My Smelly Valentine

How important to women is smell when choosing a lover? More specifically, is how a guy smells more important than how he looks?

For years, the default scientific answers were “smell is important” and “smell is more important than looks.” The answers were drawn from studies by psychologist Rachel Herz. In 1997, based on a survey of college students, she concluded that women “considered olfactory information to be the single most important variable in mate choice,” and that “female sexual interest is more affected by body odor than any other sensory stimulus.”

The underlying numbers are interesting. In response to the statement, “How someone smells can make a big difference to me,” women gave an average rating of 5.58, where 1 indicates “strongly disagree” and 7 indicates “strongly agree.” Ratings given to the other senses—how a man “looks” (5.15), “feels” (4.63) and “sounds” (4.48)—were lower, but not by much. Women rated smell as most important, but smell outscored looks by less than half a point. (For what it’s worth, men rated both sight and sound at 5.79 points.)

In 2002, Herz gave a modified version of her questionnaire to 198 college students. She concluded that when selecting a lover, “women considered a man’s smell to be more important than ‘looks,’ ‘voice,’ or ‘how his skin feels.’” Again, there was only about one-third of a point difference between women’s ratings for smell (5.59) and looks (5.26)

In 2005, a team of English psychologists led by Mark Sergeant repeated the second Herz study by putting her questionnaire online; it was filled out by 440 people. Once again, women rated how a man smells (5.73) as more important than how he looks (5.17) when they were selecting a lover. This time, smell beats looks by three-quarters of a point.

So the conventional scientific wisdom regarding female mate choice is that smell trumps looks. But this past December, social psychologist Joshua Foster completely upset the apple cart. He published a study in which women rated the actual body odor of men and actual photographs of their faces. The title of his paper gives away the result: “Beauty is mostly in the eye of the beholder.”

Foster got the materials for his experiment by taking portrait photos of 21 men and then having them wear fresh T-shirts to bed for two nights in a row. For the test session, Foster randomly paired the shirts and photos, and asked 22 women to take a look and a sniff. They rated the pleasantness, sexiness and attractiveness of the shirt smell and the photo. Half the women rated shirts and photos separately; the other half rated the photo and shirt simultaneously and gave a single score to the combination.

Foster found that the pleasantness, sexiness, and attractiveness ratings were all similar, so he collapsed them into single scores for body odor attractiveness, facial attractiveness, and combined BO/face attractiveness. He was interested in how well BO and facial scores by themselves predicted the combined BO/face scores. Statistical analysis showed that “visual cues influenced [combined] attractiveness judgments more than olfactory cues.” Foster concludes that

sight is more important that smell when women
judge the attractiveness of men.

So who are we to believe: Herz and Sergeant or Foster? Foster’s experiment better represents the real-world conditions in which women evaluate potential mates: they look at actual faces and smell actual BO. In contrast, the Herz and Sergeant studies rely on “retrospective self-report methods,” an inherently less reliable research method that asks women to recall the importance of attractiveness cues. Foster argues that

when judgments of attractiveness are made without
the filter of memory, visual cues are significantly more
important than olfactory cues.

My inclination as a scientist has always been to trust behavioral observation over questionnaire data, and contemporaneous judgment over recollection. Therefore my money’s on Foster.