Scientists Claim Homosexual Men Have Different Faces

In the journal that brought you “Is a Woman’s Preference for Chest Hair in Men Influenced by Parasite Threat?”1, “Accurate Identification of a Preference for Insertive Versus Receptive Intercourse from Static Facial Cues of Gay Men”2, and an “original” paper by Garrett Prestage, Michael Hurley, and Graham Brown which I refuse to name but if I did would cause you to lose sleep, comes “Shape Differences Between the Faces of Homosexual and Heterosexual Men” by Jaroslava Varella Valentova and a slew of others.

Yes, the Archives of Sexual Behavior has been especially fecund in producing eye-popping research, some of which even makes the press. Huffington Post itself peered into Valentova’s study and asked “Could Face Shape Reveal If Someone Is Gay Or Straight?” Guess their answer.

No, guess. Do not click through to their site, or the Archives site before finishing reading this, because we’re going to attempt to duplicate the research, which went like this.

Valentova trolled the gay bars and websites in Prague and sent emails to what they hoped where some normal men and so snagged participants for his study. They asked the collected crew to rate on a “7-point” scale, from “0 to 7” (which is an 8-point scale, but let it pass, let it pass), their sexual preference, the higher the more homosexual.

Then came the geometric morphometrics. Sure, Valentova took pictures of the guys’ faces and digitized them, then beat the pixels severely with repeated applications of unadulterated statistics. Voilà! Wee p-values revealed essential characteristics of homosexual versus average versus normal faces.

Here they are, in random order: A, B, and C. Your task is to guess which is which. No cheating! Don’t spoil it for the rest of us by telling us that you looked at their answers, either. Give everybody a chance to guess. That makes it science.

Which is which?
Which is which?

Oh, after this test, using the men left over, they corralled 40 female and 40 male students from Charles University (the poor man must be reeling) and asked these folks to rate the sexuality of the pictures of the men on a real 7-point scale (1 to 7), with higher being more homosexual as before.

Turns out they couldn’t guess. Men rated both sets an average 3.5 (middle of the road), and women, ever hopeful, rated both sets an average just under that (towards the type of men who would want them). So much for picture-based “gaydar” (their word, their scare quotes).

As readers of science papers know, the best part comes at the end, where pretty much everything in the paper which came before is left behind and the authors launch themselves gleefully into the Land of Theory. This allows calls for “new research,” i.e. continued employment for the writers.

This paper is no different. The author theorize “Theoretically, there are two extreme head shapes in the Caucasian population”, dolichocephalic and brachycephalic which are two mighty scientific sounding words. The hypothalamus makes an entrance, too. Strangely, there were no words about the amgydala.

Perplexing the researchers was the “finding” that “the homosexual men in our sample also had some typical masculine facial traits, which might have caused the negative result in Study 2.” Dems, as they say, are the breaks.

Thanks to the Carolina Cowboy for alerting us to this study.

Update Reminder, this is a family blog with a sensitive spam filter.


1Prokop et al., October 2013, Volume 42, Issue 7, pp 1181-1189.

2Tskhay et al., October 2013, Volume 42, Issue 7, pp 1217-1222.

3Prestage et al., October 2013, Volume 42, Issue 7, pp 1347-1356.


  1. Gary

    It’s been a while since phrenology poked its head up. Nice to see a comeback. Just for entertainment value, of course.

  2. ace

    If you’re going to go down that alley, why stop at faces? This will be a growth industry, for sure …

    Hey, get that spam filter away from me!

  3. Sheri

    Actually, this looks more like one of those “can you spot the subtle differences” games. If you print out the images (and those from Huffington), then hold them up to a light and check the differences, my guess is no one could identify any traits because the differences are so miniscule. From Huffington’s page, the only difference in the gay/average/hetero seems to be a slightly wider face and higher eyes than the average/hetero ones. To tell the differences, you pretty much have to print them and then compare them overlaid. Getting any wee P value seems miraculous.

    I might not lose sleep over the papers written by Hurley and company, but the images for brachycephalic might do it. I must remember to be cautious with Google.

  4. Mike B.

    I’d love to see the results of some control surveys, like rate on a scale of 1 to 7 the telepathic abiltiy of these facial types. Or their likelihood of living in California or playing in the NFL. Come up with 20 different surveys, and you’d expect to get a wee p-value on one of them. Be fun to see what the most ridiculous one was.

  5. Scotian

    You are right Mike B., this is exactly the kind of control that the researchers should have done, at a minimum. I suggest space alien from Omicron Persei, in disguise of course. I think B looks suspicious.

  6. Jonathan S

    Sorry for over-zealous pedantry, but I believe Jaroslava Varella Valentova belongs to the female of the species.

  7. GaryL

    … on a scale of seven, from 1 to 4 including 1/2 and 3_1/2 to allow nuances at the extremes.

  8. brownthunder

    Of course the answer is C. I can spot a homosexual from 1000 yards!


  9. Pepe

    Can’t fool me. Its the one with the “ass eyes”.

  10. cminca

    And yet you use the same kind of exacting standards to “prove” the existence of God and to explain your theology.

  11. Briggs


    I’ll give you a brand new crisp American dollar if you can given the Latin name of the fallacy you just used.

  12. I would agree that they have a different way of purported themselves to the world that they are homosexuals, but hat they have a face that would distinguish them as such I would contest.

Leave a Reply

Your email address will not be published. Required fields are marked *