The Physical Attributes of a Beautiful Face
May 17, 2003 10:02 PM   Subscribe

The first time ever I saw your face: Is beauty perhaps not entirely in the eye of the beholder? [Via LinkFilter.]
posted by Carlos Quevedo (25 comments total)
Fascinating stuff, Carlos, thanks! The entire site is well worth reading.

What we regard as 'beautiful' varies a good bit from culture to culture, and over time: big eyes, small eyes; big noses, small noses; plump or thin. The only real constants I know of are symmetry and youth, both of which always seem to be considered beautiful.

The site's authors have cheated a little, in comparing Miss Germany to their idealized computer model: all the contestants here seem to have been photographed in very harsh light with a bluish or greenish tint, with the light coming from the woman's lower right. This is guaranteed to make anyone look terrible.
posted by Slithy_Tove at 10:51 PM on May 17, 2003

seems a little thin - at best.
posted by specialk420 at 11:37 PM on May 17, 2003

Interesting to see how Betty Crocker has changed through the years, culminating in her 1996 visage (adding a hint of indistinct ethnicity). Frankly, Betty-1986 was the one who made my soufflé rise ... (Yeah, she’s no Sandra Bullock, but boy can she cook!)
posted by RavinDave at 11:57 PM on May 17, 2003

This study seems like the most convoluted collection of pseudoscience i've read in recent memory. I remember seeing an episode of "Understanding" on the Science Channel that really put forth a more believable foundation as to the hierarchy of beauty. It also covered what many mathematicians, musicians and architects consider the real basis for aesthetic perfection: the golden section, also known as phi.

RavinDave: Agreed - the 1986 Betty is by far the most attractive. But was that really the most important design factor -- to make a spokesperson that guys would consider hot [at least mildly?]
posted by phylum sinter at 1:23 AM on May 18, 2003

My understanding is that all the recent Crockers have all been made from composites of several models (thus, the reason it sprang to mind in the context of this thread). I'm not sure I'd call Betty-86 "hot" per se -- but the image does strike an odd chord in me I can't quite articulate. It does look oddly familiar -- and I think that is at the whole heart of "what is beauty" and why it is indeed in the subjective eye of the beholder.
posted by RavinDave at 1:33 AM on May 18, 2003

why is it pseudoscience? it's just some kind of a survey or whatever. they're just trying to measure people's attitudes and suggest some kind of conclusion. dismissing it as "pseudoscience" is suggesting it's quackery. but the only difference between this an other scientific studies is that they are measuring people's opinions rather than something quantifiable. it could be seen as a psychologically-based experiement, investigating the factors which contribute to a person's conception of an attractive face. though i think they should go more deeply into why particular features might be thought of as attractive, in terms of psychology. one of the statements they make about people trying to improve their appearance with plastic surgery does sound rather like one of jerry springer's final thoughts.
posted by mokey at 1:43 AM on May 18, 2003

First problem: ethnicity. These women look to me as though they were all Mediterranean, maybe Spanish, maybe Italian, though some of the blondes look English. What are the similarities of a beautiful Asian face to a beautiful Caucasian face? To a beautiful Negro face? Now what about heavy-featured faces, like those of Australian Aboriginals, Samoans, or New Guineans? Are these faces, objectively ugly?

Second problem: species. It is entirely possible to rank, say, cats' faces on a scale of beauty. Same goes for dogs, rats, horses; even one crocodile can be judged prettier than another. What characteristics does a pretty dog have in common with a pretty human?

Third problem: reality. Faces which have never existed, and could not exist, faces which are obviously cartoons as opposed to computer-generated or artist-drawn nearly-real, can be ranked for beauty. Consider the "Archie" comic characters: I always thought Betty was a bit hotter than Veronica, for instance, and either one is objectively better looking than Mrs Grundy. Of the big superhero comic publishers, Image has far and away the prettiest characters, especially the females; their artistic style is apparently designed to emphasise physical beauty.

So, I think it's like trying to define "good taste", in the food sense. Pineapples taste nothing like prawn curry, but some people would rate both as very tasty.
posted by aeschenkarnos at 1:59 AM on May 18, 2003

aeschenkarnos: those aren't problems, they're just areas for further study.
posted by Slithy_Tove at 2:15 AM on May 18, 2003

I know I heard or read somewhere that the 'symmetry as a determinant of judging beauty' claim had been disproved, or at least the study that was the basis for it had some serious errors, but damned if I can find the source now.

Though I think symmetry does express itself very clearly in nature, so I personally think there is a link.
posted by lychee at 4:37 AM on May 18, 2003

Awesome link! I remember doing some light research into the subject back when the combined images were fuzzy, black and white messes. It's nice to see how far things have come!
posted by oissubke at 5:49 AM on May 18, 2003

I think the conclusions are half-assed at best, even if we're just talking about physical attractiveness here. Sure there are features that seem to be nearly-universally attractive but if you look at the prototypical sexy faces, you'll notice that while pleasant they're kind of bland. What truly makes for an alluring face is Yes attractiveness, but combined with uniqueness and character. I think I once said to someone that when a girl calls you "cute" as opposed to "handsome" she means pleasingly odd. At the time it used to irk me a bit, but now I dig it a bit more.
posted by jonmc at 8:05 AM on May 18, 2003

The "most beautiful" faces on this site are just bland. Not unattractive, not particularly pretty...just blah.
posted by Su at 8:09 AM on May 18, 2003

I found the "most beautiful" faces quite beautiful indeed, myself.

Interesting to note that when comparing the "sexy face" guy and the "unsexy face" guy, I realized I'd be more willing to date the "unsexy face" guy - he's actually pretty cute in my estimation. The "sexy face" guy just looks like he would be too arrogant to me. Heh.
posted by beth at 9:02 AM on May 18, 2003

This is fascinating. Thanks Carlos.
posted by plep at 9:12 AM on May 18, 2003

The beautiful faces are pleasing - but they did look fake to me even before I realized they were computer generated.

This site is good as an analysis of what most people find beautiful in a photographed face, but it doesn't begin to touch on the subject of what beauty really is.
posted by orange swan at 9:21 AM on May 18, 2003

nice to see aryan studies coming along nicely in the fatherland. perhaps they can figure out scientifically exactly why facial features such as the "hook nose" are...

etc. etc. etc.

this sort of "science" is repellant. i bet they rate their own children on their little beauty scale, but pride themselves on not letting it affect how they treat them.
posted by mitchel at 10:45 AM on May 18, 2003

Can the histrioinics. The average is clearly attractive, just like McDonald's fries are clearly tasty. It hits a platonic ideal of beauty, but falls short of what makes real, living and breathing people especially attractive to each other. So the average can reach the cosmopolitan airbrushed ideal, but it really can't go any further.

But what would expect of a concept of average attractiveness? These are crowd-pleasers, faces that none of us can argue are horrendously ugly, but which may not spark any special fire in any one of us. It's an aesthetic purgatory, like muzak.
posted by condour75 at 11:30 AM on May 18, 2003

Is beauty perhaps not entirely in the eye of the beholder?

That's exactly what studies like this are measuring, though -- you're looking at an average composed of input from some number of beholders. Even if intersubjective agreement were really high, this is still just a description of an average subjective preference, with no mention (that I saw during my brief review) of how much variation is observed around that average. And outliers on that continuum are no more or less "right" with regard to what they find attractive.

Interesting link, Carlos! :)
posted by boredomjockey at 12:26 PM on May 18, 2003

this sort of "science" is repellant.

Yeah, i hate it when we learn more about how we think.
posted by kindall at 2:02 PM on May 18, 2003

ok, maybe i was totally overreacting. but i looked at the pages where they compare their perfect fakes against their ugly fakes, and the uglies [the men especially] looked distinctly more east european than the perfs. that was just my first take on it, maybe i'm way off. but that's kind of beside the point, although it disturbs me at least [although this could be my problem not theirs].

i grant you that there is likely to be a certain amount of overlap from culture to culture as far as what type of person looks beautiful. but a study like this is just crap, crap, crap. total pseudoscience: simply a pile of meaningless correlation.

does it look at how people actually act? does it note whether people of these traits are more successful in the mating game and thus eventually we will all look like the perfect ones? do they just average out scores or what?

i mean, take skin color. if you ask 100,000 people about skin color, your average preference will be somewhere between gold and brown, because the people who like black and white cancel. how bout nose shapes. they prefer thin noses, apparently. why is that? is there some sort of hypothesis getting answered here? nose thinness correlates, in their view, with healthy bodies, which correlate with attractiveness to a broad swath of their "test" subjects? or is this just a survey? it's not science at all. do they really think they can learn something by trying to isolate similarities of the Miss-wahlen contestants?

or how about taking the same picture, making it more babylike to test your hypothesis [a very old hypothesis, at least back to sanger or somebody like that]. now, is there double-blind as they tinker with the images, to make sure they don't decide it's more babylike once they think it's more attractive? or is some guy sitting there, thinking, "that's pretty good, pretty babyface-ish" and at the same time finding it more attractive?

not that even double-blinding that would've kept that "test" of their hypothesis from being hopelessly polluted by bias.

ask yourself what rubens or anyone else in his century would've thought about their findings. doesn't that imply some random drift, here? of course it does. we've had similar drifts in what we think are sexy clothes, sexy sounds, sexy dancing styles, etc.

ask yourself what these guys could do if they wanted to figure out what kind of suit looks good on a man. i bet they could do a pretty good job, all right! they could use their sophisticated computers to generate a completely non-existent, composite suit; show examples of how really expensive suits resemble that computer generated perfect suit, etc.

would that tell us anything at all about anything?

now then think about what their site looks like - does it look like it's trying to present a scientific finding, or does it look oh, perhaps, like it's trying to convince you of something? it's pretty slick isn't it? what about all that press? the first article is from a popular science mag, then it's mostly newspaper style sections. finally we come to the links page, guess where some of those take you. plastic surgery, hm, who could be bankrolling this stuff?

now i could be way way off in that last para. obviously these kids are getting technical support from some computer companies too, some helpful people who want to show off their composite imagery software. but for goodness' sake, the whole thing is just completely ridiculous. could you do a study of which voices in commercials are the most appealing? of course you could. would that tell you something about anything? no. but would somebody try to concoct some hogwash about it? of course they would: "the male voice that is most attractive is centered on G below C, has rich vibrato because people with rich vibrato have healthy larynxes..." etc.

i'm totally serious. not science. panders to magazines. crap. just my off-the-cuff opinion! [god, isn't it strange how a silly internet rant every now and then can be so relaxing].
posted by mitchel at 4:22 PM on May 18, 2003

Mitchel - you're first point could be right on the mark - as you can see from this explanation of face averaging, the technique starts with a sample of faces that has been rated on some quality by a group of people. If the faces you show are all white then your end result will be white. Similarly, if the raters are all of one racial group and they prefer people of their own race, that may be reflected in the scores. Incidentally, this site came up on Metatalk last month.
posted by jamespake at 5:09 PM on May 18, 2003

Yeah, but is it Science?
posted by dgaicun at 7:01 PM on May 18, 2003

Word to mitchel. You've crystallized my opinions precisely. It's junk science like this that frustrated me out of a career in psychology.
posted by moonbiter at 7:31 PM on May 18, 2003

Ask some women, they knew about that (certain looks are perceived as being better then others) from day zero.

Take a look at this well know stars without makeup pictures.

Also as Slithy_Tove pointed out, light is extremely important when taking a photograph and can make
or destroy a photo, as seen in above site.

The interesting point about the article is , I think, that could make many realize beauties as exposed on glossy covers and in skillfully retouched movies and photos do not exist in reality as often as one would think.
posted by elpapacito at 7:49 PM on May 18, 2003

Apparently I'm not the only one who's of the opinion that this isn't science at all [and as a human who, naturally, fears being the minority (sic) this is a relief]. Put me in the same boat as mitchel and moonbiter. Well said, mitchel.

Honestly, I think this sort of study really only quantifies our desire to build systems out of everything - the human mind is basically a pattern/statistic generator that often goes overboard and builds models that have little coherence in "reality". More of a philosophical statement, but it's the bulk of my reasoning here.

So to reiterate: Nothing is proven, everything is permitted.
posted by phylum sinter at 3:45 AM on May 19, 2003

« Older The Shallowing of American Taste   |   The Daguerreian Society Newer »

This thread has been archived and is closed to new comments