Lead author Joshua Tabak, a psychology graduate student at the University of Washington, and Vivian Zayas at Cornell University, found so-called gaydar persisted even when the subjects -- college students -- saw the photos upside-down.
The study, published in the journal PLoS ONE, found gay versus straight judgments were more accurate for women's faces than for men's.
Tabak said the notion of an ability spontaneously to assess sexual orientation based on observation or instinct conflicts with the assertion that if people just kept their sexual orientation to themselves then no one else would know, and discrimination wouldn't exist -- an argument frequently used by opponents of anti-discrimination policies for lesbian, gay and bisexual people.
The study involved 129 college students, who viewed 96 photos each of young adult men and women who identified themselves as gay or straight. The photos were cropped so only faces, not hairstyles, were visible.
Study participants were 65 percent accurate in telling the difference between gay and straight faces when photos of women flashed on a computer screen, while the study participants were 57 percent accurate in differentiating gay men from straight men.
However, there was always a small number of people with no ability to distinguish gay and straight faces, Tabak said.