Marilyn Monroe's measurements (37C-23-36, for inquiring minds) were extolled by many of her contemporaries as perfect. But a new study appearing in this week's issue of Proceedings of the National Academy of Sciences suggests that her hourglass figure—and waist to hip ratio of 0.64—alone was not enough to ensure her allure; it also had to be coupled with the way she moved.

"For centuries people have been trying to understand what makes people appreciate beauty," says Kerri Johnson, a psychologist at New York University and co-author of the new report. "Finding a single parameter that defines beauty has been difficult." To wit, she notes, whereas men, especially in Western cultures, generally prefer women like Monroe—and more shapely contemporaries like actress Penelope Cruz and rail-thin model Kate Moss (whose waist to hip ratio is 0.66)—men in some remote African cultures have been found to prefer women with waist to hip ratios as high as 0.9 to 1.0.

The point is, say Johnson and her colleague Louis Tassinary, a cognitive scientist at Texas A&M University, it is virtually impossible to cite a single factor as a universal symbol of beauty without putting it in context of gender. And that includes, Johnson says, waist to hip ratio, which is "widely cited as a magical cue that signifies attractiveness.

"The judgment of biological sex," she notes, is something that takes place early on in sizing up another person. "Once the biological sex of a target is known, any other cues that are gendered will be perceived within that context."

Johnson and Tassinary propose there is, in essence, a two-step process to making social appraisals: First an immediate determination is made as to whether a target is male or female, which Johnson says is essentially automatic, one of many "fundamental social perceptions that people are likely to immediately make about others," along with variables such as race. Then, the authors hypothesize, you check out other factors, such as gait, waist to hip ratio and hair length, and judge them according to a set of gender-based rules (for instance, from extremely feminine to extremely masculine).

The researchers tested their proposal throughout five separate constructs; 370 participants observed animations that "walked" in place at different gaits as well as static, line-drawn women or outlines of human walkers, and then rated the degree of masculinity or femininity as well as attractiveness. In some cases, the gender of the outlines was specified; in others, it had to be gleaned from the shape and gait.

The results: the scientists noted that a female target was viewed as more attractive if its gait, waist to hip ratio, etcetera, were found to be more toward what is perceived as the feminine extreme. Likewise, a male target showing more masculine cues would be favored. For instance, a subject deemed to be a woman with a particularly favorable waist to hip ratio but a lumbering gait would generally be perceived as less attractive than the same figure with a pronounced hip sway or sashay.

Frank Marlowe, an anthropologist at Florida State University in Tallahassee, says that the new two-pronged model seems unnecessary given that discrepancies in preferences for certain waist to hip ratios often sprout from genetic disparities among populations and not because men have inherently different ideas of attractiveness. "I sure do not understand the psychologists not beginning by first looking around at the physical variation geographically in the shape of women," Marlowe explains. "Before jumping to why men's preferences might vary, why not first see if they simply vary with mean female waist to hip ratio in their own populations?"