ai
The Effect of Perceived Race and Gender on Police Language Use: Experimental Evidence from VR Simulations
- Source
- arXiv — Computers and Society
- Published
- Last verified
- 6 Aug 2026
- Confidence
- High
- Evidence
- Original document retained
- Reading time
- 1 min
- Country
- International
- Relevant to
- Research & Evidence, Technology & Data
Executive summary
What happened, and why should leadership care?
Research from arXiv, titled 'The Effect of Perceived Race and Gender on Police Language Use: Experimental Evidence from VR Simulations', indicates a systemic issue in police interactions. The study, using virtual reality simulations, found that most police officers exhibited less deferential language towards characters perceived as Black adult males. This pattern was observed across a majority of officers, with a notable exception being White, biracial, and multiracial female officers, who did not show this decreased deference.
Why this matters
Why is this strategically important?
This research highlights potential biases in communication within critical public interactions, which can significantly impact public trust, safety, and operational effectiveness. Understanding and addressing these nuanced communication patterns is crucial for fostering equitable and just interactions, and for strategic planning in public service sectors.
Key insights
What should be noted from the evidence?
- Most police officers used less deferential language when interacting with virtual characters depicted as Black adult males.
- The study employed VR simulations and a causal inference lens to evaluate the effect of perceived race on officer communication.
- The 'average treatment effect' (ATE) measured the social impact of the character's depiction on officer deference.
- White, biracial, and multiracial female officers were an exception, speaking with similar deference to Black male characters.
Evidence and confidence
How far can this assessment be trusted?
High confidence. Named institution, original document retained and analysis corroborated.
Analysis is prepared by the AZIZ OS Intelligence Engine. The original publication remains the authoritative record, and executive judgement remains entirely human.
Source
Where does this originate?
Reported by arXiv — Computers and Society · International. This briefing summarises the publication for executive use; the document itself is not reproduced here.
Read the original publication