We recognize less attractive faces best | Hacker News Reader

We recognize less attractive faces best

sciencedaily.com

17 pointsjamesbritt23 comments

I wonder if, rather than storing an "image" of a face, we store a sort of error correction from a standard model. This is akin to probabilistic compression, where you have a bitstream generator and then store a list of where that bitstream is in error.

Nobody knows how brain works exactly, but I believe that some sort of compression is there. This has been done in computer vision some time ago and there is a nice application which essentially represents a face by 100 dimensional vector: www.facegen.com. Starting form average face, its possible to get any face you like by varying those dimensions.

"Beautiful people all look alike. Each ugly person is ugly in his own way."

- Tolstoy

For the record, that's not actually what Tolstoy said. He actually said, “All happy families are alike; each unhappy family is unhappy in its own way.”

I only post this because some people who don't read very much might propagate the prior non-quote and allow it to assume an authority through repetition that it doesn't have.

For the record, that's not actually what Tolstoy said either. He actually wrote, «Всѣ счастливыя семьи похожи другъ на друга, каждая несчастливая семья несчастлива по своему».

To be very pedantic, the quote is what you'd quoted. In more modern russian it would be: "Все счастливые семьи похожи друг на друга, каждая несчастливая семья несчастлива по-своему."

I happily stand corrected. :)

I was going to accuse you of being a pedantic, Captian Obvious, but I'll let this one pass...this time.

I only posted because if no one objects, non-quotations often take on a life of their own.

You didn't let it pass, though... You were mean to someone who

1) did something good 2) seems to care what you think and is defensive

Shame on you.

I was just making a little joke. I sincerely didn't intend to be mean.

Interesting contribution to psychology, since unknown faces are usually considered most attractive when they are average but...

If you consider a small (or not so small) social group with a common knowledge of a person considered as attractive, then you may find that people attribute attractiveness with the _non-average_ merits displayed by that person - that contradicts the theory that attractiveness with be _just_ average!

It is also sound to say that recognizable people that evoke a positive reaction may feed the "attractiveness" recognition loop stronger...

BTW Which bright statistician discovers what is flawed in both the prevalent "averaging" theory and the research mentioned in a link?

Hint: look for difference between "alien" or "unknown" faces, and faces of those that are known - and thus whose attractiveness matters more to us.

Interesting contribution to psychology, since unknown faces are usually considered most attractive when they are average but...

My understanding it's that the average of various traits that tend to be found attractive, not "average faces." Faces that are average in every measured trait are in fact very rare.

So that's why everyone seems to recognize me.

That makes sense though. The more closely faces represent the golden ratio, the more attractive they're judged to be. Therefore, the more similar they are in that aspect, the more attractive they are. So there are a lot more ways to be ugly that there are to be attractive.

That's interesting because familiarity also increases attractiveness, which is why 'unattractive' people (think actors) might be judged somewhat attractive despite their unattractive features...which in turn make them more memorable, thus more attractive.

> The more closely faces represent the golden ratio, the more attractive they're judged to be.

Citation needed. The number of bogus references to the aesthetic power of the Golden Ratio is astonishing, most of them turn out to be false.

Online searches reveal the some articles claiming a relation between facial beauty and the Golden Ratio do so by simply redefining the Golden Ratio to equal any number that test subjects choose as most aesthetically pleasing:

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2814183/

Quote: "We found that although different faces have varying attractiveness, individual attractiveness is optimized when the face’s vertical distance between the eyes and the mouth is approximately 36% of its length, and the horizontal distance between the eyes is approximately 46% of the face’s width. These “new” golden ratios match those of an average face."

Ah -- "new" Golden Ratios. How convenient, and how utterly unlike the real Golden Ratio, 1.618.... I caution my readers that the authors of the linked paper are social psychologists, a field with only a superficial connection to the world of science.

Watch out for the fibonacci flim-flam http://www.lhup.edu/~dsimanek/pseudo/fibonacc.htm

Less attractive faces carry more information (visually) because they are not as symmetric, have distinguishing blemishes, malformed shapes, or whatever, and therefore trigger more neural pathways than the "average" face does, and therefore makes them easier to recall. Each abnormal aspect of a face triggers a neural pathway to remember that attribute, and the worst faces therefore trigger the most pathways. And each pathway improves the statistical probability of a future recognition (recall).

Seems to check out, at least anecdotally. I seem to find that extremely good looking people are initially very hard to identify. For example if I had just met such a person briefly, I wouldn't be able to spot them in a crowd 5 minutes later. I've always wondered by this happens.

This article goes against my own collected empirical evidence. I find that I have a much easier time recalling pretty faces - and especially those of the opposite sex. I think, from an evolutionary perspective at least, that this makes sense.

In day to day life, the most attractive faces tend to be rare. They're memorable simply for being attractive, meaning the face itself doesn't have to be memorized.

True story: I almost never recognise Madonna when I see her on TV or on a poster.

Frankly I find it hard to believe based on empirical data.

there goes my hope of making a career in crime :/

Reply on news.ycombinator.com