She explores inner workings of bias

On the surface, there's a lot going on with this presidential election. What gets Mahzarin Banaji excited is the stuff that's going on below that, deep inside the minds of voters. Because whether we admit it to ourselves or not, she says, it matters that the two leading candidates are an older white man and a middle-age black man.

"There's this old liberal notion of being colorblind," Banaji, a Harvard professor who studies the psychology of bias, said recently while sitting on the couch in her Cambridge home. "We need to acknowledge that we're not."

For two decades, Banaji has been a leading researcher into the nature of our implicit, unconscious biases, particularly as they unfold in a social context. Bias, she has found through her experiments into memory and its associations, is a part of being human. Every person divides up the world in certain ways.

Most of the time, she said, people show an unconscious preference toward their social group. By participating in her own experiments, Banaji has found that she favors women over men, and Harvard over MIT. But there are exceptions to these findings, which is what makes this election so intriguing to her.

"There is an aspect of our minds that operates largely in unconscious mode," she said. "We can say one thing and behave in a way that's completely opposite."

In experiments designed to test our unconscious biases, the experimental psychologist has found that 80 percent of whites show a white preference, while 40 percent of blacks show a black preference. But blacks, much more so than whites, are more vocal in saying that they prefer blacks.

So if we say one thing and do another, what will really happen come Election Day? Is Senator Barack Obama at risk for the so-called Bradley Effect, where people will say they're going to vote for a black candidate and then won't? (The term is named for Tom Bradley, an African-American who lost the 1982 governor's race in California despite being ahead in the polls.)

"I don't think so," she said as she fiddled with her tortoise-shell eyeglasses. This election year has already flown in the face of expectations, she said.

"We might even see some reverse-Bradley, said Banaji. "Who would have thought we'd have an election where the younger white men all dropped out early?" With a tone of cautious optimism, she adds: "That tells us that, in our conscious minds, we've come a long way." But our unconscious, she has found, is often unpredictable. Banaji uses something called an Implicit Association Test - rapid-fire quizzes based on word and image associations - to study a person's bias. Every time she develops a new test, she and her colleagues take bets on what they think the results will say. "I'm continually wrong," she said.

In a recent study, she asked participants to choose a quiz team mate from two candidates: an overweight person and a thin person. The overweight person, they were told, had a higher IQ. The majority chose the thin person.

"Our biases are not rational," she said. "They lead us to do things that are not even in our own best interest."

Banaji came to her line of work naturally and, not surprisingly, somewhat unconsciously. She grew up a Zoroastrian - a small religion and philosophy with only about 100,000 followers left in the world - in Hyderabad, India. "It was an unusual upbringing," she said. "You're a part of a community of 6,000 living in a city of 6 million and nobody knows about you."

Still, she said, "It never struck me consciously that I was interested in these questions because of my own experience of being different."

People like understanding their biases, she believes, just as medical patients want to know their risk factors. Over 7 million tests have been completed on Project Implicit, a Harvard website she helped create to allow people to participate in the Implicit Association Tests - and gives the test-taker an immediate analysis of their biases. (It can be found at https://implicit.harvard.edu/implicit/)

For all her years of research, and her outsider background, Banaji admits that she is not without her own biases. But she thinks that by acknowledging them, people can better attempt to counter them.

She uses the screensaver on her office computer to display images of people from far-flung places, or in unfamiliar roles (a female construction worker, say), in an effort to rewire her associations.

"You can change a behavior even if your attitude doesn't change," she said. One example? She encourages people to smile at a random old person on the street.

"And maybe the change in behavior will provoke a change in attitude."

Fact sheet

Hometown: Grew up in Hyderabad, India; lives in Cambridge

Education: Studied English literature, philosophy, and psychology as an undergraduate at Nizam College in India, where she graduated in 1976; got her master's in psychology from Osmania University in India in 1978; earned her PhD in psychology from Ohio State in 1986.

Family: Husband, R. Bhaskar, is a computer scientist and a lawyer who is working on a book about the Supreme Court.

Hobbies: "I'm a very boring person. I really don't have anything. I have a job to fill many lifetimes, so I am happy doing nothing else."