Skip to main content Skip to secondary navigation

Practice Reduces Prejudice

Main content start
Black man's hands up, in left hand "don't" is written, in right "shoot"


Although Black men make up only 6% of the U.S. population, they are the victims of 40% of unarmed fatal police shootings.


Training people not to rely on Black stereotypes can save lives.

The Details

Social psychologists E. Ashby Plant, Michelle Peruche, and David Butz randomly assigned 123 college students to complete one of two first-person shooter computer games. Participants in the training condition game pretended they were police officers shooting at suspects on the computer screen. They saw a series of individual Black or White faces paired with either a weapon or a non-weapon (such as a cell phone or camera), and their job was to press the "shoot" button only when they saw a weapon. The researchers carefully designed the game so that equal numbers of White faces and Black faces were paired with the weapons. In other words, they broke the association between "Black" and "violent" that many Americans implicitly make. 

Meanwhile, participants randomly assigned to the control condition game pretended they were gardeners swatting at insects on flowers.

One day later, participants in both conditions came back to the lab. This time, everyone played the training condition game. The researchers discovered that participants who had played the training condition game on Day 1 shot fewer unarmed Black faces than did participants who had played the control condition game. In other words, playing a game that broke the link between "Black" and "weapon" trained participants to make better decisions in this shoot-don't shoot paradigm. 

Although the participants in this study were college students, research by Joshua Correll and colleagues shows that the training game can reduce racial bias among police officers, too.

Why This Works

On the first day, participants who played the training condition game learned that seeing a Black face would not help them identify a weapon, as White faces appeared with weapons just as often as did Black faces. In other words, participants updated their stereotypes about Blacks. Once they removed the stereotypical association between "Black" and "violent," participants became more accurate at distinguishing weapons from other objects.

When This Works Best

Demonstrating that race isn’t linked to crime likely helps all people adjust their stereotypes, but this technique works best when people are allowed to figure it out for themselves. In a similar study, instructing people to ignore race actually made them more likely to associate weapons with Black facesa paradoxical effect that researchers sometimes find when they ask people to suppress their thoughts. 

The Original Study

Plant, E. A., Peruche, B. M., & Butz, D. A. (2005). Eliminating automatic racial bias: Making race non-diagnostic for responses to criminal suspects. Journal of Experimental Social Psychology, 41,141-156. 

Related Studies

Correll, J., Park, B., Judd, C. M., & Wittenbrink, B. (2007). The influence of stereotypes on decisions to shoot. European Journal of Social Psychology, 37, 1102-1117.

In The Press

The Guardian

Related Topics

Bias Criminal Justice


Text by Lily Zheng

Photo CC by Christopher Camp

Thumbnail image via Clay Banks/Unsplash