Policing BiasBy Alex Dixon | August 1, 2009 | 1 comment
Studies show that police officers are quicker to shoot black suspects than white ones, reports Alex Dixon. As a result, scientists and law enforcement are teaming up to confront a dangerous threat: racial bias.
In 1988, Sgt. Vernon Gudger was a rookie in Washington, D.C.‘s Metropolitan Police Department. He was off-duty one night at his mother’s house when he noticed two black men about his own age attempting to steal the 1964 Chevrolet Corvair parked in his mom’s backyard.
Gun in hand, Gudger ventured outside dressed in a baseball cap and a puffy red coat. He ordered the two men to the ground. Holding them at gunpoint, he told his mother to call the police to say he’d apprehended two suspects. But Gudger would soon learn that a neighbor had already called the police—reporting Gudger as the threat.
Three officers pulled up in a squad car behind Gudger and the men. Out came two black officers—Washington and Truesdale—and one white officer. The officers drew their pistols, placing Officer Gudger—who himself is black—in a pickle.
“It was Washington and Truesdale who shouted ‘Don’t shoot, don’t shoot! That’s Gudge, that’s Gudge! He’s police!’” Gudger recalls. Gudger shouted back, “I’m police!” then followed his colleagues’ directions, putting his gun down on the ground without turning toward the officers, despite his close proximity to the criminals.
After the suspects were cuffed, the white officer approached Gudger. “You’re lucky they were there,” he said, referring to the black officers, “or you’d be pushing up daisies.”
Gudger was shocked—and he couldn’t help but think that the incident would have played out differently had he been white. To white officers, he contends, “every black person holding a gun is a suspect.”
Indeed, cases like Gudger’s have elicited accusations of racism against police departments across the country. Over the last two decades, similar confrontations have ended in the deaths of innocents, like Omar Edwards, the African-American New York City police officer shot and killed in May by a white colleague, and Amadou Diallo, the West African immigrant fired upon 41 times by NYPD officers in 1999 after they mistook his wallet for a gun. As the recent furor over the arrest of Henry Louis Gates Jr. has reminded us, the role of racial bias in policing is still a highly contentious subject.
In response, cognitive scientists have set their sights on the psychology of police work, looking at how unconscious racial bias may inform the snap judgments officers have to make. Their results shed light on the deep cognitive roots of racial bias—and how these biases can complicate officers’ decisions about when to pull the trigger. They’re also informing a new wave of police trainings, attempts to reduce the odds that we see more cases like the Edwards or Diallo shootings.
Three blinks of an eye
Joshua Correll has been a leader in the psychological study of racial bias and policing. Correll, an assistant professor of psychology at the University of Chicago, first got interested in the subject after the Diallo shooting. Correll was a graduate student at the time; the incident inspired him to see whether the emerging science of racial bias might help explain the officers’ reaction.
Correll conducted a study using a relatively simple computer simulation. In the simulation, a series of photographs flash on the screen. Some just depict different settings—the countryside, a city park, the facade of an apartment building. In other photos, black or white men are shown holding something in their hands—either harmless items, such as cell phones or wallets, or a gun. The participant is supposed to shoot the men with the guns and avoid shooting men without guns. Participants have 850 milliseconds—about three blinks of an eye—to press a button labeled “shoot” or one labeled “don’t shoot.”
When Correll ran this study with white college students, he found that participants were more likely to shoot unarmed black men than unarmed white men and were less likely to shoot armed white men than armed black men. Their reaction times differed as well: They were quicker to shoot black men with guns than white men with guns, and they were slower to press the “don’t shoot” button when an unarmed man was black than when he was white. These findings were even stronger for those participants who reported having more contact with African Americans. When Correll ran this study with white and African-American community members in Denver, he found that the two groups showed the same racial bias.
Tracie Keesee, the Denver Police Department’s chief of research, was intrigued by Correll’s study. The DPD had recently been at the center of controversy over the shooting of Paul Childs, a black, mentally disabled 15 year old killed by Denver police officers.
“One of the questions that kept coming up in a lot of the community meetings was whether or not the Denver Police Department somehow trained their officers to focus on killing young African-American males,” says Keesee. “Of course our initial response was, ‘No, of course not.’ But we never really knew if our training may have been having some adverse or inverse impact on what we were trying to do.”
Keesee contacted Correll, and the DPD became the first police department willing and able to work with him on his research.
For a 2007 study, Correll and his colleagues recruited roughly 100 officers from the DPD, 100 other residents of Denver, and 100 or so additional police officers from the across the country who had attended a Denver police conference. The participants were predominantly white, though they also included some African-American and Latino participants. All of the participants went through a simulation like the one in Correll’s earlier study, where they had a split second to decide whether to shoot or not shoot white or black characters in a computer simulation, some armed with guns, some unarmed.
Similar to Correll’s earlier study, the researchers found both police and community members were quicker to shoot blacks with guns than they were whites with guns, regardless of the participant’s own race. And it took all participants more time to decide against shooting the non-threatening photographs of blacks, the ones who were holding wallets or cell phones. “What that suggests is that when they see a black target on the screen, the idea of injury or threat may pop into their minds,” says Correll.
Research by many other psychologists and neuroscientists supports Correll’s interpretation. In one study, NYU neuroscientist Elizabeth Phelps and colleagues found that participants showed more activity in a brain region known as the amygdala when they saw black faces than when they saw white faces. Amygdala activity spikes when we feel threatened or afraid, suggesting that participants might have been experiencing fear or even aggression when they saw black faces.
And in a study lead by psychologist Keith Payne, a professor at the University of North Carolina, Chapel Hill, participants looking at a computer screen were asked to press one button when they saw a picture of a gun and another button when they saw a picture of a tool. But before they saw either kind of object, a photo of a black male or a white male flashed on the screen. The participants were faster at identifying the gun after first catching a fleeting glimpse of a black man’s face—and they were more likely to mistake a tool for a gun after seeing a black face than a white one.
Findings like these suggest to researchers that the brain is wired to respond quickly to possible threats, and in American culture at least, people may have been socially conditioned to see black male faces as one of these threats. This process can even affect people who consciously shun racial bias in any form, police officers who swear to uphold the law without prejudice, and people of color themselves. “Unless one is socially isolated, it is not possible to avoid acquiring evaluations of social groups,” Phelps has written. “Yet such evaluations can affect behavior in subtle and often unintentional ways.”
For police, researchers fear, this phenomenon might have fatal consequences, perhaps making them more likely to pull the trigger when they see a black suspect, especially when they think their own life could be on the line. This may help to explain why in the NYPD’s 165-year history, there have been four incidents, including the Edwards case, where a white officer shot a black officer, but never has a black cop shot a white cop. During Sgt. Gudger’s 22 years as a D.C. police officer, three black cops have been shot by white officers. Two were killed, and one is paralyzed.
Training the trigger finger
But it might not be so simple. In Correll’s study, the time it took community members and police officers to shoot or not shoot seems to have been motivated by some unconscious bias. But when it came to their final decisions—that is, whether or not they actually pulled the trigger—police officers didn’t appear to be influenced by race: They did not shoot unarmed blacks by mistake as often as ordinary people did, nor did they shoot them any more than they shot unarmed whites.
“That suggests that police are doing something,” says Correll. “They demonstrate a pattern that we really don’t fully see with any other group we’ve tested.”
One reason police may have been more accurate, says Correll, could lie in the extensive firearm training programs police must undergo before they become officers. As they endure these trainings—from 360-degree computer simulations to trainings with paint guns—officers improve the way that their brains talk to their trigger fingers, learning when to show restraint.
“We’re doing so much better at training police,” says Lorie Fridell, a criminology professor at the University of South Florida. “So it could be that because use-of-force training has become as strong as it has, it’s countered or offset the implicit racial biases of police.”
Does this contradict the prevailing assumptions about cases like the Edwards and Diallo shootings, that they were the products of racial bias?
Not necessarily. Fridell maintains that though police firearm trainings may be effective in countering some racial bias, research shows that unconscious biases run deep in most people, and these programs don’t really get to the root of the problem.
“Even the best officers, because they’re human, might practice biased policing,” she says. Indeed, national statistics seem to back up Fridell’s claim: According to an analysis by the Justice Policy Institute, for instance, white youth are significantly more likely to use or deal drugs than African-American youth, yet African-American youth are arrested for drug offenses roughly twice as often.
Correll adds that though some of his research suggests police officers can keep unconscious racial biases in check, their ability to do so may be compromised when they’re under stress, or at the end of a long workday. His current research is exploring how stress and fatigue may affect officers’ decisions about whether to pull the trigger.
Fridell is now developing a training program with funding from the U.S. Department of Justice that is specifically tailored to combating unconscious racial bias. The program relies on two key strategies.
Fridell calls her first strategy “consciousness raising.” The idea is to teach police officers that racial biases lurk beneath everyone’s conscious minds, as the psychological research suggests. That way, police in the field will be more likely to catch themselves when their behavior may be unwittingly influenced by subtle biases. For instance, officers will engage in simulations in which they must make snap judgments about suspects, then step back and review how those judgments may have been swayed by the suspects’ race.
Her second strategy relies on what’s known in psychology as the “contact hypothesis,” and it’s a phenomenon that goes far beyond policing. This idea holds that if someone has positive experiences with members of another racial or ethnic group, that person is less apt to be prejudiced. It may seem like common sense with respect to overt racism, but it also may affect unconscious bias.
“The contact hypothesis has great implications,” Fridell says. “Police have many factors to consider when hiring, and I certainly wouldn’t want them to focus in on just one, but all things being equal you might want the person who has had the diverse, positive experiences versus the one who has not.” (The NYPD officer who shot and killed Omar Edwards grew up in a section of Long Island that is 85 percent white.)
A 2006 study by Florida State University psychologists Michelle Peruche and Ashby Plant supports this idea. Using a computer simulation very similar to Correll’s, Peruche and Plant found that officers who reported having positive contact with black people were less likely to shoot unarmed black suspects than were officers who had negative attitudes toward black people.
Fridell’s program is part of a broader movement to build on research like Correll’s and bridge the gap between law enforcement and scientific research. The Policing Racial Bias Project, a program led by Jennifer Eberhardt, a psychology professor at Stanford University, has developed partnerships between police agencies and social psychologists, allowing agencies to participate in cutting-edge research into some of the unanswered questions about racial bias and its role in policing. In 2004, Eberhardt organized an unprecedented conference that brought together researchers and law enforcement officials from 34 agencies, spanning 13 states. The conference was designed not only to help research inform policing, but to have policing inform social science, making its studies more relevant to real-world predicaments.
UCLA’s Consortium for Police Leadership in Equity is another such “matchmaker,” helping police departments team up with world-class researchers. Already, nine police departments across North America have dedicated themselves to working with researchers—Correll, Keesee, and Eberhardt among them—as part of this consortium.
There are other signs that law enforcement is taking this research seriously. Following Edwards’ death, the NYPD hired Correll to investigate the role that race may have played in the shooting. And in 2007, after the San Francisco Chronicle reported that San Francisco arrested African Americans at a rate higher than any other California city, Fridell made 28 recommendations to the San Francisco Police Department about how to guard against racially biased policing. She was encouraged to find that the department was willing to follow many of these recommendations, such as allowing her to conduct bias training with department leaders. She now hopes that the SFPD will adopt the program she is developing for the Justice Department—and will be the first among many police departments to do so.
“The reality is that any department who hires human beings needs to be proactive in promoting fair and impartial policing,” says Fridell. “That gets us away from pointing fingers at who is bad and who is good.”
Greater Good wants to know:
Do you think this article will influence your opinions or behavior?
About The Author
Alex Dixon is a frequent contributor to Greater Good. His essay will appear in the forthcoming Greater Good anthology, Are We Born Racist?, to be published by Beacon Press in the spring of 2010.