Is there a biological component to being gay beyond the chemical composition of the brain?

Is there a biological component to being gay beyond the chemical composition of the brain? Maybe that's not clear. Is being gay written into the DNA of the body?

In dog breeds certain characteristics such as enhanced smelling capabilities, sight, hunting, herding etc. have been known to express themselves physically. For example, a famous breeding program in Russia (if I remember correctly) demonstrated that breeding for heightened sense of smell in canines resulted in dogs with similar physical features even though their lines came from different stock. It explains why herding breeds all have similar features as well as gun dogs, fighting dogs, etc.

Does this translate over to human beings? Are there physical traits that show up in confident people? Funny people? Natural hunters? PhDs?
https://www.theguardian.com/technology/2017/sep/07/new-artificial-intelligence-can-tell-whether-youre-gay-or-straight-from-a-photograph

Comments

John Bump said…
Sometimes, with dogs, when we selectively breed for a characteristic, what we're actually doing is breeding for a hormonal mutant that happens to result in the characteristic, plus quite possibly several others. The whole process of converting wolves to dogs was (in part) one of selecting for individuals that didn't express the full gamut of hormones in adult wolves, and stayed in a puppy-like state for their whole lives.
There was a time when people made strong claims that being gay was genetic, and that was interrupted pretty seriously when a certain Austrian madman started rounding them up and killing them, under the idea that he could then remove them from the genetic stock of humanity, at which point there was at least somewhat of a movement towards it being either a choice or a consequence of early childhood experience, rather than genetic.
Alex Kudlick said…
This seems like a clear case of over fitting the data
Salvador Melo said…
Near as I can tell, being gay is genetic, but it is coded in the base DNA of every human and requires a trigger to activate during foetal development. It is also not simply an on|off switch. There is a gradient upon which people fall.
Alex Kudlick said…
Jason ON yea, this story should be being pitched as a pitfall of machine learning methods. There's clearly some other correlation in the input data the algorithm learned. "Learned to identify" should always be stated as "developed an algorithm with false-positive x and false-negative y"

This is like the machine learning researchers in China that claimed they could identify a tendency for criminality from the face.

Popular posts from this blog

I'm shutting down Google+ for the night and quite possibly for the weekend.