Could it be that the brain arrives at its ideal of facial beauty by averaging?

I think we like symmetry as an indicaton of genetic health.

Averaging a lot of pictures, asuming any a-symmetries are random, would lead to symmetric faces with the a-symmetries averaged out. I think that’s the main driver, here.

I think it is more complicated than that. A wide range of faces and features are beautiful and what is surprising is how much agreement there is on what is beautiful and what is not. Blue eyes, brown eyes, blonde hair, dark hair, light skin, dark skin can all be beautiful. A lot has to do with carriage and posture and attitude and the appearance of vibrant good health.

@whitenoise I agree that symmetry is a factor, but I doubt that you could get results this uniformly agreeable just by taking faces and digitally correcting for bilateral symmetry. I’m sure that doing that would make a face more appealing, but I don’t think that alone would push it up into this league.

Did the photographer choose as subjects the faces of women who were individually young, beautiful, and well-proportioned? I don’t see anyone there who resembles anyone I know, much less what I see in the mirror, and I strongly doubt that an average of the faces I see about in the world would fall into the “gorgeous” category. Rather, when I look at any assemblage of ordinary faces (a crowd shot, a yearbook, a musical choir), I am amazed at how many seem somehow “funny-looking.”

However, the variation among characteristic ethnic “looks” is interesting.

Here’s an example of what correcting for symmetry does.

@Jeruba I wish I knew. Let me try to find that out.

Ah, some very interesting stuff here.

It would seem that my thesis has already been tested and scuttled.

Well, I’d give you points for an interesting thesis, anyway.

I’m wondering if it’s not averageness per se that contributes to the perceived attractiveness (or even the averaging of the most attractive) but rather the tendency of averaging to offset individual disproportions by smoothing. If one face has a left eye misaligned and another a right eye, but most are better aligned than either of those, the misalignment will tend to be erased by the aggregate. Right?

I’ve just been reading up on tronies in conjunction with viewing a current exhibit at San Francisco’s De Young museum. Vermeer’s famous Girl with a Pearl Earring is considered to be a painting of this type—an idealized study of a face, intended to be more generic than an individual portrait. I wonder if these composite photos too aren’t more like tronies.

Different culture value different aspects of what beauty is.
Too bad beauty is determined by the advertising and Marketing agencies.
( Modelling?)