"Own-race bias" revisited: Do other races really look more alike?
Recently in your on-line archive I read your column on why white people don't look more alike, which interested me from multiple angles. The first stems from the experiences of people with Asperger's syndrome, a neurological disorder which has "face blindness" as one of its symptoms (and which I myself have). Face blindness refers to both difficulty in reading faces and in recognizing people by their faces. In other words, most people look alike to us. It has taken me years of concerted practice to be able to distinguish people by their facial features, and a lot of "Aspies" are never able to do so. I can point out from my own and others' experience that it is no easier to tell white people apart by nonfacial characteristics than people of other colors. By "nonfacial characteristics" I mean hairstyle and color, skin color, height and body shape, style of dress, places one habitually is found, and all the other things that are so much easier to tell apart than facial features.
The second point re: white people not looking alike comes from experiments in social psychology. [The writer quotes a social psych text that in turn cites studies suggesting that people of other races invariably look more alike than people of one's own race, regardless of what race we're talking about.] I'm appending the journal references.
Hm. That archived column really addressed two questions: the one that was asked (why do white people display more physical variety than other races?) and the one that was in the back of everyone's mind (why do all those people look alike?). I answered question number one, in the process challenging but not denying its premise, namely that white people do show greater variation in certain superficial traits — hair and eye color, for instance. No doubt this led those who felt question number two was the real issue to conclude: You racist pig! The fact that you would even think that members of other races look more alike proves you can't transcend your racial conditioning, etc. You sees the box this puts us in, debatewise. We plunge on nonetheless.
Let's begin by conceding that conditioning has a lot to do with it. Following up on the cites you helpfully supplied, Jennifer, we learn that most studies show that: (a) whites have an easier time recognizing white people than black people; (b) blacks have an easier time recognizing black people than white people; and (c) blacks and whites have an equally hard time recognizing Japanese people. The near universality of "own-race bias" suggests that the perception of difference has more to do with the observers than the observed — i.e., it's all in your head, and no one race actually demonstrates more facial uniformity than any other.
But come on. That's what we're supposed to think. Even once we grant that own-race bias exists, how can we assume that all races have exactly the same range of facial variation? Setting aside the black/white issue for the moment, surely Asians show more uniformity than Europeans and Africans, no?
Judging from the research, no. One study (A.G. Goldstein, "Race-related Variation of Facial Features," 1979) found that Japanese women's faces show more variation than those of, say, Irish men.
But wait a minute. Studying Goldstein's paper more carefully, we find that while he compared a boatload of facial features — head circumference, forehead height, "transverse nose protrusion," and so on — he omitted some much more obvious (albeit ephemeral) characteristics, such as hair color, style, and length and skin tone. Another social scientist makes much the same point: "[Facial measurements] may not be relevant for perceptual discrimination, and hair colour and hairstyle may be used more readily as discriminating cues" (J.W. Shepherd, "Social Factors in Face Recognition," 1981). Duh.
So where does that leave us? While I despair of distilling 30 years of complex and often contradictory research into a sentence or two, there's reason to believe that both subjective and objective factors figure into the widespread impression that people of other races look alike. One study found that, when describing faces, Europeans tend to mention hair color, length, and texture as well as eye color — characteristics in which Europeans show wide variation — while Africans single out hair position, eye size, and the appearance of the eyebrows, chin, and ears. In other words, when identifying faces, we tend to look for the features in which our own race shows the most variation — which means that people of a race that shows less variation in those features can be hard for us to tell apart. Confused? It's a confusing business. The point is this: Racial equality is poorly served by insisting that differences between groups don't exist. Even a seemingly elementary matter like identifying faces involves a subtle interplay between what we see and what we perceive.