Almost everyone knows the problem. In the United States, black and Hispanic people are more likely to be stopped, searched, arrested, beaten, and shot by the police (National Academies of Sciences, Engineering, and Medicine, 2018).

The reasons for these disparities are less well understood. A popular explanation is the implicit bias hypothesis, which makes two claims. First, that police officers as a group harbor unconscious biases against blacks and other minorities. Second, these biases, which exist outside of consciousness, lead police officers to treat blacks and Hispanics less favorably than whites.

To reduce racial disparities in policing, many police departments across the United States have held seminars on implicit bias. These trainings are unregulated and largely untested, although three recent studies shed light on their effectiveness.

Three studies on implicit prejudice training

As part of one study, researchers conducted a four-hour training on implicit bias for 85 police officers (63 percent male, 39 percent white) who completed an online survey in the days before the workshop and again immediately after the workshop.

The researchers found that most officers changed their minds about the value and importance of implicit bias education. They believed that bias education was “in their best interest and could improve their relationship with the public” (Vitriol, Banaji, & Lowe, 2024). Unfortunately, the researchers only measured self-reported attitudes, which may not predict actual behavior.

In another study, researchers examined the effects of a one-day training on implicit bias on the knowledge, attitudes, and intentions of 3,764 police officers (Lai & Lisnek, 2023). Officers who completed the training had more knowledge and were more concerned about racial bias. They were also more motivated to address bias in their behavior as police officers. Unfortunately, the effects of the training were short-lived. After one month, concerns about bias and intentions to police in a less biased manner returned to pre-training levels.

In a third study published last month, University of Albany researcher Robert Worden and his colleagues examined the effects of a one-day training on implicit bias on the frequency of police stops, arrests, searches and use of force. The training was administered to 14,471 police officers (81 percent male, 44 percent white) in New York City.

The goal of the training was to (1) increase officers’ knowledge of the harmful effects of unconscious bias and (2) teach them specific techniques to help them manage their own thoughts and behavior. Officers were taught to perform self-checks, avoid profiling, use “slow” responses, and thoroughly assess situations to avoid ambiguity.

When Worden and his team examined policing outcomes over time, they found that implicit bias training had no impact on racial and ethnic disparities. Before training, 55 to 60 percent of police stops in New York City involved black citizens and 30 percent involved Hispanics. These numbers did not change after training (Worden et al., 2024).

A second pattern was also unaffected by the training – and that may be good news. Although blacks and Hispanics are more likely to be stopped by police in New York City, those stopped were not more likely to be searched, beaten, or shot because of their race or ethnicity. In other words, racial differences are limited to police stops. After a suspect is stopped, police officers do not treat blacks, Hispanics, and whites differently, according to this study.

In short, implicit bias management training in New York City had no impact on officers’ policing decisions and failed to reduce racial disparities in policing outcomes.

Why don’t diversity training programs work?

Careful reviews of the effectiveness of anti-racism and implicit bias programs have reached a discouraging conclusion: They appear to have little or no positive long-term effects (Singal, 2023). Why don’t they work?

First, most programs are not designed to reduce implicit bias per se; they aim to make individuals aware of possible prejudices.

Second, mandatory workshops often lead to defensive reactions and what psychologists call “reactance.” When a new rule or policy threatens our autonomy, we are motivated to restore our independence by ignoring or even contradicting the rule.

Third, implicit bias is a controversial concept regarding its definition, measurement, and (in)ability to predict actual behavior. Some behavioral scientists argue that implicit bias is not a useful scientific concept and should be abandoned. (Explicit bias, on the other hand, is measurable and predicts discriminatory behavior.)

Finally, racial disparities in police stops, arrests, and use of force may not necessarily be the result of racial bias. They may also exist for reasons that have nothing to do with a police officer’s mindset. Studies have found that violent crime rates are higher in economically and socially disadvantaged neighborhoods. These neighborhoods are more likely to have Black or Hispanic residents (Worden et al., 2024).