Of course, Todorov’s years of research showed that inferences people made based on appearances alone—about who was trustworthy, who was competent, et cetera—were grossly inaccurate. Still when we “go with our gut,” these sorts of judgments are often what that comes down to. And in their new study, Todorov and colleagues found that across cultures, these biases are inconsistent. The research team showed people hundreds of faces and asked them to judge people (in terms of trustworthiness, attractiveness, competence, et cetera). The scientists found that the true underlying factor was our own experience of normalcy. We see faces more negatively the more they deviate from a learned idea of a “typical face.”
“Although there is no ‘average’ human face, you like faces that are closer to your own definition of a typical face,” Todorov told me. “It’s like a tribal bias. I like people that look like the people around me.”
That might not be especially revelatory, but this is: The idea of a “typical face” is malleable, and rapidly so. When his experimental subjects stared at faces for just a half hour, their sense of normalcy was partially reprogrammed.
“The idea is it’s not just about who looks like you, or your lifelong learning about— say, who looks masculine or feminine, and what that means,” Todorov said, “but it’s a lot about who you’ve been around in the very recent past. The positive finding that we can overcome biases by exposure more easily than once thought.”
Exposure to different faces not only shifts what faces people perceive as typical, but also what faces they evaluate more positively. The more segregated we become, the more we’re likely to narrow our understanding of normal appearances––to appraise The Others more negatively.
The maps of where people voted for Clinton and Trump are clear. In urban areas where people live in close proximity to great diversity, the vote was overwhelmingly Democratic. Trump’s support came from monolithic non-urban areas.
If you have a diverse friend/colleague group in a multicultural city, I asked Todorov, do you have a broader set for judging normalcy?
“I would think so,” he said. “Of course, in real life, it’s much more complicated. You can live in a diverse city where there are strong divisions between groups, and that would affect how you evaluate people. But assuming that you have kind of egalitarian relationships and a big social network that's diverse, and you live in a place like New York City, yes, your notion for typicality would be very different than if you grew up and spent most of your time in a rural area that has primarily one particular race or ethnicity.”
The traditional “contact hypothesis” in sociology is that if you put people together from different social and economic backgrounds, they get to know each other, and they rely on each other, and unity follows. Classically that came with qualifying conditions: You can put people together, but they have to be equal status. And they have to be dependent on each other, as in the army. When one group has more than the other, stereotypes are more likely to be confirmed than disconfirmed. (This has recently been disputed, with some experts arguing that contact alone is sufficient.)