Gaydar: It Takes 50 Milliseconds to Identify a Man’s Sexual Orientation.

Gaydar is the colloquialism used to describe the supposed ability to detect whether someone is gay or straight simply by looking at them. Does gaydar truly exist or is it merely folk belief?

Nicholas O. Rule and Nalini Ambady recently investigated this exact issue. In the first study, they downloaded photos of men (who had identified their sexual orientations) from online personal advertisements that had been posted in many American cities. The researchers standardized the photos in an attempt to remove any confounding factors (e.g., photos were placed against a white background; only photos wherein the faces contained no facial alterations, such as beards, were retained; the photos were cropped such that only the faces could be seen; photos were standardized to the same size; photos were gray-scaled; etc.). Ninety photos were retained (45 of each sexual orientation).

Ninety individuals, 68 of whom were women, participated in the study. The experimental task simply involved viewing the photos and deciding whether any given man as represented in the photo was straight or gay. There were six experimental conditions that corresponded to the length of exposure to the various photos: 33 milliseconds (ms), 50 ms, 100 ms, 6,500 ms, 10,000 ms, or self-paced. Fifteen participants were randomly assigned to each of the six conditions.

A second study was also conducted wherein the photos were in this case taken from Facebook rather than from online personal ads. Another difference in the second study is that hairstyles were cropped out from the photos (as yet another control). Finally, in the second study, only the 50 ms condition was operative. That said, since the key conclusions are similar across the two studies (although the accuracies were smaller in study 2), I shall only report the key findings of the first experiment.

Here are the accuracy rates across the six conditions:

33 ms: 0.50
50 ms: 0.57
100 ms: 0.62
6,500 ms: 0.58
10,000 ms: 0.60
Self-paced: 0.62

All of the latter proportions, short of the one for 33 ms, are statistically different from 0.50 (i.e., the proportion that one can obtain simply by guessing).

Since the base rate of homosexuals in the population is not equal to that of heterosexual men, the researchers conducted an additional analysis wherein they adjusted for that fact. The adjusted accuracies for 33 ms, 50 ms, 100 ms, 6,500 ms, 10,000 ms, and self-paced were 0.49, 0.62, 0.69, 0.64, 0.67, and 0.70 respectively. Again, all of the adjusted accuracies other than that of 33 ms were statistically greater than chance. Also, the adjusted accuracies were statistically greater than the unadjusted accuracies. Of note, men and women raters were equally accurate. Finally, for both the unadjusted and adjusted accuracies, the statistically significant accuracies (i.e., all those other than that for 33 ms) are no different from one another, meaning that participants were equally accurate irrespective of exposure time.

The bottom line is that people can judge a man's sexual orientation accurately (i.e., above chance level), and can do so within as little as 50 milliseconds. That said the authors did not venture into identifying specific cues that might be used in arriving at such accurate classifications.

An obvious extension of this work would be to conduct the same study but use homosexual and heterosexual women (for the photos). Also, an interesting issue is whether one's "Gaydar" ability is contingent on sexual orientation. I shall take up this point in a future post.

Ciao for now.

Source for Image:
http://larrylafond.com/images/430_GAYDAR_DVD_COVER.JPG