Twin studies consistently show that sexual orientation is influenced by genetics but not determined by it. When one identical twin is gay, the other is substantially more likely to be gay than the general population would predict, but the match is far from guaranteed. Concordance rates for identical twins across major studies range from roughly 20% to 65%, depending on how the sample was recruited. That gap between shared DNA and shared orientation has driven decades of research into what else shapes sexual development, from prenatal hormones to epigenetics to the unique social experiences that diverge even between twins who grew up in the same household.
What Twin Concordance Rates Tell Us
The basic logic of twin research on sexual orientation is straightforward: identical twins share virtually all their DNA, while fraternal twins share about half. If genes were the whole story, identical twins would almost always match. If genes played no role, identical and fraternal twins would match at similar rates. The reality falls in between, and the exact numbers have shifted as research methods improved.
Early studies in the 1990s, which recruited participants through gay-community publications, found strikingly high concordance. One widely cited study of male twins reported that about 52% of identical co-twins of gay men were also gay, compared with 22% of fraternal co-twins and 11% of adoptive brothers.1JAMA Psychiatry. A Genetic Study of Male Sexual Orientation A parallel study of female twins found that roughly 48% of identical co-twins of lesbian women were also lesbian, versus 16% of fraternal co-twins.2JAMA Psychiatry. Heritable Factors Influence Sexual Orientation in Women Another report on 61 twin pairs found an even higher concordance of about 66% in identical twins and 30% in fraternal twins.3PubMed. Homosexual orientation in twins: a report on 61 pairs and three triplet sets
These numbers are high partly because of how participants were found. When you advertise in gay publications for twins to volunteer, you tend to attract pairs where both twins are gay, inflating concordance. Later population-based studies, which drew from national twin registries instead of self-selected volunteers, found lower but still meaningful concordance. A U.S. national twin sample confirmed that resemblance for sexual orientation was greater between identical twins than between fraternal twins, and modeling suggested a substantial genetic contribution alongside possible family-environment effects.4PubMed. Sexual orientation in a U.S. national sample of twin and nontwin sibling pairs The best current estimate is that genetics account for somewhere between a quarter and a third of the variation in sexual orientation, with the rest coming from environmental factors that researchers are still working to pin down.
Why Identical Twins Can Have Different Sexual Orientations
The fact that identical twins frequently differ in sexual orientation is one of the most revealing findings in this area. These are people who share essentially the same genome, usually grew up in the same household, and yet developed differently. That tells us something important: whatever non-genetic factors matter, they are not about how your parents raised you in any broad sense, because both twins got the same parenting. Instead, the relevant factors seem to be what researchers call “non-shared environment,” meaning experiences and exposures unique to each twin.
A systematic review of twin studies on sexual identity found that even identical twins raised together develop different sexual identities partly because of individual differences in their reactions, social circles, hobbies, and other experiences that diverge over time.5Journal of BioMed Research and Reports. Comparison of Sexual Identity in Identical Twins: A Systematic Review Finnish twin data reinforced this picture, finding that while genetics contributed to both childhood gender-atypical behavior and adult sexual orientation, nonshared environmental effects explained the remaining variation.6PubMed. Common genetic effects of gender atypical behavior in childhood and sexual orientation in adulthood: a study of Finnish twins
Importantly, “non-shared environment” is a statistical category, not a specific mechanism. It could include anything from differences in prenatal hormone exposure (even identical twins do not experience the womb identically) to childhood friendships to random biological variation during brain development. It is a reminder that when people ask “is it nature or nurture,” the answer for sexual orientation is genuinely both, and the “nurture” part is not what most people assume.
The Prenatal Hormone Window
One of the most studied non-genetic influences is the hormonal environment in the womb, and twin research has been central to investigating it. Even identical twins can differ in their exposure to androgens during fetal development, and several lines of evidence suggest these differences matter for sexual orientation.
The ratio between the lengths of the second and fourth fingers, known as 2D:4D, is commonly used as a rough marker of prenatal androgen exposure. A lower ratio suggests more androgen exposure; a higher ratio suggests less. In studies of identical twin pairs where one twin is gay or lesbian and the other is straight, the non-heterosexual female twin consistently showed a lower, more masculinized 2D:4D ratio than her heterosexual co-twin.7PubMed. Finger-length ratios in female monozygotic twins discordant for sexual orientation This pattern held across multiple studies. A Japanese twin sample found the same left-hand difference in discordant female pairs, while non-heterosexual male twins actually showed more feminized ratios than their straight co-twins.8PubMed. The second to fourth digit ratio (2D:4D) in a Japanese twin sample: heritability, prenatal hormone transfer, and association with sexual orientation A later study of 18 discordant female pairs confirmed the masculinized pattern in non-straight female twins’ left hands, while discordant male pairs showed no significant difference.9PubMed. Finger Length Ratios of Identical Twins with Discordant Sexual Orientations
Because identical twins share the same DNA, these finger-ratio differences point directly at the prenatal environment rather than genetics. The evidence is stronger and more consistent for women than for men, which fits a broader pattern in this field: the biological pathways to male and female homosexuality may not be the same.
Another well-studied prenatal mechanism involves the maternal immune system. Gay men tend to have more older brothers than straight men do, and research has identified a plausible biological explanation. Some mothers appear to develop antibodies against a protein called NLGN4Y, which is encoded on the Y chromosome and plays a role in male brain development. With each successive male pregnancy, the maternal immune response may grow stronger, potentially influencing sexual development in later-born sons. A direct test of this hypothesis found that mothers of gay sons, especially those with older brothers, had significantly higher levels of anti-NLGN4Y antibodies than mothers of straight sons.10PubMed Central. Male homosexuality and maternal immune responsivity to the Y-linked protein NLGN4Y This mechanism would not apply to firstborn sons or to women, and it probably accounts for only a fraction of male homosexuality, but it illustrates how biological processes outside the genome can shape orientation.
The Search for Genetic Variants
If twin studies show that genes matter, the natural next question is: which genes? The search has been long, and the honest summary is that no single “gay gene” exists, but many small genetic influences likely contribute.
The story began in 1993 with a report that DNA markers on a region of the X chromosome called Xq28 were linked to male sexual orientation in a study of 40 families with two gay brothers. About 64% of brother pairs shared the same Xq28 markers, significantly more than chance would predict.11PubMed. A linkage between DNA markers on the X chromosome and male sexual orientation A follow-up study two years later confirmed the Xq28 linkage in new families with gay male siblings but found no such linkage in families with lesbian sisters, suggesting sex-specific genetic pathways.12PubMed. Linkage between sexual orientation and chromosome Xq28 in males but not in females
The Xq28 finding generated enormous public interest but proved difficult to replicate consistently. The field eventually moved toward genome-wide approaches, which scan the entire genome rather than focusing on one region. The largest such study to date analyzed genetic data from nearly half a million people and identified five spots in the genome significantly associated with same-sex sexual behavior. But the key finding was how little any of them mattered individually. All tested genetic variants combined accounted for only 8 to 25% of the variation in same-sex behavior, the genetic effects partially differed between men and women, and the variants could not meaningfully predict any individual’s sexual orientation.13PubMed Central. Large-scale GWAS reveals insights into the genetic architecture of same-sex sexual behavior
More recently, a study in Han Chinese populations identified two additional genetic loci associated with male sexual orientation, and a combined analysis across Chinese and European samples revealed three more genome-wide significant locations.14Cell Discovery. Discovery of new genetic loci for male sexual orientation in Han population The picture that emerges is of a highly polygenic trait, one shaped by many genetic variants of tiny individual effect, similar to height or personality traits. There is no master switch to find.
The Epigenetic Hypothesis
One model attempts to bridge the gap between the substantial heritability seen in twin studies and the failure to find major DNA variants. It centers on epigenetics, the chemical modifications that sit on top of DNA and regulate which genes get turned on or off without changing the underlying genetic code.
The idea, proposed in 2012, focuses on sex-specific epigenetic marks that normally help buffer developing fetuses from the wrong level of androgen exposure. These marks boost testosterone sensitivity in male fetuses and dampen it in female fetuses. Ordinarily they are erased between generations. The model proposes that when particularly strong marks from one parent carry over to an opposite-sex child, they could shift sexual development, including sexual orientation.15PubMed. Homosexuality as a consequence of epigenetically canalized sexual development The researchers have proposed a testing protocol for this model, suggesting that the relevant epigenetic marks could be measured in specific tissues and compared between concordant and discordant twin pairs.16PubMed Central. Homosexuality via canalized sexual development: a testing protocol for a new epigenetic model
This remains a theoretical model. Direct evidence that specific epigenetic marks transmit sexual orientation across generations has not yet been produced. But the framework is appealing because it explains why identical twins can differ (epigenetic marks vary even between twins), why heritability is substantial (epigenetic patterns can run in families), and why no major DNA variants have been found (the relevant variation may sit on top of the DNA, not in it).
Brain Processing Differences in Twins and Beyond
Research on the brain has revealed that sexual orientation correlates with measurable differences in how the brain responds to certain stimuli, and some of these differences align with the prenatal-hormone story.
A brain imaging study found that when gay men were exposed to a chemical compound derived from testosterone (AND), their hypothalamus activated in a pattern resembling that of heterosexual women rather than heterosexual men. Common odors, by contrast, were processed identically across all groups.17PubMed Central. Brain response to putative pheromones in homosexual men A companion study showed the reverse pattern in lesbian women: they processed the testosterone-derived compound through ordinary smell pathways rather than the hypothalamus, and they partly shared heterosexual men’s hypothalamic response to an estrogen-derived compound.18PubMed Central. Brain response to putative pheromones in lesbian women
These findings do not prove that brain wiring causes sexual orientation, and they do not tell us when the differences arose. But they do confirm that sexual orientation is accompanied by real, measurable differences in how the brain is organized, particularly in regions tied to sexual behavior. And the cross-sex pattern, where gay men’s brains resemble straight women’s and lesbian women’s brains resemble straight men’s, is consistent with the idea that prenatal hormones play a role in shaping both brain function and orientation.
Even physiological arousal patterns differ in the expected direction. A study that measured genital arousal in identical twins with discordant sexual orientations found that these twins differed from each other in the same way unrelated gay and straight people differ, confirming that self-reported discordance is not just a matter of labeling but reflects genuine differences in sexual response.19Scientific Reports. Sexual Arousal Patterns of Identical Twins with Discordant Sexual Orientations
Visible Differences From Childhood
If you could look at childhood photos of identical twins who later turned out to have different sexual orientations, could you tell them apart? Surprisingly, the answer leans toward yes. Researchers who had independent raters evaluate childhood photographs of discordant identical twin pairs found that the twin who later identified as non-straight was visibly more gender-nonconforming from mid-childhood onward. The difference was smaller than the gap between pairs where both twins were straight and unrelated pairs where both were gay, but it was detectable.20PubMed. Gender nonconformity of identical twins with discordant sexual orientations: Evidence from childhood photographs
Handedness offers another subtle physical marker. An early study found an increased rate of left-hand preference in homosexual women, with a similar trend in homosexual men, suggesting a link between sexual orientation and brain lateralization.21Psychoneuroendocrinology. Left-handedness in homosexual men and women: Neuroendocrine implications Handedness is established prenatally, which again points back to the womb rather than to social learning as a source of the difference.
Studies of the auditory system reinforce this picture. Faint sounds produced by the inner ear, called otoacoustic emissions, tend to be stronger in women than in men. Homosexual and bisexual women were found to produce emissions that fell between those of heterosexual women and heterosexual men, consistent with partial masculinization of the auditory system.22PubMed. Comparison of the auditory systems of heterosexuals and homosexuals: click-evoked otoacoustic emissions No such difference was found between homosexual and heterosexual men. A follow-up study broadly confirmed the pattern, finding correlations most consistent with undermasculinization in gay men and overmasculinization in lesbian women.23PubMed. Otoacoustic emissions, auditory evoked potentials, and traits related to sex and sexual orientation
The Evolutionary Puzzle
If sexual orientation has a genetic component, a natural question is why natural selection has not eliminated genes associated with homosexuality, given that gay individuals tend to have fewer biological children. One long-standing idea, the sexually antagonistic genes hypothesis, proposes that genetic variants that reduce reproduction in one sex might boost it in the other, keeping those variants in the population. For example, a gene variant that makes a man more likely to be gay might make his sisters more fertile.
A recent test of this hypothesis in over 2,000 individuals from Czechia and Slovakia confirmed that gay men and lesbian women had significantly fewer children than straight people, as expected. But it found no fertility advantage among their opposite-sex relatives. Sisters of gay men were not more fertile than sisters of straight men, and brothers of lesbian women were not more fertile than brothers of straight women.24Evolution and Human Behavior. An evolutionary perspective on homosexuality: Testing the sexually antagonistic genes hypothesis through familial fertility analysis The sexually antagonistic model, while elegant, did not hold up in this sample.
Whatever the evolutionary explanation turns out to be, same-sex sexual behavior is hardly unique to humans. A large-scale analysis found it documented in 261 mammalian species across more than half of all mammalian families. Both males and females engage in it, and over half the species with recorded same-sex behavior show it in both sexes.25Nature Communications. The evolution of same-sex sexual behaviour in mammals That breadth suggests same-sex behavior is not a quirk of human culture or psychology but something deeply embedded in mammalian biology, making a purely social explanation for human homosexuality hard to sustain.
Why the Male-Female Distinction Keeps Coming Up
One pattern that runs through nearly all of this research is that male and female homosexuality may not share the same biological story. The Xq28 linkage was found in gay men but not lesbian women. The large genome-wide study found genetic effects that only partially overlapped between the sexes. Finger-ratio data are stronger and more consistent for lesbian women than for gay men. The otoacoustic emission findings appeared in women but not men. Even the maternal immune hypothesis applies only to men with older brothers.
This does not mean that male and female homosexuality are unrelated, but it does mean that lumping them together as a single phenomenon and looking for a single explanation is probably a mistake. The biological inputs likely differ by sex, even if the resulting experience of same-sex attraction feels broadly similar. For researchers, this has practical implications: a study that mixes men and women without analyzing them separately may miss patterns that are real in one group and absent in the other. For everyone else, it is a reminder that “why are people gay” is probably not one question but at least two, and the answers may only partly overlap.