Speaker Explains How Implicit Bias Leads to Biased Policing

 Keynote Speaker Michael Roosevelt talks about unconscious bias at the SF Justice Summit in San Francisco

Keynote Speaker Michael Roosevelt talks about unconscious bias at the SF Justice Summit in San Francisco

Michael Roosevelt, an expert in implicit/unconscious bias, who has trained thousands of attorneys, judges and police officers, gave the keynote address at the San Francisco Public Defender’s Justice Summit.

He opened his remarks stating, “It is critical at this juncture that we let no one off the hook. It is time to address critical issues around the issue of race and implicit, now what we call, unconscious bias.”

He explained that a lot of time is spent on explicit bias, which he said “are the things that we hear, we see, we observe, and we call it racism or white supremacy.” However, he said, “The fact of the matter is that we’re all hardwired in our unconscious minds, that we are not aware of, that we have those unexpressed in a verbal way, but they get expressed in other kinds of ways.”

Mr. Roosevelt explained that the brain is really primed to make a lot of quick decisions based on a lot of information. “So the research shows, the people that study the brain, it shows that we tend to categorize. This is a normal process.”

“We have a preference for people based on group identity,” he continued. “This is a fundamental process. That we look at something and we place it in a category.  Our ability to categorize,” he said, “is really indispensable. We can’t do without it.”

We have to find these shortcuts and a way to describe our social reality. Mr. Roosevelt cited three categories that we immediately identify – race, age, and gender.

Race, he said, is really important for something called schemas. He explained schemas in this way: “Our brains encode information about groups of people into our memories along with favorable or unfavorable impressions and values. These are called constructs.”

Michael Roosevelt likened our brains to computers in this sense that they are programmed. This programming, he explained, comes from all our experiences – the media, culture, background, family – and “it informs the information that goes into our brains.” This encoded information “becomes part of the associations that we make.”

He next discussed in-groups and out-groups. “There are in-groups and out-groups that have negative associations,” Mr. Roosevelt explained. “When you see those groups of people, people sometimes walk across the street at night.”

“I call it ‘Michael with a hoodie on.’ When I have my nice suit on, there is a different reaction to me,” he continued. “But after 5 pm, I have my hoodie on, implicitly I become a danger. I become somebody in the out group. No longer the in-group.”

“We tend to look at the other people as all the same and we look at our group as unique,” he said. “This is critical. Sometimes we make in the brain, what we call exceptions to the rule.” Or, as he put it, imitating the effect: “You are not like the other black people. I don’t think of you as being black.”

Stereotypes are important concepts. “These are things that we generalize about a group, based on some shared characteristics,” he said. “Often times stereotypes involves projection – what we want to believe about that group.” This is often related to the fact that we have had limited experience of interactions with that group.

“These stereotypes can be dangerous,” he warned.

Michael Roosevelt talked about a game called “shoot/no shoot.” The participants were shown people holding an object that is ambiguous as to whether it is a weapon, and asked to determine if they are a threat and therefore should be shot.

“What they found is that people shoot more quickly if the person holding the object is black than when he is white,” he said. “They shoot more quickly. There’s something about the brain that says black man, danger. It triggers an immediate response.”

Mr. Roosevelt explained that we saw that in action with the Tamir Rice situation, where the young African-American kid had a toy gun. According to the report, the officer said “stop drop it, three times,” over about a two and a half second period of time. “The car was still rolling when they shot and killed him,” he told the audience.

“They made a quick association, shot and killed him,” he said.

Interestingly enough, Mr. Roosevelt said, the study found that police are actually a little better at shooting the right targets with the gun than the average citizen. A little bit better, but not much better.

However, he said, “Officers do poorly when it comes to figuring out gangs. They tend to shoot more quickly and accurately when they’re in an area that’s sort of gang related.”

How did they learn this? he asked. Their training, he explained. Mr. Roosevelt told the story of how a Florida police department used mugshots of black men for target practice.

The mugshots that were used for target practice by a Florida police department
The mugshots that were used for target practice by a Florida police department

“The only reason this became public is that one of the persons who was in the mugshot, his sister was in the military and she somehow saw and was shocked” to see her brother’s image used for target practice, Mr. Roosevelt said.

“We had cops who were trained for years and years and years, about shooting real black bodies,” he said.

The chief defended the department “but denied racial profiling and said officers used  images of people of all races.”

Michael Roosevelt cited other research that showed that blacks with darker skin tones, the longer the sentence you are likely to get. A Stanford study showed “that the darker you are, the more likely you are to get the death penalty.”

“Studying this, it’s critical that if you don’t change the mindset, the moving that brain to thinking about people differently,” he said we will end up with more shootings of African-American men and they will get more time in prison.

Implicit bias, Mr. Roosevelt explained, really doesn’t change based on intelligence level. He explained, “Cognitive scientists also believe that these early stereotypes or beliefs (i.e. implicit biases) about groups of people continue to exist at an unconscious level, despite our brain’s increasing ability to reason, understand, and exercise judgment as we mature.”

These biases he explained, are also environmental. For example, “if we are in a certain neighborhood, we know we need to be more ready to shoot. In a certain kind of environment, we’re more prone to act in an aggressive way.” He said, “We call this affective realism.”

Michael Roosevelt then talked about the text messages, which he said are not “implicit bias, but rather San Francisco bias.” These are police officers, he said, pointing to the transcripts on the screen, exchanging racist messages on the public dime “and calling African Americans, monkeys.”

“One would say, this wouldn’t happen in San Francisco,” he said. “Well let me tell you – it happens in San Francisco. That’s not implicit bias, but it comes from a place where they learn unconsciously to devalue groups of people.”

He spoke about Yulanda Williams, he read the explicit portion of that exchange and pointed out, “This is a colleague. They thought they could express that form of racism and bias and there would be no accounting for it.”

“People often say this, our young people are less biased today than we were… the stats don’t support that,” he said. “Witness what happened on that bus when they were all singing the song about ‘no niggers can join this club.’ A fraternity. These were young college students. They weren’t born in 1957.”

He said, “It is not about age, it’s about how the environment, how the world that you’re in, how your culture shapes this.” He said, “It’s critical that we understand this.”

Michael Roosevelt gave an example of a member of the Oakland Police Department who has killed four African-American subjects. “Now how does that happen?” he said, noting that the problem goes beyond San Francisco. “Most officers never pull their guns. You have a trigger happy, wild eyed murderer out in the police department.”

And yet he’s still in the police department. “What does he get? A slap on the wrist,” he said. “A slap on the wrist does not address bias.”

In order to address bias, he said, we need to first have accountability. He said, without accountability, you cannot address bias. That means, he said, “Community people, policing the police.” He said they can’t rely on the “fox to guard the chicken, we know the outcome.”

He would add that “training without accountability is useless.”  We need to get different results, if we don’t get different results, he said, we need to revisit what the training has or has not accomplished.

—David M. Greenwald reporting

About The Author

David Greenwald is the founder, editor, and executive director of the Davis Vanguard. He founded the Vanguard in 2006. David Greenwald moved to Davis in 1996 to attend Graduate School at UC Davis in Political Science. He lives in South Davis with his wife Cecilia Escamilla Greenwald and three children.

Related posts


  1. Davis Progressive

    this is probably the most important piece of the day.  we constantly hear the rejoinder that the police aren’t racist.  and several people have noted here that they don’t have to be overtly racist – the implicit bias theory i think is a powerful explanatory theory for how people who do not believe they are biased or racist are capable of biased policing.

  2. Frankly

    I urge everyone everywhere to reject this new narrative from the race obsessed left of “implicit bias”.  This is them reaching for that never before thought to be attainable goal of reading people’s minds.  Well, not so much reading peoples minds, but finding them guilt of thoughts that they absolutley have just because of the group they belong to.

    How about instead of trying to stamp out implicit bias that we have to guess exists, we instead stamp out explicit ignorance that we clearly see?

    1. Don Shor

      Maybe we could get further if we focused on behaviors and outcomes, and worked less to impute motives to the people involved. Focus on police tactics, rather than on their bias. I think you’d get little opposition to proposals for better police training, better communication, more transparency, possibly even community oversight. Trying to decide if someone or some organization is biased just becomes a distraction.

      1. Davis Progressive

        if people are being treated differently based on race, then you have to address that issue – though i agree better police training, some of that better training includes ways to address their unconscious biases.  that’s what the school district did with regards to the achievement gap.

      2. Miwok

        Let’s see – perp sees cop, perp runs. Toward or away, runs. Implicit bias says “might be a threat”. Who cares what race? The act of a sudden physical movement is an alert to another person. One black guy in the middle of a crowd of black guys runs, he gets the attention. What “implicit Bias” is there?

    2. Tia Will


      OK, so which part of implicit bias do you not believe in ? Do you not believe that our ability to discern color, gender, and relative age is so rapid as to be automatic ?  Do you not believe that people make judgements based on what they see and automatically register ? Then how would you account for stopping a child about to step into the street, while letting an adult go if there is not a clear and present danger that they have not seen. How would you label a single heterosexual females automatic decision of potential interest in an age appropriate male across the room if not implicit bias since she would not show the same behavior to a child or an adult woman ? You yourself have frequently noted “tribalism” as an inherent human precept. So how do you see that as any different ?

      Is it so very hard to believe that there could be negative automatic judgements and actions on those just as their are neutral or even life saving automatic behaviors that humans engage in as in the above examples ?



      1. Frankly

        “Implicit bias” is a term that is used by those stuck in a victim mentality and racism template to accuse, convict and blame white people for the continued plight of black people.   You pull a lot of examples that detract from this point.  We are talking about race here.  So why not call it “implicit race bias”?

        One thing that separates other animals from humans is our ability to reason and contemplate.  Maybe you yourself worries about your own personal implicit race bias, but not me.   I think things through.  I weigh pros and cons, costs and benefits, risks and probabilities.  None of this is implicit, it is all calculated.

        It is a huge slippery slope to give in to this BS that someone is guilty of racial bias just because he exists.  It is a sad, sad situation that lacking evidence of bias, instead of turning introspective and inward, the race-obsessed crusaders just double-down on this new narrative that all white people are just racist despite what you say or do.

        Sort of like Hitler’s opinion of Jews having this implicit racial bias against non Jews…

        The black-haired Jewish youth lies in wait for hours on end, satanically glaring at and spying on the unsuspicious girl whom he plans to seduce, adulterating her blood and removing her from the bosom of her own people. The Jew uses every possible means to undermine the racial foundations of a subjugated people.

  3. TrueBlueDevil

    The devil is in the details. Who crafted these “studies”, what were the questions, who were the participants, who funded them?

    Two recent on-the-ground studies show the exact opposite in regards to police officers. The first – mentioned previously – showed that police officers were actually reluctant to engage or shot back suspects, as they knew the intense scrutiny such encounters would spur. They feared for their jobs.

    In a newer study, Peter Moskos, assistant professor at the John Jay College of Criminal Justice at the City University of New York, used data from the website “Killed by Police”. 

    “If one adjusts for the racial disparity in the homicide rate or the rate at which police are feloniously killed, whites are actually more likely to be killed by police than blacks,” said Mr. Moskos, a former Baltimore cop and author of the book “Cop in the Hood.”

    Read more: http://www.washingtontimes.com/news/2015/apr/21/police-kill-more-whites-than-blacks-but-minority-d/#ixzz3ZBoGvqsG 

    Note that New York City police officer Brian Moore was just shot in the face and is in critical condition. The officer is white, the suspect, Demetrius Blackwell, is black. The past three police officers who have been shot in New York have all been shot by black men, and there have been 5 police officers shot (targeted) in New York the past 5 months.

    1. Davis Progressive

      i think these kinds of studies miss the core issue that is addressed by mr. roosevelt – are police officers more likely under some circumstances – such as being in a high crime area – to shoot a black than a white.  the study that mr. moskos performs doesn’t get to that question – it assumes that the homicide rate differences answer the problem.

      btw, the police officer in new york died this morning.  but that doesn’t have a bearing on this issue.

  4. TrueBlueDevil

    Mr. Roosevelt apparently works for the city, but I’m not sure what makes him an “expert”. I saw where he has 2 psychology degrees, but it didn’t say from where. I tried to google the Oakland police officer who has allegedly shot 4 suspects, but I couldn’t find the individual, though I did find the crime were four police officers were killed in Berkeley / Oakland in 2009 by Lovelle Mixon.

  5. Tia Will


    showed that police officers were actually reluctant to engage or shot back suspects”

    I suspect that what was actually demonstrated was a “lag” time in the shot taken when the individual portrayed was black. The remainder is speculation about why this lag time may have been observed.

    I think that this is moving from an observation to an uncertain conclusion of why the effect is observed. Data from a study is objective. Interpretation of that data frequently is not.  In making this assumption, the far bigger point is missed. Bias is bias regardless of whether it leads to a greater or lesser likelihood of shooting. Bias on the basis of race implies differential treatment based on race. It does not in and of itself state whether that bias is for, or against a given group, merely that it exists and race is the cause of the differential treatment.

    Let’s step outside the race issue again for a moment. Let’s take a  gender based bias. A man recognizes that another human being on the street is a woman. If he reacts towards her differently than he would if that human were a man, he is demonstrating gender based bias. Now this may work in her favor, if he for instance he offers to hold a door for her or carry a heavy object, or it may work against her if he decides to pay her unwanted sexually based “compliments” or harass her. Whichever is the case, if he would not behave the same with a man, he is acting on bias.

    Bias is not necessarily a dirty word. But to deny that it is occurring when it clearly is present, for good or bad, is a very simplistic notion of how the human brain processes information before we are even aware that it has done so.


Leave a Reply

X Close

Newsletter Sign-Up

X Close

Monthly Subscriber Sign-Up

Enter the maximum amount you want to pay each month
Sign up for