How sexual orientation develops has been intensely studied for decades. Researchers seek to understand the relative contributions of genetics and various environmental factors in shaping this complex human trait. Twin studies emerged as a primary research method because they offer a natural experiment for separating genetic influences from environmental ones. By comparing individuals with varying degrees of genetic similarity who share a common environment, scientists can generate estimates for the heritability of a trait. This approach suggests that while genes play a role, they do not tell the whole story of how a person’s sexual orientation is established.
The Research Model Using Twins
The twin study model compares two types of twins who develop under different genetic conditions. Monozygotic (MZ), or identical, twins originate from a single fertilized egg that splits, resulting in individuals who share virtually 100% of their DNA. Dizygotic (DZ), or fraternal, twins develop from two separate eggs fertilized by two different sperm, sharing about 50% of their DNA. Both types of twins typically share the same prenatal environment and are raised within the same family structure, allowing researchers to control for shared environment.
The core measurement in this research is the concordance rate, which is the probability that if one twin exhibits a certain trait, the other twin will also exhibit it. If a trait is purely genetic, the concordance rate for MZ twins should theoretically be 100%, and significantly higher than the rate for DZ twins. Conversely, if a trait is purely environmental, the concordance rates for both MZ and DZ twins should be approximately the same. By mathematically comparing the difference in concordance rates between MZ and DZ pairs, researchers can estimate the degree to which a trait is influenced by genetic factors.
Findings on Genetic Influence
Early studies on male sexual orientation reported high concordance rates. These studies, which often recruited participants through advertisements in specialized publications, found that if one identical twin was gay, there was an approximately 52% chance the co-twin would also be gay. The concordance rate for fraternal twins was significantly lower, about 22%, providing strong evidence for a heritable component in male sexual orientation. However, the non-100% concordance rate for identical twins immediately indicated that genetics alone are insufficient to determine sexual orientation.
More recent population-based studies, which draw samples from twin registries to minimize recruitment bias, have yielded lower, but still significant, estimates. For example, a large Swedish study found that genetic factors accounted for approximately 34% to 39% of the variance in same-sex sexual behavior for men. The estimate was lower for women, with genetic factors explaining about 18% to 19% of the variance in female same-sex behavior. These findings reinforce the conclusion that there is a moderate genetic contribution to sexual orientation in both sexes, appearing somewhat stronger in men than in women.
The fact that identical twins, who share all of their DNA, do not always share the same sexual orientation is highly informative. This discordance suggests that the remaining influences must be attributed to non-shared environmental factors, which encompasses unique life experiences and biological factors that are not strictly genetic. The shared family environment, or how twins are raised together, generally accounts for a very small proportion of the variance. This realization shifted the focus of research toward other biological and developmental factors contributing to the non-genetic portion of the trait.
Non-Genetic Biological Factors
One prominent area of investigation concerns the influence of the prenatal environment, specifically the balance of hormones a fetus is exposed to. The hormonal theory suggests that the organization of brain regions related to sexual preference is affected by prenatal exposure to androgens during critical developmental windows. Evidence supports this, such as studies on women with Congenital Adrenal Hyperplasia (CAH), a genetic condition causing high androgen exposure in utero associated with higher rates of non-heterosexual orientation. This suggests that variations in the uterine chemical environment can biologically influence sexual orientation independently of shared genes.
Another non-genetic mechanism involves epigenetic modifications, which are changes in gene activity that do not involve alterations to the underlying DNA sequence. These modifications, such as DNA methylation, can be influenced by environmental factors, including the prenatal environment, and can mediate the long-term effects of hormonal exposure on the developing brain. While the specific epigenetic patterns linked to sexual orientation remain difficult to pinpoint, this mechanism provides a plausible biological pathway for environmental factors to create differences between genetically identical twins.
The Fraternal Birth Order (FBO) effect is a unique non-genetic biological phenomenon observed exclusively in men. The FBO effect describes the statistical finding that the probability of a man being gay increases with the number of older biological brothers he has. Researchers hypothesize this is due to a maternal immune hypothesis, where the mother’s immune system progressively develops antibodies against a male-specific protein with each successive male pregnancy. These antibodies are thought to cross the placental barrier and alter the sexual differentiation of the brain in later male fetuses. This effect accounts for a substantial percentage of gay men, providing a strong example of a biological influence that is neither genetic nor related to the postnatal social environment.
Methodological Limitations
One of the most significant challenges in early twin studies was sampling bias, often referred to as ascertainment bias. Studies that relied on recruiting participants through advertisements in gay publications tended to find a higher concordance rate, likely because concordant twin pairs were more likely to volunteer. Population-based studies, which draw from general twin registries, addressed this limitation and resulted in lower heritability estimates, highlighting the importance of representative sampling.
Another methodological constraint is the difficulty in defining and consistently measuring sexual orientation across different studies. Sexual orientation is a multi-faceted trait encompassing attraction, behavior, and identity. Some studies rely on a single self-identification question, while others use more detailed scales, making direct comparisons between research findings challenging. The reliance on retrospective self-reporting, particularly regarding childhood experiences, also introduces potential inaccuracies due to the fallibility of human memory.
The twin model itself is based on certain assumptions that can be difficult to satisfy in practice, such as the “equal-environment assumption.” This assumption posits that identical twins are not treated more similarly by their parents or peers than fraternal twins, which would artificially inflate the concordance rate for identical pairs. Furthermore, the model has trouble completely separating the effects of non-shared environment from measurement error, meaning that the large portion of variance attributed to “unique environmental factors” includes many unmeasured or unknown biological and psychological influences.