What can genes tell us about a person's personality? Twins may resemble each other strongly without being identical, and no single gene dictates extraversion or self-control. To understand those similarities and differences, researchers study many genetic variations with small associations.
The tempting question about a child's future
A child is quiet at a birthday party. A relative says, 'It must be in the genes.' The remark sounds like an explanation, but it skips several questions. Is the child usually quiet or tired today? How is quietness measured? Which genetic differences are being considered, and how much can they tell us about one person? Ted Schwaba, Michel Nivard, Elliot Tucker-Drob and Dorret Boomsma examine what studies of DNA now add to the older study of personality and families. Their answer is more informative, and much less deterministic, than a label attached to a child.
Personality traits describe relatively enduring differences in how people tend to think, feel and act. They are patterns across many situations, not verdicts drawn from one afternoon. The Big Five framework, for instance, groups recurring differences under broad labels such as extraversion and conscientiousness. Questionnaires are useful but imperfect: who answers, in what language and under which circumstances matters. If the trait measure is noisy, a genetic study built on it inherits some of that noise.
Genomics examines variation in DNA sequence across many people. A large association study asks whether particular variants occur a little more often among people with higher or lower measured trait scores. That is a population comparison. It does not find an 'introversion gene' or read a child's future from a sample. The distinction between explaining average variation and predicting an individual is the thread that holds this review together.
From twins to measured DNA
Before researchers could scan huge numbers of genetic variants, they compared twins, adoptees and other relatives. If identical twins tended to resemble one another more than fraternal twins in a trait, that pattern helped estimate a genetic contribution to differences in the studied population. It did not identify the DNA involved. Nor did a heritability estimate mean that a particular person's trait was that percentage genetic. Heritability depends on the people and environments being compared; it is a property of variation in a group.
Measured DNA adds another route. Genome-wide association studies compare variants spread across the genome with personality scores from large samples. Such studies can test whether specific patterns of DNA are associated with a trait and whether those patterns also appear with another outcome. They can examine questions that family resemblance alone cannot settle. But sample size and careful measurement are crucial, because each individual association is tiny. Early studies lacked the statistical power to distinguish many genuine small signals from chance.
The authors treat the older and newer approaches as complementary. Twin studies anticipated that genetic effects would be widespread; genome-wide studies can investigate their distribution and overlap. Agreement between two methods is useful, yet it should not erase their assumptions. Twins may share environments in ways that matter. DNA comparisons can be distorted by population structure or family context. The scientific gain comes from asking where methods converge and where one exposes the other's blind spots.
Thousands of small contributions
A central finding is polygenicity. Differences in common personality traits are associated with very many genetic variants, each contributing a minuscule amount to the average pattern. There is no useful search for one switch that turns extraversion on. A variant associated with a trait usually changes a probability by so little that it says almost nothing about the person carrying it. Even combining variants into a score leaves much of a person's personality unexplained. The score describes statistical tendency in particular data, not a fixed destiny.
A second finding is pleiotropy: some genetic variants associated with personality are also associated with other characteristics, including health and social outcomes. This overlap can suggest shared pathways, but it does not tell us which direction any causal arrow points. A trait may influence a later outcome; an outcome may influence how a trait is reported; both may share influences; or an apparent relation may reflect the sample and the measure. Calling a variant a 'gene for success' because it overlaps with conscientiousness and education would be a dramatic overreading.
The authors also discuss generality. Effects associated with one broad trait can appear across related measures or groups, but the degree of transfer has to be tested rather than assumed. A questionnaire item about being sociable may not mean exactly the same thing in different languages or ages. Research samples have also been unevenly drawn from the world's populations. A genetic score developed in one ancestry group may perform poorly in another. That limitation is scientific and ethical: a model that works mainly for those already well represented cannot be presented as universal.
Why family background can confuse an association
Suppose a genetic variant is more common in one geographical population, and that population also differs in schooling opportunities. An association between the variant and an adult personality score might partly reflect those histories rather than a direct biological effect. Population stratification is the name for this possible mix-up. People also tend to choose partners who resemble them in certain traits or backgrounds, and children inherit both genes and a family environment. These processes can make unrelated variables travel together in a data set.
The review explains ways to probe this problem, including comparisons within families. Siblings share much of their background but differ in which copies of parental variants they inherited. Comparing siblings can therefore reduce some confounding that appears when unrelated people are compared across many families. It is not a magical causal experiment: siblings may experience different environments, and measurement error remains. But a result that survives a within-family check is harder to dismiss as merely a difference between families.
Many recent associations are not dramatically altered by familial environmental confounding, the authors report, although they still consider the issue essential to test. This is a more careful claim than saying that every observed association is a direct biological cause. A genetic variant can influence many steps along a life course. Social responses to a person's behavior can feed back into personality development. Understanding those pathways calls for designs that follow people over time, better measures and more diverse samples, not merely ever larger lists of associated DNA sites.
Association is not a forecast or a cause
A polygenic score sums information from many variants according to associations found in earlier studies. It can be useful for research on broad patterns, but its precision for a single person's personality is limited. A person with a score near one end of a distribution might still show a wide range of behavior. Experiences, age, relationships, opportunities and the way a trait is measured all matter. Even when average group differences are statistically robust, two individuals with different scores can be more alike than the labels suggest.
Nor does a shared genetic association settle whether personality causes a medical or economic outcome. Researchers can use genetic information in designs that ask sharper causal questions, but every design requires assumptions. For example, a variant used as a proxy for one trait may also affect an outcome through a different pathway. If that happens, a causal estimate can mislead. The review values genomic data as a tool for testing models of personality, not as a shortcut past the difficult work of defining mechanisms.
There is also an important practical boundary. A child's DNA should not be used to assign an identity or close off opportunities based on a group-level association. That conclusion follows from the evidence's limited individual prediction and from the fact that traits develop in changing environments. The review does not say genes are irrelevant. It says their many small contributions must be interpreted alongside how people live, what researchers measure and which question the data are capable of answering.
What a better science of personality needs
Genomics can improve personality research when its evidence is joined with good psychology. Broad trait frameworks help scientists compare results instead of studying disconnected questionnaire items. Repeated observations show whether a trait is stable, changes with age or behaves differently across settings. Family and longitudinal designs help separate inherited variation from experiences that travel alongside it. More representative samples test whether a result extends beyond the populations in which it was first discovered.
The authors describe a productive exchange in both directions. Genetic studies give psychologists new ways to examine long-standing questions about trait overlap and development. Personality theory gives genomics meaningful categories and warnings about what an observed score actually represents. Without the theory, a very large data set could produce precise associations with a poorly understood label. Without the data, a plausible personality theory could ignore biological contributions or overestimate what family resemblance tells us.
Return to the quiet child at the party. The honest answer is not 'all genes' or 'no genes.' Genes contribute to differences in personality across populations, usually through many individually tiny associations. They do not tell us why this child is quiet this afternoon, how they will change or what opportunities might help them flourish. The question worth asking is what a particular study measured and whether its method can support the conclusion we are tempted to draw from it.
What to keep
- Personality is highly polygenic.
- Effects overlap with many outcomes.
- Prediction and explanation are different goals.
The nuance
What this story does not let us claim
Each variant has a tiny effect, samples remain unevenly representative, and genetic association is not destiny.
Go deeper: concepts and optional reading practice
Reversible cards
Test the concept
Topic assessment
Check whether you can explain the idea, not merely recognize it.
Every answer is grounded in the explanation above. Some questions introduce new cases to test whether you can apply and analyze the ideas, not merely recall them.
Source & scope
Personality Genomics. Ted Schwaba, Michel G. Nivard, Elliot M. Tucker-Drob & Dorret I. Boomsma. Annual Review of Psychology, vol. 77 (2026).
This editorial guide is an independent explanation. Consult the original publication for the studies, references and full context.
Open the original review ↗