BOB GARFIELD:
This is On the Media. I'm Bob Garfield.
BROOKE GLADSTONE:
And I'm Brooke Gladstone. Now that the Democratic Convention is over, the Republicans are preparing for their grand old party in the Twin Cities. Both parties realize that they have to reckon with the impact of race in this election, but they don't really know how we feel about race, and the fact is neither do we.

Psychologist researchers have spent more than a decade searching for the answer, testing millions of people using something called the Implicit Association Test. And they found that nearly 80 percent of whites respond more positively to whites, and so do 30 percent of blacks.

Here’s how the test works. On a computer you are asked to place, quickly as you can, positive words, like “love” or “success” next to black faces, and negative words, like “agony” or “failure” next to white faces. Then you are asked to switch negative for positive, white for black. Your reaction time is then measured. If your reaction time is shorter when you, say, match the positive words with the white faces, that’s your natural association.

Over and over, millions of times, the test demonstrates hidden biases when it comes to not just race but age, weight, gender, etc. University of Washington psychologist Tony Greenwald created the Implicit Association Test. He explains why even one-third of black respondents react more favorably to whites.
TONY GREENWALD:
Black respondents were evenly divided, with about a third favoring white, a third favoring black and a third right in the middle, with no preference at all. I think this is a consequence of the black respondents who have taken our test. They are reflecting the dominant culture in their, and they are probably reflecting a lot of favorable exposures to white people.

Whites in our culture, on the other hand, do not have, often, a lot of experience with black people. And most of their experiences with blacks are through entertainment and news media, where the presentations are often not favorable.

White people who have close friendships with black people do not show this pattern. They may show a favorability to black on our test.
BROOKE GLADSTONE:
You took the test on race and the results have changed over time. What did you find?

TONY GREENWALD:
Well, I created the test about 12 or 13 years ago and the result I got from it so surprised me that I immediately concluded this was going to be a very interesting test [BROOKE LAUGHS] and started to do research on it. And I've probably taken it a few hundred times since then.

I know the faces that are on the test like they're old family friends, and I do respond much faster to the test than I used to. And my automatic preference, which I won't confess but you can guess, shows up more mildly than it used to. But it’s remarkable that it nevertheless is still there. It’s something I don't want, but I've got.
BROOKE GLADSTONE:
You know, I read a piece today in The Concord Monitor about a columnist who wrote that she showed a rare preference for African-Americans over whites, and she’s white. She guessed that was because she’d spent the morning researching Barack Obama and pondering the impact of race in the election.

Also, I read that respondents who were shown a picture of Martin Luther King before taking the test generally showed less bias against African-Americans. So this has to do with immediate prior exposure before taking the test.
TONY GREENWALD:
That’s true. What you described was something done in a laboratory experiment that I was involved with, but we don't think that will last too long. If one had those experiences over and over again, yes, there might be an enduring change, but we actually don't get those experiences over and over again in our media.
BROOKE GLADSTONE:
So generally I think that the media are given way too much credit for moving the culture. I generally think it’s much more of a reflection of the culture. But you’re suggesting that the media may be able to fundamentally change our implicit biases if they offer us enough favorable images.
TONY GREENWALD:
The media can do this, but I actually don't think the media are prepared to do this. They're not prepared to say - okay, we're not going to present crime as a black phenomenon, we're not going to put on entertainment programming in which the criminals are more likely to be black than white. That’s just not going to happen.

On the other hand, if we get a black president, the media will have no choice but to present the black race in a very different way than it’s been presented in the past.
BROOKE GLADSTONE:
Is there anything that political strategists can take away from your research?
TONY GREENWALD:
First, let me say whenever you hear a politician say, you know, it’s the policies that are going to influence the voters - taxation, health care, education, Social Security, etc. - you know, my research doesn't take me there at all. I think it’s not the policies. I think it’s the simple associations.

For example, what do we know about Biden? Well, some people know details about him. But if there’s a sound bite that he came away with, it’s “goes home to Wilmington every night.” That’s a great sound bite in terms of associating him with good things, like home. And notice, he’s not associated with Washington, D.C., which some politicians want to distance themselves from, but rather Wilmington.

And I heard a phrase at the convention that I am sure is going to come up a lot, and that is “a race for the future.” For Obama to get an association of himself with the future and try to link McCain to the past, those are simple associations that could work very well. And if you’re going to have a race, you actually want a young guy to be doing the running. [LAUGHS]
BROOKE GLADSTONE:
So what do you think we can take away from this research, that most of us are, on some subconscious level, just immovably racist? I mean, that’s depressing.
TONY GREENWALD:
I wouldn't say that. I do think that we have associations with race that will influence our behavior even though we don't want those associations to influence our behavior. In fact, I'm convinced that race is importantly involved in this presidential election. Just how it’s going to play out we don't exactly know. We are in uncharted territory. We have never had a campaign and an election like this.
BROOKE GLADSTONE:
Tony, thank you very much.
TONY GREENWALD:
Well, you’re very welcome. Thanks for being interested, Brooke.
BROOKE GLADSTONE:
Tony Greenwald is a professor of psychology at the University of Washington in Seattle. He created the Implicit Association Test with colleagues at the University of Virginia and Harvard University. We'll link to the test from our site, onthemedia.org.