How to improve policing with evidence-based training

Print More
Getting your Trinity Audio player ready...

The killing of George Floyd on May 25, 2020, by Minneapolis police officer Derek Chauvin, ignited widespread protests and a renewed focus on systemic issues within U.S. policing. The incident highlighted longstanding concerns about racial discrimination, excessive force, and accountability within law enforcement, particularly against Black Americans. It led to mass demonstrations under the banner of the Black Lives Matter movement, demanding police reform, the defunding of police departments, and a reevaluation of the criminal justice system.

Amitrajeet A. Batabyal

The killing of Floyd—and locally, Daniel Prude’s death after he was restrained by Rochester Police Department officers and placed in custody—has served as a painful reminder of the need for comprehensive reforms to address police brutality and to ensure that all individuals are treated with fairness and dignity under the law.

Generally speaking, there are two views about the causes of adverse policing outcomes. One view is that the problem is caused largely by a small group of problem officers who either use excessive force or make arbitrary arrests. In fact, there is some evidence to suggest that approximately 2 percent of officers are responsible for 50 percent of cases of misconduct. The second view is that the problem is largely the result of poor police department regulations.

These two views are reasonable in the sense that they focus either on individual officers or police department regulations. Even so, these views completely overlook a vital aspect of policing: the fact that police work requires an officer to make complex decisions in circumstances that cause stress, trigger multiple emotions, and necessitate quick action. A vast body of research in psychology shows that high cognitive demands undermine decision-making, resulting in humans making quick, intuitive judgments (System 1 thinking) and spending a lot less time on considered responses (System 2 thinking). The key point to grasp is that high cognitive demands can lead humans to interpret the situations they confront very narrowly.

This matters greatly in a policing context because when faced with high cognitive demands, officers are likely to act without thinking through alternate interpretations of a situation they confront, and this may well lead to mistakes, negative interactions, and adverse outcomes. So, for instance, a Black teenager who runs in response to a police officer’s command to stop is not necessarily guilty of some crime but may simply fear the police officer.

It is, of course, impossible to remove the high cognitive demands that are a part of police work, but a significant research question is this: Is it possible to improve policing outcomes by training police officers to navigate these high cognitive demands? Important new research by Oeindrila Dube and her colleagues addresses this question comprehensively. These researchers develop and evaluate a new training tool called Situational Decision-making (Sit-d), which combines a thorough understanding of routine policing with insights from behavioral science to see if police officers can be trained to more deliberately process information and make decisions. Specifically, the Sit-d tool first trains officers to recognize circumstances that are likely to cause stress and impose cognitive demands. Next, it teaches officers about particular cognitive biases that they may confront in these circumstances. The evaluation was conducted with a so-called randomized controlled trial with more than 2,000 officers from the Chicago Police Department.

Three results emanating from this research are thought-provoking. First, we learn that relative to the so-called control group of police officers, across several metrics, the Sit-d trained officers considered a wider range of potential motivations behind an individual’s behavior, they recalled more information that went against their initial assumptions, and they were more likely to update their responses as circumstances changed. Second, the Sit-d trained officers reduced their use of non-lethal force and discretionary arrests by 23 percent. Finally, even though the focus of Sit-d training was not on racial bias, officers with Sit-d training made 11 percent fewer arrests of Black subjects.

Even though the effects of Sit-d training do tend to fade over time, what is gratifying about this approach is that it costs about the same as other kinds of training used by large police departments, its benefits are large, and hence Sit-d training easily passes a “benefit-cost” test.

Police departments and policymakers need to concentrate a lot more on this kind of evidence-based training if they genuinely want to improve policing outcomes.

Amitrajeet A. Batabyal is a Distinguished Professor, the Arthur J. Gosnell professor of economics, and the Interim Head of the Sustainability Department, all at Rochester Institute of Technology, but these views are his own.

The Beacon welcomes comments and letters from readers who adhere to our comment policy including use of their full, real nameSee “Leave a Reply” below to discuss on this post. Comments of a general nature may be submitted to the Letters page by emailing  [email protected]

3 thoughts on “How to improve policing with evidence-based training

  1. I would note that historically Rochester area police departments have for decades been a leader across the country in terms of quantity and quality of academy and in-service police training. In my opinion, the question of whether improving police training will produce better outcomes may not be the most relevant issue for Rochester. I would suggest that the more interesting question is, “providing that current police training is valid and sufficient would changing the environment in which police officers must respond and navigate complex matters produce better outcomes?”
    Being understaffed, overworked, vilified and unsupported by elected and community leaders is going to do more to undermine what any established or additional training philosophies may seek to improve.
    I echo Gary’s reminder that the Prude matter was not police brutality, nor was it a few bad apples. The issues stemmed from an insufficiency of necessary resources in one sector and suboptimal leadership decisions within both city hall and the police department, particularly regarding transparency and timeliness.
    The common trend for the last few years in improving policing has been in addressing the issues from the ground up instead of from the top down.
    Apparently while there is a shortage of police officers across the country there is not a shortage of expert police chiefs and administrators judging by the revolving door of leadership that we have seen here and across the country.

  2. The idea that evidence based training for police is worthy of consideration and implementation.

    An area of further evidence based research would be to take a look at the culture of police departments in terms of norms and attitudes, specific training in descalation strategies, community policing, and supervision. The culture is heavily influenced by the chief and the command structure.

    The bad apple theory of police misconduct has been discredited. This theory begs the question of where did the bad apple come from and why is it still on the force?

    Also, the theory that regulations and “standing orders” mitigate dysfunctional behavior by participants in the system is flawed because of the informal norms and attitudes in the department which is often taught and acted on “under the radar” as part of the informal organizational dynamics of “how things are really done here.” This dynamic of the hidden organization is a management responsibility.

    I appreciate the ideas in this article. Sit-d is a step in the right direction and needs to be amplified, expanded, and supported by leadership and command.

  3. Agreed, this is what the activists, like City Council member Stanley Martin do not seem to understand, training matters; and training costs money. If you want better trained officers it will require resources , and resources cost. Instead of defunding the police the better response is to increase funding for training. (Just as an aside, the police did not brutalize Daniel Prude. Even AG James had to agree to that. It’s unfortunate many continue to repeat this distortion.)

Leave a Reply

Your email address will not be published. Required fields are marked *