ANN ARBOR – A new study based on body-worn camera footage capturing police-community interactions in Oakland, Calif., provides empirical evidence that police officer trainings can improve their interactions with the communities they serve. 

Findings published today in PNAS Nexus showed that police officers communicated more respectfully with drivers during traffic stops after they were trained in procedural justice. 

And more broadly, “the study demonstrated that body camera footage that goes largely unobserved can serve as a source of data to affect and measure change,” said Nick Camp, a social psychologist at the University of Michigan who led the study with Rob Voigt of Northwestern. 

Footage as Data

Body-worn cameras have been widely adopted over the past decade. The 2014 police killing of the black teenager Michael Brown in Ferguson, Missouri helped spark a national call for police reform and accountability to which body cameras have been a visible answer. As of 2016, almost half of U.S. law enforcement agencies and 80% of large police departments had adopted body cameras– figures that have only grown since, according to the report. But while police interactions with the public have been recorded at an unprecedented scale, only a small subset of this footage is ever viewed for accountability or evidentiary purposes.

For researchers like Camp and Voigt, whose previous work on traffic stop footage found police officers speak less respectfully to Black than to White drivers, however, body cameras offer a rich source of data for observing broad, agency-wide patterns in police-community interactions.

Does police officer training actually improve their interactions with community members?

“Until now, that question would have been challenging to answer,” said Camp. “Administrative records offer a limited window into police behavior. Ride-along observers are costly and logistically challenging, and could lead to unreliable results for assessing the effects of an intervention because the presence of a physical observer might lead officers to change their behavior.”

Measuring Respect

For the Oakland study, the researchers used recorded footage to examine changes in officer language following an 8-month procedural training conducted by the Oakland Police Department. The training included a module the research team helped develop to instruct officers in empirically-based ways of communicating respect during traffic stops: Greeting drivers; stating the reason for the stop early on; offering reassurance; expressing concern for safety, and using formal, rather than informal titles of address.

The researchers applied natural language processing tools and expert annotations of the traffic stop reportings to see if officers enacted these five recommendations.

Compared with recordings of stops that occurred prior to the training, they found that officers used more of these techniques after training; in particular, they were more likely to express concern for drivers’ safety, offer reassurance, and provide explicit reasons for the stop. These findings were consistent across demographic groups.

“It is meaningful to be able to observe how and when policy-community interactions change, or do not change,” said Camp. “These interactions play an important role in how community members perceive the police, and have implications for addressing systemic racism and the building of police-community trust.”

Camp also said he hopes this study will be the first of many investigations that will use body cam recordings to examine behavior change: “Reformers and agencies nationally are trying to affect change, and this study shows an opportunity to measure it.”

Group Dynamics

In related work, Camp is currently conducting a study at the Research Center for Group Dynamics (RCGD) HomeLab at the University of Michigan Institute for Social Research, assessing the physical and psychological impact of police disrespect across race. 

In this study, Black and White community members listen to clips that other participants have rated as more positive or negative interactions. The study team then observes how they react, measuring their physical stress and how their impressions spread to how they think about broader departments.

“What I like about this project is that we’re combining the real-world observations from body cameras with rigorous, controlled research on mechanisms. So far, we’re finding that Black and White participants agree in their perceptions of specific clips, but Black participants are more likely to generalize those perceptions to the police department,” said Camp.

Camp is currently the organizer of a weekly seminar series on the “Social Psychology of Systemic Racism,” offered by RCGD this fall; he presented his work on the psychology of policing and being policed in the series yesterday.

RCGD fills a niche in the social sciences by observing human behavior in social contexts, with research that can often directly address real-world problems.

Camp is an affiliate of RCGD and the Department of Organizational Studies at the University of Michigan. The body cam research team also included MarYam G. Hamedani, Dan Jurafsky, and Jennifer L. Eberhardt, of Stanford University.

Contact: Tevah Platt