Detecting sexual orientation from 50 millisecond face stimuli

If there were no differences between the faces of hetero- and homosexuals, then we'd expect a 50% accuracy in classification of the faces in the two categories. This accuracy was exceeded for both male and female target faces, and irrespective of whether or not the faces were presented upright or inverted (accuracy was smaller with inverted faces).


The authors of the paper also have a piece in the NY Times on the science of Gaydar. The authors write:

Should you trust your gaydar in everyday life? Probably not. In our experiments, average gaydar judgment accuracy was only in the 60 percent range. This demonstrates gaydar ability — which is far from judgment proficiency.

I find it fairly remarkable that this was achieved with 50ms static stimuli. Could a better accuracy be possible with longer stimuli or even moving ones? I'd wager that "real-life" gaydar which uses motion and sound in addition to a quick visual impression may be even more potent. So, I wouldn't be quick to dismiss gaydar's utility in a real-life setting.

An interesting finding of the paper is that women's sexual orientation was assessed more accurately than men's. I personally don't find that very surprising; the authors write:

The prospect of distinct processes for extracting sexual orientation from women’s and men’s faces is intriguing, yet not entirely surprising. The face is assumed to reflect experiences. Men and women differ in their subjective experiences and overt expressions of romantic love and sexual desire, as well as their biological (neurophysiological and hormonal) underpinnings, e.g., [25], [26], [27], [28], [29], [30], [31]. The current findings suggest that facial expressions of sexual orientation also differ by gender.

It would be quite interesting to create facial composites out of the stimuli used in this study, i.e., to create average male/female homosexual/heterosexual faces, and then have these measured on both objective grounds (= anthropometric differences between them) and subjective ones (= attractiveness, masculinity/femininity, etc.).

My personal guess is that the differences between females based on sexual orientation will be much more pronounced (e.g., greater masculinity, lower attractiveness), which might be consistent with both (i) the greater ability to correctly detect female sexual orientation, as evidenced in this study, and (ii) the smaller numbers of female vs. male homosexuals in the general population, which may suggest that the former are a more "idiosyncratic" population than the latter. But, in any case, one ought to carry out such an experiment.

PLoS ONE 7(5): e36671. doi:10.1371/journal.pone.0036671

The Roles of Featural and Configural Face Processing in Snap Judgments of Sexual Orientation

Joshua A. Tabak, Vivian Zayas

Abstract

Research has shown that people are able to judge sexual orientation from faces with above-chance accuracy, but little is known about how these judgments are formed. Here, we investigated the importance of well-established face processing mechanisms in such judgments: featural processing (e.g., an eye) and configural processing (e.g., spatial distance between eyes). Participants judged sexual orientation from faces presented for 50 milliseconds either upright, which recruits both configural and featural processing, or upside-down, when configural processing is strongly impaired and featural processing remains relatively intact. Although participants judged women’s and men’s sexual orientation with above-chance accuracy for upright faces and for upside-down faces, accuracy for upside-down faces was significantly reduced. The reduced judgment accuracy for upside-down faces indicates that configural face processing significantly contributes to accurate snap judgments of sexual orientation.

Link