A description of the images above made by Harvard psychologist Richard Russell, who won third prize in the 5th Annual Illusion of the Year Contest for them:
In the Illusion of Sex, two faces are perceived as male and female. However, both faces are actually versions of the same androgynous face. One face was created by increasing the contrast of the androgynous face, while the other face was created by decreasing the contrast. The face with more contrast is perceived as female, while the face with less contrast is perceived as male. The Illusion of Sex demonstrates that contrast is an important cue for perceiving the sex of a face, with greater contrast appearing feminine, and lesser contrast appearing masculine.
I found the following explanation much more useful and interesting though:
What you’re looking at isn’t an optical illusion, but is a play on the basic expected traits of men and women’s faces. The flusher lips of the left pic coincide with our expectations for women’s faces, as does the fairer skin. And it’s not just the illusion of lipstick; even without lipstick, we expect women’s lips to be more red than men’s. The difference in skin tone also brings to mind a recent a study suggesting that, on the whole, men’s faces are more red complected, while women’s are more green. Thus, even in the B&W photo, we infer that the darker complected face has the deeper reddish tone of masculinity; the lighter, the paler, greenish tone of femininity.
Obviously there’s much that’s debatable in that, especially whether those “expected traits” are universal or culturally-determined, but in the meantime I can’t deny that contrast is an important cue for determining the sex of a face, and that this provides more evidence for Korean women’s mania for lightening their skins being influenced by much more than merely wanting to emulate the wealth and sophistication represented by Caucasians.
Update) There is an 11-page PDF about these images available here, and you can find out more about Richard Russel and his research interests here.
(Thanks very much to reader Nicolas for passing this on)
The Grand Narrative 
I cannot recall the links or even the publication, but I recall something about men preferring dark-complected woman because dark skin expresses health. Is there an evolutionary basis, or are we just trading cultural perspectives from the vantage of different racial groups?
LikeLike
I’d be very interested in reading whatever it is you got that from(!), but in the meantime, can you think of any evolutionary reason why dark skins would express health? None is coming to mind, nor for light skins for that matter. Either may have a role as a secondary sexual characteristic though.
LikeLike
Just curious, is the person pictured a man or a woman?
LikeLike
I recognised the faces immediately as it is a woman’s face taken from a study published recently. According to the reports, the paper (I haven’t read) showed that a woman is more attractive during ovulation. To simulate that, a photograph was modified to display the “characteristics of ovulation” on the face.
Personally, I think that this kind of studies, which seem to have multiply in recent years, is void of contents at best. Evolutionary sociology is not a science. It makes me sad that so many people misuse darwinism (which was mainly about proving natural selection, not evolution), as evolutionary sociology is none other than social darwinism in new clothing.
Some people in some countries like some kind of faces because the media widely portray this kind of faces as attractive. Media are prescriptive in terms of values. That’s a simple explanation based on culture that is more believable than any biological fantasy. It is as if people do not recognise their own lack of critical thinking and, worse, marvel at how Nature is the source of what they feel. A narcissistic trip, with a grandiose detour.
As I also think of the wide ignorance conveyed by the media about genetics (the impact of the environment being always underplayed or fully neglected), I can’t help thinking that we are living a return of ontological determinism, but a materialistic fate as opposed to the moral and spiritual fate of ancient times, and an active murder of culture as a source of freedom, i.e., greatness and mistakes altogether.
LikeLike
JB–It’s a man.
Christian–You’re entitled to your opinions, but if you want to change other people’s, then you can’t criticize something for being “void of contents at best” while providing something so vacuous yourself.
I’m quite familiar with evolutionary psychology (similar to, and with considerable overlap with evolutionary sociology), and am well aware of the tendency for proponents of both to see some contemporary psychological or social phenomenon and ipso facto see some evolutionary justification or logic for it, whereas both their perceptions of the phenomenon and of evolution are heavily influenced by their cultural, racial and/or sexual baggage, and thereby possibly completely wrong. But most researchers in the field are well aware of these problems and acknowledge them and try to take them into account, so to describe either as “social darwinism in new clothing”? I’m going to need a lot more evidence of that than just your word for it!
If you do want to critique either than I suggest being much more focused when you do so, and on something other than on a woman being much more attractive when she’s ovulating (like in this image, which I discuss in this post), which there is such a wealth of evidence for that I seriously doubt you’ve looked at much of it.
Certainly “people in some countries like some kind of faces because the media widely portrays those kind of faces as attractive,” but on the other hand Richard Russel mentions that in every racial group women’s bodies and faces as a whole have lighter skin than men’s. Dammed if I can see a “simple explanation based on culture” for that.
LikeLike
I’m nearly certain that article was an Economist Science and Technology article, but the Economist search function is lousy if one doesn’t know the article before searching for it. It did concern secondary characteristics, as I recall, which, as Christian points out broadly, is a weak point in Darwin’s work. The Descent of Man is not as rigorous as The Origin of Species. As far as I understand it, though, a Social Spencerist claim would connect a claim, like a preference for a secondary characteristic with some progressive goal, like group selection. I’m almost certain James is not making such a claim. Even if one could make an empirical claim about preferences on a species-level, it would still not be a Social Spencerist claim.
I offer this lecture to referee this discussion:
LikeLike
It wasn’t my intention to post a video, just a link. What plugin does that, James?
LikeLike
Will respond to your first comment later (at work on next post sorry!), but in the meantime wordpress.com does that automatically for linked youtube videos these days.
LikeLike
Is anyone else getting an optical illusion of the faces moving? When I look at one, I feel like the features on the other face shift. Then when I look at the other the same thing happens.
-On another note, doesn’t this picture just prove dark thick eyebrows look better on men?
LikeLike
I haven’t noticed myself sorry, nor has anyone mentioned it in the comments to all the various posts about the picture I gave.
I’d disagree that “all” the picture proves is that dark eyebrows look better on men actually, or even at all.
While I wouldn’t go so far as to say that the person on the left looks like a woman, I’d imagine that the vast majority of people – regardless of sex or ethnicity – would say that he does look much more androgynous than the distinctly masculine person on the right, thereby as a minimum pointing to a culturally and/or biologically-determined link we have between light skins and femininity, albeit not “proving” one per se.
As for eyebrows and other facial features? I haven’t read it myself yet (am about to go out shopping sorry!), but I believe Richard Russell does examine and discuss that aspect of his experiments in this PDF.
(Before I forget, when I do read it I’ll check what proportion of people agreed with the L-female, R-male contrast also, and if results differed by their sex and ethnicity or not)
LikeLike