Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2013 Oct 30;105(4):474–485. doi: 10.1111/bjop.12055

Do assortative preferences contribute to assortative mating for adiposity?

Claire I Fisher 1, Corey L Fincher 1, Amanda C Hahn 1, Anthony C Little 2, Lisa M DeBruine 1, Benedict C Jones 1,*
PMCID: PMC4282125  PMID: 24168811

Abstract

Assortative mating for adiposity, whereby levels of adiposity in romantic partners tend to be positively correlated, has implications for population health due to the combined effects of partners' levels of adiposity on fertility and/or offspring health. Although assortative preferences for cues of adiposity, whereby leaner people are inherently more attracted to leaner individuals, have been proposed as a factor in assortative mating for adiposity, there have been no direct tests of this issue. Because of this, and because of recent work suggesting that facial cues of adiposity convey information about others' health that may be particularly important for mate preferences, we tested the contribution of assortative preferences for facial cues of adiposity to assortative mating for adiposity (assessed from body mass index, BMI) in a sample of romantic couples. Romantic partners' BMIs were positively correlated and this correlation was not due to the effects of age or relationship duration. However, although men and women with leaner partners showed stronger preferences for cues of low levels of adiposity, controlling for these preferences did not weaken the correlation between partners' BMIs. Indeed, own BMI and preferences were uncorrelated. These results suggest that assortative preferences for facial cues of adiposity contribute little (if at all) to assortative mating for adiposity.

Background

Assortative mating for adiposity, whereby levels of adiposity in romantic partners tend to be positively correlated, has frequently been reported in studies of population health (see Di Castelnuovo, Quacquaruccio, Donati, De Gaetano, & Iacoviello, 2009 for a meta-analytic review). Moreover, this correlation between levels of adiposity in romantic partners appears to be robust (Di Castelnuovo et al., 2009). For example, although correlations between levels of adiposity in romantic partners have typically been shown using body mass index (BMI; e.g., Jacobson, Torgerson, Sjostrom, & Bouchard, 2007; Silventoinen, Kaprio, Lahelma, Viken, & Rose, 2003), they have also been demonstrated using other measures of adiposity, including dual-energy X-ray absorptiometry (Speakman, Djafarian, Stewart, & Jackson, 2007) and skinfold thicknesses (Ginsburg, Livshits, Yakovenko, & Kobyliansky, 1999). Moreover, assortative mating for adiposity does not appear to be an artefact of potential confounds. For example, although age and adiposity tend to be positively correlated (e.g., Pasco, Nicholson, Brennan, & Kotowicz, 2012), as are romantic partners' ages (e.g., Watson et al., 2004), assortative mating for adiposity is not simply a by-product of the combined effects of these correlations (e.g., Speakman et al., 2007). Similarly, although socio-economic status and adiposity tend to be negatively correlated (e.g., Moore, Stunkard, & Srole, 1997) and romantic couples tend to be from similar social backgrounds (e.g., Schwartz & Mare, 2005; Smits, 2003), combined effects of these correlations do not fully explain assortative mating for adiposity (e.g., Silventoinen et al., 2003). That the within-couples correlation between levels of adiposity appears to change very little as the duration of the relationship increases also suggests that assortative mating for adiposity is not simply due to the effects of the shared environment on adiposity (see Di Castelnuovo et al., 2009 for a meta-analytic review).

Importantly, assortative mating for adiposity could have serious consequences for population health. For example, as there is a substantial genetic component to adiposity (reviewed in Speliotes et al., 2010), assortative mating for adiposity may increase the proportion of individuals with high levels of adiposity (i.e., obese individuals) in the population (Hebebrand et al., 2000; Speakman et al., 2007). Indeed, some models of the effect of assortative mating for adiposity on the prevalence of obesity in a baseline population suggest that switching from random to completely assortative mating could more than double the percentage of obese individuals in the population within just two generations (Speakman et al., 2007). The combined effects of men's and women's adiposity on the fertility of couples (i.e., the tendency for infertility to be particularly common in couples where both the man and woman have high levels of adiposity) may counteract, to some extent, this effect of assortative mating on rates of obesity (Ramlau-Hansen et al., 2007). However, the combined effects of men's and women's adiposity on fertility also present additional evidence that assortative mating for adiposity can have negative effects on important aspects of population health (in this case, rates of infertility). In the light of findings such as these, many researchers have emphasized the importance of establishing why assortative mating for adiposity occurs (Courtiol, Picq, Godelle, Raymond, & Ferdy, 2010; Di Castelnuovo et al., 2009; Hebebrand et al., 2000; Speakman et al., 2007).

Assortative preferences for cues of adiposity, whereby leaner people show stronger preferences for leaner individuals, are a potential explanation for the positive correlation between romantic partners' levels of adiposity (e.g., Courtiol et al., 2010; Speakman et al., 2007; Zietsch, Verweij, Heath, & Martin, 2011). Indeed, assortative preferences for other physical characteristics have been reported in several studies of human mate preferences (reviewed in, e.g., Havlicek & Roberts, 2009) and assortative preferences are thought to play a critical role in assortative mating for physical characteristics in several non-human species (e.g., Møller, 1994). Furthermore, men's and women's preferences for leaner body shapes in silhouettes of opposite-sex bodies are negatively correlated with their actual romantic partner's BMI (Courtiol et al., 2010), although the cross-sectional design used in this work means that the causal direction of the relationship is unclear. Market-value-contingent preferences, whereby more attractive individuals demonstrate stronger preferences for attractive characteristics in images of potential mates (e.g., Little & Mannion, 2006), also suggest that assortative preferences for cues of adiposity could occur. However, while some work has suggested that preferences for cues of adiposity in body images are correlated with the perceiver's own BMI (Tovée, Emery, & Cohen-Tovée, 2000), this effect of own BMI on attractiveness judgements appears to be largely due to atypical perceptions in individuals with eating disorders (e.g., anorexia nervosa, Tovée et al., 2000). Indeed, some other studies also suggest that measures of own adiposity, including BMI, do not predict mating-related perceptions of body images, such as perceptions of potential mates' health or youth (Han, Morrison, & Lean, 1999) and attractiveness (Price, Pound, Dunn, Hopkins, & Kang, 2013). Although these results are not necessarily conclusive, findings such as these suggest that own adiposity may have little effect on body adiposity preferences, at least in healthy individuals (Speakman et al., 2007), raising the possibility that assortative mating for adiposity is at least partly due to the additional constraints on the mate choices of individuals with higher levels of adiposity (see, e.g., Zietsch et al., 2011).

Despite the fact that there have been no direct empirical tests of the role of assortative preferences in assortative mating for adiposity, the lack of evidence for a robust relationship between own adiposity and mating-related perceptions of bodies has led some researchers to conclude that assortative preferences contribute little to assortative mating for adiposity (e.g., Speakman et al., 2007). However, focusing exclusively on body attractiveness may limit the conclusions that can be drawn about the role of mate preferences in assortative mating for adiposity. For example, although facial characteristics (including cues of adiposity) were obscured in the stimuli used in the above studies, facial cues of adiposity communicate information that is known to be important for human mate choice. For example, facial cues of adiposity communicate information about peoples' physical attractiveness (Coetzee, Perrett, & Stephen, 2009; Hume & Montgomerie, 2001), perceived health (Coetzee et al., 2009), actual physical health (Coetzee et al., 2009; Tinlin et al., 2013), life expectancy (Reither, Hauser, & Swallen, 2009), immunocompetence (Rantala et al., 2013), psychological condition (Tinlin et al., 2013), and hormonal profile (Tinlin et al., 2013). Indeed, perceived facial adiposity (i.e., the perception of fatness in the face) conveys information about health over and above that which is explained by BMI (Tinlin et al., 2013) and some aspects of health (e.g., reported frequency and duration of respiratory infections) are more strongly correlated with facial adiposity than they are with BMI (Coetzee et al., 2009). In addition, some studies suggest that facial cues are more important than body characteristics for judgements of men's and women's attractiveness, especially, but not exclusively, when bodies are clothed and when participants judged the attractiveness of potential mates for long-term relationships (e.g., Confer, Perilloux, & Buss, 2010; Currie & Little, 2009; Peters, Rhodes, & Simmons, 2007). Together, these results raise the intriguing possibility that individual differences in preferences for facial cues of adiposity may play an important role in assortative mating for adiposity. However, no study to date has addressed the relationships between preferences for facial cues of adiposity and either own or actual partner characteristics. Perhaps more importantly, it is also not known whether individual differences in preferences for facial cues of adiposity contribute to assortative mating for adiposity.

In the light of the above, we investigated the contribution of individual differences in preferences for cues of adiposity in opposite-sex faces to assortative mating for adiposity (measured using BMI). If controlling for the possible effects of individual differences in preferences for facial cues of adiposity weakens the positive correlation between romantic partners' BMIs, it would suggest that assortative preferences contribute to assortative mating for adiposity. However, if controlling for the possible effects of individual differences in preferences for cues of adiposity does not weaken the predicted correlation between romantic partners' BMIs, it would suggest that assortative preferences for facial cues of adiposity contribute little to assortative mating for adiposity. We assessed individual differences in preferences for facial cues of adiposity in two ways. One method measured participants' preferences for perceived facial adiposity. The other measured participants' preferences for facial characteristics associated with actual BMI. Given the correlations between perceived facial adiposity and measured BMI that were reported in previous studies (Coetzee et al., 2009; Tinlin et al., 2013), these two measures of adiposity preference were expected to be highly correlated.

Methods

Participants

Sixty-two heterosexual couples were recruited for the study. The mean age of the men was 21.8 years (SD = 1.96 years) and the mean age of the women was 21.2 years (SD = 1.94 years). The average duration of these couples' relationships was 18.4 months (SD = 15.1 months). Following Courtiol et al. (2010), the man and woman in each couple were tested at the same time, but were separated during testing.

Stimuli

The stimuli that we used to assess preferences for cues of adiposity in opposite-sex faces were full-colour images of 50 white men (mean age = 24.2 years, SD = 3.99 years) and 50 white women (mean age = 24.3 years, SD = 4.01 years). All images were taken under standardized lighting conditions and against a constant background. All individuals photographed posed with neutral expressions and direct gaze. Images were standardized on pupil position and masked so that clothing was not visible. Height and weight measurements were taken from each of these 50 men (mean height = 180.2 cm, SD = 6.62 cm; mean weight = 77.3 kg, SD = 12.4 kg) and 50 women (mean height = 168.6 cm, SD = 6.48 cm; mean weight = 57.2 kg, SD = 11.4 kg).

The height and weight measurements were used to calculate each of the photographed individuals' BMI (men: M = 23.7 kg/m2, SD = 3.13 kg/m2, range = 17.7–31.0 kg/m2; women: M = 20.1 kg/m2, SD = 3.66 kg/m2, range = 16.2–38.4 kg/m2). According to the World Health Organization's (WHO) classifications (World Health Organization, 2000), 28% of the women and 2% of the men were in the underweight BMI category (<18.5 kg/m2), 68% of the women and 68% of the men were in the normal category (18.5–24.99 kg/m2), and 4% of the women and 30% of the men were in the overweight category (>25 kg/m2). None of these individuals were extremely underweight (i.e., none had BMI < 15 kg/m2, Bray, 1978) and only two individuals (both women) had BMI < 17 kg/m2.

Methods for collecting ratings of facial adiposity were identical to those used in previous studies (Coetzee et al., 2009; Tinlin et al., 2013). The 50 male face images were rated for adiposity using a 1 (very underweight) to 7 (very overweight) scale by 60 heterosexual raters (30 women, 30 men; mean age = 22.08 years, SD = 3.53 years). The order in which images were presented was fully randomized. A different group of 60 heterosexual raters (30 women, 30 men; mean age = 23.18 years, SD = 3.04 years) rated the 50 female images for adiposity using the same scale. Inter-rater agreement was extremely high for adiposity ratings of both men's (Cronbach's alpha = .98) and women's (Cronbach's alpha = .98) faces. Consequently, we calculated the average adiposity rating for each face image (male faces: M = 3.73, SD = 0.76; female faces: M = 3.79, SD = 0.80). Men's and women's average ratings were highly correlated for both men's faces (r = .97, N = 50, p < .001) and women's faces (r = .96, N = 50, p < .001). Consistent with prior work (Coetzee et al., 2009; Tinlin et al., 2013), BMI and rated facial adiposity were positively correlated (men: r = .55, N = 50, p < .001; women: r = .69, N = 50, p < .001).

Procedure

Height and weight measurements were taken from each of the 62 men (mean height = 180.6 cm, SD = 6.55 cm; mean weight = 77.1 kg, SD = 11.2 kg) and 62 women (mean height = 166.3 cm, SD = 5.35 cm; mean weight = 64.0 kg, SD = 12.6 kg) who made up our romantic couples. These measurements were used to calculate BMI (men: M = 23.6 kg/m2, SD = 3.13 kg/m2, range = 16.4–31.7 kg/m2; women: M = 23.1 kg/m2, SD = 4.19 kg/m2, range = 18.0–37.1 kg/m2). According to the World Health Organization (2000) classifications, 5% of the women and 5% of the men were in the underweight BMI category (<18.5 kg/m2), 72% of the women and 69% of the men were in the normal category (18.5–24.99 kg/m2), and 23% of the women and 26% of the men were in the overweight category (>25 kg/m2). None of these individuals were extremely underweight (i.e., none had BMI < 15 kg/m2, Bray, 1978) and only one individual (a man) had BMI < 17 kg/m2.

Each of the 62 men in our study rated the attractiveness of the 50 female faces described in our Stimuli section. In addition, each of the 62 women in our study rated the attractiveness of the 50 male faces described in our Stimuli section. Attractiveness ratings were made using a 1 (much less attractive than average) to 7 (much more attractive than average) scale. Following previous studies of preferences for facial cues of adiposity (e.g., Coetzee et al., 2009), the order in which images were presented was fully randomized and each image remained on screen until the participant had entered their rating. Inter-rater agreement, as measured by Cronbach's alpha, was high for both sets of ratings (both >.96).

Results

Calculating preference scores

For each couple, we calculated the Pearson product-moment correlation between (1) the woman's attractiveness rating for each of the 50 men's faces and those 50 men's rated facial adiposity (mean r = −.16, SD = .17, p = .27); (2) the woman's attractiveness rating for each of the 50 men's faces and those 50 men's BMI (mean r = −.13, SD = .13, p = .37); (3) the man's attractiveness rating for each of the 50 women's faces and those 50 women's rated facial adiposity (mean r = −.23, SD = .14, p = .11); and (4) the man's attractiveness rating for each of the 50 women's faces and those 50 women's BMI (mean r = −.27, SD = .10, p = .058).2010 Note that this procedure produces two correlation coefficients for each participant (one being a measure of their preference for perceived facial adiposity and the other a measure of their preference for facial cues of BMI). These correlation coefficient scores served as the dependent variables in subsequent analyses. For each of these preference scores, larger positive values indicate stronger preferences for facial characteristics associated with higher BMI and larger negative values indicate stronger preferences for facial characteristics associated with lower BMI. This method for calculating preference scores by deriving the correlation coefficients for the relationships between each participant's attractiveness ratings of individual faces and another variable (in this study, either rated facial adiposity or BMI) has been used in previous studies to assess individual differences in face preferences (e.g., Fisher, Fincher, Hahn, DeBruine, & Jones, 2013).

Next, we analysed women's preference scores (i.e., their preferences for perceived facial adiposity and their preferences for facial cues of BMI) using factor analysis. This analysis produced a single factor that explained 87% of the variance in women's preference scores and was highly correlated with both of the original variables (both r = .93). We labelled this factor women's preference for cues of BMI in men's faces. A corresponding analysis for men's preference scores also produced a single factor. This factor explained 89% of the variance in men's preference scores and was highly correlated with both of the original variables (both r = .94). We labelled this factor men's preference for cues of BMI in women's faces. On both of these factors, higher scores indicate stronger preferences for facial characteristics associated with higher BMI. Romantic partners' preferences for cues of BMI in opposite-sex faces were positively correlated (r = .31, N = 62, p = .016).

Assortative mating for BMI in our sample

We first tested for evidence of assortative mating for BMI in our sample. As predicted, romantic partners' BMIs were positively correlated (r = .49, N = 62, p < .001). Subsequent partial correlation analyses showed that this correlation between romantic partners' BMIs remained significant when we controlled for the possible effects of women's age (partial r = .38, p = .003), men's age (partial r = .45, p < .001), or both men's age and women's age simultaneously (partial r = .38, p = .003). These results show that assortative mating for BMI in this sample is not due to the combined effects of older individuals tending to have higher BMI and couples tending to be similar in age. Similarly, results for a partial analysis controlling for the possible effects of relationship duration (partial r = .45, p < .001) suggested that relationship duration had little effect on the strength of assortative mating for BMI in this sample.

Assortative preferences and assortative mating for BMI

To investigate whether individual differences in preferences for cues of BMI in opposite-sex faces contributed to assortative mating for BMI, we conducted a second set of partial correlation analyses. These analyses showed that the correlation between romantic partners' BMIs (r = .49, N = 62, p < .001) changed very little when we controlled for the possible effects of women's preferences for cues of BMI in men's faces (partial r = .50, p < .001), men's preferences for cues of BMI in women's faces (partial r = .47, p < .001), both men's and women's preferences for cues of BMI in opposite-sex faces simultaneously (partial r = .50, p < .001), or the average of men's and women's preferences for cues of BMI in opposite-sex faces (partial r = .47, p < .001). These results suggest that individual differences in preferences for facial cues of BMI contributed little (if at all) to assortative mating for BMI.

Preferences for cues of BMI and own/partner's BMI

Although our results suggest that individual differences in preferences for facial cues of BMI contribute little to assortative mating for BMI, it is still possible that participants' preferences for cues of BMI in opposite-sex faces are correlated with either their own BMI or their partner's BMI. Thus, we investigated the intercorrelations among men's BMI, women's BMI, men's preferences for cues of BMI in women's faces, and women's preferences for cues of BMI in men's faces (Table 2010). Men's and women's preferences for cues of BMI in opposite-sex faces predicted their romantic partner's BMI (men's preferences: r = .30, N = 62, p = .017; women's preferences: r = .30, N = 62, p = .017), indicating the individuals who showed stronger preferences for cues of low BMI in opposite-sex faces had leaner romantic partners. In contrast with these results for partner's BMI, own BMI predicted neither men's nor women's preferences for facial cues of BMI (men's preferences: r = .17, N = 62, p = .19; women's preferences: r = .04, N = 62, p = .77).

Table 1.

Correlations among men's body mass index (BMI), women's BMI, men's preferences for cues of BMI in women's faces, and women's preferences for cues of BMI in men's faces

Women's BMI Men's preference for cues of BMI Women's preference for cues of BMI
Men's BMI .49** .17NS .30*
Women's BMI .30* .04NS
Men's preference for cues of BMI .31*
Open in a new tab

Note N = 62 for each correlation

NS

p > .05

*

p < .05

**

p < .001 (all non-significant correlations were p > .19).

Finally, we conducted regression analyses to test whether preferences for cues of BMI in opposite-sex faces and own BMI independently predicted actual romantic partner's BMI. Partner's BMI was entered as the dependent variable and own BMI and preferences for cues of BMI were entered simultaneously as predictors. Separate regression analyses were conducted for men and women. For men, their own BMI (t = 4.04, standardized beta = .45, p < .001) and their preference for cues of BMI in women's faces (t = 2.03, standardized beta = .23, p = .047) each independently predicted their romantic partner's BMI. This pattern was also observed for women; own BMI (t = 4.43, standardized beta = .48, p < .001) and preference for cues of BMI in men's faces (t = 2.63, standardized beta = .28, p = .011) each independently predicted their romantic partner's BMI. Stepwise versions of these analyses showed that adding preferences to a model in which only own BMI was a predictor significantly increased the variance in partner BMI explained by the model (men's preferences: R2 change = .05, p = .047; women's preferences: R2 change = .08, p = .011).

Discussion

As in many previous studies that have reported assortative mating for adiposity (see Di Castelnuovo et al., 2009 for a meta-analytic review), romantic partners' BMIs were positively correlated. Partial correlation analyses showed that this correlation between romantic partners' BMIs was not due to the possible effects of age or relationship duration (see also, e.g., Di Castelnuovo et al., 2009; Speakman et al., 2007).

Consistent with previous work showing that men's and women's preferences for leaner body images were correlated with their actual romantic partner's BMI (Courtiol et al., 2010), men's and women's preferences for facial cues of BMI in opposite-sex faces were correlated with their romantic partners' BMIs; people with leaner partners showed stronger preferences for cues of low BMI in opposite-sex faces. In addition, romantic partners' preferences for cues of BMI in opposite-sex faces were concordant; the romantic partners of people who had particularly strong preferences for cues of low BMI also tended to demonstrate stronger preferences for these facial cues. Despite these significant correlations between BMI preferences and partner's BMI and between romantic partners' BMI preferences, we found no evidence that individual differences in preferences for facial cues of BMI contributed to assortative mating for BMI. Indeed, controlling for the possible effects of assortative preferences for cues of BMI had no discernible effect on the correlation between romantic partners' BMIs. Additional analyses indicated that this pattern of results was due to the independence of own BMI and preferences for facial cues of BMI; own BMI and preference for facial cues of BMI were not significantly correlated in our sample and independently predicted partner's BMI in both men and women. In other words, although preferences for cues of BMI in opposite-sex faces explained some of the variance in the adiposity of men's and women's romantic partners, this variance was wholly independent of that which was explained by assortative mating for BMI. Together, these results then suggest that assortative preferences contribute little (if at all) to assortative mating for adiposity. That own BMI and preference for facial cues of BMI were not significantly correlated in our sample is consistent with other recent work that observed no significant correlations between measures of participants' own adiposity and their preferences for cues of adiposity in opposite-sex bodies (Price et al., 2013).

Although factors not considered in this study will almost certainly contribute to assortative mating for adiposity (e.g., social homogamy, Silventoinen et al., 2003), our results are consistent with the proposal that assortative mating for adiposity is not due to assortative preferences for cues of adiposity, but is likely to be due (at least in part) to the additional constraints on the mate choices of individuals with higher levels of adiposity (Speakman et al., 2007; see also Zietsch et al., 2011). Additional constraints on the mate choices of individuals with higher levels of adiposity may arise because the pool of people who are willing to choose mates with higher levels of adiposity will be smaller (and include a higher proportion of fatter individuals) than the pool of people who are willing to choose relatively lean mates (Speakman et al., 2007). Indeed, individuals with higher levels of adiposity do report having had fewer sexual partners in the previous year, consistent with the proposal that their mate choices are more constrained (e.g., Bajos, Wellings, Laborde, & Moreau, 2010). Our results are also consistent with recent research on the genetic basis of assortative mating for BMI, which suggests that it is more likely to be due to the heritability of BMI than heritability of preferences for cues of BMI (Zietsch et al., 2011).

Findings for attractiveness judgements of opposite-sex faces are often assumed to give insight into the factors that influence human mate choice (for reviews see Gangestad & Thornhill, 2008; Little, Jones, & DeBruine, 2011; Rhodes, 2006) and are frequently interpreted as evidence that sexual selection has been an important factor in the evolution of human face preferences (Gangestad & Thornhill, 2008; Little et al., 2011; Rhodes, 2006). However, many researchers have noted that few studies have investigated the possible correlations between face preferences and actual partner choice (e.g., Penton-Voak, 2011; Puts, Jones, & DeBruine, 2012). Our data show that preferences are linked to real-life mate choice as they demonstrate correlations between partner BMI and both men's and women's face preferences, revealing a pathway through which sexual selection could have influenced preferences for facial cues of adiposity. Although the conclusions that can be drawn on this point from our data are limited to conclusions relating to preferences for facial cues of adiposity, similar tests involving preferences for other facial characteristics (e.g., sex-stereotypical shape cues) may provide converging evidence for links between face preferences and romantic partner choice (see, e.g., DeBruine, Fincher, Watkins, Little, & Jones, 2012).

There are several limitations to this study that should be acknowledged. First, the cross-sectional design of this study means that the causal direction of the relationship between face preferences and partner BMI is unclear. Studies using longitudinal designs to investigate this issue may clarify whether mate preferences directly influence mate choices, mate choices directly influence mate preferences, or both. Second, the ranges of BMIs represented in our stimuli and our participant couples were relatively narrow (e.g., did not include many obese individuals). Further studies with a greater proportion of overweight and obese individuals could yet implicate assortative preferences in assortative mating for adiposity. Such studies could also use face stimuli that were more closely matched in age to the participants than was the case in this study where the faces were, on average, between 2 and 3 years older than the participants. Third, we used a subjective measure of facial adiposity that, although positively correlated with actual BMI, may still be subject to perceptual biases (e.g., attractiveness halo effects, Dion, Berscheid, & Walster, 1972). Subsequent studies exploring individual differences in preferences for facial cues of BMI may consider employing more objective measures of facial adiposity. However, although some facial metric measures of adiposity have been developed that correlate reasonably well with actual BMI (see Coetzee, Chen, Perrett, & Stephen, 2010), these measures can be subject to systematic errors (see Schneider, Hecht, & Carbon, 2012). Further work is needed to develop more robust, objective measures of facial adiposity. Fourth, although our analyses reveal individual differences in preferences for facial correlates of BMI, it is unclear whether these individual differences reflect variation in motivation to obtain mates with low BMI or variation in motivation to obtain mates who, for example, lead particular lifestyles or possess particular hormonal profiles that are correlated with BMI. Studies investigating the determinants of individual differences in preferences for cues of BMI may clarify this issue (see, e.g., Fisher et al., 2013).

This study directly tested the contribution of assortative preferences for cues of BMI in opposite-sex faces to assortative mating for adiposity in a sample of romantic couples. Analyses suggested that individual differences in preferences for facial cues of BMI contribute little (if at all) to assortative mating for BMI. However, both men's and women's preferences for facial cues of BMI were positively correlated with their actual romantic partners' BMIs. Thus, our data potentially not only implicate preferences for cues of BMI in partner choices but also show that partner choice is not redundant with face preferences, at least with regard to preferences for cues of BMI. Indeed, some differences between preferences for cues of BMI and actual mate choices would be expected, given that mate choices are likely to be constrained in ways that mate preferences are not. Importantly, the causal direction of the relationship between preferences for facial cues of BMI and partner BMI is unclear. One possibility is that preferences for cues of adiposity directly contribute to partner selection (i.e., preferring cues associated with lower BMI causes people to choose leaner partners). Another possibility is that having a leaner partner causes people to prefer cues of lower levels of facial adiposity. For example, people may realign their preferences to match partner characteristics to reduce cognitive dissonance and experiments have demonstrated that increasing participants' recent visual experience with images of the bodies of individuals with higher levels of adiposity increases their preferences for facial characteristics that are correlated with higher BMIs (Re et al., 2011). That these possibilities are by no means mutually exclusive may have important implications for the mechanisms and processes through which individuals' mate preferences develop. For example, people typically have more than one romantic partner in the course of their lives (e.g., Brown & Sinclair, 1999). Consequently, if current partner choice influences preferences, these preferences may influence future partner choice, establishing a feedback loop that amplifies the effects of early mate choices on partner choice later in life. Consistent with this proposal, recent work has found that characteristics of participants' sexual experiences that occurred early in adulthood predicted their physical and emotional satisfaction with their current sexual interactions, even when the effects of global sexual satisfaction were controlled for statistically (Smith & Shaffer, 2013). Although work on the development of human mate preferences has typically focused on experiences in early life (e.g., imprinting-like effects in childhood, Perrett et al., 2002), the possibility that early mate choices are another important factor for the development of mate preferences has received relatively little attention. We suggest that studies directly testing the role of previous mate choices in future mate choice and mate preferences may provide important insights into the ontogeny of mating behaviour.

Acknowledgments

Parts of this research were funded by ESRC grant ES/1031022/1 awarded to LMD and BCJ, and by ERC Starting Grant 282655 (OCMATE) awarded to BCJ.

Footnotes

One exemplar in the sample of women's faces had particularly high BMI (38.4 kg/m2) and was identified as a potential outlier who could unduly influence preference scores. However, men's BMI preference scores when this exemplar was included and excluded were highly correlated (r = .80, N = 62, p < .001). More importantly, the patterns of results in subsequent analyses were identical when men's BMI preference scores were calculated with this exemplar excluded. Consequently, we have not excluded this exemplar from the main analyses reported here.

References

  1. Bajos N, Wellings K, Laborde C. Moreau C. Sexuality and obesity, a gender perspective: Results from French national random probability survey of sexual behaviors. British Medical Journal. 2010;340:c2573. doi: 10.1136/bmj.c2573. doi: 10.1136/bmj.c2573. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Bray G. Definition, measurement, and classification of the syndromes of obesity. International Journal of Obesity. 1978;2:99–112. [PubMed] [Google Scholar]
  3. Brown NR. Sinclair RC. Estimating number of lifetime sexual partners: Men and women do it differently. Journal of Sex Research. 1999;36:292–297. doi: 10.1080/00224499909551999. [Google Scholar]
  4. Coetzee V, Chen J, Perrett DI. Stephen ID. Deciphering faces: Quantifiable visual cues to weight. Perception. 2010;39:51–61. doi: 10.1068/p6560. doi: 10.1068/p6560. [DOI] [PubMed] [Google Scholar]
  5. Coetzee V, Perrett DI. Stephen ID. Facial adiposity: A cue to health? Perception. 2009;38:1700–1711. doi: 10.1068/p6423. doi: 10.1068/p6423. [DOI] [PubMed] [Google Scholar]
  6. Confer JC, Perilloux C. Buss DM. More than just a pretty face: Men's priority shifts towards bodily attractiveness in short-term versus long-term mating contexts. Evolution and Human Behavior. 2010;31:348–353. doi: 10.1016/j.evolhumbehav.2010.04.002. [Google Scholar]
  7. Courtiol A, Picq S, Godelle B, Raymond M. Ferdy J-B. From preferred to actual mate characteristics: The case of human body shape. PLoS One. 2010;5:e13010. doi: 10.1371/journal.pone.0013010. doi: 10.1371/journal.pone.0013010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Currie TE. Little AC. The relative importance of the face and body in judgments of human physical attractiveness. Evolution and Human Behavior. 2009;30:409–416. doi: 10.1016/j.evolhumbehav.2009.06.005. [Google Scholar]
  9. DeBruine LM, Fincher CL, Watkins CD, Little AC. Jones BC. Preference versus choice: Do face preferences predict actual partner choice? New Mexico: Human Behavior and Evolution Society Conference in Albuquerque; 2012. [Google Scholar]
  10. Di Castelnuovo A, Quacquaruccio G, Donati MB, De Gaetano G. Iacoviello L. Spousal concordance for major coronary risk factors: A systematic review and meta-analysis. American Journal of Epidemiology. 2009;169:1–8. doi: 10.1093/aje/kwn234. doi: 10.1093/aje/kwn234. [DOI] [PubMed] [Google Scholar]
  11. Dion K, Berscheid E. Walster E. What is beautiful is good. Journal of Personality and Social Psychology. 1972;24:285–290. doi: 10.1037/h0033731. [DOI] [PubMed] [Google Scholar]
  12. Fisher CI, Fincher CL, Hahn AC, DeBruine LM. Jones BC. Individual differences in pathogen disgust predict men's, but not women's, preferences for facial cues of weight. Personality and Individual Differences. 2013;55:860–863. doi: 10.1016/j.paid.2013.07.015. [Google Scholar]
  13. Gangestad SW. Thornhill R. Human oestrus. Proceedings of the Royal Society of London B. 2008;275:991–1000. doi: 10.1098/rspb.2007.1425. doi: 10.1098/rspb.2007.1425. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Ginsburg E, Livshits G, Yakovenko K. Kobyliansky E. Genetics of human body size and shape: Evidence for an oligogenic control of adiposity. Annals of Human Biology. 1999;26:79–87. doi: 10.1080/030144699282994. doi: 10.1080/030144699282994. [DOI] [PubMed] [Google Scholar]
  15. Han TS, Morrison CE. Lean ME. Age and health indications assessed by silhouette photographs. European Journal of Clinical Nutrition. 1999;53:606–611. doi: 10.1038/sj.ejcn.1600816. doi: 10.1038/sj.ejcn.1600816. [DOI] [PubMed] [Google Scholar]
  16. Havlicek J. Roberts C. MHC-correlated mate choice in humans: A review. Psychoneuroendocrinology. 2009;34:497–512. doi: 10.1016/j.psyneuen.2008.10.007. doi: 10.1016/j.psyneuen.2008.10.007. [DOI] [PubMed] [Google Scholar]
  17. Hebebrand J, Wulftange H, Goerg T, Ziegler A, Hinney A, Barth N. Remschmidt H. Epidemic obesity: Are genetic factors involved via increased rates of assortative mating? International Journal of Obesity and Related Metabolic Disorders. 2000;24:345–353. doi: 10.1038/sj.ijo.0801135. doi: 10.1038/sj.ijo.0801135. [DOI] [PubMed] [Google Scholar]
  18. Hume DK. Montgomerie R. Facial attractiveness signals different aspects of “quality” in women and men. Evolution and Human Behavior. 2001;22:93–112. doi: 10.1016/s1090-5138(00)00065-9. doi: 10.1016/S1090-5138(00)00065-9. [DOI] [PubMed] [Google Scholar]
  19. Jacobson P, Torgerson JS, Sjostrom L. Bouchard C. Spouse resemblance in body mass index: Effects on adult obesity prevalence in the offspring generation. American Journal of Epidemiology. 2007;165:101–108. doi: 10.1093/aje/kwj342. doi: 10.1093/aje/kwj342. [DOI] [PubMed] [Google Scholar]
  20. Little AC, Jones BC. DeBruine LM. Facial attractiveness: Evolutionary based research. Philosophical Transactions of the Royal Society B. 2011;366:1638–1659. doi: 10.1098/rstb.2010.0404. doi: 10.1098/rstb.2010.0404. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Little AC. Mannion H. Viewing attractive or unattractive same-sex individuals changes self-rated attractiveness and face preferences in women. Animal Behaviour. 2006;72:981–987. doi: 10.1016/j.anbehav.2006.01.026. [Google Scholar]
  22. Møller AP. Sexual selection and the barn swallow. New York, NY: Oxford University Press; 1994. [Google Scholar]
  23. Moore ME, Stunkard A. Srole L. Obesity, social class, and mental illness. Obesity Research. 1997;5:503–508. doi: 10.1002/j.1550-8528.1997.tb00678.x. doi: 10.1002/j.1550-8528.1997.tb00678.x. [DOI] [PubMed] [Google Scholar]
  24. Pasco JA, Nicholson GC, Brennan SL. Kotowicz MA. Prevalence of obesity and the relationship between the body mass index and body fat: Cross-sectional population-based data. PLoS One. 2012;7:e29580. doi: 10.1371/journal.pone.0029580. doi: 10.1371/journal.pone.0029580. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Penton-Voak IS. In retreat from nature? Successes and concerns in Darwinian approaches to facial attractiveness. Journal of Evolutionary Psychology. 2011;1:1–21. doi: 10.1556/JEP.9.2011.2.5. [Google Scholar]
  26. Perrett D, Penton-Voak IS, Little AC, Tiddeman BP, Burt DM, Schmidt N. Barrett L. Facial attractiveness judgements reflect learning of parental age characteristics. Proceedings of the Royal Society of London B. 2002;269:873–880. doi: 10.1098/rspb.2002.1971. doi: 10.1098/rspb.2002.1971. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Peters M, Rhodes G. Simmons LW. Contributions of the face and body to overall attractiveness. Animal Behaviour. 2007;73:937–942. doi: 10.1016/j.anbehav.2006.07.012. [Google Scholar]
  28. Price ME, Pound N, Dunn J, Hopkins S. Kang J. Body shape preferences: Associations with rater body shape and sociosexuality. PLoS One. 2013;8:e52532. doi: 10.1371/journal.pone.0052532. doi: 10.1371/journal.pone.0052532. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Puts DA, Jones BC. DeBruine LM. Sexual selection on human faces and voices. Annual Review of Sex Research. 2012;49:227–243. doi: 10.1080/00224499.2012.658924. doi: 10.1080/00224499.2012.658924. [DOI] [PubMed] [Google Scholar]
  30. Ramlau-Hansen CH, Thulstrup AM, Nohr EA, Bonde JP, Sørensen TIA. Olsen J. Subfecundity in overweight and obese couples. Human Reproduction. 2007;22:1634–1637. doi: 10.1093/humrep/dem035. doi: 10.1093/humrep/dem035. [DOI] [PubMed] [Google Scholar]
  31. Rantala MJ, Coetzee V, Moore FR, Skrinda I, Kecko S, Krama T. Krams I. Adiposity, compared with masculinity, serves as a more valid cue to immunocompetence in human mate choice. Proceedings of the Royal Society of London B. 2013;280:20122495. doi: 10.1098/rspb.2012.2495. doi: 10.1098/rspb.2012.2495. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Re DE, Coetzee V, Xiao D, Buls D, Tiddeman BP, Boothroyd LG. Perrett DI. Viewing heavy bodies enhances preferences for facial adiposity. Journal of Evolutionary Psychology. 2011;9:295–308. doi: 10.1556/JEP.9.2011.4.2. [Google Scholar]
  33. Reither EN, Hauser RM. Swallen KC. Predicting adult health and mortality from adolescent facial characteristics in yearbook photographs. Demography. 2009;46:27–41. doi: 10.1353/dem.0.0037. doi: 10.1353/dem.0.0037. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Rhodes G. The evolutionary psychology of facial beauty. Annual Review of Psychology. 2006;57:199–226. doi: 10.1146/annurev.psych.57.102904.190208. doi: 10.1146/annurev.psych.57.102904.190208. [DOI] [PubMed] [Google Scholar]
  35. Schneider TM, Hecht H. Carbon C-C. Judging body weight from faces: The height-weight illusion. Perception. 2012;41:12–124. doi: 10.1068/p7140. doi: 10.1068/p7140. [DOI] [PubMed] [Google Scholar]
  36. Schwartz CR. Mare RD. Trends in educational assortative marriage from 1940 to 2003. Demography. 2005;42:621–646. doi: 10.1353/dem.2005.0036. doi: 10.1353/dem.2005.0036. [DOI] [PubMed] [Google Scholar]
  37. Silventoinen K, Kaprio J, Lahelma E, Viken RJ. Rose RJ. Assortative mating by body height and BMI: Finnish twins and their spouses. American Journal of Human Biology. 2003;15:620–627. doi: 10.1002/ajhb.10183. doi: 10.1002/ajhb.10183. [DOI] [PubMed] [Google Scholar]
  38. Smith CV. Shaffer MJ. Gone but not forgotten: Virginity loss and current sexual satisfaction. Journal of Sex and Marital Therapy. 2013;39:96–111. doi: 10.1080/0092623X.2012.675023. doi: 10.1080/0092623X.2012.675023. [DOI] [PubMed] [Google Scholar]
  39. Smits J. Social closure among the higher educated: Trends in educational homogamy in 55 countries. Social Science Research. 2003;32:251–277. doi: 10.1016/S0049-089X(02)00049-2. [Google Scholar]
  40. Speakman JR, Djafarian K, Stewart J. Jackson DM. Assortative mating for obesity. American Journal of Clinical Nutrition. 2007;86:316–323. doi: 10.1093/ajcn/86.2.316. [DOI] [PubMed] [Google Scholar]
  41. Speliotes EK, Willer CJ, Berndt SI, Monda KL, Thorleifsson G, Jackson AU. Loos RJF. Association analyses of 249,796 individuals reveal eighteen new loci associated with body mass index. Nature Genetics. 2010;42:937–948. doi: 10.1038/ng.686. doi: 10.1038/ng.686. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Tinlin RM, Watkins CD, Welling LLM, DeBruine LM, Al-Dujaili EAS. Jones BC. Perceived facial adiposity conveys information about women's health. British Journal of Psychology. 2013;104:235–248. doi: 10.1111/j.2044-8295.2012.02117.x. doi: 10.1111/j.2044-8295.2012.02117.x. [DOI] [PubMed] [Google Scholar]
  43. Tovée MJ, Emery JL. Cohen-Tovée EM. The estimation of body mass and physical attractiveness is dependent on the observer's own body mass index. Proceedings of the Royal Society of London B. 2000;267:1987–1997. doi: 10.1098/rspb.2000.1240. doi: 10.1098/rspb.2000.1240. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Watson D, Klohnen EC, Casillas A, Simms EN, Haig J. Berry DS. Match makers and deal breakers: Analyses of assortative mating in newlywed couples. Journal of Personality. 2004;72:1029–1068. doi: 10.1111/j.0022-3506.2004.00289.x. doi: 10.1111/j.0022-3506.2004.00289.x. [DOI] [PubMed] [Google Scholar]
  45. World Health Organization. Obesity: Preventing and managing the global epidemic. Geneva, Switzerland: World Health Organization; 2000. [PubMed] [Google Scholar]
  46. Zietsch BP, Verweij KJH, Heath AC. Martin NG. Variation in human mate choice: Simultaneously investigating heritability, parental influence, sexual imprinting, and assortative mating. The American Naturalist. 2011;177:605–616. doi: 10.1086/659629. doi: 10.1086/659629. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from British Journal of Psychology are provided here courtesy of Wiley

RESOURCES