Detecting Institutional Dialog Acts in Police Traffic Stops

Vinodkumar Prabhakaran,Prateek Verma,Camilla Griffiths,Jennifer L Eberhardt,Nelson Morgan,Hang Su,Dan Jurafsky

doi:10.1162/tacl_a_00031

Abstract

We apply computational dialog methods to police body-worn camera footage to model conversations between police officers and community members in traffic stops. Relying on the theory of institutional talk, we develop a labeling scheme for police speech during traffic stops, and a tagger to detect institutional dialog acts (Reasons, Searches, Offering Help) from transcribed text at the turn (78% F-score) and stop (89% F-score) level. We then develop speech recognition and segmentation algorithms to detect these acts at the stop level from raw camera audio (81% F-score, with even higher accuracy for crucial acts like conveying the reason for the stop). We demonstrate that the dialog structures produced by our tagger could reveal whether officers follow law enforcement norms like introducing themselves, explaining the reason for the stop, and asking permission for searches. This work may therefore inform and aid efforts to ensure the procedural justice of police-community interactions.

Highlights

Improving the relationship between police officers and the communities they serve is a critical societal goal
We propose to study this relationship by applying NLP techniques to conversations between officers and community members in traffic stops
We propose automatically extracting dialog structure from body camera footage to contribute to our understanding of police-community interactions

Summary

Introduction

Improving the relationship between police officers and the communities they serve is a critical societal goal. Our models extend this work by drawing on the notion of institutional talk (Atkinson and Drew, 1979), an application of conversational analysis to environments in which the goals of participants are institution-specific. Actions, their sequences, and interpretations during institutional talk depend on the speaker (as speech act theory suggests) or the dialog (as conversational analysts argue), but they are inherently tied to the institutional context. Whalen and Zimmerman (1987) studied dialog structure in transcripts of citizen calls for help They observed that the “regular, repetitive and reproducible features of calls for police, fire or paramedic services [...] arise from situated practices responsive to the sequential and institutional contexts of this type of call”. Such recurring patterns in language and conversation exist across different institutional contexts such as doctor-patient interactions, psychological counseling, sales calls, court room conversations, as well as traffic stops (Heritage, 2005)

Objectives

Results

Discussion

Conclusion