Proceedings of the Royal Society B: Biological Sciences
You have access Review article

Humans as a model species for sexual selection research

Michael Lawrence Wilson

Michael Lawrence Wilson

Department of Anthropology, University of Minnesota, 395 Humphrey Center, 301 19th Avenue South, Minneapolis, MN 55455, USA

Department of Ecology, Evolution and Behavior, University of Minnesota, 1987 Upper Buford Circle, Saint Paul, MN 55108, USA

[email protected]

Google Scholar

Find this author on PubMed

,
Carrie M. Miller

Carrie M. Miller

Department of Anthropology, University of Minnesota, 395 Humphrey Center, 301 19th Avenue South, Minneapolis, MN 55455, USA

Google Scholar

Find this author on PubMed

and
Kristin N. Crouse

Kristin N. Crouse

Department of Anthropology, University of Minnesota, 395 Humphrey Center, 301 19th Avenue South, Minneapolis, MN 55455, USA

Google Scholar

Find this author on PubMed

Published:https://doi.org/10.1098/rspb.2017.1320

    Abstract

    Ever since Darwin, questions about humans have driven sexual selection research. While studies of other organisms are often justified as useful for improving understanding of humans, humans themselves can be useful models. Although humans present some drawbacks as model organisms (complicated societies, slow reproduction and strong ethical constraints on experimental options), humans nonetheless offer many advantages (being abundant, accessible and having detailed historical records for some populations). As an additional challenge, humans exhibit a rather puzzling combination of traits. Some traits (pair-bonding, biparental care and modest sexual dimorphism in body size) suggest selection for monogamous mating, while other traits (including sexual dimorphism in body composition and appearance) suggest selection for polygyny. Such puzzles have motivated research on other species, resulting in a rich set of comparative data that provides insights into humans and other species. Recent studies of visual trait dimorphism suggest that human appearance reflects adaptation for multi-level societies, rather than high levels of polygyny. In addition to biological traits, human cultural traits have undergone rapid evolution. Changes in subsistence strategies profoundly affect opportunities for sexual selection. The enormous variability of human behaviour and ecology provides abundant opportunities to test key hypotheses, and poses challenging puzzles for future research.

    1. Introduction

    Sexual selection research has focused on humans from the very beginning. Darwin considered a full treatment of sexual selection ‘indispensable’ to understanding human evolution, and published his first detailed accounts of both human evolution and sexual selection together in The descent of man and selection in relation to sex [1]. He argued that sexual selection drove variation in traits such as skin and hair colour, and also shaped many differences between men and women. He argued that such traits help, not with the struggle for survival, but with the struggle for reproduction. Darwin noted that in many species, showy, sexually assertive males compete to be chosen by drab, choosy and coy females, while in humans and several other exceptional cases, female ornamentation suggests an important role for male choice [1]. Darwin's contemporaries accepted that males competed for mates, but were sceptical that arbitrary aesthetic preferences could be an important force in evolution [2,3], and the theory remained relatively neglected for a century. However, since the 1970s, sexual selection has generated tremendous interest, becoming a central topic in behavioural ecology [4] and studies of human behaviour and evolution [58]. Numerous studies have examined topics including the extent to which different human traits function as ornaments [6,915], armaments [1618] and tools for post-copulatory competition (an arena little suspected by Darwin) [1921]. Sexual selection has been proposed to explain not just features of human bodies, but a vast realm of behaviour, including clothing choices [22], art [23], music [24], humour [25], sports [26], indeed, ‘all cultural practices’ [27] and even human intelligence itself [6]. Whether or not the strongest of these claims are ultimately substantiated, sexual selection helps make sense of many human traits, and has become a dominant theoretical framework for evolutionary psychology [6]. The rapidly growing number of human studies raises the question: should humans be considered a model species for sexual selection research, suitable for making theoretical advances with broad application to other species?

    Here, we consider humans as a model species, focusing on two aspects that present both challenges and opportunities. First, humans exhibit an unusual combination of traits that defy easy categorization as adaptations to a particular mating system. Considering humans in isolation leads to confusion, which is resolved to some extent by comparison with other primates. Second, human behaviour varies greatly within and among societies. Such variation makes humans difficult to pin down, but provides abundant opportunities to test hypotheses.

    2. Humans as model organisms?

    Researchers use models—simplified representations—to increase understanding of more complex systems. Model organisms are typically small, experimentally tractable organisms that grow and reproduce rapidly, such as fruit flies (Drosophila melanogaster) [28] and guppies (Poecilia reticulata) [4]. Good model organisms improve the efficiency of research, because researchers can build on a wealth of previous studies. For example, Drosophila first attracted interest as a model for genetics, but has proved useful for a wide range of topics in biology, including sexual selection. The many known genetic markers in Drosophila made possible Bateman's experiments on sex differences in reproductive potential [29], which suggested that for males, but not for females, increasing the number of mates increased the number of offspring. Similarly, Maynard Smith obtained early experimental evidence of female mate choice in Drosophila subobscura, where females discriminated against ‘less athletic’ inbred males during courtship [30]. Working with flies, Maynard Smith directly arranged and observed mating opportunities, something not easily done with human subjects.

    In contrast with classic model organisms, humans are large, grow and reproduce slowly, and ethically cannot be subjected to certain kinds of experiments. Far from being simple, humans live in the largest, most complex of animal societies. Additionally, humans are unique in many respects, having cumulative culture [31], language [32] and an exceptional capacity for abstract reasoning [33]. Nonetheless, several features make humans amenable study subjects. Human subjects are abundant and accessible, particularly on the university campuses where most scientists work. While certain kinds of experiments are unethical or impractical with human subjects, many other kinds of experiments can be conducted ethically and without great difficulty. The ease of studying humans, combined with people's great interest in our own species, has resulted in an enormous number of studies. A search on the Web of Science for publications using the search term ‘sexual selection’ found 27 033 publications, 98.5% of which were published in 1990 or later. Adding search terms for various taxa found that humans were among the most commonly represented taxon, with over 8% (2242) of the publications, which was nearly as many as birds (3735), Drosophila (2648) and fish (2293), and much more than primates (678), even though these other taxa each include many species.

    Language enables human subjects to explain their thinking to researchers (even if the actual mechanisms motivating action may not be accessible to introspection). We cannot directly ask a chimpanzee or a stalk-eyed fly what they find attractive in a mate, and instead have to make inferences about other species from observation and experiments. By contrast, we know, as individuals, what traits of prospective mates we find attractive, and we can readily learn what traits other people find attractive from informal conversation, the arts, including visual representations [34] and stories [35], formal surveys [36,37] and patterns of online behaviour [38]. Historical documents [3941] and records of marriages, births and deaths [42] can provide insights into the lives of people long dead. Studies using detailed genealogies can provide a time-depth of centuries [43], far longer than any primate field study. Human studies can take advantage of enormous sample sizes, such as a database of 500 000 genotyped individuals [44], greater than the number of chimpanzees on the planet [45]. Human cultural diversity presents opportunities to test a broad range of hypotheses relevant to many species, such as factors affecting fertility [46].

    Sexual selection affects everyone on a personal level. People choose mates, or choose not to mate, and everyone must navigate a gendered social world, whatever their own gender identity. While intuition alone can prove misleading, insights from introspection and observation of daily life can suggest hypotheses that lead to theoretical advances. For example, considerations of human mating behaviour, plus observing that ‘male pigeons practiced the same type of sexual double standard, and had the same kind of attendant psychological problems (or at least behavior ones) as did human males’, [47] led Trivers to an explanation for Darwin's generalization that males compete for mates and females choose. Trivers built on Bateman's findings to argue that patterns of sexual selection depend strongly on sex-specific parental investment [48]. The difference between male and female reproductive potential that Bateman found for Drosophila reflects a general pattern: whichever sex invests more in offspring should become a limiting resource, and thus the target of competition by members of the other sex. Because females (by definition) produce larger gametes, they begin with a bias towards parental investment, which is often amplified through evolutionary lineages, though in some cases, this can be reversed, as in seahorses [48].

    Subsequent studies have refined and challenged this framework [49], suggesting, for example, that the intensity of sexual selection is better measured using the operational sex ratio [50] or potential reproductive rate [51]. A recent attempt to replicate Bateman's experiments indicated that the results of the original study were biased by not taking into account the viability of different crosses [52]. In contrast with expectations from Bateman, females in many species mate with multiple males [53]. Nonetheless, meta-analysis of studies across animals indicates that males generally do have higher variance in reproductive success, and this is related to both mating system and parental effort [54].

    Despite the various advantages of studying humans, cultural views about how humans ought to be can make it difficult to discern how humans actually are. Surveys, commonly used in human studies, can be problematic, because people often lie, providing answers based on what they perceive to be socially expected, and/or what they wish to believe about themselves [55]. Excessive focus on university students in ‘WEIRD’ societies (Western, educated, industrialized, rich, democracies) may yield biased results [56]. Efforts to explain humans from an evolutionary point of view have often generated controversy, partly out of concern that explaining behaviour such as sexual coercion, rape or murder from an evolutionary perspective may serve to excuse or justify such behaviour [57].

    Despite these challenges, studies of sexual selection in humans have shed important light on the evolution of human traits. The value of human studies for illuminating other species depends to a large extent on whether such studies can convincingly explain human traits. We therefore review selected examples from this now-voluminous literature in the sections below. Additionally, we do not wish to argue that studying humans alone is sufficient for understanding sexual selection; humans represent only a fraction of life's diversity, and understanding any single species requires comparative study. This is particularly the case with humans, because humans have a puzzling set of traits whose significance continues to be debated, such as: are humans highly sexually dimorphic, or only moderately so? We cannot know without comparative study, and therefore discuss humans in comparison with other primates.

    3. Humans: a puzzling combination of traits

    As anthropologists have long noted, humans in all documented societies form enduring pair-bonds [5860]. In most societies, most marriages are monogamous, even when social rules permit marrying multiple spouses [61,62]. Others, however, have drawn attention to the great variability in human mating behaviour. Cross-cultural surveys consistently find that men, to a greater extent than women, demonstrate an interest in mating with a variety of partners during their lives [63,64], but both men [65] and women [21,66] mate with multiple partners under some circumstances. Given the variability in human mating behaviour, and the likelihood that the behaviour of contemporary humans is shaped by evolutionarily novel conditions, researchers have often turned to human anatomy to provide insights into what mating patterns are likely to have predominated over the course of human evolution [6,67,68]. Even considering anatomy alone, however, people considering essentially the same set of data come to strikingly different conclusions: humans are either basically monogamous [59,69], polygynous [18,70], promiscuous [71] or adapted for both polygyny and polyandry [72].

    4. Dimorphism in visual appearance

    Both men and women display sex-specific traits that are attractive to potential mates, including broad shoulders, beards, low voices and masculine facial features in men, and broad hips, breasts and feminine facial features in women [18]. While some have claimed that humans are ‘relatively monomorphic’, systematic comparison with other primates reveals that men and women differ substantially in appearance, particularly compared with primates classified as having monogamous mating systems [73]. Dixson et al. [73] developed a rating system for visual trait dimorphism in a range of conspicuous traits displayed by primates, including hair tufts, capes, manes, baldness, colourful skin patches and fleshy protuberances. They scored all sexually dimorphic traits on the head, trunk and limbs on a six-point scale, from no difference between the sexes (0) to maximum dimorphism (5), being fully developed in one sex but absent in the other. Total scores ranged from 0 to 32 [73]. In most primates classified as monogamous, males and females looked similar, with a median score of 0 (range = 0–11.5, n = 41 [73]), whereas humans received a score of 10. Among monogamous primates, humans tied with white-faced saki monkeys (Pithecia pithecia), and were exceeded only by the pileated gibbon (Hylobates pileatus) [73]. Recent observations have found that both white-faced saki monkeys and several species of gibbon have greater diversity in mating behaviour than previously assumed [74,75], further supporting the view that high visual trait dimorphism is unusual for monogamous primates.

    Grueter et al. [76] recently refined the visual trait dimorphism analysis to take into account the growing recognition that humans and several other primates are best classified as having multi-level societies, in which social groups consist of nested sets of modular units, such as ‘one-male units’ (one breeding male and several females) combining in higher levels such as band or herd [77,78]. Ten of 140 species categorized had multi-level societies: some Asian colobines, some African papionins (baboons and their kin) and humans [76]. Primates living in multi-level societies are more strikingly ornamented than closely related species living in other societies [76] (figure 1). For example, in contrast with the rather dull appearance of male olive baboons (Papio anubis), male hamadryas baboons (Papio hamadryas) have long capes of white fur and bright red paracallosal skin (which in olive baboons occurs only in pregnant females). Similarly, men and women differ more in their appearance than do male and female chimpanzees (Pan troglodytes) (figure 1). In multi-level societies, the prevailing threats of mate poaching and mate-switching may favour the evolution of conspicuous ornamentation to display condition, to maintain the interest of current mates, attract higher quality mates and/or to deter rivals [76,77].

    Figure 1.

    Figure 1. Contrasting patterns of sexual dimorphism in (clockwise, from the top right, with visual trait dimorphism scores [73,76] given after each species name): western gorillas (Gorilla gorilla: 10), humans (Homo sapiens: 10), chimpanzees (Pan troglodytes: 0), olive baboons (P. anubis: 4), hamadryas baboons (P. hamadryas: 13), and geladas (Theropithecus gelada: 13). In each species, the female is on the right. Gorillas mate polygynously, chimpanzees and olive baboons mate polygynandrously, while humans, hamadryas baboons and geladas live in multi-level societies. In humans, biological traits are commonly modified by cultural practices, such as removal of male facial hair, use of jewellery and other ornaments, and wearing clothing that may exaggerate some features while concealing others. Non-human primate photos by M. L. Wilson; paintings of Charles and Emma Darwin by George Richmond approximately 1839 (public domain).

    Multi-level societies are not, of course, the only ones that produce ostentatious ornaments in primates. Mandrills (Mandrillus sphinx) scored highest in visual trait dimorphism, but live in large groups rather than multi-level societies [76]. Additionally, while group size was found to predict visual dimorphism scores (figure 2), orangutans (Pongo spp.) scored nearly as high as mandrills, despite being solitary [76]. Grueter et al. [76] argued that visual ornamentation is particularly important in species in which individuals are relatively anonymous, and thus display cues of condition and competitive ability to individuals that have not directly observed or experienced their fighting ability. Such anonymity may occur both in extremely large groups, such as mandrill hordes, and among solitary individuals that rarely encounter one another. Social anonymity is surely an even greater issue for human societies, which may number in the millions of people, and may promote the cultural elaboration of ornamentation.

    Figure 2.

    Figure 2. Visual trait dimorphism index versus (a) log10 of group size and (b) body mass dimorphism (d = male : female body mass). Data are from [76], except that we show two points for humans: one with total body mass (d = 1.14), and one with fat-free mass (d = 1.41), because humans have an unusual dimorphism in body composition: women have more fat, while men have greater muscle mass [79]. Dichromatism is scored as a sexually selected trait.

    5. Human mating systems

    The diversity of human mating systems provides abundant opportunities to test evolutionary hypotheses for how mating behaviour relates to socioecology. Humans exhibit a wide range of marriage systems and mating practices within and among societies, including monogamy (one individual of each sex), polygyny (one male and multiple females), polyandry (one female and multiple males), polygynandry (multiple males and females) as well as same-sex relationships. In the 186 societies in the standard cross-cultural sample (SCCS), 82% are coded as polygynous, 17% monogamous and only 1% polyandrous [80]. Such broad characterizations, however, obscure substantial variation within each society. Additionally, as in other species, the social mating system for humans (marriage) may differ from the sexual mating system. Because humans mate in private, direct evidence of departures from the social mating system is difficult to obtain. Recent reviews of data for contemporary and historic societies reveal generally low rates of extra-pair paternity (around 1–2%) but with considerable variation among populations [20,81,82].

    Polygyny occurs widely, yet even in societies that permit polygyny, only some men—those who are especially adept at hunting, or who have acquired wealth through skill or good fortune, or have otherwise gained high status—marry multiple wives. For example, among 31 forager populations in the SCCS, the percentage of married men with more than one wife varied considerably (range: 0–70%) [80] (figure 3). Across these societies, a mean of 12.4% of married men had more than one wife [80]. This mean is influenced by the few highly polygynous societies; calculating the median reveals that only 5% of married men had more than one wife.

    Figure 3.

    Figure 3. Per cent of married males with more than one wife, versus per cent of family subsistence provided by the male in 31 forager societies in the SCCS (adapted from [83]).

    Although few societies are coded as polyandrous, such relationships occur in many societies, particularly where polygyny among high-status men leads to a shortage of wives for lower-status men [65,80,84]. Additionally, in societies with ‘partible paternity’ (the belief that semen from multiple men can contribute to the formation of a child), polyandrous mating by women is publically acknowledged in the attribution of multiple fathers to some children [85].

    Polygynandry, in which mating occurs among multiple males and females, is known mainly from subgroups of larger societies, such as intentional communities and advocates of polyamory [86]. Polygynandry also may occur as a life stage among people who expect to marry monogamously eventually, but are delaying reproduction to focus on their education, as in the ‘hook-up culture’ described for American high school and college students [87].

    Nonetheless, despite tremendous variation within and among societies, and despite the occurrence of extra-pair copulations, the modal human marriage system is socially monogamous [61]. In this respect, humans are distinct from other primates [60,88]. Humans do not typically mate like orangutans (intermittent couplings between isolated individuals), gibbons (isolated pairs), gorillas (isolated one-male, multi-female units), or chimpanzees and bonobos (promiscuous groups). Instead, humans typically form enduring breeding bonds between men and women (marriage), and live with other families within multi-level societies [88,89].

    6. Paternal effort and monogamy

    While humans resemble some other primates in having multi-level societies, they are unique among primates in having mostly monogamous, rather than polygynous, families within larger societies [61]. Humans also practice an unusual sexual division of labour: in hunter–gather societies, men do most of the hunting, while women gather plant foods and do most of the food preparation and cooking [90]. Provisioning by males represents a striking reversal in the mammalian trend towards female specialization in parental effort. Whether male provisioning promoted the evolution of monogamy in humans, as some have argued [59,91], or emerged following the evolution of monogamy for other reasons (as appears to be the case in other instances of paternal care in mammals [92]), parental investment theory predicts that males should experience trade-offs between parenting and mating effort [48,80,93]. By providing food, a man can improve the survivorship and growth rates of his offspring, but doing so constrains his ability to provision multiple wives [91]. Consistent with this prediction, Marlowe found that among foragers, more marriages are polygynous when men contribute less to the diet (figure 3; [83]). In the tropics, where abundant plant food enables women to provide much of the family's sustenance, polygyny is more common [83]. At extremely high latitudes, such as among the Inuit, where men provide most or all of the diet, rates of polygyny drop to zero [83]. Similarly, considering all 186 societies in the SCCS, the male contribution to the diet corresponded closely to the mating system. Men contributed the most to the diet in societies coded as polyandrous, less in monogamous societies and least in polygynous societies [80].

    7. Coevolution of subsistence and mating strategies

    The transition from food collection to food production had profound effects on human societies, including cultural coevolution between subsistence strategies and marriage practices. Mobile foragers have little opportunity to accumulate wealth. Sedentary societies with food storage and/or food production (through domesticated plants and animals) create wealth and thus inequality. If a few males are able to monopolize many females, then some males will have no reproductive success, leading to high variance in lifetime reproductive success (LRS) [94].

    Data from hunter–gatherers, herder/gardeners and intensive farmer/fishers illustrate the dramatic effects of subsistence strategies on reproductive skew (table 1 and figure 4). Such measures of LRS are of great interest in evolutionary studies, but are difficult to obtain for natural populations, especially long-lived species such as primates. Only two studies have reported estimates of LRS for non-human primates (mandrills [96] and rhesus macaques, Macaca multatta [95]), and both of these are for free-ranging but not wild populations, and thus should be treated with caution. Comparable data, however, have been obtained from multiple human societies, thanks to painstaking efforts by anthropologists to piece together life histories through ethnographic interviews [94] and historical records [42].

    Figure 4.

    Figure 4. (a) Maximum reproductive success and (b) standardized variance in reproductive success for males and females in two non-human primates and a range of human societies. Adapted from table 1.

    Table 1.Measures of reproductive success in humans and other primates.

    subsistence society maximum LRS (male) maximum LRS (female) mean (male) mean (female) variance (male) variance (female) IM IF source
    hunter–gatherer !Kung 12 9 5.14 4.69 8.6 4.87 0.326 0.221 [94]
    hunter–gatherer Meriam 12 11 3.63 2.06 11.7 6.43 0.887 1.52 [94]
    hunter–gatherer Aché 13 12 6.4 7.84 15.0 3.57 0.367 0.058 [94]
    hunter–gatherer Aka 14 11 6.34 6.23 8.64 5.20 0.215 0.134 [94]
    hunter–gatherer Hadza 16 12 4.55 4.58 14.3 7.70 0.691 0.367 [94]
    herder/gardener Pimbwe 12 12 5.99 6.14 9.00 7.27 0.251 0.193 [94]
    herder/gardener Tsimane 22 15 9.06 8.91 19.9 12.7 0.243 0.160 [94]
    herder/gardener Xavante 23 8 3.6 3.6 12.1 3.90 0.934 0.301 [94]
    herder/gardener Yomut 30 18 5.12 3.87 8.07 7.09 0.308 0.473 [94]
    herder/gardener Yanomamö 43 14 5.59 3.67 39.6 8.02 1.27 0.595 [94]
    herder/gardener Kipsigis 80 12 12.42 6.6 85.6 5.9 0.555 0.135 [94]
    intensive farmers/fishers Finland 17 15 3.90 3.89 13.7 11.1 0.898 0.736 [42]a
    non-human primates Mandrillus sphinx 41 17 3.5 4.7 68.2 19.8 5.57 0.896 [95]
    non-human primates Macaca mulatta 47 16 8.7 7.7 98.7 15.4 1.30 0.260 [96]

    aCalculation of Finnish data for those surviving to age 15. Courtesy A. Courtiol 2017, personal communication.

    In hunter–gatherers, the maximum LRS for men (median = 13; range = 12–16) only slightly exceeds that for women (median = 11; range = 9–12; n = 5). In herders and gardeners, however, the highest LRS for men (median 26.5; range = 12–80) greatly exceeds that for women (median = 13; range = 8–18; n = 6). Thus, the concentration of resources in food-producing societies greatly increases the potential for reproductive success in men, but not women. This tendency is further amplified in early empires. Betzig [94] notes that ‘in every one of the six pristine civilizations—in Mesopotamia, Egypt, India, China, Mexico, and Peru— emperors collected hundreds of women and had hundreds of children’. Strikingly, though, in more recent history, the reproductive options for high-status men have been sharply curtailed. However, in some societies with intensive agriculture, monogamy has become more prevalent.

    Among intensive farmers and fishers living in Finland before the demographic transition to low fertility, both men and women had relatively high fertility (maximum = 17 and 15 offspring, respectively) probably due to better nutrition from food production and storage, but the difference between male and female LRS was small, as in hunter–gatherers [42]. Debate continues as to whether monogamy in such agricultural societies results mainly from social coercion [9799] or through individuals pursuing their inclusive fitness by, for example, ensuring that inherited wealth is not split among children from different wives [100].

    Maximum LRS focuses attention on a few extreme cases. To compare the intensity of selection across populations while controlling for overall differences in fertility, biologists calculate a standardized variance, or ‘opportunity for selection’, I, calculated by dividing the variance in LRS by the square of the mean LRS [95,101]. Comparison with other mammals suggests that the opportunity for selection is low if I < 1, moderate if 1 < I < 3 and high if I > 3 [95]. (The data in table 1 underestimate I, because they are from individuals that survived to reproductive age, but nonetheless illustrate patterns in reproductive success across populations.) Estimates for I in men generally exceeded those for women in hunter–gatherers (IM: median = 0.37, range = 0.21–0.89; IF: median = 0.22, range = 0.13–1.5; n = 5), herder/gardeners (IM: median = 0.43, range = 0.31–1.27; IF median = 0.25, range = 0.14–0.6, n = 6) and farmer/fishers (IM = 0.898, IF = 0.736). For most human societies, the values were lower than 1. Only among societies such as the Yanomamö (IM = 1.27) did the opportunity for selection approach that for rhesus macaques (IM = 1.3), which was modest in comparison with mandrills (IM = 5.57). Even for the Kipsigis, where men had up to 80 children, the standardized variance in LRS was low (IM = 0.55), because men in general had high fertility (mean = 12.4 offspring; this fertility estimate is probably too high, though, as it excludes emigrant men, who likely had low LRS [102]). In summary, these data support the view that men generally have higher variance in reproductive success than women; cultural innovations such as herding and farming can increase that variance; and even in mainly monogamous societies, substantial variation in reproductive success occurs.

    8. Showy traits, monogamy and extra-pair copulations

    In humans, sexual selection shapes not only anatomical traits, but also cultural practices, such as clothing, jewellery, tattoos and other ornaments. Cultural practices can evolve much more rapidly than anatomical traits, and provide fascinating opportunities for testing hypotheses relating to the evolution of ornamentation more broadly. Just as human anatomical traits are more visually striking than would be expected for a strictly monogamous species, human cultural ornamentation can be surprisingly ostentatious for both men and women. Hamilton puzzled over this, as part of the larger puzzle of why showy traits evolved in presumably monogamous species, such as many songbirds:

    Why did Beau Brummell in Regency England dress up as he did? Was it to find a wife, or to find an ‘affair’? And, why did the women who admitted themselves attracted to such men dress themselves in equal finery? Many were married; but we may note a possibility that in a time when marriages were often arranged for financial advantage, their mates may have been more rich than attractive or healthy. Of course there is a very great deal of complexity in such a human situation that birds could never have. Particularly there is the transferable wealth, which to my mind overshadows even the role of rational thought [103].

    Hamilton & Zuk [104] argued that in birds and other animals, a primary function of showy displays was to exhibit genetic quality, particularly resistance to parasites. Hamilton noted that it made little sense for individuals in long-term pair bonds to exhibit showy displays of quality; each individual knows very well the other's health, strength, resource acquiring potential and parenting skills [103]. Nonetheless, in many songbirds, and in many human societies, one or both sexes exhibit showy traits such as brightly coloured feathers or expensive, fashionable clothing. Hamilton argued that such showy displays serve to attract extra-pair copulations [103], and indeed, genetic testing revealed that songbird species with brightly coloured males had higher rates of extra-pair paternity [105]. Whether a similar correlation exists in human societies remains to be tested. Alternatively, in a multi-level society abounding with opportunities and risks of extra-pair mating and mate-switching, both sexes may face intense pressure to maintain the attraction of their current mate to reduce the risks of cuckoldry and abandonment.

    9. Conclusion

    While humans are unwieldy study subjects compared with Drosophila, they present a wealth of opportunities for studies that advance our understanding of sexual selection. Humans pose many fascinating puzzles, which have inspired sexual selection theory from the beginning, and which have motivated comparative studies that have greatly expanded our understanding of humans and other species. Humans are abundant, accessible and amenable to a broad range of observational and experimental studies. Human cultural artefacts, including visual arts, literature and historical records, provide a richness and time depth that is simply not available for other species. Human patterns of mating, marriage, ornamentation and display vary greatly within and among societies, providing a wealth of opportunities for future studies.

    Data accessibility

    This article has no additional data.

    Authors' contributions

    M.L.W. drafted the manuscript. C.M.M. and K.N.C. helped draft the manuscript. All authors gave final approval for publication.

    Competing interests

    We declare we have no competing interests.

    Funding

    While writing this paper, the authors were supported by funding from the University of Minnesota, including a Talle Faculty Research Award and an African Studies Initiative Faculty Travel Grant to M.L.W., and a Leakey Foundation grant to C.M.M.

    Acknowledgements

    Many thanks to Sarah Brosnan and Erik Postma for the invitation to contribute to this issue, and thanks to Anthony Collins, Laura Betzig, Nisarg Desai and Anthony Massaro for helpful discussion, and to Virpi Lummaa for permission to publish unpublished data.

    Footnotes

    Special feature paper: Humans as a model for understanding biological fundamentals.

    Published by the Royal Society. All rights reserved.

    References