If you're a member of America's anxious middle class, you can feel downtrodden one minute and privileged the next, just watching the news. Here's some super-rich guy planning his run for President, way above you on the social ladder. Next, a story about destitute refugees, which reminds you that you're many rungs above the worst-off. Does that matter? Yes, if this study in October's Psychological Science is right. It suggests people's "emotional intelligence" is keener when they feel humble and low-class, while feeling elite makes people less emotionally alert.

Michael W. Kraus and his co-authors offer three experiments to back up their argument that upper-class people are less sensitive to others' feelings than are people of lower social class. (You can get a pdf of the whole paper here.) The most striking involved 81 college students (59 women and 22 men) who had to place themselves on a 10-rung ladder of social status. Before they did so, as Stephanie Pappas reports here, half were asked to think about a super-elite figure like Bill Gates. The other group was asked to think about someone at the very bottom of society's heap. When they were all tested for their ability to read emotions from photos, one bunch scored better: The students who'd been reminded that their social status was lower than other people's.

In another experiment, the researchers gave 200 university employees a test of emotional perceptiveness—identifying emotions in photographs—and found that those who had ended their education with high school scored 7 percent better than those with higher education. In the third study, two-person teams went through a mock job interview and then each member reported what s/he had felt, and what s/he thought the partner had. People who had rated themselves lower on the class ladder read their partners' feelings better than did those who reported themselves more privileged.

All three studies, write Kraus et al., support their hypothesis that upper-class people, having more control over their lives and less need to rely on relationships, are less empathic. Which is certainly interesting, (though the notion that poor people empathize more than the upper crust is hardly new).

But what's remarkable about the paper, I think, is the first experiment I mentioned. There, the researchers induced the phenomenon, apparently making some students more emotionally alert by manipulating how they saw themselves in the class system, right there, right then.

That result, if it holds up, would tell us that social-class effects are genuinely psychological—not driven by some hidden historic force that acts on social groups, which can only be seen by noting correlations among traits (like "high school education" and "higher empathy score"). Instead, psychologists can study the cause and effects of the phenomenon by looking at individuals in real time.

That matters because, important though they be, social class and its effects can be difficult to pin down. Researchers use readily available data—education level, income and occupation, for example—as a proxy for "socioeconomic status." But these aren't perfect. If you're a Ph.D working a low-paying job, are you upper-class or lower? If your paying work is menial but you're the all-powerful president of the PTA, where are you on the social ladder? Moreover, it can be difficult to distinguish the effects of, say, occupation, from the effects of other traits. If you're underpaid and frustrated, is it because of you're working class, African-American or female?

If it's people's own perceptions of their social status that alters their thoughts and feelings, though, then it may not be so important to triangulate their social class from their personal information, and then tease the effects of class out of the mix. Instead, psychologists can just, you know, ask.

Not that corroboration does any harm, or that this is an either/or choice. (In fact, write Kraus et al., in the job-interview experiment, students' self-rating on the class ladder correlated pretty well with their parents' education levels and even more with family income.) It is, though, a reminder that social class can be psychically important without being an objectively defined physical measurement.

Kraus, M., Cote, S., & Keltner, D. (2010). Social Class, Contextualism, and Empathic Accuracy Psychological Science, 21 (11), 1716-1723 DOI: 10.1177/0956797610387613