Heritability is a measure of how much the variation in a trait, across a population, can be attributed to genetic differences between people. It’s expressed as a number between 0 and 1 (or 0% and 100%), where higher values mean genetics explains more of the variation. A heritability of 0.80 for human height, for example, means that 80% of the differences in height between people in that population trace back to genetic differences, while the remaining 20% comes from environmental factors like nutrition and health during childhood.
That single sentence definition is simple enough, but heritability is one of the most commonly misunderstood concepts in genetics. Getting it right changes how you interpret headlines about “the gene for” intelligence, depression, or obesity.
What Heritability Actually Measures
The key distinction is between a trait being genetic and a trait being variable because of genetics. Every human is born with two eyes. That’s genetically determined. But because almost everyone has two eyes, there’s virtually no variation in that trait across the population, which means the heritability of “number of eyes” is essentially zero. The trait is completely inherited, yet it has no heritability. That paradox makes sense once you understand that heritability only cares about differences between people, not about what causes the trait itself.
As the National Institutes of Health puts it: a heritability of 0.7 does not mean a trait is 70% caused by genes. It means 70% of the variability in that trait across a population is due to genetic differences among people. This is a population-level statistic, not an individual-level one. You can’t look at a person who is 6 feet tall and say “genes account for 80% of their height.” Heritability describes patterns across groups, not the breakdown of causes within any single person.
Broad-Sense vs. Narrow-Sense Heritability
Geneticists use two versions of this concept. Broad-sense heritability (written as H²) captures the proportion of trait variation attributable to all genetic causes, including interactions between genes on the same chromosome, interactions between genes on different chromosomes, and the straightforward additive effects of individual genes. Narrow-sense heritability (h²) is more restrictive: it only counts the additive effects, meaning the portion of genetic influence that parents reliably pass to offspring.
Narrow-sense heritability matters more in practical contexts like plant breeding or predicting traits across generations, because only additive genetic effects transfer predictably from parent to child. The complex gene interactions captured by broad-sense heritability can reshuffle with every generation, making them less useful for prediction.
How Scientists Estimate Heritability
The classic method uses twins. Identical (monozygotic) twins share 100% of their DNA, while fraternal (dizygotic) twins share about 50%. If identical twins are more similar for a given trait than fraternal twins, the extra similarity points to genetics. The standard calculation, known as Falconer’s formula, works by subtracting the correlation in fraternal twins from the correlation in identical twins, then multiplying by two. If identical twins correlate at 0.90 for height and fraternal twins at 0.45, the heritability estimate is 2 × (0.90 − 0.45) = 0.90, or 90%.
Modern genetics has added a second approach: scanning the genome directly. Researchers look at hundreds of thousands of common genetic variants across large groups of people and calculate how much of a trait’s variation those variants collectively explain. This method consistently produces lower estimates than twin studies, a gap scientists call “missing heritability.” In dairy cattle, where this has been studied extensively, pedigree-based heritability for milk yield is about 0.25 while genomic heritability is only 0.14, a gap of 43%. For milk protein percentage, the gap shrinks to just 11%. The missing portion likely reflects rare genetic variants that current gene chips don’t capture, interactions between genes, and interactions between genes and the environment.
Why Heritability Changes With the Environment
A heritability estimate is always specific to one population in one environment at one point in time. Change the environment, and the number shifts. If everyone in a population ate exactly the same diet and had identical access to healthcare, the environmental contribution to height variation would shrink and heritability would rise. Conversely, in a population with extreme nutritional inequality, environmental factors would explain more of the variation, pushing heritability down.
This is why heritability estimates for the same trait differ across countries, time periods, and socioeconomic contexts. It also means you can’t take a heritability number from one setting and apply it to another. A heritability of 0.80 for height in well-nourished Northern European populations doesn’t necessarily hold in populations facing food insecurity. The overall range of twin and family estimates for human height spans 30% to 90%, with most estimates in well-resourced populations clustering toward the upper end.
Heritability of Intelligence
Intelligence is one of the more striking examples of how heritability shifts across a lifetime. In infancy, the heritability of intelligence is only about 20%. It climbs to roughly 40% in childhood, 50% in adolescence, and 60% in young adulthood. Some evidence suggests it reaches as high as 80% in later adulthood before declining to around 60% after age 80.
This increasing pattern seems counterintuitive. You might expect genes to matter most early in life, before experience accumulates. But the opposite appears to be true, likely because as people age, they gain more freedom to seek out environments that match their genetic predispositions. A child’s environment is largely chosen by their parents, but an adult selects their own career, hobbies, and social circle, amplifying the influence of innate tendencies. Interestingly, twin studies show strong genetic continuity across ages: the genetic factors influencing intelligence at age 7 correlate about 0.75 with those at age 12, even though heritability itself increases and the brain undergoes significant structural changes.
Heritability of Mental Health Conditions
Psychiatric conditions tend to have substantial heritability. Twin studies estimate heritability for schizophrenia at around 80%, and bipolar disorder falls in a similar range. When researchers use national population data from Scandinavian family and adoption records instead of twin studies, the estimates come in somewhat lower, around 60%. The consensus places both conditions in the 60% to 80% range.
High heritability doesn’t mean these conditions are inevitable for people with a family history. It means that within the population, genetic differences account for most of the variation in who develops the condition and who doesn’t. Environmental triggers, life experiences, and random developmental processes still play a meaningful role for any given individual.
Heritability of Personality
Twin studies estimate that 40% to 60% of the variation in the Big Five personality traits (openness, conscientiousness, extraversion, agreeableness, and neuroticism) is heritable. When researchers try to account for this using common genetic variants identified through genome scanning, the numbers are much smaller. Only openness (21%) and neuroticism (15%) showed significant heritability from common variants in one large study. Extraversion, agreeableness, and conscientiousness did not reach statistical significance through this method. The gap between twin-based and genomic estimates follows the same missing heritability pattern seen in other traits.
What Heritability Does Not Tell You
Heritability says nothing about how fixed or changeable a trait is. A trait can be highly heritable and still respond dramatically to environmental intervention. Nearsightedness is substantially heritable, yet it’s easily corrected with glasses. Height is one of the most heritable human traits, yet average heights have increased dramatically over the past century thanks to improved nutrition, not genetic change.
Heritability also cannot explain differences between groups. If two populations differ in average scores on some trait, that difference could be entirely environmental even if the trait is highly heritable within each group. Imagine two fields of genetically identical seeds: one planted in rich soil, the other in poor soil. Within each field, height differences between plants would be 100% genetic (since the environment is the same for all plants in a field). But the difference between the two fields would be 100% environmental. Heritability within a group tells you nothing about the causes of differences between groups.
Finally, heritability is not destiny at the individual level. Knowing that intelligence has a heritability of 60% in adulthood tells you something useful about the population-level role of genetics, but it cannot predict any one person’s cognitive trajectory. Your genes set a range of possibilities. The life you live determines where in that range you land.