Jennifer Richeson has a sweet tooth. She likes jelly beans—especially green jelly beans. "I could eat them ad nauseam—and I do," she tells her students in the "Stereotyping and Prejudice" course she teaches at Northwestern University. If she were to pick only one jelly bean from a pack, it would probably be green. But if she were to scoop up a handful, she wouldn't put the other colors back. "Because it's rude, and because it just doesn't seem right. It's called a variety pack for a reason."

Taking jelly beans one at a time, you can easily fail to realize that you favor a single color. See all your green selections at once, though, and it's obvious. The anecdote relates to what she tells her students: if you want to understand prejudice, don't look only at conscious thoughts and spoken words. Look at what people feel and do without realizing it.

That's where the action is in today's research on discrimination, and Richeson, 35, is at its forefront. A social psychologist, she peers into the unconscious world of race relations, using computers to measure microsecond differences in reaction times, for example, and functional magnetic resonance imaging (fMRI) to look at how the brain reacts to interracial encounters. The methods allow her to examine the "they aren't like us" feeling—which can be about gender, age, religion, language, sexual orientation or even obesity. Richeson works on race relations, she says, because "race is particularly marked" for Americans—that is, we pay a lot of attention to it. But her true subject is not a particular kind of identity, but identity in general.

Richeson's tests indicate that everyone has measurable, often unconscious, preferences for some social groups over others. For example, a computer-based procedure called the Implicit Association Test, or IAT, measures the fraction-of-a-second differences in how quickly people associate stereotypically "white" names (such as "Chip") with positive words like "heaven" versus how quickly they associate "black" names (such as "Jamaal") with the same words. Most white Americans, despite their conscious beliefs, are measurably faster to pair the white names with the positive words—and that holds true even for some African-Americans.

In other words, prejudice is not a trait, like baldness or brown eyes, that some have and some don't. Rather, it is  a state of mind to which nobody is immune. Forty years ago social psychologists tried to figure out what made prejudiced people tick. Nowadays, Richeson says, they try to understand prejudice itself, which is a part of what makes all of us tick.

Aside from not recognizing our own prejudice, we often aren't aware of the extra work we do to cope with it. For example, Richeson and her collaborators recently used an fMRI scanner to capture images of brain activity in white student volunteers as they looked at photographs of black men. Two brain regions were unusually active: the right prefrontal cortex and the anterior cingulate cortex, both of which are known to be hard at work when people have to evaluate and shape their own behavior—a process some psychologists call "executive function" and the rest of us might call "self-control."

The brain scans help explain why whites did less well on a puzzle (sorting words flashed on a computer screen) after a brief encounter with a black interviewer than whites who had a similar encounter with a white interviewer. Richeson and a colleague, J. Nicole Shelton, found that the more strongly biased the white volunteer appeared—according to the Implicit Association Test—the worse he or she did on the puzzle after being interviewed by a black person. (In a later study, the same held true for black students who interacted with white interviewers.)

The reason, Richeson posits, is the laudable desire not to look like a bigot. Faced with someone of another race, the heavily biased person devotes more mental effort to self-control—to behaving in an unbiased way. That effort, unconscious though it may be, leaves the white volunteer with less mental capacity for the test.

Richeson even found—counterintuitively—that whites who scored high on a measure of racial prejudice tended to get more favorable ratings from black research volunteers they talked to than whites who were actually less biased. She thinks this is probably because people with greater bias work harder to conquer it, and thus come across, to the African-American volunteers, as more careful and polite.

For Richeson, the subject of identity and its effects has fascinated her since childhood. She grew up in Baltimore, where her father was a businessman and her mother was a school principal. In her predominantly white elementary school, she was content to be an average student, in the shadow of her older brother, David.

In middle school, though, she encountered a new set of teachers and a more diverse student body, and she gained confidence in herself. "My IQ didn't change," Richeson says. "Yet my trajectory was completely different—from a C student to an A student." She cites her own story as an example of how situation affects self-perception, which in turn affects performance. She also had a racially mixed group of friends, and "having a truly diverse space, not a token space, was incredibly important," she says. "All of my friends, black and white and Jewish and Asian, we all felt like we belonged."

Though her schools were 80 percent black, she found that students taking advanced classes with her were disproportionately non-African-American—a fact that led her to become a student activist and aspiring politico (when she wasn't going to ballet classes, another childhood passion).

After high school, Richeson traded her ballet dreams for Brown University. "Again, a flip-around," she recalls: now she was one of only a few minority students. A course in the psychology of race, class and gender turned her focus from politics to psychology.

In graduate school at Harvard, one of the faculty members in her department had written a book claiming that blacks were, on average, less intelligent than whites. "I was like, ‘Oh, man, I don't belong here. Look, even some of my own professors say I don't belong here,'" she says. Still, she was determined to stick it out. "I worked liked hell the first year."

In her office after class, Richeson makes it clear she's still working like hell, planning more experiments and deciding how to use a 2006 MacArthur Foundation grant. Her energy is a potent mix of a scientist's passion to know and an activist's passion to change the world. "We talk in class about Jim Crow, and my students sometimes say ‘that was so long ago.' I tell them look, my mother couldn't try on clothes in a Baltimore department store. This isn't ancient history. People who lived this are still alive."

David Berreby is the author of  Us and Them: Understanding Your Tribal Mind. He lives in Brooklyn.