Despite the long history of police violence against racial minorities in the United States, recent high-profile shooting incidents of unarmed racial minorities have gained national attention, such as the shooting deaths of African Americans Trayvon Martin and Michael Brown. The resulting social movements, led by Black Lives Matter, have refocused media attention on the roles that race and racial stereotypes play in police behavior. What causes both officers and community members to shoot unarmed racial minorities? What role do racial stereotypes play in this process? These are some of the questions that my Gender, Race, and Sexual Prejudice (GRASP) lab studies, trying to parse apart what might be responsible for the disproportionate number of African Americans killed by police each year.

Social psychologists have studied the causal role of racial stereotypes on decisions to shoot by taking an experimental approach and using shoot/do not shoot video simulations. In these studies, participants view images of Black and White individuals who are holding either neutral objects, such as a wallet or cell phone, or a gun. Participants must make the quick decision of whether or not to shoot the individual, depending on what object they are holding. Consistent with racial differences found in real-world police shootings, participants showed a bias in which unarmed Blacks are more likely to be mistakenly “shot” compared to unarmed Whites, and armed Whites are more likely to be mistakenly “not shot” compared to armed Blacks. This phenomenon has been termed shooter bias. Implicit racial stereotypes linking African Americans with criminality and danger altered perceptions of threat, leading to a higher tendency to see Blacks as threatening enough to shoot. Both police and community members can hold implicit stereotypes and demonstrate shooter bias.

Our recent research expanded on the shooter bias effect by studying how other factors in the situation interact with suspect race to impact decisions to shoot. Specifically, it focused on how neighborhood context and suspect clothing combine with suspect race to influence individuals’ decision-making. Our first study used real neighborhoods that are generally viewed as dangerous versus safe. We hypothesized that shooter bias would be higher in perceived dangerous neighborhoods by making the racial stereotype of Blacks being dangerous more accessible, leading to more mistaken shooting decisions against Blacks. Oppositely, the perceived safer neighborhood would counteract the racial stereotype and reduce bias.  Because our sample was in Los Angeles, we chose Beverly Hills as the “safe” neighborhood and South Central as the “dangerous” neighborhood. Student participants were subliminally primed with each neighborhood name while completing the shooter bias task. As we predicted, shooter bias was higher in the perceived threatening neighborhood and reduced in the perceived safe neighborhood. These results suggest that Black individuals in areas considered “dangerous” are more likely to be mistakenly shot when compared to White individuals in those same areas.

Our next study examined how shooter bias is influenced by the style of clothing worn by the suspect. After the death of Trayvon Martin, some individuals speculated that Trayvon’s hoodie, combined with his race, made him seem more physically threatening which may have lead him to be shot. Thus, in our second study, we examined the causal role of perceived “threatening” versus “safe” attire in decisions to shoot Black and White suspects. We predicted that Black suspects in perceived threatening attire would have an increased risk of being mistakenly shot, compared to those in perceived safe attire. This would be because participants’ stereotype of Blacks as dangerous criminals would be more accessible when the suspect’s clothing primed them to think of physical threat. In the shooter bias task, we had Black and White subjects holding neutral objects or guns appear in either attire stereotypically associated with criminality and danger (e.g., baggy pants, sweatshirts) or attire stereotypically associated with professionalism and safety (e.g., a business suit). We found that, consistent with our expectations, Black suspects faced more shooter bias when they wore stereotypically threatening attire compared to when they wore stereotypically safe attire. There was no effect of suspect clothing on decisions to shoot White suspects. These data suggest that the type of clothing worn by a suspect increases their risk of being mistakenly shot, but only when the shooter holds the stereotype that members of the suspect’s racial group are more prone to violence (i.e. when the suspect is Black).

While we cannot directly say what caused any specific police or community member shooting of an unarmed individual, these studies identify the conditions that increase the likelihood of biased decisions. More specifically, these studies show that racially biased shooting of Blacks is more likely if the Black person is wearing a hooded sweatshirt or lives in a predominately Black neighborhood, which is often viewed as less safe. Efforts to change shooters’ decision-making biases may involve training, intervention, and education. In light of these goals, our GRASP lab is currently working with police departments to deliver implicit bias training to counteract this bias.


Dr. Kimberly Barsamian Kahn is an Associate Professor of Social Psychology at Portland State University, and leads the Gender, Race, and Sexual Prejudice (GRASP) Lab.