Dec 302012

Three researchers have proposed a new hypothesis for a biological cause of homosexuality. Attempts to find a “gay gene” have so far been inconclusive. This group of researchers theorized that it was epigenetics instead.¬†Epigenetics are the meta level of genetics. Where genetics is the raw components of DNA (the C’s, T’s, G’s and A’s of the DNA strand), epigenetics is the packaging (e.g., histones) and/or temporary modification (e.g., DNA methylation) of DNA.

Essentially, these researchers hypothesize that epigenetics influences the responsiveness of a fetus to hormones they are exposed to in the womb (fetal hormones). Fetal androgens, like testosterone, masculinize the fetus at high levels (or feminize the fetus by their absence). The masculinization/feminization then somehow leads to being gay, maybe through effects on brain structure.

Technically speaking, it’s a good hypothesis; it’s testable and it’s disprovable. They even go so far as to list specific predictions based on their hypothesis. Good science. However, I’m troubled by their hypothesis. Namely, I don’t think there’s sufficient evidence that fetal masculinization/feminization results in non-heterosexuality.

It is known that fetal androgen exposure is associated with gender non-conforming play in childhood. That is, fetuses exposed to higher levels of androgens are more likely to become children who participate in “masculine” play (e.g., rough-and-tumble physical games) than fetuses exposed to lower levels of androgens. Studies of children with congenital adrenal hyperplasia are some of the strongest evidence I know of for this.

Is gender non-conforming play in childhood associated with adult homosexuality? Some studies have found that it is, but those studies were retrospective.¬†That means that researchers had the participants report on their play preferences as children (through memory, videos, records, etc). Retrospective studies are very subject to confirmation bias! A much stronger study would be prospective (e.g., assign a cohort of children into groups based on their play preferences, and follow them until they’re 18 and ask them to report their orientation). Prospective studies don’t have the whole “20/20 hindsight”/confirmation bias problem.

The idea that fetal androgens levels affect sexual orientation also ends up reinforcing the stereotype of “gay men are feminine and lesbian women are masculine.” There’s very little evidence to support those stereotypes on the biological level. Besides, if the fetus’s brain is masculinized/feminized by fetal androgens, wouldn’t you expect a transgender child and not a gay one?

There’s a lot more to this study, so if you want to read through it yourself, the full article is available. It was published in The Quarterly Review of Biology, Dec 2012 issue.

 Leave a Reply

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>