R1. Expanding the research

Many commentators praised our target article, recognized the value of our approach, accepted our basic premises and conclusions, and suggested future avenues for research on attractiveness-related biases. We welcome these suggestions and hope that both our target article and these commentaries will stimulate and guide future research in this area. Agthe & Maner cite additional studies that support our conclusions (a) that financial and prosocial biases in favor of attractive adults are stronger for individuals who are potential mating partners, (b) that physically attractive mating partners have higher reproductive value, and (c) that the psychological, neural, and neuroendocrine mechanisms underlying favorable biases toward attractive individuals have likely been shaped by natural selection. They recommend that future research take into consideration boundary conditions for mating-related biases in favor of attractive individuals, such as the moderating influence of being in a committed relationship, one's tendency for sexual promiscuity, women's phase of the menstrual cycle, individuals' ethnic background, culture, family expectations, kinship rules, and the extent of individual choice that is allowed in personal relationships.

Lee, Adams, Li, & Gillath (Lee et al.) make many similar points, highlighting the importance of considering the influence of relationship status, sex ratio, pathogen prevalence, and individualistic versus more collectivistic cultures and settings. Agthe & Maner also highlight the importance of considering both mate attraction and mate competition, as physical attractiveness sometimes leads to negative, rather than positive, interpersonal judgments and outcomes (Roberts makes a similar point, too). They point to research showing that negative reactions toward attractive same-sex individuals are displayed particularly by people who are likely to fear intrasexual competition.

Similar to Agthe & Maner, Little and Buunk recommend that future research on attractiveness-related biases should further explore both same-sex mate competition, in which attractive individuals are perceived as sexual rivals, and same-sex sexual collaboration, in which an individual strategically associates with a better-looking same-sex individual to enhance his or her potential to attract highly valued mates. Buunk also underscores the importance of examining physical attractiveness of potential mates in relation to other characteristics such as dominance (Ravina makes the same suggestion). Le Lec, Alexopoulos, Boulu-Reshef, Fayant, Zenasni, Lubart, & Jacquemet (Le Lec et al.) recommend that the influence of attractiveness on decision making be further explored in the context of a general model of courtship behavior, which takes into account not only the benefit of mating with an attractive individual, but also the probability of doing so in relation to one's own attractiveness, competition, and other variables. Ravina suggests that financial biases in favor of attractive individuals should be examined both in contexts in which the stakes are high and those in which they are low, with the hypothesis that they may be stronger in the latter than in the former case. Ravina also suggests that future research should investigate the experience and the expertise of decision makers, as this variable may have an important moderating influence on attractiveness-related biases, especially in the labor market.

Stephen, Burke, & Sulikowski (Stephen et al.) suggest that the attractiveness of high-quality individuals as potential mates should also be investigated from a comparative perspective, for example, by studying chimpanzees, as this analysis could elucidate both the evolutionary precursors of the human attractiveness bias and the more recent selection pressures that may have shaped it. They also advocate more cross-cultural studies of attractiveness-related biases in industrial and non-industrial societies. Finally, future studies should explore whether attractiveness-related biases in the labor market vary in relation to different professional fields or types of jobs. Building on the notion that prosocial behavior toward attractive individuals can be interpreted as sexual courtship, Farrelly suggests that research on attractiveness-related biases could be expanded with more detailed analyses of the role of prosocial behavior in courtship and mate choice. For example, he cites recent research showing that courtship involving prosocial behavior is particularly important in the context of long-term relationships. Both Farrelly and Roberts suggest that further research is needed to better understand the signaling function of prosocial behavior in the context of sexual courtship. According to Hurlemann, Scheele, Maier, & Schultz (Hurlemann et al.), future research should also address the role of oxytocin as a possible physiological mechanism underlying the attractiveness-related biases in prosocial behavior.

We agree with Barclay that biological market theory can provide a powerful theoretical framework that could guide future research on prosocial and financial biases in favor of attractive adults. Barclay makes four important predictions that could be empirically tested: attractiveness should matter less in environments where everyone is attractive; attractiveness biases should be greater in environments where physical attractiveness is in high demand or low supply; mating-related attractiveness biases should be greater in single men than in men who are married, raising children, sexually unreceptive, or chaste; and these biases should be especially strong in individuals who are themselves highly attractive (a prediction for which there is already some supporting evidence). Barclay also points out that just as physical attractiveness gives adults high value in the mating market, other characteristics such as high status, wealth, competence, and trustworthiness give individuals high value in other social markets in which partner choice is regulated by the laws of supply and demand.

Similar to Barclay, Eisenbruch, Lukaszewski, & Roney (Eisenbruch et al.) advocate a broader partner choice framework for future research on attractiveness-related biases. They argue that physical attractiveness can make individuals valuable not only in mating markets, but also in markets for business or coalition partners, friends, leaders, or followers. Little and Ronay & Tybur make the same point. Eisenbruch et al. argue that attractiveness may be associated with longevity, continued ability to extract resources from the environment, and low risk of transmitting pathogens, all of which are valuable in multiple types of social markets. We agree with Eisenbruch et al., Little, Ronay & Tybur, and Stephen et al. that evolutionary studies of attractiveness-related biases should not limit their focus on its role in mating partner choice, and that characteristics such as attractiveness, health, longevity, status, and strength, which make individuals valuable in different social markets, are often interrelated. For example, Little notes that tall men are found to be both attractive and dominant, and a bias toward other attractive men by heterosexual men may be predicted because tall, attractive, dominant men are likely to make powerful coalitional partners and be favored as friends and allies. We believe, however, that although the value of attractiveness in mating markets is clear, the role of attractiveness in the choice for coalition partners, leaders, or cooperative partners remains to be elucidated. In markets for coalition partners or leaders, physical strength, high status, and resources are clearly crucial, whereas in markets for cooperative partners, personality and behavior are crucial (e.g., whether an individual is trustworthy and generous). Although attractive individuals may sometimes possess these other valuable attributes, many studies reviewed in our target article indicate that attractive individuals tend to behave in a selfish and manipulative way in economic games.

Little suggests that attractive individuals might make popular leaders because of the association between attractiveness and health. However, if attractive individuals were not also physically and psychologically strong, high status, and in possession of material and social resources (e.g., political support), it is not immediately obvious that their health alone would make them appealing for a leadership position. A man may be extremely healthy and have a beautiful face; however, if he is very short, does not have well-developed musculature, is financially poor, has low status, has no social or political support, and is extremely shy, he would not be appealing as a coalition partner or a leader.

Becker highlights the need for further research on the cognitive processes underlying attentional and memory biases toward beautiful faces, emphasizing that our functional, mating-related perspective can provide a powerful framework for guiding this research. According to Becker, our functional perspective can explain the finding that although physically attractive male and female faces both garner more attention, the long-term encoding of individuating features favors attractive women but not men. Becker suggests that cognitive biases toward the attractive faces of potential mating partners “could be largely automatic and could readily produce biases in hiring, and so forth, even when proximity/mating is not an explicit goal” (para. 6). He continues:

We agree with Becker. Olivola & Todorov recommend that research on the influence of facial attractiveness on decision making be linked to the large body of work investigating face-based social attributions. We briefly mentioned this research in our target article, but believe that it is only indirectly related to the main questions we addressed in our article. We agree with Olivola & Todorov that “Although the biasing effects of physical attractiveness may be well explained in terms of evolutionary mating motives, the same is not true of face-based trait inferences” (sect. 3, para. 1).

R2. The issue with children

The notion that attractive individuals make valuable partners also in nonmating contexts is relevant to the discussion of biases in favor of attractive children. We addressed this issue in our target article, and it was also raised by several commentators (Stephen et al., Chen, Rennels & Verba, and Schein, Trujillo, & Langlois [Schein et al.]). We reiterate that biases in favor of attractive infants and children (and adolescents), as well as those shown by infants, children, and adolescents, are consistent with and do not contradict the theoretical stance taken in our target article (as argued by some commentators, e.g., Schein et al.). We argued that in infants and children, as in adults, physical attractiveness is probably an indicator of genetic or phenotypic quality.

Many cases of adult biases in favor of attractive children can be explained with the differential parental solicitude model, which predicts that parents should invest more in higher-quality offspring. We also hypothesized that the same neural and neuroendocrine mechanisms that regulate perceptual and behavioral biases toward attractive children from a parental motivation perspective also regulate perceptual and behavioral biases toward attractive adolescents and adults from a mating motivation perspective. The same can be said about the biases shown by prepubertal adolescents toward same-age peers or adults. It is very likely that the same neural mechanism in the human mind is responsible for positive biases toward attractive others at any age.

Consistent with our view, Becker argues that mating-related biases in favor of attractive adults can arise from cognitive processing biases that develop early in life. Attractiveness-related positive biases shown at different ages are all essentially the same phenomenon – human beings are predisposed to respond to attractiveness as a marker of quality – but this phenomenon acquires different functional significance in different contexts (e.g., survival, parental investment, mating, and reproduction), with the mating function becoming preponderant in adulthood. This view is consistent with those of many of our commentators, including those who have discussed children such as Chen, Stephen et al., Rennels & Verba, and Becker.

Rennels & Verba cite and discuss research showing that attentional and affective biases toward attractive females are stronger and more consistent than those toward attractive males already in infancy and childhood. Similar to what we suggest in our target article, Rennels & Verba propose that these early biases contribute to the strength of adult men's prosocial and financial biases toward attractive women, which from puberty on acquire functional significance in a mating context. Similarly, Becker suggests that attentional vigilance to signs of genetic fitness such as facial attractiveness should emerge “even in children without mature mating motivations, because such attention facilitates the building of internal representation systems that will later form the multidimensional ‘mating space’” (para. 3). He states:

R3. Misunderstandings and disagreements

Some commentators misinterpreted what we wrote in our target article and/or made statements in their commentaries with which we do not agree. A case in point is the commentary by Dang. Dang writes that “there was plenty of evidence suggesting no relationship between attractiveness and biological fitness” (para. 2). We are skeptical of statements like this, which suggest that the null hypothesis can be proved. We also believe that the association between physical attractiveness (e.g., as indicated by symmetry) and biological fitness is well established in evolutionary biology, with supporting data coming from plants, nonhuman animals, and humans. Dang contends:

Dang misunderstands the theories we discussed: The explanations provided by economists and social psychologists indeed rely on the assumptions that people perceive attractive individuals to be more intelligent, trustworthy, competent, and so forth. Our functional evolutionary hypothesis, however, does not imply that people perceive attractive individuals as having higher biological fitness. Our preference for attractive individuals as potential mates has likely been shaped by natural selection because of the correlation between attractiveness and fitness (i.e., individuals who were genetically predisposed to be attracted to good-looking individuals and mated with them produced more and healthier offspring). Organisms are predisposed to respond to cues of fitness, not to consciously perceive the fitness value of other individuals, just as they are predisposed to behave in an adaptive way without consciously perceiving the adaptive value of their own behavior.

Dang writes that “the latent variable (i.e., mating goal) assumed by Maestripieri et al. seems like a river without headwaters or a tree without roots, given the insignificant effect even for explicit and direct manipulation of this variable” (para. 4). When we speak of “mating motives” in our target article, we emphasize that the functional significance of attractiveness-related prosocial and financial biases is related to mating. Although in some cases individuals who express positive biases in favor of attractive individuals are actually motivated to mate with them, in other cases such biases can be expressed in the absence of conscious mating motivation (Becker makes this point too), thus explaining why manipulating mating goals may not always necessarily influence attractiveness-related biases. For example, a happily married man may have no motivation to mate with an attractive woman and risk compromising his marriage, yet he might still behave prosocially toward this woman because his mind is predisposed to do that. Similarly, we respond to visual and gustatory cues of attractive food even in the absence of motivation to eat.

Dang states: “Maestripieri et al. relied on the evolutionary explanation to predict higher biases in favor of attractive same-sex individuals for homosexuals. This seems problematic because there is no reason to assume a relationship between homosexuality and gene passing” (para. 5). The minds of homosexual individuals are predisposed to respond to cues of fitness in potential mates just like those of heterosexuals. Men become sexually aroused and ejaculate in response to pornography even though there is no relationship between masturbation with pornography and gene passing. Dang writes that “facial attractiveness is highly dependent on sociocultural factors such as mass media influence and cultural transmission” (para. 6). These sociocultural influences exist, yet beautiful faces are universally recognized as such.

Dang argues that attraction to beautiful faces is the result of perceptual and cognitive mechanisms that have nothing to do with the signaling of genetic quality. He writes that “because of the rewarding property of attractiveness, in the labor market or economic games, attractive individuals would generally be treated better, especially when there is a potential chance for treaters who are in pursuit of a stimulating sexual experience” (para. 6). Here Dang is confusing issues of mechanisms and function. We agree that from a mechanistic perspective beautiful faces have rewarding properties, which drive individuals' preferences for these faces. This is perfectly compatible with a functional explanation of these rewarding properties and preferences in terms of mating and fitness.

Hafenbrädl & Dana argue that mating motives are not sufficient to explain the persistence of attractiveness-related biases in the labor market because, in a competitive market, firms without these biases will outcompete biased firms and eventually the beauty premium effect will be wiped out. We do not share Hafenbrädl & Dana's optimism about the power of the market to self-correct any biases in hiring on the basis of competitive mechanisms alone. Attractiveness-related biases in decision making reflect a basic expression of human nature; they are widespread, presumably convey advantages to the individuals who express these biases, though not necessarily to their firms, and are unlikely to be wiped out by competitive market forces. Take nepotism in hiring practices as a similar case; although nepotism often results in the hiring of related but lower-quality individuals over unrelated higher-quality individuals, nepotism has been around for thousands of years in all markets around the world, and it is not disappearing anytime soon.

Hafenbrädl & Dana remark that the idea that attractiveness is correlated with desirable but unobserved qualities such as genetic quality coincides with the economic concept of statistical discrimination. They then argue that just as people can reliably assess that attractive individuals have greater genetic quality and health, they should also be able to reliably assess that they are better workers. We disagree. In one case, the attractiveness-quality link and the preference for attractive individuals as mating partners have been established by natural selection over millions of years. In the other case, human beings try to establish the competence of potential employees based on a quick look at their faces. Human beings are more likely to be wrong in their perceptions of other human beings than natural selection is more likely to have made mistakes over millions of years. The question of whether attractive individuals are better workers is an empirical one; the empirical studies we reviewed in our article suggest that they are not. We agree with Hafenbrädl & Dana that “they need not even be right in their beliefs that attractive workers are, on average, better for statistical discrimination to operate” (sect. 2, para. 2). This, however, begs the question of why people have the mistaken belief that attractive individuals are better workers. We believe that our target article provides the most likely answer to this question.

We agree with Hafenbrädl & Dana that mating motives cannot supplant all other explanations of the beauty premium, including economic ones. In fact, the evolutionary explanations we discussed in our article are broadly compatible with the economists' taste-based discrimination model. They both assume that people have a preference for attractive individuals regardless of their personality or behavior. However, the taste-based discrimination model is so general and vague as to be useless when trying to understand the origins of preferences for attractive individuals.

Similar to Hafenbrädl & Dana, Le Lec et al. believe that we were too quick in rejecting economic and stereotype-based explanations of attractiveness-related biases. For example, they argue that as stereotypes operate most of the time on an unconscious level and their influence cannot be captured through explicit self-reports, evidence from studies that failed to demonstrate a causal role for stereotypes in attractiveness-related biases should be taken with caution. We are all in favor of caution, but stereotype-based explanations of attractiveness-related biases are accepted uncritically by many social psychologists and economists in the absence of any scientific evidence supporting these theories and without even considering alternative explanations.

We disagree with almost everything Roberts has written in his commentary. He writes that “Maestripieri et al. consider ‘mating motives’ as central to the evolutionary approach. Aside from this being an unfortunate term in invoking proximate mechanisms rather than adaptation, this is not a term used in EP [evolutionary psychology]” (para. 2). In reality, a central assumption of evolutionary psychology is that adaptation occurs at the level of psychological mechanisms (e.g., perception, emotion, motivation, and cognition) rather than behavior. “Mating motives” is an expression commonly used in evolutionary psychology (e.g., see commentary by Agthe & Maner).

Roberts also states: “It is notable that sexual selection is not referred to at all in the review, so the theoretical basis of the article must be considered weak” (para. 2). Sexual selection is an evolutionary process, whereas our target article focuses on adaptation, the product of sexual or natural selection, and not on the process itself. Roberts also misunderstands the three evolutionary hypotheses we discuss in our target article. He writes:

We clearly stated that according to the functional evolutionary hypothesis, attractiveness-related biases serve to maintain proximity to attractive individuals, which in turn may increase the probability of mating with them. Roberts continues: “The second, referred to as a ‘nonfunctional hypothesis,’ turns out to be a misunderstanding of how adaptation works on characteristics that are on average beneficial” (para. 3). No, the by-product hypothesis is not a misunderstanding of how adaptation works. By-product explanations are clearly described in evolutionary psychology textbooks. Roberts goes on to say that “both are therefore in fact corollaries of their third hypothesis, which they refer to as the ‘sexual signaling hypothesis’” (para. 3). No, the functional evolutionary hypothesis and the by-product hypothesis are not corollaries of the sexual signaling hypothesis because they do not assume that attractiveness-related biases have a signaling function.

Roberts contends: “There are too many misunderstandings to mention here. For example, Maestripieri et al. say that a ‘preference for attractive individuals … may not in itself increase an individual's biological fitness’ (sect. 2.1.3, para. 2), whereas in reality such a preference should have on average increased fitness, hence the preference persists” (para. 5). There is no misunderstanding on our part. We wrote:

Roberts claims that we repeatedly conflate mechanisms and function. We do not. We do agree with Roberts on one point though: that economists often ignore, underestimate, or misunderstand the value of the evolutionary approach in understanding the behavioral phenomena they study and that our target article can play an important role in improving communication and exchange of ideas between evolutionary psychologists and other behavioral scientists.

Schein et al. argue that research on attractiveness-related biases cannot determine whether the effects are driven by a positive, beauty-is-good response, as the authors argue, or by a negative, ugly-is-bad response. In reality, there is evidence that beauty-is-good and ugly-is-bad are not independent phenomena and that they both operate simultaneously. For example, research on prostitution has shown that there is a linear relationship between the physical attractiveness of the prostitutes and the amount of money their clients are willing to pay to them. Ugly prostitutes are paid less, and beautiful ones are paid more than the average (reviewed in Hamermesh Reference Hamermesh2011). Schein et al. write: “Research on ugly-is-bad bias, infant research, and research on same-sex preferences for attractive others are not consistent with most versions of mating strategy” (para. 3). We disagree. Schein et al. reject all evolutionary explanations for attractiveness-related biases and propose, as an alternative explanation a domain general information processing system, cognitive averaging, which results in preferences for attractive faces. First, we emphasize that an explanatory cognitive mechanism is not necessarily alternative to the functional evolutionary explanations we discuss in our target article, because these are explanations at different levels of analyses. Second, although it is true that average-looking faces are generally perceived as attractive, the proposed mechanism, cognitive averaging, cannot explain the phenomenon of attractiveness-related biases for many reasons. One is that these biases are not limited to exposure to faces, but also apply to situations in which (female) attractiveness is measured with waist-to-hip ratios. Another one is that the brain imaging studies we reviewed in our target article indicate that different neural circuits are activated when expressing aesthetic judgments about same-sex beautiful male and female faces and when experiencing mating motivation triggered by beautiful opposite-sex individuals. Finally, objections have been raised by other researchers to the argument that the attractiveness of beautiful faces is entirely accounted for by their prototypicality. For example, in his commentary, Becker reviews studies showing that the attention-grabbing properties of beautiful faces cannot be simply reduced to symmetry detection and that memory is more sensitive to beautiful female faces than male faces. He remarks that the morphed prototypical average of attractive female faces is noticeably different from the morphed average of all female faces, suggesting that these faces have additional features for which recognition could be vigilant. Taken together, these findings suggest that attractiveness-related prosocial and financial biases are not triggered by beautiful, average-looking faces in general, but specifically by the beautiful, average-looking faces of potential mating partners, particularly females.

Maguire & Racine's commentary and their critique of our ideas revolve around a misinterpretation of what we wrote in our target article. They quote a statement we made in section 4.1, paragraph 5, of the target article, which described not our opinions, but the conclusions of a study by Lemay et al. (Reference Lemay, Clark and Greenberg2010). Many social psychologists assume that stereotypes about attractive individuals cause people to express favorable behavioral biases toward them. The study by Lemay et al. clearly showed that stereotypes do not play a causal role. People are motivated to affiliate with attractive individuals, and as a result, they construct images of these individuals as interpersonally receptive and responsive. Lemay et al. suggested that people's stereotypes about attractive individuals, if they ever occur, are a post hoc rationalization of their desire to establish bonds with them. Unlike what Maguire & Racine wrote, our functional explanations for the occurrence of attractiveness-related biases do not imply the existence of a psychological construct about attractiveness or attractive individuals that is in any way similar to a stereotype. We suggest that attractive potential mates grab people's attention and increase their motivation to affiliate with them the way sexually explicit images grab men's attention and increase their motivation to view them, or the way the taste of sweet food in the mouth stimulates our appetite for it and increases our motivation to continue ingesting such food. In none of these cases is it necessary or helpful to postulate a psychological construct as an intervening variable mediating our preferences.

We also find it unnecessary to postulate, as Oda does in his commentary, that expressing prosocial biases in favor of attractive individuals requires a special imagination or an optimistic delusion that allows people to believe that they can mate with very attractive individuals when the probability of doing so is very low. If prosocial or financial biases in favor of attractive people increase one's probability of mating with them relative to other individuals who do not express such biases, natural selection will favor any genetically based tendencies to express such biases. To clarify also what we perceive as misunderstandings by other commentators such as Stephen et al. and Oda, our account of attractiveness-related biases provides both a functional explanation and a mechanistic explanation for these biases. Our hypothesized perceptual, cognitive, neural, and neuroendocrine mechanisms are different from those implied by stereotype theories and other explanations favored by social psychologists and economists, including the “cognitive averaging” mechanism idea proposed by Schein et al. in their commentary.

Unlike what Stephen et al. write, some of the brain imagining studies we reviewed, which show activation of neural reward circuitry in response to exposure to attractive faces, are consistent with our hypothesis but not with stereotype theories, because these effects are observed only with faces of potential mating partners (opposite-sex individuals for heterosexuals) and not with the faces of all attractive individuals as predicted by stereotype theories. Clearly, we do not provide a comprehensive account of all mechanisms responsible for attractiveness-related biases. For example, although we acknowledge that social and cultural influences probably contribute to the expression of these biases, we do not examine such influences in our target article. Rennels & Verba, however, provide an interesting and useful discussion of how social experience acquired during the formative years reinforces the tendency to be positively biased toward attractive faces that is present already in infancy, particularly in the case of attractive female faces.

The contrast between our explanations, which address both function and mechanisms, and those favored by social psychologists and economists, which focus on mechanisms, does not simply reflect a difference between levels of analyses as suggested by Oda and Ronay & Tybur. We do not offer a contrast between disciplines that see each other as foes, as noted by Ronay & Tybur. We do not see evolutionary psychology, to use Ronay & Tybur's expression, as the “wolf at the door” that threatens to swallow other behavioral disciplines. We believe that the claims that evolutionary behavioral disciplines will take over the social sciences are unfounded. Evolutionary psychology is no wolf, but social psychology and economics are no lambs either. Ronay & Tybur write that “by themselves, positive stereotypes and taste preferences cannot explain the relationship between attractiveness and prosociality at any fundamental level” (para. 7); however, many social psychologists and economists believe that they do. Although in some cases the explanations provided by evolutionary psychologists are compatible with those of other behavioral scientists, in others they are not. This is the case especially whenever the evolutionary explanations address not only issues of functional significance, but also those of mechanisms, and the hypothesized mechanisms are different than those proposed by other researchers in the same or another discipline. We have already remarked that our evolutionary account of attractiveness-related biases, which addresses both function and mechanism, is compatible with the taste-based discrimination model used by economists. We do believe, however, that the mechanistic explanations of attractiveness-related biases proposed by many social psychologists and economists are substantively different from those proposed in our target article.

The contrast is not between the disciplines but between the specific hypotheses that researchers working within these disciplines have proposed. In our target article, we use the expression “evolutionary explanations” in reference to the three specific evolutionary hypotheses. We agree that they are not the only evolutionary explanations and that future research on attractiveness-related biases should explore other evolutionary aspects of this phenomenon that are not directly related to mate choice but are more broadly relevant to the issue of social partner choice. We are all in favor of researchers from disciplines working together as allies in a common enterprise. Indeed, this is one of the main motives that prompted us to write our target article. We hope that future research on attractiveness-related biases will benefit from the theoretical and methodological tools contributed by different disciplines. We hope that this interdisciplinary research will establish which explanations of attractiveness-related biases are most supported by the empirical evidence, regardless of the discipline from which these hypotheses originated.

R4. Scholarship issues

The authors of two commentaries (Feingold and LaFrance & Eagly) accuse us of poor scholarship. Feingold describes our article as an old-school qualitative review of the literature and warns that “researchers who conduct qualitative reviews of the attractiveness literature may draw tendentious conclusions that are inconsistent with the findings from past or future meta-analyses” (para. 3). If we wanted to conduct a meta-analysis of the effects of attractiveness on social decision making, we would have done it and would have written a very different article. Our goal instead was mainly to write an article of ideas, which reviewed the different conceptual perspectives used in previous attractiveness research and critically discussed and integrated the empirical findings of different disciplines. To our knowledge, ours is the first article of this kind to be written about attractiveness research, a field of inquiry in which researchers working within a particular discipline have traditionally ignored, misinterpreted, or dismissed the theoretical and empirical contributions of other disciplines. We hope that our article will generate more comprehensive and integrative research on attractiveness biases, in which the strengths and weaknesses of different disciplinary approaches are recognized. This is an important way in which scientific research advances. Science is not just about P values. If conceptual papers are considered “old- school,” we are happy to be considered old-school researchers.

As for Feingold's criticism that our conclusions contradict those of previous meta-analyses of the literature (e.g., Eagly et al. Reference Eagly, Ashmore, Makhijani and Longo1991; Feingold Reference Feingold1992a; Jackson et al. Reference Jackson, Hunter and Hodge1995; Mazzella & Feingold Reference Mazzella and Feingold1994), these meta-analyses were conducted more than 20 years ago. Even more recent meta-analyses (e.g., Hosoda et al. Reference Hosoda, Stone-Romero and Coats2003; Langlois et al. Reference Langlois, Kalakanis, Rubenstein, Larson, Hallam and Smoot2000) are now outdated, given that attractiveness-related biases are an extremely active field of research, with new studies on this topic being published every month. In our target article, we cited approximately 100 articles published after 2000, many of which have provided new empirical data on attractiveness-related biases, and others that have presented conceptual advances that help us better understand this phenomenon. Finally, previous meta-analyses of this literature were selective in that they focused mainly on studies conducted by social psychologists. Studies of the labor market and experimental studies involving economic games were largely ignored. Since 2000, there has been an explosion of studies of the effects of attractiveness on decision making in economics games. These studies using carefully controlled and highly standardized experimental conditions represent an optimal paradigm for investigating attractiveness-related biases. Our review of these studies highlights some clear trends in their results, which make it necessary to re-evaluate the conclusions of previous research and previous reviews of the literature.

With agree with LaFrance & Eagly that both men and women value physical attractiveness in their sexual partners very highly, particularly in the context of short-term mating. It is important to remember, however, that most research on the role of attractiveness in mate preferences has focused on people's “ideal” preferences. In the real world, given that women are generally choosier about their sexual partners than are men, it is likely that highly attractive women are less accessible to average-looking men than highly attractive men are accessible to average-looking women. In the labor market and other real-life contexts, this may result in men's biases in favor of attractive women being stronger than women's biases in favor of attractive men. In our target article, we also acknowledged that the stronger attractiveness biases for women than for men in the labor market likely reflect, at least in part, the fact that more men are in positions of power than women and therefore have more opportunities to express their biases. The hypothesis that physical attractiveness is an indicator of good genes is well accepted by evolutionary biologists, based on evidence from both animal and human research. Whether physical attractiveness is an honest marker of fertility is largely irrelevant to the point in our article. It is well known that for men fertility is generally not an issue, whereas for women the best marker of fertility is age, which explains the universal men's preference for young women.

Contrary to what was stated by LaFrance & Eagly, we did not neglect the relation between trustworthiness and attractiveness. In fact, we reviewed studies using experimental trust games, which show that attractive people are trusted more than unattractive ones. That people often infer trustworthiness, dominance, and other traits from faces independent of attractiveness is something we mentioned in our article, but only briefly because it is not strictly relevant to our article, which is focused on attractiveness and not on faces in general. Contrary to what was hypothesized by LaFrance & Eagly, preferences for attractive partners are not a modern human phenomenon. They are widespread in animals as well. It is not astonishing that the overlap in the reference list of our target article and LaFrance & Eagly's commentary is limited to two articles, given that most of the studies they cite are irrelevant to our article. Given that LaFrance & Eagly's commentary does not directly address any of the explanations for attractiveness-related biases we discussed in our article, the statement made in the commentary's title (“Omitted Evidence Undermines Sexual Motives Explanation for Attractiveness Bias”) is unwarranted.

R5. Policy implications of research on attractiveness-related biases

Two commentators suggested that the conclusions drawn in our target article, and more generally the findings of research on the determinants of financial and prosocial biases in favor of attractive people, have potentially important implications for society and policy. Ravina suggests that “depending on the causes of the positive bias toward attractive people, and the relationship between attractiveness and productivity, prosocial behaviors, and personality traits, we should either ignore the bias or make sure that our employees/decision makers are made aware or protected from it and from the ‘mistakes’ to which it leads” (para. 7). She also remarks that “understanding the mechanism will also help us identify the people more prone to the bias, the contexts in which it is stronger, and possibly the best devices to protect the decision makers from it when stakes are high, if they do not do so already by themselves” (para. 7).

Similar to Ravina, Minerva suggests that it is important to conduct research to understand the origins of attractiveness-related biases. She writes: “Although lookism is often considered a politically incorrect topic to discuss, it is only by learning more about it that we will be able to find ways to deal with such a phenomenon and, we hope, alleviate its negative impact” (para. 13). We agree. Minerva offers a number of suggestions about how to limit the impact of attractiveness-related biases in the labor market and even suggests that legal measures could be introduced to protect individuals who are penalized by these biases.

We thank both Ravina and Minerva for their insightful comments and helpful suggestions. Although we did not discuss any policy implications of the research on attractiveness-related biases reviewed in our target article, we strongly believe that government officials, policy makers, and legal experts should be aware of the results of scientific research on human behavior, especially with respect to how biologically based behavioral tendencies can influence social and economic processes. Nonscientists should also be educated about avoiding the naturalistic fallacy. The fact that some behavioral phenomena have a biological basis and an evolutionary origin in no way implies that they are socially or ethically acceptable or that they are immutable and inevitable. We hope that our target article and the discussion it has already stimulated will generate new knowledge of human behavior, and that this new knowledge will increase our awareness of our biases and give us the tools to control or eliminate them, if we as a society choose to do so.