But with respect to the sort of facial beauty which yields a physical, sexual attraction, the road may fall rather more narrowly. Here, the researchers speculate, attraction is all about a nagging desire to reproduce, and to do so in the way which will generate the most advantaged offspring. So here is where it gets interesting. After all, what exactly are the physical features most likely to yield offspring who will live to adulthood and outcompete their acquaintances in the reproduction contest? Why ought an oddly shaped nose or too much of a chin be off-putting in terms of the whole 'survival of the fittest' racket. Well it turns out that what makes somebody attractive tends to be their.... averageness.
Averageness and symmetry, indeed, go hand in hand on that score. Or, to be more exact, what tends to cause somebody to be deemed unattractive is their deviation from averageness and symmetry. Now this tends to be surprising because we often think of 'average' as a synonym for 'plain, but what is average and what is plain are not so aligned as one might think. There's an historical story here intertwined in the history of evolutionary biology. Charles Darwin, you see, was not the only scientifically minded member of family in his generation. Oh, you might think of his grandfather, good old Erasmus Darwin, but Charles had a cousin as well -- Francis Galton -- who invented 'composite photography' (that is, the laying of photographic images over one another to demonstrate a final image averaging the features of the initial ones). And in 1883, Galton set out to employ this technique in what he thought would be a crime-fighting endeavour. Specifically, Galton supposed that by taking many images of criminals and overlapping them, he could ultimately produce the image of the 'average criminal.' This, in turn, he thought would aid society in finding those with criminal features and identifying them before they engaged in nefarious deeds (all of this being in the age of phrenology, when it was believed that the bumps and knots of a person's skull provided details about their personality and characteristic tendencies). But as Galton put together more and more of these faces, he discovered the composite result to be not somebody looking more and more 'criminal,' but more and more conventionally attractive. And, indeed, further experimentation by Galton and by generations of later experimenters has repeatedly confirmed this, that if a random grouping of a few dozen or more randomly picked people from the population are image-composited, the 'average' person's face turn out a beauty.
The simple explanation of all this is that each face deviates from the average in some aspect or another, but most every faces deviates in a different way; and so, as more and more faces are composited together, any especial deviation displayed by one will be subdued by comparison to the others, which are free of that deviation. So, if one face has a very long nose, one too wide a brow, and one sunken cheekbones, the two faces with average noses will subdue the long nose, the two with average brows will subdue the wide brow, and the two with sharper cheekbones will subdue the sunken example. And indeed it goes beyond that, for a dozen faces with too-long noses composited with a dozen faces with too-short noses will yield one face with that perfectly average nose (and approaching perfect symmetry as well, for the same reasons). And in the same breath, since wrinkles and other signs of aging are not equally distributed amongst the aged, a composite of many people (even of many old people) will subdue all of these features, and end up looking youthful, smooth-skinned and vibrant. (Note that this effect is restricted by gender; mixing of genders yields androgynous faces, which are themselves a deviation from gender norms, and have generally been deemed unattractive by those surveyed).
So having gotten to that point, the question returns, what is the advantage of being most attracted to the person with the most average of faces? Well, it is always possible that any deviation from the average evidences a defect, a parting from the healthy biological norm. And so, though adhering to average faces causes the reproducing person to pass upon potentially beneficial evolutionary changes, it does so in the cause of keeping to the safest position, that of reproducing with the person least likely (based on outward appearance) to carry some harmful genetic quirk. And, just to provide an illustration to this principle, I went to faceresearch.org and generated some 'average' faces (try it yourself -- pick the twenty faces from one gender which you would find least attractive, and then see how those unattractive faces average out); and then uploaded those average specimens to morphthing.com, a site which allows you to create composite images of celebrities (and upload your own), and morphed dozens of Hollywood starlets and women similarly considered famously beautiful, and then morphed that into the other 'average' faces; then I put this in Photoshop and tinted the whole thing green for no especial reason, and got this face, for your consideration.
One of the startling discoveries man has made is that in taking a wide variety of faces -- all sorts of faces, the plain and typical and unremarkable, the quirky and unusual and memorable -- and 'morphing' them together to average out their features yields what people generally recognize as a strikingly attractive face. This is perhaps in part a consequence of the averaging out of asymmetry and the gradual elimination over every generation of morphing of every distinctive blemish, every outlier feature.
Consider this woman. She does not exist. She is instead a carefully constructed average of sixty-four different actresses and models and other famously "attractive" women. Black, White, Asian, Latina, all went into the mix. Confessedly, none of them was green (that element was simply added to further remove the generated image from common notions of "race").
