Racial profiling

Print Friendly
Martin case: What role for implicit bias?

Racial profiling, implicit bias, and the Martin-Zimmerman case

Protester holds aloft a photo of Trayvon Martin
A rally at the Criminal Justice Building in Sanford, Fla, on March 19, 2012.

The acquittal of George Zimmerman on July 13, on charges of murder and manslaughter in the Trayvon Martin case, has polarized the nation, sparking demonstrations and firing a national debate on civil rights, self-defense laws and racial profiling.

 Official portrait of Barack Obama with flag in the background
Photo: Pete Souza
On July 19, President Obama told White House reporters, “Trayvon Martin could have been me 35 years ago.” Racial profiling, he added, is “one area where I think there are a lot of resources and best practices that could be brought to bear if state and local governments are receptive. And I think a lot of them would be.”

The jury, wrote the New York Times “rejected the prosecution’s contention that Mr. Zimmerman had deliberately pursued Mr. Martin because he assumed the hoodie-clad teenager was a criminal and instigated the fight that led to his death.”

So there was no legal proof that Zimmerman had racially profiled Martin, the unarmed black teenager who he fatally shot during a scuffle on Feb. 26, 2012.

In racial profiling, judgments and actions are based largely or solely on race.

But outside the courtroom, Martin’s fatal meeting with Zimmerman, a volunteer neighborhood watchman, continues to raise questions about race and racial profiling.

On July 19, President Obama weighed in, saying that African-Americans remember “there is a history of racial disparities in the application of our criminal laws — everything from the death penalty to enforcement of our drug laws.”

Implicit bias

Racial profiling can have conscious or unconscious roots. Explicit bias — the perception, accurate or not, that certain people, neighborhoods or activities are dangerous — can shape behavior during confrontations.

But how does unconscious bias shape day-to-day judgments? Specifically, what is the role of a phenomenon called “implicit bias” in racial profiling?

As the name implies, implicit bias operates without notice, and is1

not accessible by introspection, but may nevertheless have significant effects on behavior. For example, people with higher implicit pro-white bias make economic decisions that are more disadvantageous to black people, prescribe fewer medical treatments to black people seeking health care, and have less friendly social interactions with black people.”

Women are another common target of implicit bias, says Patricia Devine, chair of the psychology department at University of Wisconsin-Madison. “Even if I’m at the chair’s desk, talking with a man across the desk, people will direct questions to him. They are not intending to assume I don’t have leadership skills, but they have an expectation that is stereotypic in nature that they learned honestly by being socialized in a culture where these expectations are pervasive, and influence the assessments, judgments and behavior we engage in.”

A female scientist looks at a microscope.
British scientist Rosalind Franklin produced the X-ray images that led James Watson and Francis Crick to lay out the structure of DNA in 1953. As implicit bias continues to hamper the entry and progress of women in math and science. How many potential Rosalind Franklins are declining to enter science?

According to Devine, 85 to 90 percent of whites show implicit racial bias.

How much role?

Joshua Correll, an associate professor of social psychology at the University of Colorado, who has spent years researching bias, stresses that we will never know whether and how implicit or explicit bias played a role in the deadly face-off between Zimmerman and Martin. “Interpreting any particular event is next to impossible,” he says. “We can talk about things that might have an influence, but we don’t know what happened on that particular night with those particular people.”

“We can talk about things that might have an influence, but we don’t know what happened on that particular night with those particular people.”

Instead, Correll says, studies of implicit bias are intended to explore “factors that impact people in general. I would be cautious about over-interpreting a specific case or any particular finding. However, it’s also true that we have lots of evidence that people tend to associate African Americans with threats, and that is particularly true for young male African Americans. That’s a pretty widespread stereotype, that colors a lot of judgments.”

Implicit bias is certainly not the only possible reason why Zimmerman was interested in Martin. Explicit bias, says Michael Inzlicht, associate professor of social psychology at the University of Toronto, reflects “a fully conscious belief that there is an association, in this example, between black young men, violence, and guns: ‘I think it makes sense for me to think these people are especially likely to be dangerous.'”

Even without explicit bias, implicit bias can play a role, Inzlicht says, “and when you see a young black male with a hoodie, that association bubbles to surface.”

Detecting implicit bias

The study of implicit bias emerged from the Implicit Association Test, which was developed during the 1990s by Anthony Greenwald of the University of Washington. Based on the principle that we respond faster to stimuli that jibe with our preconceptions, the IAT “asks you to pair two concepts (e.g., young and good, or elderly and good). The more closely associated the two concepts are, the easier it is to respond to them as a single unit. So, if young and good are strongly associated, it should be easier to respond faster when you are asked to give the same response … to these two. If elderly and good are not so strongly associated, it should be harder to respond fast when they are paired. This gives a measure of how strongly associated the two types of concepts are.”

Left: Black screen with Shanice in the middle, White/Pleasant in the top left-hand corner and Black/Unpleasant in the top right-hand corner. Right: Black screen with Happiness in the middle, Black/Pleasant in the top left-hand corner and Black/Unpleasant in the top right-hand corner.
In the implicit association test for race bias, participants categorize black and white names and pleasant and unpleasant nouns. A white person with implicit race bias would respond faster to black/unpleasant and white/pleasant, and slower to black/pleasant and white/unpleasant.

Want to take a sample test? If the black face is paired to the positive emotion, a test subject with implicit bias is slower to press the correct button.

There are other ways to detect implicit bias. For example, Inzlicht showed a white or black face to the subject for one-tenth of a second, followed by an unknown and therefore meaningless symbol, which the subject was supposed to evaluate. “When a white person is exposed to a black face, they tend to judge the symbol less favorably than after seeing a white face,” Inzlicht says. “Your feeling about the initial stimulus spills over to your judgment of the symbol.”

“I would be cautious about over- interpreting a specific case, but we have lots of evidence that people associate young, male African Americans with threats.”

Fixit squad

Researchers2 have also measured implicit bias by looking at blood flow in the brain. As white subjects looked at a white or a black face, more implicit bias is associated with larger regional differences in flow.

Even people with the best of intentions evince implicit bias, Inzlicht says. “To some extent, all people have biases that lead to automatic behavior responses that we are more or less aware of. Those who are less aware may deny that they act in a biased way, but it’s very rare to find people who are completely free from bias.”

So what can be done? In a tantalizingly simple study, Inzlicht found that having subjects mimic black and white actors during a 140-second video could significantly reduce implicit bias.

During seven, 20-second clips, an actor lifted a cup and drank from it. Some subjects were asked to just watch the video, others copied the actions, drinking from a real cup. Afterward, a test showed a reduction in implicit bias among the copiers only.

From left to right: black person, Glagolitic character and question “”pleasant or unpleasant?”
Portrait from Shutterstock, pictogram from MacedonianBoy.
In the “affect misattribution paradigm,” participants briefly see a “priming” image or a word that may elicit a positive or negative attitude, followed by a neutral stimulus (such as this Glagolitic letter), which participants are asked to rate as more or less pleasing than average. The misattribution occurs because the feeling from the priming image is attributed to the neutral stimulus.

The explanation, Inzlicht says, may reside in “mirror neurons,” which are motor neurons that, oddly, fire as we observe someone else performing an action. Previous tests have shown that mirror neurons were less active when people watched movement by a person of a different race. “The more prejudiced they were, the less motor cortex activity we saw,” Inzlicht says.

The change can be due to “self-other overlap,” or, in common language, empathy. The overlap is “the extent to which I see you overlap with my own sense of self or my self overlaps with you,” Inzlicht says. These overlaps can be based on personal characteristics such as personality, looks, culture or religion: “anything that leads me to see a little bit of myself in you.”

Going to de-bias school

Testing implicit bias with a first-person shooter game

Undergraduates are faster to 'shoot' armed Black than White, Latino, or Asian but slower to 'not shoot' unarmed Black than other races while police officers are faster to shoot both armed Black and Latino.
Adapted from Melody Sadler et al3
The graphs show how quickly undergraduates (top) and police officers (bottom) responded to images showing white, Hispanic, Asian and black men holding guns, or a non-lethal item like a phone. The college students were faster to correctly “shoot” a black armed target and slower to correctly “not shoot” a black unarmed target — both evidence of implicit bias. The officers had similar responses, but also shot whites faster than Asians, and were slower to refrain from shooting whites. (Vertical scale is highly exaggerated)

While Inzlicht tested a momentary intervention for implicit bias, Devine and colleagues devised a 12-week course, premised on the idea that prejudice is a type of bad habit. “When these stereotypes are pervasive, when we see a woman or a member of a particular group, a spontaneous reaction pops to mind automatically,” Devine says. “That’s an automatic response, and we have to use the power of the conscious mind to override those automatic reactions.”

The course introduced a series of tactics, such as substituting a benign image for a negative stereotype, or putting yourself in the other’s shoes. “If you have awareness and the motivation to overcome prejudice, you can draw on the toolkit we offered,” Devine says. “If you took advantage, and practiced, that would reduce risk of having that habitual response.”

After 12 weeks, the intervention produced a significant reduction in implicit bias, and also a burgeoning motivation to combat it, Devine says. “They showed an increase in concern about bias, and thought this is something important I should be working on.”

This type of training might interest judges, police officers or teachers, Devine says. “There are dramatic differences in discipline rates for minority students, compared to majority ones, in a lot of school districts. A lot of this could be a function of a classroom climate created by implicit bias. If we are able to address those, maybe the kind of aggressive behavior you see on the playground won’t be differentially interpreted as a function of race, and would lead to similar types of discipline.

If you don’t have 12 weeks, perhaps popping a pill would suffice? In a recent study, propranolol, a drug for hypertension and anxiety, reduced or eliminated implicit, but not explicit, bias.4. While nobody is seriously considering treating implicit bias as a medical problem, the results did suggest that the hormone norepinephrine, which is blocked by propranolol, plays a role in implicit bias.

Mural depicts Amadou Diallo
Photo: Hobo Matt
This mural is a few doors from the site where Amadou Diallo, an unarmed 23-year-old immigrant from Guinea, was killed by four police officers in the Bronx, New York, in 1999. The Diallo case raised issues of police brutality and racial profiling, but the officers were acquitted on charges of second-degree murder. The case was the reason that Joshua Correll of the University of Colorado chose to focus his research on bias.

Calling the cops

The on-the-street outrage that greeted the exoneration of Zimmerman, a volunteer neighborhood watchman, reflected a feeling in the black community that police — and now volunteers — can get away with murder when they shoot black people.

A 2012 study5 correlated racial bias in reaction time to “the extent to which officers overestimated the amount of violent crime in a community. As violent crime increased, bias to shoot Latino targets increased, but bias to shoot White targets decreased.”

Reassuringly, police officers from three U.S. regions showed “no evidence that target race biased a police officer’s ability to correctly shoot armed targets and to not shoot unarmed targets.”

For Correll, the studies have offered a refreshing view into the police. “I have to say that most of the officers that we have dealt with are really trying hard to do a very difficult job. It’s literally the case that they don’t know if they are going to come home; that’s such a different reality than what most of us face.”

In the wake of the Zimmerman-Martin trial, Correll thinks attitudes toward racial profiling have changed. “It’s a cultural-societal anathema to be accused of this kind of race-based decision making,” he says. “They may fall victim to certain shortcuts, including the use of racial cues, but that is a very important cultural shift.”

But bias can be sneaky, Correll says. “The tricky thing about implicit bias is that we don’t necessarily know we have it. There is an association between ‘black’ and ‘threat’ in our minds, even though we don’t endorse that consciously.”

— David J. Tenenbaum

6 7 8 9 10 11 12

Terry Devitt, editor; S.V. Medaris, designer/illustrator; Yilang Peng, project assistant; David J. Tenenbaum, feature writer; Amy Toburen, content development executive


  1. Implicit Race Bias Decreases the Similarity of Neural Representations of Black and White Faces, Tobias Brosch et al, Psychological Science February 2013 vol. 24 no. 2 160-166
  2. Implicit Race Bias Decreases the Similarity of Neural Representations of Black and White Faces, Tobias Brosch et al, Psychological Science February 2013 vol. 24 no. 2 160-166.
  3. The World Is Not Black and White, Melody S. Sadler et al, Journal of Social Issues: Special Issue: The Reality of Contemporary Discrimination in the United States: Volume 68, Issue 2, pages 286–313, June 2012
  4. Propranolol reduces implicit negative racial bias, Sylvia Terbeck et al, Psychopharmacology, 2012, 10.1007/s00213-012-2657-5
  5. The World Is Not Black and White, Melody S. Sadler et al, Journal of Social Issues: Special Issue: The Reality of Contemporary Discrimination in the United States: Volume 68, Issue 2, pages 286–313, June 2012
  6. Why we see guns that aren’t there
  7. How much does race still matter?
  8. How the brain views race
  9. The Obama Effect
  10. Test your own implicit bias
  11. Gender bias in science is real
  12. Psychological issues In Trayvon Martin shooting case