Effects of sexual dimorphism on facial attractiveness
Testosterone-dependent secondary sexual characteristics in males may signal immunological competence and are sexually selected for in several species. In humans, oestrogen-dependent characteristics of the female body correlate with health and reproductive fitness and are found attractive. Enhancing the sexual dimorphism of human faces should raise attractiveness by enhancing sex-hormone-related cues to youth and fertility in females, and to dominance and immunocompetence in males. Here the authors report the results of asking subjects to choose the most attractive faces from continua that enhanced or diminished differences between the average shape of female and male faces. As predicted, subjects preferred feminized to average shapes of a female face. This preference applied across UK and Japanese populations but was stronger for within-population judgements, which indicates that attractiveness cues are learned. Subjects preferred feminized to average or masculinized shapes of a male face. Enhancing masculine facial characteristics increased both perceived dominance and negative attributions (for example, coldness or dishonesty) relevant to relationships and paternal investment. These results indicate a selection pressure that limits sexual dimorphism and encourages neoteny in humans.
D I Perret, K J Lee, I Penton-Voak, D Rowland, S Yoshikawa, D M Burt, S P Henzi, D L Castles & S Akamatsu
Effects of sexual dimorphism on facial attractiveness
(Letter to Nature)
Nature 394, 884 (1998)
August 31, 1998
Nothing Becomes a Man More Than a Woman's Face
By NATALIE ANGIER
timeless movie scene some of us wish we could forget: Leonardo DiCaprio climbing to the bowsprit of the "Titanic" and crowing, "I'm the king of the world!"
The latest word from scientists: He may have a point. Or, rather, a fetching bit of roundness. In a new study of facial attractiveness, researchers from Scotland and Japan have found that, much to their astonishment, people of both sexes prefer feminine-looking men over rugged, manly-miened men. When shown a series of computerized photographs of young men whose images had been manipulated to make them look either more masculine or more feminine than the norm, viewers designated the artificially feminized faces as somewhat more attractive than the average faces, and more appealing by far than the masculinized versions. The results, which appeared in Thursday's issue of the journal Nature, held cross-culturally, whether the faces shown or the people passing judgment were Japanese or Caucasian.
The scientists had predicted that such traditional hallmarks of male dominance and maturity as a big jaw, square face, prow nose, and heavy brow would prove impressive to men and irresistible to women, who in theory are ever on the lookout for their alpha mate. Instead, appraisers seemed drawn to a touch of girlishness -- slender nose, cupid's lips, lightened brow, adorable chin.
In other words, Leo rules! So, too, do the likes of Johnny Depp, Matt Damon and Ralph Fiennes, and Paul McCartney when his top still mopped, and his Monkee clone Davey Jones, and the young Elvis in white and the young Brando in buff, who were both so pretty you could swear they wore Maybelline. Far from putting a man at a competitive disadvantage, it seems, femininity may be a source of sexual and social strength.
"Our team has been working on this study for four years," said Ian
Penton-Voak of the University of St. Andrews in Fife, Scotland.
"When it was found early on that there was a preference for
feminized male faces, nobody believed it, so we did it again, and
again. The preference for a feminized face keeps coming up."
Most recently, he added, "we've even replicated this work among hunter-gatherer tribes in the Amazon." Penton-Voak, an author of the report, is a graduate student in the laboratory of Dr. David I. Perrett, a professor of psychology.
In the same study, the scientists also found that viewers preferred ultra-feminized women's faces over either average female faces or those that had been slightly masculinized. That result, at least, was in keeping with their expectations.
The new report adds to a growing body of evidence that beauty may not be restricted to the eyes of a beholder, and that a sense of it may be more innate and less culturally fungible than people care to believe.
At the same time, the results underscore that the meaning of beauty is still up for grabs and open to any number of Kiplingesque Just-So interpretations. Perrett and his colleagues speculate that women might choose feminine-looking men as their mates because such men give the appearance of being sensitive, honest and intially reliable fathers
."It is quite interesting that people like feminized male faces," said Dr. Stefano Ghirlanda, a zoologist at the University of Stockholm in a telephone interview. "But the significance of the work is not clear. We understand so little about what faces mean, or what the information content of any particular feature may be."
In the latest experiment, the researchers started with the assumption that people would find most appealing those faces that displayed "sex-appropriate" signs of ostensible genetic and reproductive fitness, and that the more obvious the signs, the more attractive the face should be. Thus, an attractive female face should look unmistakably female, which means it has been shaped by sufficient stores of estrogen, the principal hormone of female fertility. Estrogen is thought to help make the bottom half of the face narrower than the top half, for example, and to make the cheeks high and round.
And because female fertility declines sharply with age, female attractiveness has also been linked to signs of youthfulness, including wide, bright eyes (the pupils of the eyes shrink with age, the corneas get ever cloudier, and the upper lids tend to droop), a small nose (being cartilage, the nose continues to grow throughout life), and full lips (lips lose their connective tissue and thin out over time).
By comparison, the male hormone testosterone is known to enlarge the jaw and to thicken the brow and upper bridge of the nose. If a woman is looking for a man who is clearly a man and who brims with male hormones, she theoretically should be drawn to a face with a comparatively square jaw and strong brow. In other species, the traits and behaviors that males use to attract females, like being able to grow a long, showy tail or to sing a complex melody, are also under the control of testosterone.
In addition, if a woman is looking for a mature man who hypothetically has the resources to invest in her and offspring, she should care little for youthfully full lips or wide eyes.
To test the degree of masculinity or femininity preferred, the researchers began by making computer composites. For each ethnic group they studied, they took photographs of 20 men and 20 women in their early 20's, digitized the images and defined 174 salient features -- for example, nose-tip -- as "facial
landmarks." Those features were then used to calculate a single image of an average white man and average white woman, or average Japanese man and average Japanese woman.
The scientists defined "masculinity" and "femininity" as the measurable ways that the composite images of the two sexes differed from one another. "This seemed the best and most logical place to start," Penton-Voak said.
A group of 50 students and university staff members in Scotland and a similar number in Japan served as judges. Sitting at a computer console, the appraisers could manipulate the images with a mouse, morphing the pictures in either a more masculine or feminine direction. All the features would change simultaneously as the mouse moved, which meant that the appraisers did not have the option of shaping features to their particular liking -- say, by enlarging the jaw of the man while keeping his nose and brow comparatively slender.
In the end, the judges preferred, on average, significantly feminized female faces for both sexes. Female faces were rated as most attractive when they were about 20 percent more feminized than the female norm, while male faces were judged most appealing when they were about 15 percent more feminine than the standard male composite. Interestingly, when people were rating the faces of their own ethnic group, they liked an even greater degree of feminization of both male and female faces than when they were ranking the faces of a different ethnic group.
When asked to associate certain characteristics with masculine or feminine faces, appraisers said that masculinized male faces did look more dominant and mature than the average male face, but they also gave the masculinized faces comparatively lower scores in qualities like warmth, honesty, emotionality, cooperativeness and parenting abilities. The feminized male faces were given top ratings when it came to all the sensitive-guy personality traits.
For the female faces, masculinization also lent the women a look of comparatively high dominance and low warmth and trustworthiness, while the feminized faces ranked high in all aspects of sweetness and light. When it came to picking out the aspect of a potentially good mother, though, appraisers preferred the average female face to that of a feminized face.
This quirky result raises the question of what, from an evolutionary perspective, a man is choosing if he designates as highly attractive an ultra-feminized face that he does not necessarily think belongs to a woman capable of rearing his brood.
Indeed, as Ghirlanda and others point out, the results of the new study raise at least as many questions as they answer. Dr. Perrett and his colleagues suggest that the preferences they have detected show there is a limit to how sexually distinctive, or dimorphic, the faces of men and women can be. In many species, competition between males for the eye of females has resulted in profound sexual dimorphism, the most famous example being the difference between the ostentatious peacock and the drab peahen.
By the Perrett scenario, social skills like cooperativeness, honesty and gentleness proved generically desirable in the early stages of human evolution. Because such nurturing traits are associated with femaleness and juvenileness, the appeal of the feminine, youthful look became pansexual, and helped to counter such standard engines of sexual dimorphism as competition between males.
But as some critics have observed, the Perrett study does not account for at least one source of considerable esthetic dimorphism in humans: facial hair. None of the men photographed for the study had beards or mustaches. The role of facial hair in human evolution, and why men have it while women do not, remains a mystery.
In theory, though, prehistoric men had neither the time nor the wherewithal to shave, and so may have looked as distinct from prehistoric women as antlered male elks do from bare-headed does.
The scientific study of beauty is rife with mysteries and contradictions. A number of reports have demonstrated that a composite face is usually deemed more attractive than any one particular face -- that is, a consolidated image of 60 people is voted more attractive than is the image of most of the individual members. The classic evolutionary explanation for the triumph of the norm is that an average-looking person conveys a comforting familiarity, and is unlikely to harbor any unusual genetic mutations.
Yet the exaggerated pretty face has been shown to be favored over the average. If the faces of the 15 people rated as most attractive of the original pool of 60 are merged into a composite image, that averaged face outranks the first composite.
And if the features of the top-15 composite are then pushed to extremes, by raising and emphasizing the cheekbones, for example, the image is reckoned more beautiful still.
What remains to be understood is why one sort of extreme signal, like prominent cheekbones, should be considered alluring, while other extreme features, like a big nose, should be rejected in favor of a mainstream profile. If the norm is the most likely to be genetically healthy, why be drawn to any sign of genetic deviance?
Extreme beauty is not necessarily a sign of superior reproductive potential. For instance, there is a genetic condition called androgen insensitivity syndrome, in which a fetus with male chromosomes lacks the capacity to respond to the male hormones its embryonic testes produce. As a result, the fetus's body develops as female. Girls with the condition often grow into unusually beautiful women, with long legs, clear skin, ample breasts and thick hair; a number of famous models and actresses are thought to have androgen insensitivity syndrome. Yet women with A.I.S. lack normal reproductive organs and are infertile. By a cold Darwinian reckoning, their beauty is deceptive.
And, hey, Leonardo DiCaprio may not be as sweet as he looks, either, but do his lusting young fans really care?