Поиск по сайту




Пишите нам: info@ethology.ru

Follow etholog on Twitter

Система Orphus

Новости
Библиотека
Видео
Разное
Кросс-культурный метод
Старые форумы
Рекомендуем
Не в тему

список статей


MHC-heterozygosity and human facial attractiveness

S. Craig Robertsab, Anthony C. Littleb, L. Morris Goslinga, David I. Perrettc, Vaughan Carterd, Benedict C. Jonese, Ian Penton-Voakf, Marion Petriea

1. Introduction

2. Methods

2.1. Genotyping of male participants

2.2. Image scoring

2.3. Skin ratings/symmetry estimates

2.4. MHC similarity

2.5. Analyses

3. Results

3.1. Study 1: effects of heterozygosity

3.2. Study 2: controlling for genetic similarity

4. Discussion

Acknowledgment

References

Copyright

1. Introduction

Sexual selection theory asserts that males maximise reproductive success through seeking multiple matings, while females achieve this goal through discrimination of mate quality, or choosiness (Bateman, 1948, Trivers, 1972). Through mate choice, females gain direct and indirect benefits, where direct benefits include resources provided by males to females or offspring, and indirect benefits refer to genetic properties that enhance fitness of resulting progeny (Andersson, 1994, Kirkpatrick & Ryan, 1991). Females may base their discrimination on phenotypic or behavioural variation among candidate males in traits that signal genetic quality (Hasselquist et al., 1996, Petrie, 1994) or genetic complementarity (usually dissimilar genotype: Potts et al., 1991, Yamaguchi et al., 1981), or both (Colegrave et al., 2002, Roberts & Gosling, 2003).

In most cases, we know little about the exact nature of the good genes involved, although Brown, 1997, Brown, 1999 has recently suggested that genetic quality, in this sense, may often be defined as heterozygosity at certain loci. He proposed that heterozygosity should be correlated with the expression of condition-dependent male traits and that females should value heterozygosity in their offspring and, in some cases, their mates. Many studies investigating influences of heterozygosity on mating patterns have focussed on genes in the major histocompatibility complex (MHC, known in humans as human leukocyte antigen loci, HLA). MHC genes encode proteins involved in immunological self/nonself recognition and are among the most polymorphic in the genome (Mungall et al., 2003).

It is now well-established that MHC disassortative mating preferences exist in several species, with beneficial effects in terms of offspring heterozygosity (Bernatchez & Landry, 2003, Jordan & Bruford, 1998, Penn & Potts, 1998), but the extent to which heterozygosity in mates plays a role in female mate preferences is less clear. Preferences for heterozygosity in mates could result in direct benefits to females, for example, through reduced risk of disease transmission or provision of high-quality paternal care (Kirkpatrick & Ryan, 1991) because heterozygous males are often less susceptible to infectious diseases (Carrington et al., 1999, McClelland et al., 2003, McClelland et al., 2003, Penn et al., 2002, Thursz et al., 1997). Although heterozygote mating advantages have been demonstrated less often than predicted by theory (Brown, 1997, Brown, 1999), higher reproductive success has been found in male rhesus macaques, Macaca mulatta, which were heterozygous at a Class II locus (Sauermann et al., 2001), and in male spotless starlings, Sturnus unicolor, of intermediate heterozygosity (Aparicio, Cordero, & Veiga, 2001). Mate choice for males with high MHC-allelic diversity occurs in sticklebacks (Reusch, Haberli, Aeschlimann, & Milinski, 2001), while in humans, Thornhill et al. (2003) found that women prefer odors of men who are relatively heterozygous at MHC loci. Such odor effects may not be independent of other cues (Cornwell et al., 2004), and this raises the intriguing possibility that MHC genes might also be involved in facial preferences because facial attractiveness plays a key role in human mate choice.

Facial characteristics are thought to advertise underlying “good genes” to potential mates (Penton-Voak et al., 2001, Perrett et al., 1994, Thornhill & Gangestad, 1999). These characteristics include indicators of healthiness and developmental stability (Fink et al., 2001, Fink & Penton-Voak, 2002, Jones et al., 2001, Thornhill & Gangestad, 1999) and, potentially, immunocompetence (Penton-Voak et al., 2001, Perrett et al., 1998, Rhodes et al., 2003). Only recently, however, has the potential link between MHC genotype and facial preferences been investigated. In addition to odor preferences, Thornhill et al. (2003) tested facial preferences for MHC heterozygosity but found no significant effects. While the reason for the discrepancy between facial and odor preferences is unclear, it emphasises the need for further investigation of the potential links between MHC and facial attractiveness.

Here, we test the hypothesis that MHC heterozygosity influences women's preferences for male faces. We presented women with photographs of men genotyped at three MHC loci and asked them to score each image for attractiveness. We compare scores of men who were heterozygous at all three loci with those who were homozygous at one or more of these loci. In addition, we compare two potential condition-dependent facial characteristics of men in each group, each of which have been previously linked to facial attractiveness: fluctuating asymmetry (Penton-Voak et al., 2001, Perrett et al., 1999) and visible skin condition (Fink et al., 2001, Jones et al., 2004, Jones et al., 2004). Finally, because MHC similarity is known to influence the attractiveness of human odor (Jacob et al., 2002, Thornhill et al., 2003, Wedekind & Furi, 1997, Wedekind et al., 1995), we tested whether these attractiveness judgements were independent of levels of MHC similarity.

2. Methods

2.1. Genotyping of male participants

Participants were students or staff at the University of Newcastle. To minimise possible confounding variables, we included only participants who were white. Ninety-two men aged between 18 and 31 (mean=21) contributed to the study.

Approximately 5 ml of blood was collected using vacuettes (Greiner Bio-One, Stonehouse, Gloucestershire, UK) lined with EDTA to prevent clotting. Genomic DNA was typed at three key MHC loci (HLA-A, -B, and -DRB1), by polymerase-chain reaction using sequence-specific primers (PCR-SSP). We used these three loci because, with over 300 known alleles each, they are the most polymorphic HLA loci, according to the HLA database (http://www.ebi.ac.uk/imgt/hla/stats.html), and because they have been used in previous studies of HLA-associated mate preferences (Jacob et al., 2002, Thornhill et al., 2003, Wedekind & Furi, 1997, Wedekind et al., 1995). Among the men, 13 different alleles were recorded at HLA-A, 27 at HLA-B, and 13 at HLA-DRB1. Sixty-nine men were heterozygous at all three loci, while 23 were homozygous at one or more loci. Of these, 8 were homozygous at HLA-A, 5 at HLA-B, 6 at HLA-DRB1, 3 at both HLA-A and -DRB1, and 1 at both HLA-A and -B.

2.2. Image scoring

Digital color photographs of the men's faces were taken under standard lighting conditions using a Nikon Coolpix 775 digital camera and a resolution of 1600×1200 pixels. Men were instructed to look directly at the camera and adopt a neutral expression. Each image was normalised on the interpupillary distance (Penton-Voak et al., 2001) and digitally masked so that only the face was visible, minimising potentially confounding information about hairstyle and clothing (e.g., Roberts et al., 2004). These images were presented to 50 female participants (age 18–49 years, mean=23) in random order, on a liquid crystal display computer screen (on-screen face size approximately 12×18 cm). Women rated the attractiveness of these faces using a seven-point rating scale (high scores=attractive).

2.3. Skin ratings/symmetry estimates

Estimates of horizontal and vertical facial symmetry were obtained using the xy coordinates of seven bilateral landmark points on the 2D facial images, as defined previously by Penton-Voak et al. (2001) and Scheib, Gangestad, and Thornhill (1999). Both studies found that symmetry, as measured using these landmarks, correlates positively with perceived attractiveness.

All women were also asked to rate the apparent healthiness of patches of skin from each man's image using a seven-point scale (high score=healthy). Skin patch images were squares of skin from the right cheek of each man, the equivalent of a 2.5-cm square, the lower edge being aligned with the bottom of the nose and the right-hand edge immediately next to the right nostril. Images were magnified by 300% to facilitate ratings (Jones, Little, Burt, et al., 2004). This process slightly blurs the images, but the degree of blurring is constant across images. Although magnification made rating skin patches easier for participants than without it, skin patches remain poor substitutes, in our view, for the assessment in vivo of either the same skin area or, indeed, skin covering the entire face. Hence, magnification should be seen as an aid towards ecologically valid levels of mate quality discrimination rather than providing unrealistically high resolution. Skin patches were presented in randomised order, in a different order than the whole faces. Cues to facial identity, shape, or facial hair were not visible in the skin patches. These methods have been successfully used by Jones, Little, Feinberg, et al. (2004) and Jones, Little, Burt, et al. (2004); example skin patch stimuli (though from different individuals than those used here) can be seen in Jones, Little, Burt et al.

All women were asked to complete both the facial attractiveness and the skin patch healthiness rating tasks. The order in which these tasks were completed was alternated between female raters.

2.4. MHC similarity

We carried out a second experiment (Study 2) to check that the effects of heterozygosity were independent of relative levels of genetic similarity between raters and men in the images. Thirty-six women participated, having been genotyped at the same loci as the men. Only 5 also took part in the first experiment. MHC similarity between females and each male was computed as the total number of alleles in common across the three loci (see also Thornhill et al., 2003, Wedekind et al., 1995).

In addition to genetic similarity, we also controlled more closely a number of additional factors. First, we controlled the country of origin (to standardise the degree of allelic diversity) by excluding 13 non-British men from mainland Europe (all 36 women were also white and of British origin). Second, we excluded three men with beards because facial hair might influence facial judgements. Of the remaining 76 men, 20 were homozygous at one or more loci (6 at HLA-A, 4 at HLA-B, 6 at HLA-DRB1, 3 at both HLA-A and -DRB1, and 1 at both HLA-A and -B). Finally, as it is known that the stage of menstrual cycle influences women's facial preferences (Penton-Voak & Perrett, 2001, Penton-Voak et al., 1999), all women raters were tested in the late follicular phase, between 10 and 14 days after the beginning of their cycle, and none of the participants were using hormone-based methods of contraception. In this smaller experiment, women rated the facial attractiveness and skin healthiness of the 20 men with one or more homozygous loci and 20 heterozygotes selected at random from the remaining 56 men (heterozygotes were selected using the “Select random sample of cases” function in SPSS version 11). All women saw the same 40 faces. Within this sample, age matching of both men and women was improved in comparison with the first experiment (range: men 18–25; women 18–34). The order in which women rated faces for attractiveness and skin patches for perceived healthiness was alternated. Within each task, images were also presented in a randomised order.

2.5. Analyses

We used standardised image scores (number of standard deviations from the mean for each rater; Sokal & Rohlf, 1995, p. 110) to control for individual differences in the use of the rating scale (e.g., Pollard, Shepherd, & Shepherd, 1999). As expected, significance levels are improved slightly after standardisation; however, the difference in facial attractiveness scores of the homozygotes and heterozygotes that we report is also significant using raw scores (p=.023; for the raw skin healthiness ratings p=.052). We calculated the mean standardised scores for each image across all raters and compared the scores for men with one or more homozygous loci against those who were heterozygous at all three loci. For illustrative purposes only, we present the mean scores of all three levels of heterozygosity in Fig. 1. We pooled men with one and two homozygous loci primarily because of the small number of men (four) with two homozygous loci. Calculating heterozygosity across several loci is a common technique and correlates with the fitness and expression of secondary sexual traits in birds (Aparicio et al., 2001, Foerster et al., 2003) and with disease progression in humans (Carrington et al., 1999, Tang et al., 1999). All data were normally distributed and analysed using SPSS Version 11 with two-tailed tests.


View full-size image.

Fig. 1. Facial attractiveness and skin healthiness ratings of men with different numbers of homozygous MHC loci. (a) Ratings are means (±S.E.) of standardised scores given by 50 women (Study 1). Sixty-nine men were heterozygous at all three loci, 19 were homozygous at one locus, and 4 were homozygous at two loci. (b) Ratings from 36 genotyped women (Study 2); n=56 men heterozygous at all three loci, 16 men homozygous at one locus, and 4 homozygous at two loci.


3. Results

3.1. Study 1: effects of heterozygosity

We found a significant effect of heterozygosity on ratings of facial attractiveness (Fig. 1a). Faces of men who were heterozygous at all three loci were judged more attractive than were the faces of men who were homozygous at one or more loci [independent samples t test, t(90)=2.29, p=.024]. Interrater reliability was high (Cronbach's α=.95), and mean attractiveness ratings for heterozygotes were higher than for homozygotes in 49 of the 50 women (sign test, p<.001), indicating a high level of agreement among women.

We also tested associations between heterozygosity and two potential health indicators that have been linked to facial attractiveness: fluctuating asymmetry (Penton-Voak et al., 2001, Perrett et al., 1999) and visible skin condition (Fink et al., 2001, Jones et al., 2004, Jones et al., 2004). We found no significant effect of heterozygosity on horizontal measures of facial symmetry [mean±S.E.: heterozygous=39.9±2.6, homozygous=42.5±4.8; t(90)=0.50, p=.62], on vertical symmetry [heterozygous=14.6±.73, homozygous=12.6±1.3; t(90)=1.38, p=.17], or on the mean of these measures (p>.9). Attractiveness scores in this sample were also not correlated with either measure of symmetry (horizontal, r=.03; vertical, r=−.067). However, the skin of heterozygous men was perceived by women as more healthy than was the skin of homozygous men [Fig. 1; t(90)=2.02, p=.046], and mean ratings of whole-face attractiveness for each man were positively correlated with mean skin healthiness ratings (Fig. 2a; r=.301, n=92, p=.004). There was again strong agreement among women: The mean healthiness ratings of skin of heterozygotes were higher than those of homozygotes in 45/50 women, and interrater reliability was high (Cronbach's α=.94).


View full-size image.

Fig. 2. Correlation between facial attractiveness and skin healthiness scores. (a) Mean scores for 92 men rated in random order by 50 women, with scores for each variable standardised within rater to control for differential use of the scale (Study 1). (b) Mean scores of 20 heterozygotes and 20 homozygotes, randomly selected from the 92 men, rated by a second group of 36 genotyped women (Study 2). Mean ratings of whole-face attractiveness for each man were positively correlated with mean skin healthiness ratings (a: p=.004; b: p=0.041).


3.2. Study 2: controlling for genetic similarity

We investigated the potentially confounding influence of levels of MHC similarity between female raters and males by recording facial attractiveness and skin healthiness ratings from a second group of women who were also genotyped. In this group, there was again a marked preference for faces of heterozygous men, with mean scores given by each rater to heterozygotes being higher than the mean homozygote scores [paired t test, t(35)=8.28, p<.001; Fig. 1b]. Women again judged skin from the heterozygous men to be healthier than was the skin of the homozygous men [paired t test, t(35)=6.38, p<.001], and the mean ratings of whole-face attractiveness for each man were also positively correlated with the mean skin healthiness ratings (r=.325, n=40, p=.041; Fig. 2b).

To test the effect of heterozygosity independently of similarity, we used a three-way ANOVA with the standardised attractiveness score for each image as the dependent variable, male heterozygosity and the number of shared alleles as fixed factors, and female rater as a random factor. This analysis revealed a significant and independent effect of heterozygosity [F(1,1194)=4.66, p=.031]. The main effect of MHC similarity also approached significance [F(5,262)=2.08, p=.069], but there was no effect of rater (p>.9) and no significant interaction terms, although the Similarity×Rater interaction approached significance (p=.051). After model simplification (removing interactions in which p>.1), the final model revealed significant positive effects on attractiveness of heterozygosity [F(1,1279)=41.36, p<.01] and the number of shared alleles [F(5,253)=2.26, p=.049], while the Rater×Similarity interaction was also significant [F(119,1279)=1.27, p=.033; no main effect of rater, p>.9].

We further investigated the independent effect of heterozygosity by calculating the mean scores given by each rater to heterozygotes and to homozygotes who shared zero, one, two, or three MHC alleles with her. For each degree of similarity, we compared these mean scores using planned paired t tests. The mean facial attractiveness scores were significantly higher for heterozygotes than for homozygotes when rater and male image shared zero, one, or three alleles [zero: t(34)=3.10, p=.004; one: t(35)=4.28, p<.001; three: t(22)=2.95, p=.007] and tended toward significance when two alleles were shared [t(35)=1.83, p=.076; Fig. 3].


View full-size image.

Fig. 3. Mean standardised attractiveness scores (±S.E.) of MHC-heterozygous (□) and homozygous (▪) men, subdivided according to the number of shared alleles with female raters; n=35, 36, 36, and 23 women for zero, one, two, and three common alleles, respectively (sample size varies because not all women share particular numbers of shared alleles with at least one homozygote and at least one heterozygote). p values are shown from paired t tests; asterisk (*) indicates that the comparison survives a Bonferroni correction for multiple tests.


4. Discussion

Our results suggest that females find the faces of MHC-heterozygous men more attractive than faces of homozygotes. They lend support to the idea that average heterozygosity at the MHC may be important in mating decisions (Brown, 1999). Arguably, this preference for heterozygosity might not be MHC specific but could rather arise through a general eschewal of relatively inbred males (or those with high average homozygosity at many loci), provided that MHC homozygosity is correlated with inbreeding and any associated deleterious effects. However, current evidence suggests that this is not always the case because Carrington et al. (1999) found no correlation in several human cohorts between homozygosity at MHC loci and homozygosity at several microsatellite loci, either within the MHC region or on different chromosomes. This indicates that the preference for heterozygous faces revealed here may indeed be specific to MHC heterozygosity and is therefore likely to be mediated by MHC-correlated phenotypic traits.

Our second study shows that high average heterozygosity carried a significant effect independently of the degree of MHC similarity between women raters and the photographed men. The absence of an interaction between heterozygosity and rater is indicative of a general preference across women. Consistent with other studies (e.g., Langlois et al., 2000), we found a high degree of agreement between women in attractiveness ratings of the men, and our results suggest that heterozygosity contributes to this agreement, along with other previously documented characteristics including masculinity, symmetry, and averageness (e.g., Penton-Voak et al., 2001, Rhodes et al., 2001). However, the number of shared alleles between the pictured males and the female raters also appears to be positively correlated with facial preferences. This is a surprising result in the light of odor preferences for MHC-dissimilar individuals (Wedekind & Furi, 1997, Wedekind et al., 1995), although it is consistent with other face studies that indicate assortative preferences and phenotypic facial similarities within human couples (Griffiths & Kunz, 1973, Hinsz, 1989, Penton-Voak et al., 1999). However, the interaction between rater and similarity indicates variability between women in the strength of their preference. While Fig. 3 appears to suggest that women give lowest scores to men of intermediate similarity (sharing one or two alleles), in fact, this nonmonotonic pattern is due to some women giving higher scores to faces of MHC-similar men while others preferred faces of MHC-dissimilar men. More work is needed to understand both this variability in preference and why the dominant preference appears to be assortative rather than for MHC-dissimilar faces, and we are actively pursuing this. Nonetheless, this result adds further evidence that MHC genotype is correlated, in some way, with facial appearance.

Although we found only four men with two homozygous loci in our sample, preventing statistical comparison with those having single-locus homozygosity, their lower mean attractiveness and skin healthiness scores (Fig. 1) suggest that homozygosity at multiple loci may compound associated negative effects. Sample sizes are small, thus, this trend should be interpreted with caution, but it is consistent with studies that find separate, additive effects of homozygosity at different loci on progression to AIDS in HIV+ patients, suggesting that multilocus homozygosity enhances negative health effects (Carrington et al., 1999, Tang et al., 1999).

While we find that male heterozygosity is associated with attractiveness, Thornhill et al. (2003) detected no such effect. Although both studies investigated heterozygosity at the same three loci, there are some methodological differences that could explain the differences in results. First, ethnic diversity of participants was greater in the study of Thornhill et al. (only 62% of men and 48% of women were reported as Caucasian), introducing a potentially large confounding effect on attractiveness ratings. An ethnically diverse sample will also promote heterogeneity in MHC alleles, potentially altering homozygote frequencies in the population (e.g., Budowle, Koons, & Moretti, 1998). Second, there was wider variation in the age of participants in Thornhill et al.'s sample than in ours, particularly in the men (range 18–54). Third, unlike Thornhill et al., we masked images to minimise the potentially confounding influences of hair style/condition and clothing. Each of these differences is likely to reduce variance (unrelated to MHC heterozygosity) in judgements by raters in our study. While Thornhill et al. did find correlations between heterozygosity and odor, perception of odor may be less sensitive to ethnicity and age than are facial judgements, and the confounding cues of hair and clothing would have been absent. Nonetheless, further work in different populations is clearly required to clarify the strength of the observed effects.

In view of the central role of MHC genes in immune response, differences in apparent healthiness are a likely explanation for the observed preference for heterozygote faces. MHC-heterozygote advantage in resistance to infectious diseases has been found in several recent studies, notably with regard to infections in mice by Salmonella and Listeria (Penn et al., 2002), and in humans by Hepatitis B (Thursz et al., 1997) and HIV (Carrington et al., 1999). In addition, of particular relevance to this study, HLA-Cw*0602 heterozygotes have a markedly lower risk of developing psoriasis, a chronic inflammatory skin disease (Gudjonsson et al., 2003). We have identified apparent skin healthiness as a potential mechanism for the effect that we describe, although it may not be the only one. Like Jones, Little, Feinberg, et al. (2004) and Jones, Little, Burt et al. (2004), we find that skin condition might, in itself, be a direct visual cue to mate quality, as it correlates with attractiveness judgements even when information about facial shape is removed. Further work is required to determine the precise nature of the cues used to assess skin healthiness, although candidates would include coloration, texture, and blemishes.

Our results are consistent with previous studies that find a correlation between perceived health and facial attractiveness (Cunningham, 1986, Jones et al., 2001, Thornhill & Gangestad, 1999). However, of three studies using measures of actual, rather than perceived health (Kalick et al., 1998, Rhodes et al., 2003, Shackelford & Larsen, 1999), only one (Shackelford & Larsen, 1999) has found a similar relationship. One reason for this may be that modern medicine will tend to disguise or weaken effects that have evolved in ancestral times (Jones et al., 2001, Thornhill & Gangestad, 1999), although the studies by Kalick et al. (1998) and Rhodes et al. (2003) used images collected before the widespread use of vaccinations and, especially, antibiotics. A second explanation is that the two studies, to find no relationship between health and attractiveness, used monochrome photographs, while the other (Shackelford & Larsen, 1999), like our own, used color images. This pattern of results suggests that color may be necessary to discriminate subtle variation in healthiness associated with heterozygosity.

In contrast to mate preferences based on MHC dissimilarity, a preference for heterozygotes is more likely to have been selected for direct benefits to females than through indirect fitness advantages for offspring. Avoidance of homozygous males could still increase offspring heterozygosity (e.g., if a male is homozygous for an allele that a female also carries), but such cases will be relatively rare, especially in view of the polymorphic nature of the MHC. In a relatively monogamous species like humans, evolutionary benefits from choosing healthy mates are likely to include a prolonged period of high-quality parental care and reduced risk of contracting disease for females and their offspring (Kirkpatrick & Ryan, 1991). Further research is needed to clarify the relative influences on the mate choice of these direct benefits compared with those indirect benefits that might result from choosing genetically compatible mates. The balance between these two selective forces may explain much of the variation in individual preferences. Although judgements of facial attractiveness typically show considerable agreement among raters, even across cultures (Langlois et al., 2000), individual differences with respect to MHC type are also likely to be evident. Thus, while studies that focus on relative genetic similarity suggest that “there is someone for everyone” (Jacob et al., 2002), a preference for heterozygotes implies wider agreement among women on attractiveness judgements of individual men.

Acknowledgments

We thank all our volunteers for their participation, staff at the National Blood Service Newcastle for doing the tissue typing, and D. Penn, G. Roberts, C. Rowe, and two anonymous referees for their comments on the manuscript. The work was funded by a Wellcome Trust grant to M.P. and L.M.G.

References

return to Article Outline

Andersson, 1994 1.Andersson M. Sexual selection. Princeton, NJ: Princeton University Press; 1994;.

Aparicio et al., 2001 2.Aparicio JM, Cordero PJ, Veiga JP. A test of the hypothesis of mate choice based on heterozygosity in the spotless starling. Animal Behavior. 2001;62:1001–1006.

Bateman, 1948 3.Bateman AJ. Intrasexual selection in Drosophila. Heredity. 1948;2:349–368.

Bernatchez & Landry, 2003 4.Bernatchez L, Landry C. MHC studies in nonmodel vertebrates: what have we learned about natural selection in 15 years?. Journal of Evolutionary Biology. 2003;16:363–377. CrossRef

Brown, 1997 5.Brown JL. A theory of mate choice based on heterozygosity. Behavioural Ecology. 1997;8:60–65.

Brown, 1999 6.Brown JL. The new heterozygosity theory of mate choice and the MHC. Genetica. 1999;104:215–221. MEDLINE | CrossRef

Budowle et al., 1998 7.Budowle B, Koons BW, Moretti TR. Subtyping of the HLA-DQA1 locus and independence testing with PM and STR/VNTR loci. Journal of Forensic Sciences. 1998;43:657–660. MEDLINE

Carrington et al., 1999 8.Carrington M, Nelson GW, Martin MP, Kissner T, Vlahov D, Goedert JJ, et al. HLA and HIV-1: heterozygote advantage and B*35-Cw*04 disadvantage. Science. 1999;283:1748–1752. MEDLINE | CrossRef

Colegrave et al., 2002 9.Colegrave N, Kotiaho JS, Tomkins JL. Mate choice or polyandry: reconciling genetic compatibility and good genes sexual selection. Evolutionary Ecology Research. 2002;4:911–917.

Cornwell et al., 2004 10.Cornwell RE, Boothroyd L, Burt DM, Feinberg DR, Jones BC, Little AC, et al. Concordant preferences for opposite-sex signals? Human pheromones and facial characteristics. Proceedings of the Royal Society of London. Series B. 2004;271:635–640. MEDLINE | CrossRef

Cunningham, 1986 11.Cunningham MR. Measuring the physical in physical attractiveness: quasi-experiments on the sociobiology of female facial beauty. Journal of Personality and Social Psychology. 1986;50(5):925–935. CrossRef

Fink et al., 2001 12.Fink B, Grammar K, Thornhill R. Human (Homo sapiens) facial attractiveness in relation to skin texture and color. Journal of Comparative Psychology. 2001;115:92–99. MEDLINE | CrossRef

Fink & Penton-Voak, 2002 13.Fink B, Penton-Voak I. Evolutionary psychology of facial attractiveness. Current Directions in Psychological Sciences. 2002;11:154–158.

Foerster et al., 2003 14.Foerster K, Delhey K, Johnsen A, Lifjeld JT, Kempenaers B. Females increase offspring heterozygosity and fitness through extra-pair matings. Nature. 2003;425:714–717. CrossRef

Griffiths & Kunz, 1973 15.Griffiths RW, Kunz PR. Assortative mating: a study of physiognomic homogamy. Social Biology. 1973;20:448–453.

Gudjonsson et al., 2003 16.Gudjonsson JE, Karason A, Antonsdottir A, Runarsdottir EH, Hauksson VB, Upmanyu R, et al. Psoriasis patients who are homozygous for the HLA-Cw*0602 allele have a 2.5-fold increased risk of developing psoriasis compared with Cw6 heterozygotes. British Journal of Dermatology. 2003;148:233–235. MEDLINE | CrossRef

Hasselquist et al., 1996 17.Hasselquist D, Bensch S, Schantz TV. Correlation between male song repertoire, extra-pair paternity and offspring survival in the great reed warbler. Nature. 1996;381:229–232. CrossRef

Hinsz, 1989 18.Hinsz VB. Facial resemblance in engaged and married couples. Journal of Social and Personal Relationships. 1989;6:223–229.

Jacob et al., 2002 19.Jacob S, McClintock MK, Zelano B, Ober C. Paternally inherited HLA alleles are associated with women's choice of male odor. Nature Genetic. 2002;30:175–179.

Jones et al., 2001 20.Jones BC, Little AC, Penton-Voak IS, Tiddeman BP, Burt DM, Perrett DI. Facial symmetry and judgements of apparent health—Support for a “good genes” explanation of the attractiveness–symmetry relationship. Evolution of Human Behavior. 2001;22:417–429.

Jones et al., 2004 21.Jones BC, Little AC, Feinberg DR, Penton-Voak IS, Tiddeman BP, Perrett DI. The relationship between shape symmetry and perceived skin condition in male facial attractiveness. Evolution of Human Behavior. 2004;25:24–30.

Jones et al., 2004 22.Jones BC, Little AC, Burt DM, Perrett PI. When facial attractiveness is only skin deep. Perception. 2004;33:569–576. MEDLINE | CrossRef

Jordan & Bruford, 1998 23.Jordan WC, Bruford MW. New perspectives on mate choice and the MHC. Heredity. 1998;81:239–245. CrossRef

Kalick et al., 1998 24.Kalick SM, Zebrowitz LA, Langlois JH, Johnson RM. Does human facial attractiveness honestly advertise health? Longitudinal data on an evolutionary question. Psychologica Science. 1998;9(1):8–13.

Kirkpatrick & Ryan, 1991 25.Kirkpatrick M, Ryan MJ. The evolution of mating preferences and the paradox of the lek. Nature. 1991;350:33–38. CrossRef

Langlois et al., 2000 26.Langlois JH, Kalakanis L, Rubenstein AJ, Larson A, Hallamm M, Smoot M. Maxims or myths of beauty? A meta-analytic and theoretical review. Psychological Bulletin. 2000;126:390–423. MEDLINE | CrossRef

McClelland et al., 2003 27.McClelland EE, Granger DL, Potts WK. Major histocompatibility complex-dependent susceptibility to Cryptococcus neoformans in mice. Infection and Immunity. 2003;71(8):4815–4817. MEDLINE | CrossRef

McClelland et al., 2003 28.McClelland EE, Penn DJ, Potts WK. Major histocompatibility complex heterozygote superiority during coinfection. Infection and Immunity. 2003;71:2079–2086. MEDLINE | CrossRef

Mungall et al., 2003 29.Mungall AJ, Palmer SA, Sims SK, Edwards CA, Ashurst JL, Wilming L, et al.. The DNA sequence and analysis of human chromosome 6. Nature. 2003;425:805–811. CrossRef

Penn & Potts, 1998 30.Penn D, Potts WK. How do major histocompatibility complex genes influence odor and mating preferences?. Advances if Immunology. 1998;69:411–436.

Penn et al., 2002 31.Penn DJ, Damjanovich K, Potts WK. MHC heterozygosity confers a selective advantage against multiple-strain infections. Proceedings of the National Academy of Sciences of the United States of America. 2002;99:11260–11264. MEDLINE | CrossRef

Penton-Voak & Perrett, 2001 32.Penton-Voak IS, Perrett DI. Male facial attractiveness: perceived personality and shifting female preferences for male traits across the menstrual cycle. Advances in the Study of Behavior. 2001;30:219–259.

Penton-Voak et al., 1999 33.Penton-Voak IS, Perrett DI, Castles DL, Kobayashi T, Burt DM, Murray LK, et al. Menstrual cycle alters face preference. Nature. 1999;399(6738):741–742. MEDLINE | CrossRef

Penton-Voak et al., 1999 34.Penton-Voak I, Perrett D, Pierce J. Computer graphic studies of the role of facial similarity in attractiveness judgements. Current Psychology. 1999;18:104–117.

Penton-Voak et al., 2001 35.Penton-Voak IS, Jones BC, Little AC, Baker S, Tiddeman B, Burt DM, et al. Symmetry, sexual dimorphism in facial proportions and male facial attractiveness. Proceedings of the Royal Society of London. Series B. 2001;268:1617–1623. MEDLINE | CrossRef

Perrett et al., 1999 36.Perrett DI, Burt DM, Penton-Voak IS, Lee KJ, Rowland DA, Edwards R. Symmetry and human facial attractiveness. Evolution of Human Behavior. 1999;20(5):295–307.

Perrett et al., 1998 37.Perrett DI, Lee KJ, Penton-Voak IS, Rowland DR, Yoshikawa S, Burt DM, et al. Effects of sexual dimorphism on facial attractiveness. Nature. 1998;394:884–887. MEDLINE | CrossRef

Perrett et al., 1994 38.Perrett DI, May KA, Yoshikawa S. Facial shape and judgments of female attractiveness. Nature. 1994;368(6468):239–242. MEDLINE | CrossRef

Petrie, 1994 39.Petrie M. Improved growth and survival of offspring of peacocks with more elaborate trains. Nature. 1994;371:598–599. CrossRef

Pollard et al., 1999 40.Pollard J, Shepherd J, Shepherd J. Average faces are average faces. Current Psychology. 1999;18:98–103.

Potts et al., 1991 41.Potts WK, Manning CJ, Wakeland EK. Mating patterns in seminatural populations of mice influenced by MHC genotype. Nature. 1991;352:619–621. MEDLINE | CrossRef

Reusch et al., 2001 42.Reusch TBH, Haberli MA, Aeschlimann PB, Milinski M. Female sticklebacks count alleles in a strategy of sexual selection explaining MHC polymorphism. Nature. 2001;414:300–302. MEDLINE | CrossRef

Rhodes et al., 2003 43.Rhodes G, Chan J, Zebrowitz LA, Simmons LW. Does sexual dimorphism in human faces signal health?. Proceedings of the Royal Society of London. Series B. 2003;270:S93–S95(Suppl.). CrossRef

Rhodes et al., 2001 44.Rhodes G, Zebrowitz LA, Clark A, Kalick SM, Hightower A, McKay R. Do facial averageness and symmetry signal health?. Evolution of Human Behavior. 2001;22:31–46.

Roberts & Gosling, 2003 45.Roberts SC, Gosling LM. Genetic similarity and quality interact in mate choice decisions by female mice. Nature Genetics. 2003;35:103–106. MEDLINE | CrossRef

Roberts et al., 2004 46.Roberts SC, Havlicek J, Flegr J, Hruskova M, Little AC, Jones BC, et al. Female facial attractiveness increases during the fertile phase of the menstrual cycle. Proceedings of the Royal Society of London. Series B. 2004;271:S270–S272(Suppl.). CrossRef

Sauermann et al., 2001 47.Sauermann U, Nürnberg P, Bercovitch FB, Berard JD, Trefilov A, Widdig A, et al. Increased reproductive success of MHC class II heterozygous males among free-ranging rhesus macaques. Human Genetics. 2001;108:249–254.

Scheib et al., 1999 48.Scheib JE, Gangestad SW, Thornhill R. Facial attractiveness, symmetry, and cues to good genes. Proceedings of the Royal Society of London. Series B. 1999;266:1913–1917. MEDLINE | CrossRef

Shackelford & Larsen, 1999 49.Shackelford TK, Larsen RJ. Facial attractiveness and physical health. Evolution of Human Behavior. 1999;20:71–76.

Sokal & Rohlf, 1995 50.Sokal RR, Rohlf FJ. Biometry. 3rd ed.. New York: W.H. Freeman; 1995;.

Tang et al., 1999 51.Tang JM, Costello C, Keet IPM, Rivers C, LeBlanc S, Karita E, et al. HLA class I homozygosity accelerates disease progression in human immunodeficiency virus type I infection. AIDS Research and Human Retroviruses. 1999;15:317–324. MEDLINE

Thornhill & Gangestad, 1999 52.Thornhill R, Gangestad SW. Facial attractiveness. Trends in Cognitive Sciences. 1999;3(12):452–460. CrossRef

Thornhill et al., 2003 53.Thornhill R, Gangestad SW, Miller R, Scheyd G, McCullough JK, Franklin M. Major histocompatibility genes, symmetry and body scent attractiveness in men and women. Behavioral Ecology. 2003;14:668–678.

Thursz et al., 1997 54.Thursz MR, Thomas HC, Greenwood BM, Hill AV. Heterozygote advantage for HLA class-II type in hepatitis B virus infection. Nature Genetics. 1997;17:11–12. MEDLINE | CrossRef

Trivers, 1972 55.Trivers RL. Parental investment and sexual selection. In:  Campbell B editors. Sexual selection and the descent of man, 1871–1971. London: Heinemann Press; 1972;p. 136–179.

Wedekind & Furi, 1997 56.Wedekind C, Furi S. Body odor preferences in men and women: do they aim for specific MHC combinations or simply heterozygosity?. Proceedings of the Royal Society of London. Series B. 1997;264:1471–1479. MEDLINE | CrossRef

Wedekind et al., 1995 57.Wedekind C, Seebeck T, Bettens F, Paepke AJ. MHC-dependent mate preferences in humans. Proceedings of the Royal Society of London. Series B. 1995;260:245–249. MEDLINE | CrossRef

Yamaguchi et al., 1981 58.Yamaguchi M, Yamazaki K, Beauchamp GK, Bard J, Thomas L, Boyse EA. Distinctive urinary odors governed by the major histocompatibility locus of the mouse. Proceedings of the National Academy of Sciences of the United States of America. 1981;78:5817–5820. MEDLINE | CrossRef

a School of Clinical Medical Sciences, University of Newcastle, Newcastle-upon-Tyne NE1 4HH, UK

b School of Biological Sciences, University of Liverpool, Liverpool L69 7ZB, UK

c School of Psychology, University of St. Andrews, St. Andrews KY16 9JU, UK

d National Blood Service, Holland Drive, Newcastle-upon-Tyne NE2 4NQ, UK

e School of Psychology, University of Aberdeen, Aberdeen AB24 2UB, UK

f Department of Experimental Psychology, University of Bristol, Bristol BS8 1TN, UK

Corresponding author. School of Biological Sciences, University of Liverpool, Liverpool L69 7ZB, UK. Tel.: +44 151 795 4514; fax: +44 151 795 4408.

PII: S1090-5138(04)00078-9

doi:10.1016/j.evolhumbehav.2004.09.002



2007:11:26