Good Cop, Bad Cop: Black and White Americans Envision Police Differently

Black Americans are roughly three times more likely to be killed by police than White Americans. Statistics like this, accompanied by recent and highly-publicized police shootings of unarmed African Americans, have spurred research attempting to understand the relationship between the police and communities of color.

Most of this research has looked at how the police view people of color.  Psychological researchers have identified a number of ways in which beliefs about and attitudes toward Black people—and Black men, in particular—bias police officers’ tendencies to use force in interactions with civilians. However, less research has focused on how people of color view the police.

A common response to instances of police violence is to question why the person ran, appeared nervous, or engaged in physical confrontation if they had nothing to hide.  Of the 47 police-caused deaths of unarmed civilians in 2018, 20 of the victims were reported to have fled from police, and 40 were reported to have either engaged or started to engage in a physical confrontation with officers (“Fatal Force: 2018 Police Shootings Database,” 2018). Such anxiety and “fight-or-flight” behaviors can be natural, and even automatic, responses to threats (Cannon, 1932).  My colleagues and I wondered, do Black Americans see the police as more threatening than White Americans do?  So, we conducted a series of experiments to examine how Black and White Americans see police officers and the implications for anxiety and behavior when imagining interacting with police.

To conduct this research, we recruited Black and White Americans to a research lab where we asked them to imagine the face of the average police officer. We then showed participants pairs of photographs where the same image had been altered to appear slightly different. Participants selected which image in each pair looked more like a police officer to them and repeated this procedure across 400 pairs of photographs. (See below for an example.) This procedure was sort of like when the eye doctor shows you two similar lenses until they find just the right one for you.

Image of two similar looking men with question "which face looks more like a police officer?"

We then used computer software to average the images that each participant chose as looking more like a police officer, and from that we were able to create the photo of an “average” police officer generated by all White and all Black Americans in the study. These typical police officers created by White and Black Americans are shown below.

image of two similar looking men one with heading "Black participants" and one with heading "White Participants"

Although they look very similar, there are subtle and important differences in the ways that Black and White participants envisioned the police.  We brought in another group of participants and asked them to rate these two images on several dimensions, without telling them what they were pictures of or who made them.  According to their ratings of the images, these participants thought that the average image of police created by Black participants (in the first study) was more negative, more dominant, more masculine, and less good than the image of police generated by White participants. In other words, Black people imagine the police as looking more threatening than White people do.

But do these differences matter for police-civilian interactions?

We examined that question by bringing in yet another group of participants.  We asked this group to imagine the following situation: “You're walking home alone at night when the person pictured above says to stop walking. They are a police officer. They are armed. They begin to approach you.” This scenario was accompanied by the picture of the typical police officer that was generated by either Black or White participants in the first study (shown above).

Participants who viewed the image of a police officer generated by Black participants reported that they would feel more anxious, have a greater desire to flee the scene, and were more prepared to physically defend themselves than participants who viewed the image of the police officer generated by White participants. Importantly, this difference in reactions to the two pictures emerged regardless of the raters’ own racial identity.  White participants were just as likely as Black participants to express these reactions to the face of the police officer generated by Black participants in the first study.

In sum, we find that Black Americans envision the faces of police as more negative, less positive, and more dominant than do White Americans. And, importantly, these mental representations of police officers seem to evoke greater anxiety and fight-or-flight tendencies in people’s interactions with police—regardless of their race.

As noted earlier, psychology has tried to understand police-civilian interactions mostly by focusing on the experiences of police.  By focusing on the experience of civilians, our research increases our understanding of this fraught relationship.  We hope that our research might aid in dismantling the fallacy that only guilty people run or fight when interacting with police. Expecting civilians, and especially Black civilians, to remain perfectly calm and composed in interactions with police fails to consider the perspective of the civilian. Personal experiences with and a longstanding history of unfair and brutal treatment erode trust in the police, increasing the likelihood of threat responses in civilians. Ultimately, decreasing the incidence of tragic interactions between the police and Black Americans involves understanding the complexity of their interactions from multiple perspectives.


For Further Reading

Lloyd, E. P., Sim, M., Smalley, E., Bernstein, M. J., & Hugenberg, K. (2020). Good Cop, Bad Cop: Race-Based Differences in Mental Representations of Police. Personality and Social Psychology Bulletin, 46, 1205-1218.

Eberhardt, J. L., Goff, P. A., Purdie, V. J., & Davies, P. G. (2004). Seeing Black: race, crime, and visual processing. Journal of Personality and Social Psychology87(6), 876-893.

Glaser, J. (2015). Suspect race: Causes and consequences of racial profiling. Oxford University Press, USA.

Goff, P. A., Jackson, M. C., Di Leone, B. A. L., Culotta, C. M., & DiTomasso, N. A. (2014). The essence of innocence: consequences of dehumanizing Black children. Journal of Personality and Social Psychology106(4), 526-545.

Lloyd, E. P., Kunstman, J. W., Tuscherer, T., & Bernstein, M. J. (2017). The face of suspicion: Suspicion of Whites’ motives moderates mental representations of Whites. Social Psychological and Personality Science, 8, 953-960.
 

E. Paige Lloyd is an Assistant Professor of Psychology at the University of Denver. She examines impression formation with a focus on how response biases and the ability to accurately read others’ cues affect discrimination toward minority and stigmatized group members.

Are Police Officers Racist? Like the Nature of Racism Itself, the Answer is Complicated

In the past couple of months, there seems to have been a sea change in the attitudes of many Americans about racism and police brutality.  In the wake of several shootings of unarmed Black men—and particularly after the slow strangulation of George Floyd on May 25, 2020 (captured in painful detail on a smartphone video)—many Americans are asking difficult questions about the American police system.  

One of the simplest and most important questions is this: Are the police racist?  This seemingly simple question has a very complex answer. The gist of it is that, in some ways, police officers do seem to be more racist than the general public—while in others, they seem to be less so.     

To my knowledge, the best survey to address the broad question of whether police are more racially biased than other U.S adults was a highly representative Pew Research Center survey of about 8,000 active-duty police officers and about 4,500 U.S. adults.  Both surveys were conducted in the spring and summer of 2016, and the details of the survey (such as identical question wording) allowed fair comparisons between the opinions of police officers and those of the general public. 

As a group, police officers often believed that racism no longer exists in America. The U.S. public—especially the Black U.S. public—often disagreed.  

Consider one of the most straightforward survey questions about race and racism. That question asked whether “our country has made the changes needed to give blacks equal rights with whites.” A surprising 92% of White American police officers endorsed this statement, suggesting that they believe racism is a thing of the past.  Only six percent of White police officers disagreed, acknowledging that there is still the need to combat racism (2% were unsure).  Among Black police officers, 69% acknowledged a lack of equal rights.  But among the White general public, 41% (still a minority, but a much larger one) acknowledged a lack of equal rights in America. Among the Black public, 84% said so.  It is hard to digest such findings and avoid the conclusion that, on average, police officers are even more racist than the rest of us. 

On the other hand, the possibility that police officers are more racist than the average American does not mean that the job of police officers should be eliminated. Psychologist Steven Pinker has cogently argued that one of the biggest reasons why human beings kill one another at much lower rates today than they did at about any other time in human history is because of modern police forces (and the associated “rule of law”).  In the 1700s, the annual homicide rate in what became the United States was about 30 per 100,000 people.  Apparently, the founding fathers found a lot of reasons to kill one another. This value dropped to 20 per 100,000 by 1800 and dropped again to about 10 per 100,000 in 1900.  In 2018, the annual U.S. homicide rate, high-profile mass school shootings included, had dropped further to about 5 per 100,000.  Most progressives thus realize that police officers do an important job.  But they foresee a better potential future in which police officers take a kinder, gentler approach to maintaining public safety.        

If there is good evidence that police officers misunderstand racism, there is also good evidence that the public misunderstands police officers. This is how the Pew Foundation’s Rich Morin and Andrew Mercer summarize the public’s view of what police officers do in the line of duty:

Many Americans believe it is common for police officers to fire their guns. About three-in-ten adults estimate that police fire their weapons a few times a year while on duty, and more than eight-in-ten (83%) estimate that the typical officer has fired his or her service weapon at least once in their careers, outside of firearms training or on a gun range…   

But the same survey that I just mentioned asked police officers whether they had ever fired their guns in the line of duty.  Only 27% reported having done so.  That’s correct. Fully 73% reported that they had never fired their guns.  Clearly, the American public believes that U.S. police officers are much more trigger-happy than they really are.  That’s a pretty large anti-police stereotype.

Getting back to police officers rather than stereotypes about them, there is some reason to believe that police officers may be better behaved than you and I might be in a life-or-death shooting situation.  Consider a study that directly compared discrimination in police officers with discrimination in the citizens who police officers are asked to protect.  In response to police shootings that occurred more than two decades ago, Joshua Correll and colleagues devised a clever way to assess “the police officer’s dilemma.” When you see a Black man holding a small shiny object that might be a handgun, what should you do?  Of course, you have only milliseconds to make your decision. 

Correll and his colleagues found that when college students played a video game that placed them in this dilemma, they shot Black men more quickly than they shot White men. They were also more likely to shoot Black men holding a cell phone than to shoot White men holding a cell phone.  They also pulled the trigger more quickly on Black targets than on White targets.  This same anti-Black shooting bias shows up in non-student populations.  Such studies reveal an unconscious, and presumably unintended, anti-Black bias. 

But in a follow-up study, Correll and his colleagues went even further.  They used this “shooter paradigm” to compare police officers with regular people—in fact, people who were from the same neighborhoods that these police officers patrolled.  Both police officers and regular citizens were racially biased in their shooting decisions: they shot Blacks more often than Whites under exactly the same circumstances. 

But, on what was arguably the most important measure—the simple shoot / don’t shoot decisions—the police officers were less biased than the community sample—not more so.  In other words, police officers do show an unconscious racially biased shooting tendency.  But on the whole, this tendency is weaker, not stronger, than the same tendency as measured in regular U.S. citizens.

On the subtler reaction time measures that assessed how quickly people fired at the suspicious person, the police officers and the non-officers were about equally biased.  Police officers don’t have a monopoly on racism.  Racism seems to be so pervasive that the citizens that police officers are asked to serve and protect are just as biased as police officers, if not more so.  

Of course, no one study can end the debate about an important question like racism. But this study does clarify that most important questions do not have simple answers. The simple answer to the question of whether racism exists in America is yes.  The more complex questions about racism in America include questions such as why, when, where, and what we are going to do about it.      

Post-script (added on September 14, 2020)

Some readers of the blog above have questioned whether social psychology has anything to contribute to the problem of police prejudice. For example, some have insisted that the answer to the title's question, "Are police officers racist?" is an obvious "yes!" and that we don’t need research to answer it.  Others have noted that research findings are too inconsistent to provide an answer anyway. I would like to make a couple of points about these reactions.

First, as a behavioral scientist, I know that the answers we get to any question depend on how we ask it.  Racism is not one simple thing.  And systemic racism, by definition, is more complex than a single racist act or attitude.  Systemic racism has many manifestations, from school funding to race-based gerrymandering and voter ID laws.  Likewise, there are people who harbor deep prejudices against some racial groups while fully accepting others. So, the answer to the question of who is more racist than whom depends largely on what one means by "racist."  Any 800-word discussion of racism will have to oversimplify this complex topic.       

This means that, even if we find that police officers as a group score in what I consider a highly racist direction on particular measures, it is not fair to say that all police officers are racist—or to say that endorsing that one racist belief means a person (officer or not) is racist through and through. Furthermore, research shows that virtually everyone harbors at least some racist beliefs and tendencies. No one can completely escape all forms of racism.

Finally, if I may make this response personal rather than empirical, allow me to note that I have a nephew who is a police officer.  This may bias my view in favor of the police.  But I also have an uncle who may have been killed by the police. (Some police officers were certainly happy to see him dead.)  I also have a brother who was beaten badly by the police—on two occasions; it is a miracle that he is alive. I grew up fearing the police, not admiring them.  So, my own attitudes about the police—like the many questions we can ask about police prejudice—are also complicated.   


For Further Reading

Correl, J., Park, B., Judd, C. M., Wittenbrink, B. (2002). The police officer’s dilemma: Using ethnicity to disambiguate potentially threatening individuals. Journal of Personality and Social Psychology, 83, 1314-1329.

Correl, J., Park, B, Judd, C.M., Wittenbrink, B., Sadler, M.S., & Keesee, T. (2007). Across the thin blue line: Police officers and racial bias in the decision to shoot. Journal of Personality and Social Psychology, 92, 1006-1023.
 

Brett Pelham is a social psychologist who studies implicit social cognition, racism, and identity. He is also an associate editor at Character and Context.  

Racism and Police Militarization

Since the 1990s, police in the U.S. have acquired billions of dollars' worth of military equipment—including mine-resistant vehicles, body armor, and semi-automatic weapons. This is perhaps the most striking example of police militarization, wherein police are adopting military culture, tactics, and equipment. Such equipment and tactics are used in many ways, from dealing with natural disasters to conducting drug raids to, as we saw in the summer of 2020, suppressing protests.

Group of Soldiers in military garb

While proponents of police militarization argue that it protects police and public safety, research often finds the opposite. For example, economist Federico Masera found that police militarization increases violence by and against the police.

Why, Then, Has Police Militarization Proliferated in Recent Years?

Researchers have typically thought of police militarization as either a reaction to rising crime or as an effort to control Black, Hispanic, and Native American communities. In support of the latter possibility, research has shown that police departments in Black communities acquire more militarized equipment than those in White communities, and that SWAT teams are more readily deployed to Black than to White neighborhoods. Importantly, these studies control for crime rates, finding that the link between racial demographics and police militarization cannot be explained by differences in crime. Building from this perspective, my research team examined recently the link between racial prejudice and police militarization.

We began by surveying 765 White Americans about their racial attitudes and opinions toward police militarization. Consistent with our predictions, people who reported more negative attitudes toward Black and Native Americans tended to express a greater preference for a militarized police force. That is, people who agreed with statements such as "if Black people would only try harder, they could be just as well off as White people" tended to also agree that "Any equipment that the military uses should be allowed to be used by police officers as well."

Does this relationship exist only because political conservatives (relative to liberals) are more likely to support police militarization, as well as more likely to hold negative racial attitudes? Testing this idea, we controlled for participant-level political conservatism, finding that this did not explain the relationship between prejudice and support for police militarization (nor did household income and perceived level of neighborhood crime).

Moving Beyond Attitudes

While this first study tells us that racially prejudiced attitudes correlate with support for police militarization, we were also interested in actual police militarization, not just people's attitudes toward it. Is racial prejudice also associated with police acquisitions of militarized equipment? To answer this question, we examined records from the 1033 Program—a federal program which distributes surplus military equipment to police for little to no cost. For each state, we computed the amount of 1033 Program equipment acquired by all police departments, which totaled over 300,000 pieces of equipment acquired by hundreds of departments. We also obtained data on state-level prejudice from Project Implicit. For each state, we computed the average level of implicit bias (measured using very subtle techniques) and explicit bias (directly self-reported by the person) against Black and Native Americans from over 3 million White Americans. As in Study 1, the results were consistent with our predictions; states relatively higher in prejudice acquired more militarized equipment, even when controlling for political conservatism.

We found similar patterns for anti-Black and anti-Native American prejudice, though some differences are worth noting. As mentioned above, we looked at implicit and explicit bias. For anti-Black bias, implicit and explicit bias were strongly correlated with each other, and both were positively associated with police acquisition of militarized equipment. However, for anti-Native American bias, implicit and explicit bias were not correlated, and only implicit bias was correlated with militarized equipment.

These findings are consistent with previous research showing that person-level prejudice is associated with positive attitudes toward the police and that region-level prejudice is associated with racial disparities in traffic stops and use of lethal force. Our studies also speak to the ongoing militarization of police, suggesting that, rather than concern about crime, its proliferation is motivated by racial prejudice.


For Further Reading

Jimenez, T., Helm, P. J., & Arndt, J. (2022). Racial prejudice predicts police militarization. Psychological Science, 33(12), 2009-2026. DOI: 10.1177/09567976221112936

Mummolo, J. (2021). Re-evaluating police militarization. Nature Human Behaviour, 5(2), 181-182. DOI: 10.1038/s41562-020-01010-7

Steidley, T., & Ramey, D. M. (2019). Police militarization in the United States. Sociology Compass, 13(4), 1-16. DOI: 10.1111/soc4.12674


Tyler Jimenez is an Assistant Professor in the Department of Psychology at the University of Washington. He researches social inequality at the macro and micro levels.

What Body Cameras Reveal About Respect and Race in Policing

The video recording turns on as the police officer gets out of his patrol car, camera bobbing as he approaches a young Black man in a sedan. “You know about your taillight?” The driver gives him a quizzical look. “Your. Left. Rear. Tail. Light. Yeah?” The officer draws out each syllable, as if he were talking to a child, before dismissing him with a curt warning. In another part of the city, a second officer has stopped a White woman, who was talking on her cell phone while driving. “Listen, ma’am,” he says, almost apologetically, “I just want you to be safe.” The officer explains the procedure for dealing with the ticket, sends the driver on her way, and turns off his camera. 

In recent years, body camera recordings have played an important role in revealing cases of police violence or misconduct. We pay less attention to the thousands of everyday encounters caught on camera, like the ones above, two of the traffic stops our research team analyzed. These routine encounters are both common—over 18.5 million people are pulled over each year—and consequential for community trust. Indeed, research on procedural justice has shown that citizens’ personal experiences in these interactions shape their willingness to cooperate with law enforcement, and even whether they follow the law themselves.

An important piece of the puzzle is missing, however: how police officers interact with the public. Police reports can tell you whether an officer issued a ticket or searched a car, but they don’t capture whether the officer treated a person with respect or condescended to them. Body cameras record these nuances, letting us observe how officers communicate with the public. In doing so, they open a window to the disparate experiences Black and White citizens have with law enforcement.

Racial Gaps in What Officers Say and How They Say It

Take language, for example. An officer can communicate the same idea in a way that communicates respect (“Could I see your license and registration, please?”) or not (“Hand over your license and registration, bro.”). Our team sampled brief exchanges between officers and White and Black drivers from transcribed body camera recordings, then asked participants to rate the respectfulness of officers’ words while keeping them blind to the driver’s race. We then taught a computer model to identify the parts of language that mapped onto these perceptions of respect, and then used those features to estimate officer respect in an entire month of traffic stop recordings.

Human raters and algorithms agreed: officers used language that was perceived as more respectful when talking to White drivers compared to Black drivers. For example, police officers were more likely to reassure White drivers and show concern for their safety, while they were more likely to address Black drivers with informal titles like “Bro” or “Dude” and pepper them with questions. These gaps in respect emerged in the earliest moments of these encounters, and persisted after controlling for factors such as the local crime rate or officer’s race.

These disparities go beyond what officers say to how they say it. We also examined police officers’ tone of voice in traffic stops of Black and White men. Our participants listened to short snippets of officer speech that were filtered to mask officers’ words but preserve their tone. Even in this subtle channel of communication, listeners perceived officers’ tone of voice as communicating more respect, friendliness, and ease with White versus Black men. And although this specific study was limited to only male drivers, we again found that racial disparities could not be explained by the officer’s race gender, or other aspects of the stop.

More than Words: The Impact of Officer Communication

There’s no law that requires police officers to be nice, so why should we care about their demeanor? The answer is that officers act not only as individuals, but as representatives of the state. As officers communicate deference or disdain towards the public, citizens learn what the state thinks of them. And our data suggest that Black and White citizens learn quite different things from their conversations with the police.

To look at the consequences of officers’ communication, we visited a Department of Motor Vehicles office, where we asked DMV customers to listen to audio clips of officer speech and give their impression of that city’s police department: the extent to which the department cared about the community and was deserving of trust, for instance. All of the clips came from a single police department, but unbeknownst to our participants, some customers heard body camera recordings of officers speaking to Black male drivers, while others heard interactions with White male drivers.

The crowd at the DMV came from all walks of life, but our findings were remarkably consistent. Those who heard Black-directed speech reported less respectful officer demeanor than those who heard White-directed speech, just as we found before. More importantly, these racial disparities lowered institutional trust. Customers who were exposed to police interactions with Black drivers had a more negative assessment of the department compared to those who were only exposed to interactions with White people.

Policing In Black And White

Racial disparities in policing extend beyond extreme uses of force: they are woven into the fabric of everyday conversations between community members and law enforcement. This matters because police stops are institutional interactions where officers literally give voice to the law. Our research shows that body cameras can play an important role in bringing such disparities to light and showing their consequences. Their greatest promise, perhaps, lies in capturing change. As our national conversation seeks to reimagine public safety, we can use body camera footage to observe how reforms change the experiences of those on the ground—and on camera.


For Further Reading

Camp, N. P., Voigt, R., Jurafsky, D., & Eberhardt, J. L. (2021). The thin blue waveform: Racial disparities in officer prosody undermine institutional trust in the police. Journal of Personality and Social Psychology. https://doi.org/10.1037/pspa0000270

Voigt, R., Camp, N. P., Prabhakaran, V., Hamilton, W. L., Hetey, R. C., Griffiths, C. M., ... & Eberhardt, J. L. (2017). Language from police body camera footage shows racial disparities in officer respect. Proceedings of the National Academy of Sciences114(25), 6521-6526. https://doi.org/10.1073/pnas.1702413114
 

Nick Camp is an Assistant Professor of Organizational Studies at the University of Michigan. He studies racial inequality at the boundaries between structures and individuals.