Social psychologist Anthony Greenwald

The Blindspot author challenges the notion that most people are fair and willing to take others as they are.

Anthony Greenwald has spent much of his career searching to understand how the mind operates in social contexts. The University of Washington professor began his teaching career at Ohio State University, has received many awards for his contributions to psychology and served on editorial boards of 13 psychological journals. His Implicit Association Test has provided the basis for numerous applications in clinical psychology, education, marketing and diversity management. In the text Blindspot, Greenwald and his co-author Mahzarin Banaji explore the hidden biases we all carry from a lifetime of exposure to various cultural attitudes.


Tavis: Most of us like to think of ourselves as fair and willing to take people as they are, but a new book titled “Blindspot: The Hidden Biases of Good People” challenges that notion.

It’s co-authored by social psychologist Dr. Anthony Greenwald. The research says that without intent, we all have ingrained habits of thought that lead to errors in how we perceive people and make decisions. Dr. Anthony Greenwald is here, and I’m honored to have you on this program.

Dr. Anthony Greenwald: Thanks. Please call me Tony.

Tavis: I’ll call you Tony. Okay, Tony. I’m going to sit back here for just a second. This is fun; I never get a chance to do this. I get to let you teach a master class, as it were, about how this works and what we learn about our own hidden biases.

So on this big screen right here, this is going to be your blackboard for a few minutes.

Greenwald: Okay.

Tavis: I’m going to sit back and let you just walk us through how this process works. So take it away, Tony.

Greenwald: Okay. I’m reliving some history. I first saw this test when I created it about 15 years ago. But it starts this way, where you see the labels in the corners, European American, African American. Could be Black, white, sometimes we use either.

Then you see a picture in the center of the screen. This is a picture of an African American face, and you’re supposed to give a right response. So on the computer keyboard, that would be the “I” key, so I would press rapidly the “I” key in response to that face.

This would be white face. I would response with the left key, the “E” key, and then there would be a series of these faces, and I would just practice learning to associate, to press the right key response for Black faces, the white key response – the E key, the left key, for white faces.

Then we switch to another task. Now, that’s an easy task. This one’s easy also. Now we’re going to see a series of words. We’ll just see a couple of them here. Laughter is a pleasant-meaning word, and so that gets the left response with the E key, so I press that.

Then I’d get another word, and okay, this is an unpleasant word, I press the “I” key, the right response. I get a series of these, and I practice associating pleasant and unpleasant words with the left response and the right response. Very easy task.

Then we move into what we call a combined task with a different screen. Now we’re just combining the two tasks, so now the white faces and the good words get a left response, the Black faces and the unpleasant words get a right response.

First time I did this, I was actually quite surprised how easy it was for me to do this. So this would get a right response, and then there would be a series of other faces, I think we’ll see a white one, but also words, pleasant and unpleasant words.

So I’m getting a mixture of words and if you try this in the book or on the website, you’ll see a series. This is, for about 75 percent of Americans, a fairly easy task, and it was for me the first time I did it.

Then along comes another – now we’re going to switch. I was responding white on the left before and Black on the right. Now, notice the instructions are changed. Now Black is on the left and white on the right. So this would now get the “I” key response instead of the “E.”

I think we’ll see a Black face coming up next, and so now I would just practice this new task, and then a new combined task. So there’d be a bunch of those trials.

Then along comes this task, and the first time I saw this one, I was amazingly poor at performing this task. I had to give the same response to Black faces and pleasant words with my left hand, white faces and unpleasant words with my right hand.

Whereas I was fast in the previous one, the combined task, I was slow in this one. I didn’t understand why. I don’t know if we have another one. We may have one more. I think this may be the last one. So this would get the left response.

But the problem was that, I later discovered, is that this was difficult for me because I must have in my head associations of more of white with pleasant than Black with pleasant, and that’s how it turned out that this test became a measure of the strength of those associations.

Tavis: So how does it feel as a researcher, as a scientist, to have to challenge, or to come face-to-face with your own biases?

Greenwald: It was a rather remarkable experience. On the one hand I was discovering something in my head that I didn’t really like, I didn’t want there, and at the same time I was discovering, wow, this produces an effect on behavior that can be measured, and it’s not a small effect, it’s a big effect.

So it’s sort of a combination of a little distress – maybe more than a little distress at what I was discovering was in my head – and a sort of scientific joy at finding something that I knew could potentially be important.

Tavis: Did I hear you suggest earlier when you were walking through these slides that 75 percent of people what?

Greenwald: Oh, so of Americans, about 75 percent show the pattern that I discovered in myself when I first took this, which was it was easier to give the same response to white faces and pleasant words than to Black faces and pleasant words.

Tavis: Okay.

Greenwald: That’s shown by about 75 percent of white Americans, 75 percent of Asian Americans, somewhat smaller percentage of Hispanic Americans, but Black Americans are quite different.

You might think they’d be just the reverse; they’d show very consistently, maybe 75 percent would show that it’s easier for them to give the same response to Black faces and pleasant words.

Well, it is easier for Blacks to do that, but it’s only about 35, 40 percent that find that easy. Another almost similar amount actually finds the same task that I found to be easy – in other words, they behave like the majority of white people.

Then there’s the 20, 30 percent in the middle that have no preference one way or the other.

Tavis: You just gave a very scientific answer that I want to ask a very unscientific question to get a more explicit answer. What does what you just said say about the humanity of Black people?

Greenwald: That’s a good question. I’m not sure I’m an expert on that. (Laughter)

Tavis: Yeah. But the fact that Black people found it almost as easy, or a good percentage of them found it as easy as you did to give the same kind of responses in reverse says something, I think, about the way they view the world, about the lens they look through, about their humanity.

Greenwald: It’s interesting, because one way to look at it is why don’t they have the benefit of the same kind of in-group preference that white people have, and the other is, my God, they are egalitarian. Wouldn’t it be great if we were all that egalitarian?

So there were these two things, and I think Black people taking the test, actually, their reactions are that they mostly are happier to get the result that’s the opposite of mine.

When they show that they can sort of affirm the value of their Blackness by giving the faster responses to Black faces and pleasant words.

Tavis: But it’s not just – I’m digging a little deeper here, because it’s not just reaffirming their love for their own kind. What I’m particularly interested and curious about is how it is that they are in sync with you when it comes to viewing those white faces, and why it is that they are not as harsh on white faces as you were on Black faces.

Greenwald: Right.

Tavis: So I say humanity, you say egalitarian; or there’s another option – Black folk are just better on keyboards than white people are, which I doubt highly. So there’s something here to wrestle with.

Greenwald: If we’re talking about a piano keyboard, many of them are. (Laughter)

Tavis: There are a lot of great white pianists, a whole lot of them, yeah.

Greenwald: Psychology, and I love all of them. But I think this reflects many Americans, Black Americans, growing up in a society that is dominant, majority white American, and they have actually incorporated some of that whiteness in themselves.

That’s the way I understand it. Actually, we need to do more studies to figure this out, and where it develops in children. We’re just starting to do this research with young children. We do it with five-year-olds and find these biases show up in kids that young. We are trying to push it even to younger ages, and see when it starts.

Tavis: Yeah, because I have these conversations all the time, and there are a number of explanations. I’m not a scientist, but I’ve been Black for 48 years and there are a lot of things that I understand about this that I don’t get from a scientific perspective, but I get it from the perspective of just being an African American in this country.

I celebrate the humanity of Black people at their best. We’ve taught this country how to love itself in spite – we’ve learned to love this country oftentimes in spite of, and not because of.

And because, to your earlier point, Black people in this country understand the white experience much better than white folk understand the African American experience, I think a lot of that comes through in these tests, but I digress on that point.

My friend Cornel West, Professor West has made the point many times that it is amazing to him that Black people in the course of this country did not start a sort of Black al Qaeda to get back at white folk for what was done to us, much like in South Africa, when Nelson Mandela came to power.

All the Black folk didn’t rise up and say, “We’re going to kill all the white folk, or maltreat them the way they did to us.” I think these tests do say something about the humanity, not just egalitarian, but the humanity of Black people. But I digress on that point.

Here’s what’s fascinating for me, though, beyond that point. Why do we call these biases as opposed to racism? I’m not suggesting that you should, but what’s the distinct different between bias and racism as you see it from a scientific perspective?

Greenwald: We very deliberately put the phrase “Hidden Biases of Good People” in this, and our first assumption is that we were measuring something that had to do with hostility of whites toward Blacks.

But so many of the people who took this test were telling us, I don’t feel hostility. I don’t really understand why I’m showing this test result, but I think I believe fully in equality. I believe the great majority of those people are telling me this accurately.

So the question is, what’s happening there? There’s something – we conclude there’s something in their head they’re not fully aware of, and we call it an automatic preference for white. We think that they are feeling a bit uncomfortable in the presence of Black people. We think what this test tells us is that, for white people, the environment in which they’re comfortable is one that has other white people in it.

Because they don’t have a great deal of experience interacting with African Americans, they feel some discomfort. Our test is measuring something that predicts that discomfort, and that discomfort is important.

I think it’s important for both white people and Black people to understand it, because interactions between white and Black can go awry, and white people can feel uncomfortable in an interaction and they will probably just want to avoid it and get out of it.

Black people will see white people behaving that way; they will assume the white person doesn’t like them. They may in turn dislike the white person for exactly that reason, and blame the white person.

So it’s probably happening that each is tending to blame the other for a poorly managed interaction that actually has to do with these hidden biases mostly inside the head of the white person. The Black person is less of a contributor to that, I think.

Tavis: I wonder, and I want to move on to cover some other areas, because race, what you walked us through was the race test, basically. There are 14 or so other tests. I want to talk about gender, for example, in a moment.

So there are other tests that one can take on the Internet courtesy of your fine research and work. But I went online myself a few days ago and played around with this so I would be prepared for this conversation, and there’s a political test that you have on here, a test as well that features a picture of Barack Obama.

Greenwald: Oh, right.

Tavis: You know the test which I’m talking about.

Greenwald: It’s the presidential preference.

Tavis: The presidential preference test. I said political, but it’s presidential preference test by its exact and accurate name. Which made me think about this just a second ago, and I’m sure there’s no connect between the two, but I want to just ask anyway, just in case there is.

I wonder if – a moment ago, when you walked us through here, you had the random faces of Black people, random Black people. If you put Barack Obama’s face in that test we just did, I suspect the response of white people might be different, or put another way, “better.”

Which leads me to this question – whether or not some of this has to do with the right kind of Black people that white folk respond to, as opposed to just random Black folk. Does that make sense?

Greenwald: Yes. We can construct a test, and actually, this has been done, in which we compare, we make the Black faces famous, admirable American Blacks; we make the white faces some of our better-known mass murderers. (Laughter)

This produces a different result, not surprisingly, so it does make a difference, and it’s for that reason that we don’t use famous people faces in this. But except a few match the faces, if you use famous white and famous Black people, you’ll still get the same result.

Tavis: Same thing.

Greenwald: We’ve done this test, and there’s one on the Web, with cute little Black and white children, and we get the same result with that test.

Tavis: Even with babies.

Greenwald: Even with five-year-olds, just the cutest you can imagine, which was very disturbing to us. It’s a property of the test. People actually – I’m not prepared to say that people treat kids the same way adults; they treat stars the same way they treat unknown white and Black people. But the test actually sort of filters all that out of the two sides are balanced and it’s just race, and race makes a huge difference in our society.

Tavis: Yeah. How do we juxtapose the reality of your research results with the Human Genome Project, which basically tells us that we ain’t different, really? Pardon my English. But how do you juxtapose those two realities?

Greenwald: I think that what we’re measuring in this test is almost entirely learned and acquired. Yes, our genes are very similar, but there’s a gene for skin color and African facial features versus European facial features; those are genes.

Those may be just a tiny portion of genes, but we have culturally grown to treat those as basic things by which we categorize people.

Tavis: Do we think, then, that they can be unlearned or unacquired?

Greenwald: It would be nice if it was that easy. What we have – we’ve been doing research, others have been doing research, trying to eradicate the associations, the implicit, the hidden biases that this test measures, but it doesn’t happen easily.

Imagine that you were – here’s an example of the difficult. All of a sudden we decide, well, women, we call she, her; men, we call he, him. Let’s just change that. We’ll call the women he and him. We’ll call the men she and her.

We couldn’t do that. Actually, in the 1970s, we did this a bit when we got rid of, or tried to get rid of male pronouns, generically. Forty years later, some of us are still having difficulty with that.

That’s the kind of difficulty, these are things so ingrained, so automatically established, that eradicating them is no simple task, perhaps impossible. We are only going to be past this when our culture has changed enough so that they are not acquired the way they are acquired now.

Tavis: We could have done the same thing we did tonight for race. We could have done the gender test, because – we didn’t, but I wanted to give people a sense of how the test actually works, because it’s pretty much the same process when you go on to take these tests.

But now that folk understand how this system is set up and how the test actually works, tell me what you have learned from these tests about gender bias.

Greenwald: Well, the interesting thing, the gender results are also a bit disturbing. We found that there’s a very strong, we call it a stereotype association, that associates men with career and work, women with family and home.

It’s a stronger association measured by the test than people will say when they just answer survey questions. When answering survey questions, women will make it clear that they don’t agree that female is associated more with family and home and not at all with work and career, but they actually show a stronger result on this test.

So women have built into their heads the same associations that men have about associating male with work and career, women with home and family, only they have it a little stronger. This is very important.

It’s important to recognize it, because it becomes a force in the workplace. Not only a force in the way people judge women in the workplace, but also the way women judge themselves.

Women in the workplace are at risk for feeling like a fish out of water. Women belong in the home. That’s what these associations tell them.

Tavis: But you’re saying that women have internalized a lot of that.

Greenwald: They can’t help it. They’re brought up from very young ages, and there may be some genetic component to this, but the learned component of it is just huge. So when women get into the workplace and are competing with men, they’re actually at a bit of a disadvantage because they have a handicap of feeling less at home in the workplace.

It’s not that they have beliefs that tell them it’s the wrong place for them to be, it’s that they’ve got these associations in their head that just tell them women, family, women, home, women, babies.

Tavis: The answer to this question I’m about to ask may be found in the two things we’ve discussed already, race and gender; maybe not. You tell me. But of these 14 or so tests that one can take online that you have helped create for us, is there one where the results have just completely surprised you or kind of blown you away?

I can’t imagine you were blown away by the race results. It impacted you personally, so I can’t imagine that the race thing was that surprising. It wouldn’t have been for me, and I’m not an expert at this.

I can’t imagine that the gender stuff would have been that surprising. I get the sense that in a patriarchal society, just like Black folk have internalized their own – Black folk have internalized white supremacy, and just like Black folk have done that, women have internalized their own, they’ve internalized patriarchy, so I’m not surprised by that.

But was there another one of these tests that just completely blew you away?

Greenwald: Within a few years I stopped being blown away (laughter) by any unexpected results of these tests, but one result did surprise me. We found a strong result, which we call an automatic preference for young.

People have a much easier time giving the same response to young-looking faces and pleasant words than old-looking faces and pleasant words. This happens for young and old people in our society. I show it as strongly as a 20-year-old does.

I went to China to talk about this test, and I thought, okay, I’m visiting a society that my stereotype tells me they have reverence for elders. I’m going to give this same test to a group of Chinese, and it’s going to come out differently from the United States. No. It came out exactly the same as in the United States.

Tavis: Wow.

Greenwald: They have as strong an automatic preference for young as Americans do.

Tavis: I’ve been to China a number of times. I’m thinking as you give me this answer why I believe that to be the case with this particular generation that you tested. Do you have any idea why you think that’s the case? Given the appreciation in their culture for –

Greenwald: I actually think the reverence for elders is a learned thing in the culture that is very traditional, and it gets, let’s say, a lot of lip service, and people believe it, and they do have reverence for elders.

What they don’t understand is that they – or what I didn’t understand is that they are exposed to many of the same things that basically communicate that youth is the good time, and youth is associated with more good things than old age is.

This will happen in any culture. It’s the old people who are getting sicker and infirmer and losing their memories.

Tavis: That’s why the older Chinese are so scared of the West. They’re scared of being Westernized with all of our values and all of our social mores, but I digress on that point.

In the end, and obviously you’re still doing great research, but in the end, what is your hope for this research? What do you want the take-away to be from this kind of research that is the kind of research that doesn’t necessarily reaffirm what we think to be the best about ourselves, we “good people?”

Greenwald: I want good people, including myself, to understand that there are things inside their head that they don’t know are there and that are causing them possibly to be less good than they think they are, and than they want to be.

They’re going to have to live with this for many years, probably, because until we learn how to change it, which we don’t really know how to do now, I think this lesson, that we need to, as we call it in the book, outsmart the machine that is our heads, that has these associations, I think if people can take that away from this work it will put them on the track of trying to do things that will allow their behavior to fulfill their goodness more than they can at present.

Tavis: Yeah. This might sound a bit hokey at the end of a conversation where you’re talking to a scientist, but when you say that we don’t know how quite to fix this, I think you’re right on a certain level. On the other hand, love wins, and I’ll leave it at that.

Dr. Anthony G. Greenwald is the co-author of “Blindspot: The Hidden Biases of Good People.” I have so enjoyed this conversation, just delighted to have you on. Thank you for your research and for your time.

Greenwald: Thank you very much for being interested, Tavis.

Tavis: I’m more than interested – just compelled by it, in fact.

That’s our show for tonight. Until next time, thanks for watching, and as always, keep the faith.

“Announcer:” For more information on today’s show, visit Tavis Smiley at

“Wade Hunt:” There’s a saying that Dr. King had, and he said, “There’s always a right time to do the right thing.” I just try to live my life every day by doing the right thing. We know that we’re only about halfway to completely eliminate hunger, and we have a lot of work to do. And Walmart committed $2 billion to fighting hunger in the U.S. As we work together, we can stamp hunger out.

“Announcer:” And by contributions to your PBS station from viewers like you. Thank you.

Last modified: March 8, 2013 at 11:11 pm