Cognitive Daily

i-eca0cf2af9fc3ac4445c7dff7d8aab70-research.gifThe recent controversial shooting of an unarmed black man in New York has generated terrible grief and perhaps justifiable anger. But if officers honestly believed the man was armed and intended to harm them, weren’t they justified in shooting?

Perhaps, but an important additional question is this: were they predisposed to believe he was armed simply because he was black? Consider this quick movie:

It will flash two pictures. One man is armed, the other unarmed. Who do you shoot? I’ve primed you to think about race, so it’s not really a fair test. If you were a police officer who believed his life to be in danger, would you respond in the same way? (You can use the slider to see which man really was armed.)

In 2003, a team led by J. Correll flashed random photos of white and black faces, some superimposed with guns, others with harmless items such as cell phones and wallets. They asked college students to press one key indicating “shoot” the suspect, and another indicating “don’t shoot.” The students were more likely to mistakenly fire at black faces that were unarmed compared to unarmed white faces.

But what about police officers? With their special training and rules about when to fire, perhaps they will do better. A study by E. Ashby Plant and B. Michelle Peruche tested police officers on a similar task.

They created composite images by superimposing guns and harmless items on the faces of 9 white men and 9 black men. The faces were previously matched for attractiveness. These faces were randomly flashed on a computer screen; each face was shown both with a gun and with a harmless object.

48 mostly white police officers who volunteered to participate in the project were told only that the study was about “decisions to shoot.” They had 630 milliseconds after each face was displayed to press the “shoot” or “don’t shoot” button on the computer running the test. As the test began, results were the same as with college students: Police officers were more likely to mistakenly shoot black suspects holding harmless items than white suspects holding the same items. They made an average of 3.63 errors over 20 trials when the suspect was black, but only 2.65 errors when the suspect was white.

But an interesting result occurred as the test continued. After another 80 trials, about 40 of which included the key situation of a person holding a harmless object, the disparity between reactions to white and black suspects disappeared. During the second half of the experiment, the average number of errors for black suspects diminished to 2.60, statistically indistinguishable from the rate for white suspects.

Plant and Peruche believe this is a hopeful result, suggesting that police officers can be trained to eliminate bias in their work. They may be right, but I do have one potential objection: police are also “trained” by what they actually see every day as the patrol the streets. Won’t this training — for better or for worse — trump whatever they learn in a few hours in the training room?

The problem in the Queens shooting of Sean Bell was, like most police incidents, more complex than the scenario presented in this study. An informant may have told the police Bell was armed, and he had attempted to ram the police with his car more than once. Also, the question was less about whether shooting at all was justified, and more about whether so many shots should have been fired. Police may have assumed other officers were justified in firing, and so that’s why they fired themselves.

Malcolm Gladwell, in his book Blink, offers another, counterintuitive solution to the problem of accidental police shootings: Police should most often work alone. With the variable of other officers removed from the problem, individuals are more likely to think for themselves. Without the option of “backup,” they are less likely to get themselves into dangerous situations in the first place.

Plant, E.A. & Peruche, B.M. (2005). The consequences of race for police officers’ responses to criminal suspects. Psychological Science, 16(3), 180-183.

Comments

  1. #1 Jenny
    November 30, 2006

    All right, I’m going to have to disagree here. If we were talking about Barney Fife in Mayberry (or Davidson for that matter), I would be more likely to agree with you that police officers should work alone. However, I will argue that cops in New York should probably not work by themselves. Cops are supposed to “get themselves into dangerous situations.” That’s why if someone breaks into your house, you call them. Solitary police officers are much more vulnerable – and much less accountable for their own actions. If something bad happens to a good cop, there’s no one there to call for help. If you’re a bad cop, it’s a lot easier for you to get away with what you’re doing if there’s no one there watching.

  2. #2 etbnc
    November 30, 2006

    “If you’re a bad cop, it’s a lot easier for you to get away with what you’re doing if there’s no one there watching.”

    Ummm…. maybe. Unfortunately, upon reading that sentence, I immediately began recalling several incidents in which groups of armed authority figures committed violent bad deeds and then tried to cover up their behavior as a group.

    I remember learning about the concept of “diffusion of responsibility”. I recently read something about “plurality of ignorance”, and it sounded very similar.

    Sometimes the behavior of social systems can be counter-intuitive.
     

  3. #3 Oran Kelley
    November 30, 2006

    “Cops are supposed to “get themselves into dangerous situations.” ”

    Personally, I’d prefer they did this less often that they do, becuase the situations seem to be more often dangerous for everyone else rather that to the police.

    When police go undercover or engage in dangerous operations in plain clothes they seem to become quite aggressive in making absolutely sure no police are hurt or killed, which means that a lot of other people end up getting hurt or killed.

    This case seems pretty much symptomatic of police in hazard mode.

    I think a lot of the “undertaking of danger” in police work is unnecessary, I think far too much work is done by plain-clothes officers that ought to be done by uniformed officers, and I think far too few precautions are taken to protect citizens from tribalistic, overprotective cops when hazards are undertaken. This has been the case again and again in NYC.

    And I don’t just mean Diallo, there’s also Ousmane Zongo and Alberta Spruill and Vivian Rodriguez and . . . A lot of folks get killed or hurt when the police go after non-existent or relatively innocuous criminals through extremely hazardous means –plain clothes, undercover, at night, etc.

    Being a policeman is not nearly as dangerous as a lot of people suppose, and one of the reasons for that is that police are very good at transferring hazard from themselves to the general public, and they seemingly have official sanction to so.

    I don’t think it’s a bad idea to have a close look at the risks police undertake (on our behalf) and how they deal with it. Policing alone may be one option.

  4. #4 Joshua
    November 30, 2006

    I’ve been commenting to a lot of these police brutality threads mentioning the de-arming of patrolling police as a method to control their behaviour. The reasoning is a lot like yours, Oran. Police are far too ready to jump into situations without exercising a lot of judgment, which they tend to compensate for by using excessive force. They need to use that force, in many cases, to protect their own, but as you point out that strategy ends up costing a lot of innocent lives.

    Without a gun or taster right there on his hip, the patrol cop will have to think twice before going in to a dangerous situation. Frankly, that’s exactly what I want my cops to be doing: thinking. And it doesn’t preclude having access to weapons in situations where they’re actually needed, either. But I would argue that fatal weapons aren’t needed for the vast majority of police officers the vast majority of the time.

    Here’s an interesting thought I had. It’s not such a good idea to make cops work alone, I don’t think. It’s much better if they have some degree of backup. And maybe there is a unique situation in America where the British solution of not issuing sidearms to patrol officers doesn’t work. (This premise is not one I accept, but let’s just run with it.) What if only one officer on a team is issued a weapon? So it’s there if needed, but overkill is reduced. As for tactics, have the unarmed officer be the main contact as far as talking to people and physically restraining suspects. That way, there’s no danger of a suspect grabbing that cop’s gun during a struggle, because the cop in the struggle doesn’t have a gun. However, his partner does, and that guy is free to keep an eye out for threats beyond the immediate since he’s not involved with the direct interaction.

    It’s just a suggestion, and I don’t know how practical it is, but it’s very clear that police brutality is a very real problem that needs to be addressed in a sensible way.

  5. #5 djexplorer
    November 30, 2006

    There’s very good reason for cops in NYC and elsewhere to be more likely to suspect blacks will commit homicide when challenged by police than whites, just as elsewhere.

    According to FBI statistics, blacks are 8x more likely (i.e. per capita) to commit murder than whites.

    That’s absolutely no reason to convict some individual black man of murder and let a white off. There is lots of individual information available about the crime committed by someone there, and that is obviously all that should be used. Our rules

    The information about the queens bridegroom shooting available so far indicates both that someone aparently among the group fleeing had shouted he was getting a gun, and more importantly, the getaway car driven by the black man who was shot dead, was either trying to kill the cop who was blocking his way in front of the car, or had indifference as to whether he did.

    In that situation, and believing with reason that men in that car were armed, the late arriving cops seeing an officer being about to be killed opened fire.

    Did the fact that it was a group of blacks who’d be in a nortorious illegal drugs and sex involved strip joint that the cops were in the process of getting enough evidence enough to close have any bearing on the cop’s split second evaluation of the situation? Yes. Should it have? Yes.

    I’m not so happy with the fifty shots. That was overkill. But shooting to incapacity (with a very high risk of killing) the black driver who seemed to be trying to vehicularly kill a cop who was barring his getaway doesn’t seem unreasonable on available evidence to me at all.

    I sure as hell would never act the way that driver did in that situation.

  6. #6 Requi34
    December 1, 2006

    As usual, when a black guy commits a crime and the white po-lice enforce the law and defend themselves, they’re branded the racists scum and the po’ innocent brotha gets a pass. What are white cops supposed to do when confronted with a black guy trying to run one of their own down? Roll over and play dead? They did the right thing. The 50 shots as overkill is ridiculous. You get the same outcome if the driver was shot a few times or 50 times: He’s dead. The threat has been elimnated. Quit the whining.

  7. #7 Requi34
    December 1, 2006

    Oh, forgot one thing. This guy was armed; with a 2000 lb weapon. Let’s not forget that.

  8. #8 KevinC
    December 1, 2006

    They were undercover cops, maybe the black guys thought they were being robbed and were just trying to get out of there. That is the problem when a police force uses large groups of undercover cops. Because none of them were in uniform they had lost their normal protection and ID for the citizens who they are serving.

    The cops (in disguise) blew somebody away who was not a criminal, not the cops job. Quit whining yourself Requi34.

  9. #9 Dave Munger
    December 1, 2006

    “po’ innocent brotha gets a pass”

    The man is dead. We have no idea whether or not he was innocent, because he wasn’t afforded a trial. This is not to say that the police didn’t do the right thing, but I think leaping to conclusions — in either direction — in this case is premature.

    “This guy was armed; with a 2000 lb weapon.”

    You could say that about anyone in a car; early reports suggest that he was indeed using it as a weapon, but it’s also possible that this is a cover-up story used by the officers after they realized what a terrible mistake they made.

  10. #11 white girl
    December 1, 2006

    wow, upon reading some of these comments I’m even more convinced that racism is still extremely rampant today. I am a strong supporter of minority rights and a fighter against racism but even I saw the black guy in the movie clip as being armed. People need to realize how damaging subtle/unconscious racism is.

  11. #12 Azkyroth
    December 1, 2006

    I saw them both as being armed. I also saw the slide for barely a tenth of a second; I’m not sure if that was supposed to happen or if the timing’s off in my browser, but I had to use the slider to be completely sure of which picture was the black guy. If I was a cop on patrol, I wouldn’t fire based on a tenth of a second evaluation of a situation where a suspect might be armed; I’d make sure I was ready to if needed, though.

  12. #13 djexplorer
    December 1, 2006

    Whitegirl–

    I am a strong supporter of minority rights and a fighter against racism but even I saw the black guy in the movie clip as being armed. People need to realize how damaging subtle/unconscious racism is.

    Seems pretty obvious to me that the test was partially ‘cooked’. It DOES show some stereotyping, but it’s designed and described in such a way as to make you think it’s ‘proving’ more than it is about us.

    First of all the time of flash is EXTREMELY short. I actually thought there was something wrong with my flash player and that nothing was playing the first time I saw it.

    There’s just enough time to be able to tell the skin color and that there’s something in front of his face. I’m not even sure I could tell I was looking at a person, rather than an armed person, if I hadn’t been primed. But the subjects were primed. They were told that some of the subjects (>that’s a person in the background) were armed (>that blob in front of the background must in some cases be a gun) and some weren’t. You have to decide to shoot or not and the implication is you’ll be ‘graded’ on how often you’re right either way. The test is further designed so that it’s nearly impossible to tell the first time you view a slide where the blob is a gun or something else.

    So what we’re down to us testing whether people have a background belief (call it prejudice if you like) as to whether people have any background belief that in the absence of any other information and where they’re REQUIRED to make a decision either way, a black man is more likely to be pointing a gun at them than a white man is.

    And surprise, surprise, college students (and cops during the first series of 20 determinations) erred on the side of a black man pointing the gun at them than a white man. Irrational predudice, right.

    Actually not. It’s an accurate application of stereotype in a situation where all tendencies of people to not decide until they’re sure are carefully excluded by the experimental design. Further the prejudice (stereotyping) is quite mild. People hardly always or almost always thought blacks had a gun. Instead:

    They made an average of 3.63 errors over 20 trials when the suspect was black, but only 2.65 errors when the suspect was white.

    Still you say, however ‘mild’ this prejudice was or wasn’t, it’s still irrational, and hence odious!!!

    Actually, no. According to crime and victim statistics compiled by the FBI and Justice Department, blacks are 8 times as likely (per capita) to commit murder than whites are.

    Actually, I bet if they substituted male and female figures instead of black and white, the effort differential or “prejudice” against males would be greater than that against blacks in this test. Again the way that test is structured, it wouldn’t be at all irrational. Men ARE far more likely to be pointing a gun at you than women are.

    Remember, there isn’t any “too little info to responsibly make up my mind” option. The results are an example of rationally applying the odds to a situation where you literally have no other info than race, and that half the flashed faces have guns, and that if you decide wrong you’re dead. It’s a similar situation to the one in which you see three late teen/young 20′s guys dressed casual urban style a block away on the same side of the deserted street as you at 3am in a big city mixed neighborhood. Are you more likely to cross the street (or get out of there) if the three are black than if they’re white? Jesse Jackson famously admitted he himself would be more like to be real worried if they were black than if they were white. It’s the rational decision on very little info for anyone.

    In the real world we nearly always have more information than in that psych experiment. What most of us have long since learned to do, is to look for the other information, since color while not a completely irrelevant predictor, isn’t remotely as accurate as the individual details. Further life and death is very rarely involved for most of us.

    Of course those who are committed to the belief that irrational white racism is at the root of all white/black differences, and none come from anything else, will find ways to dismiss or ignore all of what I’ve just said. It’s a set of beliefs (those PC ones) the core of which are held with a religious like unshakable conviction by many, evidence be damned.

  13. #14 djexplorer
    December 2, 2006

    (My post immediately above was a mess – should have previewed it. Skip the one above. Here it is cleaned up a bit.)

    Whitegirl–

    I am a strong supporter of minority rights and a fighter against racism but even I saw the black guy in the movie clip as being armed. People need to realize how damaging subtle/unconscious racism is.

    It seems pretty obvious to me that the experiment (referred to in the post in chief) was designed looking for a result, and designed to suggest it shows more than it does. It DOES show some stereotyping. It doesn’t show the pervasive and irrational prejudice in almost all of us that you seem to think it does.

    First of all the time the images are flashed before us by the video player is EXTREMELY short. I actually thought there was something wrong with my player plugin and that nothing at all was streaming the first time I saw it.

    There’s just enough time to be able to tell skin color and that there’s something in front of his face. I’m not even sure I could tell I was looking at a person if I hadn’t been primed. But the subjects were primed. They are told that some of the subjects (>that’s a person in the background) were armed and some weren’t (>that blob in front of the background must in some cases be a gun). They and we have to decide to shoot or not, and the implication is they’ll be ‘graded’ on how often they’re right either way. They probably thought it’s a quickness of perception sort of test. The experiment is further designed so that it’s nearly impossible to tell the first number of times you view a series of slides where the foreground blob is a gun or something else.

    So what we’re down to is testing whether people have a background belief (prejudice) that a black man is more likely to be pointing a gun at them than a white man is, in the absence of any other information and where the experimental subjects are REQUIRED to make a decision one way or the other.

    And surprise, surprise, college students erred more often on the side of a black man pointing the gun at them than on a white man doing so. Irrational prejudice, right?

    Actually no. It’s an accurate application of statistical probability (in the form of a fuzzy stereotype) in a situation where all inclinations of people to evaluate particularized, individual evidence and not decide until they are sure, are carefully excluded by the experimental design. Further the prejudice (stereotyping) is quite mild. People hardly always or nearly always thought the black man had a gun. Instead:

    They made an average of 3.63 errors over 20 trials when the suspect was black, but only 2.65 errors when the suspect was white.

    Still you say, however ‘mild’ this prejudice was or wasn’t, it’s still irrational, and hence odious!!!

    Actually, no. It’s not irrational at all. According to crime and victim statistics compiled by the FBI and Justice Department, blacks are 8 times as likely (per capita) to commit murder than whites are. People may not know that number but they do know the direction.

    I bet if the experiments substituted male and female figures (provided they could be distinguished in so split second a time, which would require ingenuity in designing the images) instead of black and white, the differential error rate, or “prejudice” against males would be greater than that it was against blacks. Again the way that test is structured, that prejudice (stereotype) wouldn’t be at all irrational. Men ARE far more likely to be pointing a gun at you than women are. (Which doesn’t mean all men point guns at people or that no women do.)

    Remember, there isn’t any “too little info to responsibly make up my mind” option. The results are an example of rationally applying the odds to a situation where you literally have no other info, and that half the flashed faces have guns, and that if you decide wrong you’re dead. It’s a similar situation to one Jesse Jackson once described in a public talk. He described a person who sees three late teen/early 20′s males dressed in casual urban style, a block away on the other side of a deserted street in a big city mixed neighborhood at 3 in the morning. Is that person more likely to cross the street (or get out of there) if the three are black than if they’re white? Jesse Jackson famously admitted he himself would be more likely to be real worried if they were black than white. It’s the rational decision if there’s really that little to go on and your safety really could be at stake.

    In the real world we nearly always have more information or can wait to get it. What most of us do, is to look for that other, individual, information, and ignore the stuff that predicts really weakly at best.

  14. #15 Dave Munger
    December 2, 2006

    I should point out that the demo I used as an example was not the same as the stimuli for Plant and Peruche’s experiment. In their study, the photos remained on screen for 630 milliseconds, which is quite a long time during a controlled study, and an eternity when you believe someone is threatening you.

    The purpose of the flash demo was just to show how ambiguous a “weapon” can be, not to replicate the study. I doubt study participants were ever aware their reaction times were different for black or white suspects.

  15. #16 ompus
    December 2, 2006
    “if they substituted male and female figures instead of black and white, the effort differential or “prejudice” against males would be greater than that against blacks in this test.”

    I’d like to see this test.

  16. #17 djexplorer
    December 2, 2006

    Dave Munger said–

    I should point out that the demo I used as an example was not the same as the stimuli for Plant and Peruche’s experiment. In their study, the photos remained on screen for 630 milliseconds, which is quite a long time during a controlled study

    Ok. Interesting.

    Actually very interesting. I think what you’ve in effect done, by providing us with a VERY time shortened demonstration of the experimental test, is to perform a reductio ad absurdum on it.

    That is, in the actual experiment the 0.6 second viewing time may have made the application of particularized perception while difficult, still possible. Especially after getting used to the test and adjusting for it’s requirements. This is born out by the trained cops, who after 80 viewing trials stopped making significant numbers of errors skewed in one, black suspicious, direction.

    What was impossible in what you presented to us was merely very difficult in the experiment – that is to take great care to apply only particularized individual data to interpreting the case before them, rather than a fuzzy knowledge of background statistics.

  17. #18 Healer2K
    July 1, 2008

    Interesting article, but I have to disagree with a couple of comments, and offer some criticism.

    First of all, describing the group of test participants as “48 mostly white police officers” doesn’t quite give us the information we need. How many were white? How many were latino? Were there ANY black officers? How did the non-white officers respond, compared to the white officers? Is it possible that they responded better than the white officers, or vice versa? Far too vague.

    You go on to say, “but I do have one potential objection: police are also ‘trained’ by what they actually see every day as the patrol the streets.” I disagree. There are more white criminals in America than black criminals. Yet you don’t hear about police officers killing unarmed white suspects.

    ADDITIONALLY, you also say yourself that white STUDENTS took the same test, with the same tendency to shoot black people more often. THEY don’t see the same things the officers run into every day. So is what officers “see every day as they patrol” REALLY the causative factor here? I don’t believe that.

    Few people see more black criminals than we see IN THE MASS MEDIA every since day. The mainstream news reports black violent crime disproportionately to that of white violent crime. This can be verified by looking at the FBI’s “Uniform Crime Reports” annual document, and comparing it with studies done by FAIR (Fairness and Accuracy in Reporting) and other organizations.

    I believe the truth is, no group in America is as demonized as being as dangerous, predatory, and even evil, as black males. And after them, Latino males. So the officers are predisposed to believe that – conscious and/or subconsciously – before they even enter the academy.

    We have got to end racial bias, propaganda, and hatred in the American mass-media. It is THEY who train officers to kill innocent unarmed black people.

  18. #19 Healer2K
    July 1, 2008

    Additionally, the sample test makes an interesting point, but its flawed. On the streets, officers may have to make a quick call, but its rarely a matter of a FLASH FRAME in time. And I’m certain it is also rare that the innocent black victims of these shootings are holding objects in front of their faces, as if to aim them at the police officer.

    These shooting usually happen when the officers know fully well the ethnicity of the suspect. Yes, I understand that this test is meant to test bias, but all it can test is split-second knee-jerk bias. For that, it may be very valid. But it can also cloud the issue of how offers make a decision to fire. They know the suspect is black, loooong before they pull those triggers.

    So I believe this bias [and thus the remedy] goes far deeper than even this test would suggest.