Differences in how people judge face attractiveness can be reliably measured
Individual face preferences are primarily explained by differences in environments
In contrast, face identity recognition is explained primarily by genetic variation
Although certain characteristics of human faces are broadly considered more attractive (e.g., symmetry, averageness), people also routinely disagree with each other on the relative attractiveness of faces. That is, to some significant degree, beauty is in the “eye of the beholder.” Here, we investigate the origins of these individual differences in face preferences using a twin design, allowing us to estimate the relative contributions of genetic and environmental variation to individual face attractiveness judgments or face preferences. We first show that individual face preferences (IP) can be reliably measured and are readily dissociable from other types of attractiveness judgments (e.g., judgments of scenes, objects). Next, we show that individual face preferences result primarily from environments that are unique to each individual. This is in striking contrast to individual differences in face identity recognition, which result primarily from variations in genes [1]. We thus complete an etiological double dissociation between two core domains of social perception (judgments of identity versus attractiveness) within the same visual stimulus (the face).
[...]
Selecting two participants at random produced an average of only 48% agreement (and 52% disagreement) in face preferences (see Figure 1F), even after removing apparent disagreements that could be explained away as self-inconsistency (Supplemental Information). This estimate is consistent with previous literature [11, 12] as well as with the everyday experience that on the one hand, fashion models can “make a fortune with their good looks,” while on the other hand, friends can “endlessly debate about who is attractive and who is not” [11].
[...]
Next, we estimated the contributions of genetic and environmental factors to face IP by comparing the correlation of face IP scores among MZ twins with the correlation of face IP scores among DZ twins. Although MZ and DZ twins share family environment to a similar extent, MZ twins share, on average, twice as much of their genetic variation as DZ twins. The correlations for face IP scores between MZ twins and between DZ twins can thus be used to estimate the proportion of variation in face IP that can be explained by variations in genes, shared environments, and unshared environments. We calculated a maximum likelihood correlation of 0.22 (95% CI: 0.14–0.29) for MZ twins and 0.09 (95% CI: −0.06–0.24) for DZ twins. These two correlations did not significantly differ (Fisher r-to-z transformation; p = 0.1), indicating that most of the variance in face IP is likely attributable to environmental factors.
https://www.cell.com/current-biology/fulltext/S0960-9822(15)01019-2Of that 52% (the "individual preference" of which faces an individual finds attractive), they are estimating it is ~1/4 determined by genetic instinct/blood memory, and ~3/4 by "environment":
Controlling for age and sex, the ACE model attributed 22% of variance to (A) additive genetic factors, 0% to (C) shared environmental factors, and 78% to (E) individual or unshared environment and/or measurement error (see Figure 2B and Table 1).
But does this environmental influence act specifically on face attractiveness judgments? Alternatively, it might act broadly on any judgment that involves a face or on any social judgment. As a strong test of specificity, we consider the case of face identity recognition. Face attractiveness judgments and face identity recognition both involve social evaluation of faces, in the visual domain. Moreover, both require processing of invariant face characteristics, which are known to rely upon inferior occipital and inferior temporal brain regions [23], and deficits in both have been found to coexist in patients [23, 24, 25]. If the etiology of face identity processing were to differ from that of face IP, then that would provide strong evidence that the observed environmental effect is specific not only to social stimuli in general, or to faces in particular, or even to judgments of invariant face characteristics, but rather to a particular subset of judgments of invariant face characteristics. ... Yet despite equal precision of measurement, a sample drawn from the same population, and similarly robust evidence for independence from various non-face categories, we found little to no impact of environment on face recognition ability. Genetic variation accounted for most or all of the reliable face recognition variance, in contrast with face IP (68% versus 22% heritability; p of difference < 1E−14; see Figure 2C and Table 1). Indeed, looking across the behavioral genetic literature, face IP is among the most environmental objectively measured behavioral traits, whereas face identity recognition is among the most heritable [1, 27].
Pay close attention to the second paragraph below. In other words, lower exposure to media or propaganda would allow innate genetic preferences to assert themselves more strongly:
Previous evidence has indicated that preferences for particular faces or face characteristics are shaped by a range of factors, including personality preferences [28], the rater’s own facial characteristics [29], features of the socioeconomic and cultural environment [30, 31, 32, 33, 34], previous visual experience [35, 36, 37, 38, 39], and history of social learning [19, 40, 41, 42, 43, 44].
[...]
Our study was conducted with a relatively homogeneous sample of Australian twins [45]. Given the sociocultural homogeneity of our sample, the low contribution of genetic variance to face IP is particularly noteworthy: estimates of genetic contributions tend to be higher where environments are less variable [46].
I.e. face preferences from peer pressure/media/propaganda exposure can easily outweigh genetically-determined preferences for faces, and change preferences from generation to generation (or even among siblings if they are exposed to different media):
Our results further establish that the important environments are individual specific; that is, they are not consistent across family members.
I am skeptical that they have definitively demonstrated that face preferences are primarily a product of "environmental conditions", since, you know:
Assortative mating is a mating pattern and a form of sexual selection in which individuals with similar phenotypes mate with one another more frequently than would be expected under a random mating pattern.
https://en.wikipedia.org/wiki/Assortative_matingBut this study's finding does suggest that Eurocentric beauty standards and preferences for high sexual dimorphism in the media would indeed inflict significant psychological damage on individuals' innate preferences!
Some additional considerations: this study only used individuals from Australia (i.e. a highly Westernized nation with high media exposure). Additionally, I focused on the potential media exposure has to influence people's preferences, but innocent things like a childhood crush could also be an environmental influence on an individual's preferences. (The study authors do not talk about what potential "environmental" influences would be). I imagine, on average, throughout all individuals, media exposure would be the most significant "environmental influence", although for some individuals, an influence from personal experiences like a crush or whatever could outweigh the preferences presented to them in the media.
Also, it would be interesting to examine the "low agreement participants" (see Figure 1E), whose face ratings highly differed from the average ratings. Do these individuals have particularly strong blood memory in other ways (e.g. are they highly tribalist or anti-tribalist?); do the physical traits of these individuals tend to better correspond to a certain face/body shape (afterall, the average person is just average and does not have an extreme correspondence to any face shape archetype)?
Notice even in the "high agreement participant" in Figure 1D, the average rating for one of the faces was 3.5, but they rated that face as 7 (which was the highest rating allowed). Analyzing outliers like that can be very insightful, but unfortunately it was overlooked.
(I think this pie chart suggests that 50% of facial attractiveness comes from things like geometry, proportion, health, etc. which would be universally considered attractive (labelled "common preference") and then 50% of facial attractiveness comes down to whether an individual finds the face attractive or not? If I'm interpreting that correctly.)
Lastly, the study doesn't seem to describe the faces used in the study (e.g. whether there were any real faces or if they were all uncanny valley computer-generated mesh faces with no hair, or those face composite images, etc.). Nor do they seem to describe what the test for recognizing "face identity" was.
Stimulus sources: Neutral face stimuli were taken from four sources: (1) the MIT face database, which Bronstad and Russell [S7] used to look at similarity in face preferences between siblings, friends, and spouses (we used all 74 of their images for the pilot, then reduced to 50 images for final version of test), (2) the Glasgow Unfamiliar Face Database (GUFD) [S8] (140 images selected for the pilot, reduced to 50 images for the final test), (3) GenHead software (140 new images generated for the pilot, reduced to 50 images for the final test), and (4) several databases including the Facial Recognition Technology (FERET) Database [S9], the NimStim Set of Facial Expressions [S10], and the Karolinska Directed Emotional Faces (KDEF) [S11] (140 images selected for the pilot, reduced to 50 images for the final test). Faces from these four sources were administered in separate blocks in both pilot study and twin study.