and environmental responsiveness can coexist.

That also explains why IQ score stability is a separate issue. A trait can show substantial genetic influence while individual scores still shift through development, practice, health changes, or different testing conditions.

Does Heritability Depend on Environment?

Yes, although the pattern is more complicated than a single rule.

A 2016 meta-analysis of nearly 50,000 twins and siblings found that the relationship between socioeconomic status and genetic influence differed by region. In U.S. samples, genetic differences tended to explain more variation in higher-SES environments. In Western European and Australian samples, that pattern was absent or reversed. See the cross-national Gene × SES meta-analysis.

Later large studies have also reported null or mixed results for this interaction. The safest conclusion is that heritability is context-sensitive, while the exact way socioeconomic conditions modify genetic variance is still debated.

Can You Use Heritability to Explain One Person’s IQ?

No. This is probably the most important limitation.

If someone has an IQ score of 120, a heritability estimate cannot tell you how many of those points are genetic. It also cannot tell you what that person would have scored under different educational, developmental, nutritional, or health conditions.

Heritability works at the level of differences within populations. It is useful for understanding why people vary as a group. It is the wrong tool for dividing one individual into “genetic intelligence” and “environmental intelligence.”

Does Heritability Apply Equally to Every Part of IQ?

Different cognitive abilities can have different genetic architectures and different heritability estimates. Verbal knowledge, working memory, processing speed, fluid reasoning, and spatial ability are related, yet they are not identical traits.

That matters when you look at an overall measure such as Full Scale IQ. A single composite can hide a profile with very different strengths underneath it. The Working Memory Index, for example, isolates one narrower part of the broader cognitive picture.

The Number to Remember Is: There Is No One Number

Intelligence is clearly heritable, yet the percentage depends on age, population, environment, test, and method. Twin studies often produce estimates that rise from childhood into young adulthood. Molecular-genetic studies confirm a substantial genetic contribution while capturing a different slice of that variation.

If you want to explore your own performance across several reasoning formats, Cognitive Train’s IQ-Style Test includes letter series, number series, analogies, matrix reasoning, mental rotation, and cube folding. Its score uses Cognitive Train’s own IQ-style scale and should be read as a snapshot of performance on those tasks.

For more ways to separate the abilities that sit underneath an overall score, explore our brain tests and cognitive assessments or browse Cognitive Train’s brain training and cognitive training tools.