Alex Cope, Ph.D.
June 2026
(8 Minutes)

How much does the DNA we inherit from our parents influence our lives? This is a major focus of many human geneticists. It is also a topic that will inevitably garner numerous clickbait headlines. For example, consider the article “Scientists Found the Biggest Factor in How Long You’ll Live. There’s Not Much You Can Do About It,” published in Popular Mechanics. This and many other similar headlines were written about the results of a recent article titled “Heritability of intrinsic human lifespan is about 50% when confounding factors are addressed,” by Shenhar et al., published in the scientific journal Science. 

The idea behind the Shenhar et al. study is that by controlling for extrinsic mortality (think car accidents, murders, etc.), the authors found that the heritability of human lifespan was about 50% using data from a Scandinavian twin study. This was much higher than previous estimates of the heritability of human lifespan, which were typically less than 33%. What does all of that mean? If you were just reading the headlines, you might get the sense that our DNA plays a much greater role in determining how long we live than was previously appreciated.

So, is there really not much you can do to impact your lifespan? Is your DNA really the biggest factor determining how long you live? The result is not quite as clear-cut as some of the headlines would suggest. This is because the concept of heritability is actually very unintuitive. When the average person hears the term “heritability,” a common interpretation is this represents how much of a trait is determined by your DNA. If this were true, then the Shenhar et al. study would mean 50% of your lifespan is due to your DNA. This is a very intuitive understanding of heritability because many of us have a concept of inheritance, even outside the context of genetics. Despite how intuitive this definition seems, it is wrong.

Heritability, which I will denote with the symbol H², is a common measure in the field of quantitative genetics. Heritability H² has to do with the variation in DNA and traits across a population of individuals. Without getting too much into the math, by “variation,” I essentially mean how different the individuals are within a population. (In mathematical terms, this is the “variance”). For example, if you were to estimate the variation in the number of fingers a person has on their left hand across the United States, you can expect a really small number because most people have 5 fingers. In contrast, body metrics like body weight are much more variable. DNA also varies from individual to individual, which is referred to as “genetic variation.” Heritability H² is a number between 0 and 1 (0% and 100%) that represents the amount of genetic variance in a population (specifically, genetic variation relevant to the trait of interest) divided by the amount of variation in the trait itself. If the amount of genetic variation and trait variation are the same, then the trait is 100% heritable (i.e., H² = 1). 

When H² is less than 1, then the remainder of trait variation (i.e., 1- H²) is considered to be the result of non-genetic factors, which is often broadly referred to as the “environment.” When thinking about humans, the environment can be many things. Did your family struggle to afford food while you were growing up? Were you physically active as a child? When you took your neighborhood walks during a presidential election, did you see more signs supporting the Republican or the Democrat? Was there even a neighborhood to walk around, or was your closest neighbor a mile away? All of these factors can contribute to variation in our traits.

Heritability H² was formalized in the 1930s by Dr. Jay Lush, who is considered the father of the science of animal breeding. Heritability H² was originally intended as a tool for the development of breeding programs (think farm animals, crops, etc.). In this context, where breeders can have a strong influence over which individuals reproduce (i.e., artificial selection), heritability can be a useful tool for predicting how a trait will change from generation to generation. However, heritability was soon adopted by human geneticists and social scientists to understand how genetics shapes variation in human physical (like height), cognitive (like IQ), and behavioral (like aggression) traits, with studies suggesting many human traits have a heritability H² greater than 50%. 

What does this mean about the role of genetics in determining our lifespan, or any trait, for that matter? What can a statistic that was intended for animal and plant breeding under controlled breeding programs really tell us about the role of genetics in our lives? The answer is…not much. There are active discussions amongst geneticists regarding the usefulness of heritability H² for humans (see “The heritability fallacy” by Moore and Shenk). Some scientists argue that the very notion of disentangling genetic from environmental effects is pointless, because our DNA and environment interact in complex ways (i.e., the effect of DNA depends on the environment). Even if we set that issue aside, many other issues with heritability are often miscommunicated to the public.

First and foremost, heritability H² tells us nothing about how much of an individual’s trait is determined by genetics. Remember, heritability is all about variation across a population. Even if a trait is 100% determined by your DNA, if everyone has the same DNA (specifically, the same regions of DNA relevant to the trait), then the heritability of the trait is 0%. Consider the number of eyeballs we have. This likely has a strong genetic component because nearly all humans have two eyeballs. Despite this trait being strongly determined by our DNA, its heritability is close to 0% because there seems to be little genetic variation in humans associated with the loss of one or both eyeballs (some genetic disorders can lead to birth defects in eye development, but most variation across individuals is due to injury or disease). 

Second, heritability H² is not a set-in-stone property of genetics or traits. That is to say, it is not an inherent property of biology. It is a summary statistic that varies across populations, time, and environments. For example, some studies suggest heritability of cognitive traits increases with age. Altering the environment is expected to alter the heritability. If everyone in New York City started smoking a pack of cigarettes a day, the heritability of lung cancer for New Yorkers would increase. This would not be because the genetics of lung cancer suddenly changed, but because everyone was now exposed to a risk factor for lung cancer. In this sense, the results of Shenhar et al. are not surprising: if you reduce environmental variation, then statistically, the remaining variation must be genetic.

Third, because heritability H² is a summary of a specific population rather than an inherent property of biology, it has no meaning outside of that population. Were you born and currently live in a Scandinavian country? If not, the heritability estimate from a Scandinavian twin study is probably not relevant to the population you live in. Again, this is because a key component of heritability is the environment, which can vary significantly between populations. But even if you are Scandinavian, heritability tells us nothing about the genetics of a trait for any specific individual!

Fourth, a high heritability H² is not evidence that group differences in a trait are due to genetics. A classic thought experiment by Dr. Richard Lewontin (Lewontin 1970) illustrates the issue. Imagine we have two groups of plants. All plants within a group are grown under the same conditions. However, the groups differ in their environments: one group grows in ideal conditions, while the other grows with less light and nutrients. Within a group of plants, all differences in height are genetic because the environments are the same (i.e., there is no environmental variation). That is, plant height is 100% heritable in both groups. However, the group grown in ideal conditions will be, on average, much taller than the group grown in a less favorable environment. Recent work by my colleagues Dr. Josh Schraiber and Dr. Michael D. Edge formalized Lewontin’s thought experiment, showing mathematically that within-group heritability cannot be used to separate trait differences across groups into genetic and environmental components.

There have been a few technical critiques of Shenhar et al.’s study, but that is not my intent here. It is simply a recent example of how non-experts can misinterpret heritability. Although this is usually harmless, human genetics research can be misappropriated to justify racism, sexism, and other forms of prejudice. For example, some have used the high heritability H² of IQ to argue that gaps in IQ scores between races are biological rather than the result of systematic societal biases. Such arguments are persuasive to non-experts because of the non-intuitive meaning of heritability. Understanding what heritability does, and does not, measure is the best defense against such misuse. Misunderstandings of heritability can also lead to a sense that our traits are fixed by our genetics, as implied by the Popular Mechanics headline (“There’s not much you can do about it”). Remember: non-genetic factors also play a significant role in shaping our traits! Heritability was never intended to tell us what our genes mean for our lives as individuals, and we should be skeptical of any claims to the contrary.

Acknowledgments
Thank you to Dr. Josh Schraiber, Dr. Michael “Doc” Edge, and an anonymous proofreader for helpful comments and suggestions on this article.

Footnotes
¹ A twin study essentially attempts to separate the genetic and environmental influences on a trait by studying twins adopted by different families. The logic goes that because identical twins share 100% of their DNA, whereas fraternal twins only share 50%, traits with a stronger genetic influence should be more similar between identical twins than fraternal twins. These have been subject to much critique.

² Mathematically, how much variation there is in a trait is given by the variance, which I will denote generally with the symbol 𝘝. The heritability 𝘏² of a phenotype is just a ratio of two variances: the amount of variation in their DNA (the genotypes, the genetic variance 𝘝_𝘎) and the total variance in the phenotype of interest (the phenotypic variance 𝘝_𝘗). Importantly, because a phenotype is a function of both genetics and environment (which also varies), the phentoypic variance 𝘝_𝘗 is a combination of the genetic and environmental variance i.e., 𝘝_𝘗 = 𝘝_𝘎+ 𝘝_𝘌 + 2Cov(𝘎,𝘌), where 𝘝_𝘌 is the amount of variance in the environments of the population under consideration. The term Cov(𝘎,𝘌) represents the covariance between genotype and environment within the population. This captures the fact that particular genotypes may be more or less common in certain environments than others. With all of this at hand, the heritability of a trait can be represented as 𝘏² = 𝘝_𝘎/𝘝_𝘗.

³ As Dr. Michael D. Edge pointed out to me, genetic variation in lung cancer is associated with genetic variation for smoking behavior. This makes the hypothetical smoking example more complicated because it means there is a statistical association between genetics and the environmental risk factor (i.e., smoking) for lung cancer. In this case, if everyone starts smoking, there are some scenarios where it could actually dilute the relationship between genetics and lung cancer, decreasing heritability 𝐻². This highlights how environment and genetics can interact in complex ways to lead to differences in heritability across environments.

⁴ I will note that although heritability 𝐻² estimates for IQ are often high (0.6 – 0.8) based on twin-studies, the heritability estimates of IQ (or other measures of cognitive function) from population-level DNA sequencing studies are often much lower (<20%). See Matthews and Turkheimer 2022, as well as this blog post by Dr. Eric Turkheimer.