Mate choice decisions: the role of facial beauty (PART 2)

Mate choice decisions: the role of facial beauty (PART 2)

Men's facial attractiveness

There is some controversy concerning the appearance of an attractive male. Before reading about this debate, you can examine your own preferences by completing the experiment presented below.



What do women want?

Given the central role of hormones in both the displays of male animals and the configuration of attractive female faces, several groups of experimenters have devised methods for systematically manipulating the hormone markers on male faces. One group, based in the UK, developed a method for ‘masculinizing’ or ‘feminizing’ male faces by applying caricaturing or warping transforms that exaggerate or reduce facial masculinity based on the differences between an average male and an average female face. Using this methodology, they found that both Japanese and UK women expressed a consistent preference for male faces that were more feminized than the average male. A group of US investigators manipulated masculinity by morphing an average male face towards an average female face (feminizing) or towards a male face that was rated high in perceived masculinity (masculinizing). In contrast to the UK findings, US females expressed a consistent preference for male faces that were more masculinized that the average male.



The discrepancy between the UK and US findings might have arisen from real cultural differences or from methodological differences. For example, the average male face used in the US studies was developed from a sample of only 16 randomly selected male faces. It is possible that this ‘average’ face is actually more feminized than the ‘average’ face used by the UK group. If so, the observed masculinized male preference of US participants might actually be feminized with respect to the UK average face. It is also possible that the UK group's linear extrapolation of female to male differences (caricaturing) to produce masculinized male faces does not capture the real-world hormonal interactions of androgens, estrogens and growth hormone, that underlie facial masculinity.



The four male faces shown in Figure I are designed to evaluate three theories of facial attractiveness. (b) is an average male face generated by a simultaneous morph of 117 randomly selected male faces. Votes for this face support the ‘average-is attractive’ hypothesis. Face (a) is the average face warped 20% of the distance towards an average female face. This ‘feminized’ face is similar to that found to be attractive by UK experimenters. Faces (c) and (d) are the average male face morphed 20% and 50% of the distance towards a very masculine male face. The ‘masculinized’ face (d) is similar to that found to be attractive in US studies.

Figure I. Which face is most attractive?



Although there is still no complete explanation for the apparently contradictory findings discussed above, attempts to resolve the issue have uncovered several important variables that influence women's preferences. Both the UK and US groups have found that women: (i) shift their preferences towards a more masculinized male face during the high-probability-of-conception phase of their menstrual cycle; and (ii) select more masculinized male faces for short-term mates (STMs) than long-term mates (LTMs). These are the very circumstances when females would be most interested in males with ‘good genes’.

The relationship between masculine secondary sexual traits and ‘good genes’ is supported by studies of fluctuating asymmetry (FA). FA is the measured deviation from perfect bilateral symmetry of those physical traits for which signed differences between the left and right sides have a mean of zero over the population. Across many species including humans males with low FAs enjoy better health and more mating success than asymmetrical males. Asymmetries can arise as a result of parasites, toxins or other insults encountered during the course of development, so global body symmetry is believed to be a valid index of immunocompetence. Although several studies have found that the facial attractiveness of males is correlated with their measured symmetry, Scheib et al. have demonstrated that this relationship persists even when judging pictures of one side of a male's face; a situation where cues to bilateral symmetry are absent. These authors concluded that facial masculinity serves as a proxy for bilateral symmetry. A follow-up study demonstrated a significant positive correlation between facial masculinity and body symmetry so it appears that, just like a peacock's tail, the testosterone markers on a human male's face might be attractive because they serve as ‘honest’ signals of good genetic quality.

The beholder's brain

Women and men agree on the appearance of attractive male and female faces, but electrophysiological and fMRI studies indicate that only attractive faces of the opposite sex evoke an emotional response in the brain of heterosexual observers. Such brain sex differences are presumably a consequence of the organizational effects of early androgen exposure. According to the standard model of sexual differentiation of the mammalian brain, the ‘default’ sex is female and maleness is a consequence of gonadal androgens that de-feminize and masculinize the developing embryo. In humans, studies of androgen insensitivity, congenital adrenal hyperplasia, idiopathic hypogonadotropic hypogonadism, Turner's syndrome, and homosexuality are all consistent with this model. The ‘default’ sex (female) is normally attracted to male features in adulthood but following early exposure to androgens there is a remarkable reversal in preference; the vast majority of the resulting sex (male) is now attracted to female features. Scarbrough and Johnston proposed that smaller variations in androgen exposure might account for some of the differences in mate preference within heterosexual males and females.

Women's mate preferences

To evaluate this hypothesis, Scarbrough and Johnston examined the mate choices of heterosexual females as a function of their ‘2D:4D’ digit ratio (index finger length divided by ring finger length), a sexually dimorphic trait (it is lower in males than females) that is a putative index of androgen exposure in the uterus. They found that compared with high 2D:4D females, low 2D:4D females were psychologically de-feminized (measured by the Bem Sex Role Inventory), physiologically de-feminized (measured by menstrual regularity), bonded poorly to their fathers (measured using a parental bonding instrument), reported shorter intimate relationships with males, and preferred a more masculine male face around ovulation. Although all women had a preference for a masculinized STM who was not significantly different from their attractive male choice, only low 2D:4D women desired similar masculine attributes in their LTMs.

It has been shown that women who prefer masculinized LTMs also like the odor of 4,16-androstadien-3-ol, a putative male pheromone. Furthermore, women with low 2D:4D digit ratios also score high on the Sociosexual Orientation Inventory, a measure of their willingness to engage in uncommitted sexual activity, and they might also have an impaired resistance to parasites. Taken together, these findings indicate that low 2D:4D women are more attracted to a male's secondary sexual characteristics (facial androgen markers and male pheromones), prefer such ‘good genes’ males as mates (STM, LTM, and at ovulation), but don't bond well (low paternal bonding, short relationships, and promiscuity), perhaps as a consequence of their emotional structure (de-feminized). By contrast, high 2D:4D females are stereotypically female, bond well to males, are sexually reserved, and seek less masculinized males for LTMs, or when there is a high probability of conception.

These two mate-choice strategies are best viewed as the extremes of a ‘good genes’ to ‘good dad’ continuum, with most females falling somewhere in between. A choice for ‘good genes’ directly benefits offspring through the inheritance of immunocompetence genes, whereas a ‘good dad’ choice offers indirect benefits to offspring in the form of paternal care and/or physical and psychological resources (see Figure 2).

Figure 2. Mate selection and personality traits. Using a movie that morphs a very masculine male face (frame 1 of 700) into an androgynous face, the facial pictures and vertical lines indicate the mean location of participants' dominant male (DOM), short-term mate (STM), long-term mate (LTM), average male (AVM) and androgynous face (AND) selections, with respect to experimentally assigned personality traits. F1 (‘Friend’ factor) is composed of positive attributes such as sensitive, helpful and trustworthy. F3 (‘Enemy’ factor) consists of undesirable attributes like selfish, controlling and threatening. The ‘Lover’ factor (F2) includes sexually exciting, supportive and healthy. The STM selection appears to be the best ‘good-genes’ choice (Lover factor), while avoiding the negative traits associated with high degrees of masculinity (Enemy factor). The LTM selection appears to trade off some ‘good genes’ attributes in favor of those required for a good friend and good father (included in F1).

The finding that STM and attractive faces are statistically identical suggests that STM selections are primarily based on affective ‘good genes’ considerations. By contrast, the additional cognitive inference of ‘good dad’ traits appears to influence LTM decisions. However, as noted above, this preference for ‘good dad’ traits varies with increasing digit ratio, feminization and bonding, hence implicating hormonal and affective factors in the evaluation. This observation provides additional support for Trafimow's hypothesis that cognitive evaluations depend on the affect (feeling) attached to such cognitions because affect serves as the common currency for integrating emotional and cognitive appraisals into the decision process. Indeed, examining how 2D:4D digit ratio and steroid hormone levels influence the effects of other cognitive factors involved in mate choice, or other decisions, offers a general methodology for revealing how emotional and cognitive processes might be integrated into decision processes.

Conclusions

There is now considerable evidence that (i) pubertal hormone markers influence the attractiveness of male and female faces. I am proposing, however, that the emotional impact of such sexually selected signals depends on the beholder's brain that has also been (ii) organized by hormonal events early in development, and (iii) activated by circulating hormones following puberty. For the most part, I have constructed and supported this model of attraction on the basis of experimental research on facial beauty, but I am also proposing that this hormone-mediated model of sexual selection is applicable to all sexually selected traits, and the list of these might be very large indeed