Do Stereotypes of Black Criminality Change the Way We See?

Yesterday, I noted that, “Studies show that black men and boys are already heavily associated in the American mind with criminality.”  I thought today I’d give you all an example of this, using a series of studies co-authored by Perception Institute research advisor Philip Atiba Goff and published under the title “Seeing Black: Race, Crime and Visual Processing.” 

The studies were an examination as to how stereotypes linking African Americans with criminality shape how we perceive the world using a technique called subliminal priming. When you subliminally prime a test subject, you expose them to images that they do not realize they see. Their brains process the images without them being consciously aware.

 Here’s how the five studies worked:

In the first study, researches divided subjects into three groups. One group was subliminally primed with images of black faces, the second with images of white faces and the third with meaningless line drawings. 

They were then given a task where they were asked to view a fuzzy, distorted image of an object that slowly became clearer. They were instructed to press the space bar (freezing the image) when they knew what the object was and write down the name of the object. Some of these objects were related to crime, some were not. 

The White faces group took (on average) over thirteen seconds. The neutral group took around eleven. Meanwhile, the group of test subjects primed with Black faces were able to discern crime-related objects in (on average) under three seconds

For the next study, researchers essentially ran the first study in reverse.  They subliminally exposed people to objects that were either crime-related or not and then had them complete a “dot-probe” task during which two faces (one Black and one White) appeared simultaneously on a computer screen and then swiftly disappeared, replaced by a dot in the visual location of either face. The subjects were tasked with finding the dot as quickly as possible. The crime-primed group found the Black face position dot probe roughly 300 milliseconds faster than the White one. The no-primed group found the White face position dot probe roughly 300 ms faster than the Black one. 

Like criminality, basketball is highly associated with black people but, unlike criminality, these associations are largely positive. In the third iteration of the study, the researchers repeated the second study, but with basketball related images instead of crime related images. The results were largely the same, but the basketball-primed group found both white and black faces faster.

Study 4 expands on the findings of studies 2 and 3.  The researchers set out to answer whether or not our attention lingers longer on black faces after subliminal priming, whether or not our memories of what the faces looked like is affected, and what the effects of all of this are on police officers.

Study 4 expanded on studies 2 and 3. It used law enforcement words (violent, crime, stop, investigate etc) instead of crime images, and used  five faces for each race that had been rated by a pilot group for stereotypicality. Once again, the Crime Prime group found the Black Face Position Dot faster than the White Position Dot and much, much faster than the No Prime group.

Here’s where it gets interesting, and troubling: After the police officers completed the find-the-dot task, they were asked to identify the faces they had been looking at from a lineup. Officers in the Crime Prime group were more likely to mistakenly choose more stereotypical faces of both races (with a more significant trend with regard to Black faces) when compared to the No Prime Group.  In other words: law enforcement related imagery changed the way that the police officers in the study remembered black people’s faces, making them appear in memory to be more stereotypical than they actually were. 

The final study was much simpler in design then the first four.  Police officers were divided into six groups. Half of these groups looked at slides of Black male faces, the other of White male faces. 1/3 of the officers rated the faces on stereotypicality, 1/3 on attractiveness, and the other 1/3 were asked to answer whether or not the faces appeared criminal to them or not. The results showed that police officers were more likely to rate Black faces criminal than White faces, and, the more stereotypically Black the face, the more likely it was to be rated criminal. 

  The use of police officers for the fourth and fifth iteration of the study should not be taken as an indictment of law enforcement. After all, the subjects int he first three tests were not police officers and demonstrated similar results. What tests four and five show instead is that whatever training is currently place is not adequate to the task of addressing the effects of implicit bias.

Taken together, the implications of these five studies are disturbing, and show how, despite the fact that the majority of crimes are committed by white people, stereotypes about black criminality remain quite powerful.  ”Seeing Black” shows how these stereotypes can affect both the way we see the world and the way we remember it.