Vehicle stop studies explain the importance of an officer’s first words

Eugenia Rho believes in the importance of first impressions, especially when the vehicle is stopped.

Eugenia Rho believes in the importance of first impressions, especially when the vehicle is stopped.

An assistant professor in the Department of Computer Science, Rho is the lead author of a new research paper that illustrates how a law enforcement officer’s first 45 words when a vehicle stops with a black driver can often indicate how the stop will end.

“We found that there were key differences in the way officers spoke to black drivers during the first moments of a stop that ended in arrest, handcuffs, or a search versus those that didn’t end in such an outcome,” said Rho, who leads the Society, AI, and the Language research laboratory (SAIL) at Virginia Tech. “Simply put, officers start with orders rather than reasons in escalating stops.”

Published in Proceedings of the United States National Academy of SciencesPeer-reviewed studies have also found that black men can often predict quitting outcomes simply by listening to the same 45 words, which generally last less than 30 seconds.

“There are clear linguistic cues for increased vehicle stops. It was seen by trained coders, computational language models, and perhaps most importantly, by black male citizens,” said Rho.

Rho began this research as a postdoctoral researcher at Stanford University, working with Jennifer Eberdhardt, a professor of organizational behavior and psychology, and Dan Jurafsky, a professor of computer science and linguistics.

The research team, which also included researchers from the University of Michigan, analyzed audio recordings and transcripts from police body cameras of 577 vehicle stops that occurred over a month in a medium-sized, racially diverse US city. data includes stops that result in arrest, handcuffs, or searches and those that don’t, but doesn’t include stops that use force.

One of the reasons the team decided to focus on black drivers is that they are disproportionately represented in the data, according to Rho.

“We limited the study to black drivers because less than 1 percent of the increased stoppages included non-black drivers in our sample,” said Rho. “We included male and female drivers, but the stops that increased were dominated by male drivers.”

The data was used in two studies included in the paper, one focused on the language officers use at the start of a traffic stop and the second aimed to better understand black men’s perceptions of hearing the same words. The third part of the paper includes a case study that examines the first moments of a traffic stop involving George Floyd in May 2020.

Dissect dialogue

In the first study, researchers used computational linguistics and hand annotations to analyze transcripts by identifying dialogue actions, such as greetings, commands, questions, excuses, and others.

“Dialogue actions are like conversational roadmaps, says Rho. “They not only show what the speaker is trying to do – like ask a question or give an order – but also how that part of the conversation fits into the larger discussion, helping guide what might be said next.”

When analyzing the findings, controls were established to account for factors that could influence the language used, such as why drivers pulled over, the crime rate in the area, and more.

“When the vehicle stops, the officer may ask for an ID card, explain why the driver was pushed over, or give a ticket. We were interested in how this balance of dialogue action might differ between escalated and non-escalated stops,” said Rho.


The study found that stops that ended in escalation were almost three times more likely to be initiated by the attendant issuing an order to the driver and 2 1/2 times less likely to give a reason for the stop.

“We found that stops that end up increasing, often start increasing,” Rho said.

Evaluate experience

During the second study, researchers played audio from a traffic stop in the first study to a nationally representative sample of 188 black male U.S. citizens ranging in age, region, education, and political ideology. Each participant was asked to listen to 10 random stops – five that ended in escalation and five that didn’t – from the driver’s perspective and were then surveyed about their feelings and predictions for the outcome of those stops.


Black male participants appeared to use the officer’s language as a guide as to whether they believed the stop would result in the driver’s handcuffs, search, or arrest. They estimate that 84 percent of stops involving officers giving orders without reason will increase. In addition, they were concerned about the use of force in more than 80 percent of the stops involving orders and no reason compared to only 47 percent of stops involving reasons without orders.

Pattern now at another stop?

After finding the elevating vehicle stops carried a unique “language sign” — an officer giving an order without specifying the reason for the stop — the researchers wanted to see if the same signature was present at the stops involving the force. As a case study, the team examines the initial moments between Floyd and the officer who first approached him during an encounter that was made public on May 25, 2020.


In less than 30 seconds of Floyd’s interaction with the officer, the officer uttered 57 words in nine speech turns, consisting only of physical commands. Floyd, in his 11 speech turns, offered an apology, found reasons to quit, pleaded not guilty, expressed fear, and pleaded with officers. Yet every act of dialogue from Floyd is met with a single response from the officer: an order.

Better practice, better relationship

While vehicle stops that end with the use of force often receive national attention, Rho said the team felt it was important to better understand police-citizen interactions during more common vehicle stops.

“The most common way for ordinary citizens to confront law enforcement is through vehicle stops,” said Rho. “So we really wanted to understand more about how we could improve communication between officers and residents during those meetings.”

While both studies reveal valuable insights, Rho said he hopes the observations don’t reach the reach of this paper.

“We wanted this study to really start a conversation about how we can inform training around de-escalation practices for law enforcement and possibly a better understanding of how to facilitate relations between black communities and law enforcement as well,” said Rho.

Source link

Related Articles

Back to top button