YouTube VideoGenetics and Homosexuality: Are People Born Gay?
The Biological Basis for Sexual Orientation
by Richard Deem, MSc


Born Gay?

There is a popular belief that people are born either gay or straight. Conservatives tend to believe that sexual orientation is actually sexual preference, which is chosen by the individual. This page represents a review of the scientific literature on the basis for homosexual orientation.

Rich Deem

Sexual orientation is an abstruse topic with regard to human sexuality. Biologically, human beings, like most mammals, produce two sexes (males and females) that are encoded genetically by the sex chromosomes (X and Y). From a biological standpoint, males and females are expected to be attracted to each other, ensuring reproduction and subsequent perpetuation of the species. Therefore, same sex attraction (SSA) is biologically aberrant, since it fails to produce offspring. Since the brain is the seat of emotion and sexual attraction, studies examining SSA and sexual orientation have focused on biological differences in the brains of those who exhibit homosexual behavior vs. those who exhibit heterosexual behavior. Studies have focused on brain structure, possible hormonal influences, heritability, and genetics.

Brain structure

Since sexual attraction commences in the brain, scientists initially examined the question of sexual orientation by comparing the anatomy of brains from males and females.


Studies of the brain indicated that male and female brains were sexually dimorphic in the pre-optic area of the hypothalamus, where males demonstrated a greater than two-fold difference in cell number and size compared to females (Swaab and Fliers 1985). A second study found that two of four Interstitial Nuclei of the Anterior Hypothalamus (INAH) were at least twice as large in males as females (Allen, et al. 1989). Since the INAH was involved in sexual dimorphism, Simon LeVay hypothesized that there might be differences in this region in heterosexual vs. homosexual men. Postmortem examination of the brains of AIDS patients vs. control male subjects (presumed to be heterosexual) showed that the heterosexual men exhibited INAH3 that were twice the size of both females and presumably homosexual men who had died of AIDS (LeVay 1991). The study has been criticized for its uncertainty of sexual orientation of the subjects, and potential complications caused by the AIDS virus (HIV infects the human brain), and also by lowered testosterone levels found in AIDS patients. A popularized Newsweek cover story, "Is This Child Gay?" characterized LeVay as a "champion for the genetic side," even though the study involved no genetic data at all (Gelman 1991).

A subsequent study examined the question of INAH3 size on the basis of sex, sexual orientation, and HIV status.(Byne, et al. 2001) The study confirmed large differences in INAH3 volume based upon sex, but found that IHAH3 volume was decreased in male heterosexual men who had contracted AIDS. There was no statistically significant difference between IHAH3 sizes of male heterosexuals vs. male homosexuals who had contracted AIDS (Byne, et al. 2001), suggesting that LeVay’s original finding was the result of HIV infection and not the result of sexual orientation.

A study that examined the sexually dimorphic nucleus of the hypothalamus showed a decrease in volume and cell number in females at 2-4 years postnatal (Swaab, Gooren, and Hofman 1992), suggesting that not only chemical and hormonal factors, but also social factors, might have influenced this process, casting doubt on the validity of LeVay’s findings.

Anterior commissure

A study by Allen and Gorski examined the anterior commissure of the brain, finding that females and homosexual males exhibited a larger size than heterosexual males (Allen and Gorski 1992). However, later studies using more subjects found no such differences (Bishop and Wahlsten 1997).

Sex affects brain structure

Scientists originally hypothesized that brain structure affected sexual behavior. However, studies have shown that sexual experiences themselves can affect brain structure (Breedlove 1997). So, regardless of the brain differences found between those who practice homosexual vs. heterosexual behavior, it can never be certain if the brain affects the behavior or if the behavior affects the brain.

Brain structure conclusions

Although early studies found some differences in the brain structure of homosexuals, later, better controlled studies found no such differences. The fact that sexual experience itself can affect brain structure complicates the interpretation of brain structure studies, since  it cannot be ascertained whether brain structure affects sexual orientation or where sexual orientation affects brain structure.

Hormonal influences

Sexual differentiation of the fetus occurs within the womb, as a result of hormonal influences on the development process. These hormones not only affect differentiation of the sex organs, but also affect brain development. Scientists hypothesized that variation in fetal hormones might affect brain development in such a way as to influence subsequent sexual preferences. However, since it would be unethical to obtain fetal hormone levels (too dangerous to the fetus), proxies have been used in place of actual hormone levels. These proxies include differences in skeletal size and shape, including the ratio of the long bones of the arms and legs relative to arm span or stature and the hand bones of adults (the ratio of the lengths of various fingers). However, there are no differences in circulating sex hormones between heterosexual and homosexual men (Meyer-Bahlburg 1984).

Finger length

Studies have shown that ratios of digit length are predictors of several hormones, including testosterone (an androgen), luteinizing hormone and estrogen (Manning, et al. 1998). In women, the index finger (2D, second digit) is almost the same length as the fourth digit (4D). However, in men, the index finger is usually shorter than the fourth and that this 2D:4D ratio in females is established in two-year-old children. It was hypothesized that the sex difference in the 2D:4D ratio reflects the prenatal influence of androgens on males. A subsequent study showed that the 2D:4D ratios of homosexual men vs. heterosexual men were not different (Williams, et al. 2000). However, homosexual women displayed significantly smaller 2D:4D ratios compared with heterosexual women, suggesting that women exposed to more androgens in the womb tend to exhibit SSA.

Other studies have found that the more older brothers a male child has, the more likely he is to develop a homosexual orientation (McConaghy, et al. 2006). This study also found that homosexual men had a greater than expected proportion of brothers among their older siblings (229 brothers: 163 sisters) compared with the general population (106 males: 100 females). Subsequent, larger studies found that older brothers did not affect male sexual orientation (Zietsch, et al. 2012; Bogaert 2010). Males who had two or more older brothers were found to have lower 2D:4D ratios (Williams, et al. 2000), suggesting that they had experienced increased androgens in the womb. The reason increased androgens would predispose both males and females to exhibit homosexual behavior was not explained in the study. A study examining 2D:4D ratios in twins concordant and discordant for sexual orientation found differences between discordant female twins, but not male twins, although the sample sizes were very small (Hall and Love 2003). However, a large (255,116 participants) cohort from the UK found no link between female SSA and 2D:4D ratios (Manning, Churchill, and Peters 2007).

Long bones

Another study examined the length of long bones in the arms, legs and hands. Both homosexual males and heterosexual females had decreased long bone growth in the arms, legs and hands compared to heterosexual males or homosexual females (Martin and Nguyen 2004). Accordingly, the researchers hypothesized that male homosexuals experienced decreased androgen exposure during development than male heterosexuals, while female homosexuals had greater steroid exposure during development than their heterosexual counterparts. Needless to say, this study directly contradicted the results of the Williams study above, which showed that males with multiple older brothers (who tended to be homosexual) experienced increased androgen exposure.

Congenital adrenal hyperplasia

Studies involving a rare hormonal imbalance, congenital adrenal hyperplasia (CAH), caused by defective 21-hydroxylase enzyme, suggest that extreme hormonal abnormalities can influence sexual orientation to some extent. CAH increases production of male hormones during development. In female fetuses, increased androgens resulted in development of ambiguous external genitalia. A meta-analysis of 18 CAH studies shows that 91.5% of those women suffering from CAH end up with a heterosexual orientation (Meyer-Bahlburg, et al. 2008), despite having a severe imbalance of testosterone during fetal development. These studies show that hormonal influences in utero are not the major reason for female SSA.

Childhood gender nonconformity

The behaviors of children have been hypothesized to be influenced by hormonal imbalances during development or early childhood. Such non-stereotypical childhood behaviors (e.g., boys playing with dolls or girls showing interest in rough play) have been associated with same-sex sexual orientation (Bailey and Zucker 1995). However, such studies have been done retrospectively based upon the memories of participants or family members, which could be tainted by post-childhood experiences. The association of childhood gender nonconformity with SSA has been disputed by other scientists (Gottschalk 2003) and has been shown to be a predictor of poor psychological health (Rieger and Savin-Williams 2012).

Hormonal influences conclusion

All of the studies reporting possible hormonal influence on SSA suffer from the lack of any concrete evidence that hormones actually play any role in sexual orientation. In fact, androgen gene variants do not play a role in male sexual orientation (Macke, et al. 1993). The fact that contradictory studies report increased vs. decreased androgens as a basis for SSA does not provoke confidence that the proxies are legitimate. Fairly, a study that documented actual hormone levels, as opposed to proxies, would plausibly provide more definitive data. The fact the extreme hormonal abnormality found in CAH only moderately affects SSA shows that it is unconvincing that subtle abnormal hormonal influences account for few, if any, instances of SSA.


Studies on the role of genetics have examined the prevalence of SSA in identical twin and non-twin family members, studies of chromosomal regions thought to be involved in SSA, and genome wide association studies (GWAS), which examine actual mutations (single nucleotide polymorphisms, “SNPs”) in thousands of individuals. Only GWAS measures actual genetic changes, since inheritance, based upon twin studies or specific chromosomal regions, could also be explained as epigenetic differences.

Twin studies

The observation that familial factors influence the prevalence of homosexuality led to the initiation of a number of twin studies, which are a proxy for the presence of possible genetic factors. Most of these early studies suffered from methodological flaws. Kallmann sampled subjects from correctional and psychiatric institutions—not exactly representative "normal" populations (Kallmann 1952). Bailey et al. published a number of studies in the early 1990's, examining familial factors involved in both male (Bailey and Pillard 1991; Bailey and Bell 1993) and female homosexuality (Bailey and Bell 1993; Bailey and Benishay 1993; Bailey, et al. 1993). These studies suffered from the manner in which subjects were recruited, since the investigators advertised in openly homosexual publications, resulting in skewed populations. Later, studies by the same group did not suffer from this selection bias, and found the heritability of homosexuality in Australia was up to 50 and 60% in females but only 30% in males (Bailey, Dunne, and Martin 2000).

A 2000 study examined 1,588 twins selected by a random survey of 50,000 households in the United States (Kendler, et al. 2000). The study found 3% of the population consisted of non-heterosexuals (homosexuals and bisexuals) and a genetic concordance rate of 32%, somewhat lower than that found in the Australian studies. The study lost statistical significance when twins were broken down into male and female pairs, because of the low rate (3%) of non-heterosexuals in the general U.S. population.

A Swedish study found that in male SSA, heritability accounted for 34-39% of the variance, while individual-specific environment accounted for 61-66% of the variance (Långström, et al. 2010). Heritability was much lower in females (18-19%). A 2012 Australian twin study found heritability accounted for 31% of variation in sexual orientation and 44% depression (Zietsch, et al. 2012).

Chromosomal inheritance

Since homosexual men had more homosexual male relatives through maternal than through paternal lineages, Dean Hamer examined the X chromosome to find such an association at region Xq28 (Hamer, et al. 1993). If male sexual orientation were influenced by a gene on Xq28, then homosexual brothers should share more than 50% of their alleles at this region, whereas their heterosexual brothers should share about 50% of their alleles. An analysis of 40 pairs of homosexual brothers and found that they shared 82% of their alleles in the Xq28 region, which was much greater than the 50% allele sharing that would be expected by chance. However, a follow-up study by the same research group, using 32 pairs of homosexual brothers and found only 67% allele sharing (Hu, et al. 1995), which was much closer to the 50% expected by chance. Later attempts to repeat the Hamer study resulted in only 46% allele sharing (Wickelgren 1999), and insignificantly different from chance (Rice, Friberg, and Gavrilets 2016), contradicting the Hamer results. At the same time, an unpublished study by Alan Sanders (University of Chicago) corroborated the Rice results (Wickelgren 1999). Ultimately, no gene or gene product from the Xq28 region was ever identified that affected sexual orientation. When Jonathan Marks (an evolutionary biologist) asked Hamer what percentage of homosexuality he thought his results explained, his answer was that he thought it explained 5% of male homosexuality. Marks' response was, "There is no science other than behavioral genetics in which you can leave 97.5% of a phenomenon unexplained and get headlines" (Marks 2003).

Candidate genes that have been eliminated include the androgen receptor gene (Macke, et al. 1993) and aromatase cytochrome P450 (CYP19) (DuPree, et al. 2004). A 2012 study found an association between male homosexual orientation and the human sonic hedgehog (SHH) gene is located in the 7q36 region (Wang, et al. 2012). However, the SHH signaling pathway is crucial in the development of all animals and contributes to numerous disorders, such as holoprosencephaly 3 (brain hemisphere abnormality), hypoplasia or aplasia of tibia, Laurin-Sandrow syndrome (feet and fibula disorder), microphthalmia (eye disorder) and preaxial polydactyly (finger disorder). A 2015 study found an association between variations within the COMT gene and male sexual orientation (Yu, et al. 2015). Other disorders associated with the COMT gene include panic disorder and schizophrenia.

Genome wide association studies (GWAS)

Prior to 2005, genetic studies were limited to regions of chromosomes or specific genes. Howbeit, a new technique called genome wide association was invented shortly after the completion of the Human Genome Project in 2003. GWAS can examine specific mutations (called SNPs) along the entire length of the entire human genome. A 2005 GWAS (including 456 subjects) found chromosome regions 7q36, 8p12, and Xq28 associated with sexual orientation (Mustanski, et al. 2005). Inevitably, an attempt to repeat the finding (along with ~6000 well-defined SNPs spread comparatively evenly across the human genome) failed to find any significant SNPs (Ramagopalan, et al. 2010). The largest GWAS (over 23,000 subjects) of sexual orientation was performed by scientists at 23andme in 2012. The data were presented at the American Society of Human Genetics annual meeting in San Francisco, showing that there were no loci associated with sexual orientation, including Xq28 on the X chromosome (Drabant, et al. 2012). A 2015 GWAS found the existence of genes related to male sexual orientation on chromosome 8 and chromosome Xq28 (Sanders, et al. 2015). However, the study had a relatively small sample size (908) and, in order to gain significance, had to employ multipoint LOD analysis, which is known to lead to increased false-positive results (Ott, Kamatani, and Lathrop 2011).


Since no genes have ever been consistently associated with sexual orientation, scientists have thought that that certain epigenetic marks (which control gene expression) might explain the phenomenon. First in 2012 and then again in 2016, a research group hypothesized that canalizing sexually antagonistic epigenetic marks that are produced during early ontogeny might influence development and possibly sexual orientation (Rice, Friberg, and Gavrilets 2012; Rice, Friberg, and Gavrilets 2016), although no evidence directly supporting the hypothesis was provided. Since epigenetics is cell-specific and sexual orientation is likely to be a function of the brain, it is unlikely that epigenetics in the brain will ever be directly measured, due to ethical concerns (it is unlikely any human subjects review board would authorize a study to take brain biopsies of living human beings). So, this question, in all probability, will fail to be scientifically affirmed.

Persistence of homosexuality

Since male homosexuals tend to have, on average, five times fewer children than heterosexual males (Chaladze 2016), it seems unlikely that any homosexual genetic trait would persist in a population of non-reproducing individuals. In order to get around this incontrovertible problem, scientists have hypothesized that the reduced reproductive capacity of homosexuals is offset by increased reproductive capacity of genetic relatives of those homosexuals. However, mothers of homosexual men reproduce only 1.16 times higher (Camperio-Ciani, Corna, and Capiluppi 2004), 1.31 times higher (Iemmola and Camperio Ciani 2009), and 1.38 times higher (Vasey and VanderLaan 2007), which is insufficient to compensate for the minimal reproductive capacity of homosexual men. A new hypothesis states that an X-linked, sexually antagonistic model of homosexuality requires that more than half of the females and half of the males are carriers of genes that predispose a male brother to homosexuality (Chaladze 2016). Of course, such high concordance rates are not observed in identical twin studies, so would be unlikely to explain the persistence of homosexuality.

Genetics conclusions

Although twin studies suggest a type of ancestry for male homosexual orientation, the majority affect is environmental. Attempts to identify specific genes related to male homosexual orientation have failed to consistently identify such genes and have often been contradictory. The best predictor of genetic association is through the new technique of GWAS, which have also failed to confirm any specific genetic associations with homosexual orientation. If a gene or genes were ever identified, it would be arduous to resolve how it could persist within the population, given the severely reduced reproductive capacity of homosexuals.

Non-biological factors

Family of origin and environmental factors play a role in homosexuality. Factors that increase the probability of becoming homosexual include having divorced parents, absent fathers, having older mothers, being the youngest child, and being a city (as opposed to rural) dweller. For women, maternal death during adolescence and being the only or youngest child or the only girl in the family increased the likelihood of homosexuality (Frisch and Hviid 2006). In particular, paternal overprotection plays an important role in the development of male homosexuality (Lung and Shu 2007). In addition, having homosexual parents also increased the likelihood of becoming homosexual up to 16-57% of such children adopted a homosexual lifestyle (Cameron 2006; Schumm 2010).

Abusive childhood experiences

A study of 13,000 New Zealand adults (age 16+) examined sexual orientation as a function of childhood history. The study found a 3-fold higher prevalence of childhood abuse for those who subsequently engaged in same sex sexual activity (Wells, McGee, and Beautrais 2011). However, childhood abuse was not a major factor in homosexuality, since only 15% of homosexuals had experienced abuse as children (compared with 5% among heterosexuals). So, it would appear from this population that only a small percentage of homosexuality (~10%) might be explained by early childhood abusive experiences. A twin-based analysis found that childhood experiences of sexual abuse and risky family environment accounted for 8.5% and 7.7% of the covariance between sexual orientation and depression (Zietsch, et al. 2012). Another study found that a history of sexual abuse predicted increased prevalence of same-sex attraction by 2.0 percentage points (Roberts, Glymour, and Koenen 2013). A 2011 review of 46 studies found that the prevalence of childhood sexual assault for homosexual males ranged from 4.1% to 59.2% (median 22.7%) and from 14.9% to 76.0% for females (median 34.5%) (Rothman, Exner, and Baughman 2011). The overall rate of childhood sexual assault in the general population is 5.1% for boys and 26.6% for girls (Finkelhor, et al. 2014). Studies have found that childhood sexual abuse can affect brain structure (Sheffield, et al. 2013; Teicher, et al. 2016).

Orientation vs. preference

If homosexual orientation were completely genetic, one would expect that it would not change over the course of one's life. However, sexual preference, especially for females, does seem to change over time. A 5-year study of lesbians found that over a quarter of these women relinquished their lesbian/bisexual identities during this period: half reclaimed heterosexual identities and half gave up all identity labels (Diamond 2003). In a survey of young minority women (16-23 years of age), half of the participants changed their sexual identities more than once during the two-year survey period (Diamond 2000). In another study of subjects who were recruited from organizations that serve lesbian/homosexual/bisexual youths (ages 14 to 21 years) in New York City, the percentage that changed from a lesbian/homosexual/bisexual orientation to a heterosexual orientation was 5% over the period of just 12 months (the length of the survey) (Rosario, et al. 2006). A 2015 study found that among young adults, sexual fluidity in attractions was reported by 64% of women and 52% of men, with 49% of those women and 36% of those men reporting sexual fluidity in sexual identity based on experiencing changes in attractions (Katz-Wise 2015). A 2011 study of Christian homosexuals who wanted to change their sexual orientation found that 23% of the subjects reported a successful "conversion" to heterosexual orientation and functioning, while an additional 30% reported stable behavioral chastity with substantive dis-identification with homosexual orientation (Jones and Yarhouse 2011). Prominently, for a portion of individuals, being homosexual or heterosexual is something they can choose.

Conclusion Top of page

The origin of homosexual orientation has been the subject of much press, with the general impression being promoted that homosexuality is largely a matter of genes, rather than environmental factors. Whereas, if one examines the scientific literature, one finds that the data is not quite as clear as the news bytes would suggest. The early studies that reported differences in the brains of homosexuals were complicated by HIV infection and were not substantiated by larger, better controlled studies. Since sexual behavior can actually affect brain structure, it remains unclear whether differences in brain structure are the result of experience or the cause of sexual orientation. Numerous studies have reported that possible hormonal differences affect homosexual orientation. However, these studies were often directly contradictory, and never actually measured any hormone levels, but just used proxies for hormonal influences, without direct evidence that the proxies were actually indicative of true hormone levels or imbalances. Twin studies showed that there likely are genetic influences for homosexuality, although heritability (what was actually measured) does not necessarily represent changes in DNA sequences. In fact, DNA microarray technology, the gold standard for true genetic studies, have largely failed to pinpoint any specific genes as a factor in sexual orientation. Several familial and environmental factors (which represented the majority affect in twin studies) have been shown to influence development of SSA. Early childhood abuse has been associated with homosexuality, but, at most, probably explains only about 10% of those who express homosexual orientation. The reality that sexual orientation is not constant for many individuals, but shows vicissitude over time, suggests that in many cases, sexual orientation is de facto, sexual preference. The quintessential determination is that for the majority, sexual orientation is not the result of biology or genetics, but achieved through experiences and behaviors.

La Genética y la Homosexualidad: ¿Nace la gente, homosexual?

Share on Facebook

References Top of page
Last updated January 3, 2017


Rich's Blog