Related Experiment Video
Updated: Jan 12, 2026

Creating Virtual-hand and Virtual-face Illusions to Investigate Self-representation
Published on: March 1, 2017
Almost no one feels free to leave an interrogation room: Findings from a virtual reality study of custody perceptions
Hayley M D Cleary1, Lucy A Guarnera1
1Department of Criminal Justice, L. Douglas Wilder School of Government and Public Affairs, Virginia Commonwealth University.
Objective:
Police custody is a critical legal determination because it triggers constitutionally mandated Miranda advisements and, in many states, electronic recording requirements. However, what constitutes "custody" is both legally and empirically nebulous, and few empirical studies have examined custody perceptions.
Hypotheses:
We hypothesized that participants' custody perceptions would vary across time and interrogation technique within our virtual interrogation. We also predicted that an array of individual characteristics (e.g., Black racial identity, educational problems, trauma symptoms, stereotype threat) would be associated with custody perceptions.
Method:
This study exposed 168 participants (ages 18-25) to a realistic police interrogation using virtual reality technology. Virtual reality participants perceived themselves as seated in an actual interrogation room next to a young suspect interrogated about a homicide. The scripted and filmed interrogation systematically varied three different interrogation techniques (rapport building, maximization, minimization).
Results:
After only 3 min of rapport-building interactions between the detective and suspect, about half of participants believed the suspect was not free to leave, and that percentage significantly declined over time and after the detective initiated confrontational interrogation techniques. Mixed-effects logistic regressions revealed that time point significantly predicted objective custody perceptions (R²m = .14; R²c = .78), with an intraclass correlation coefficient of .74. Perspective-taking custody ratings (i.e., how free to leave the suspect felt) averaged 2.5 on a 1-10 scale after the rapport-building segment and declined to near-floor levels thereafter. Mixed-effects cumulative link regressions revealed that time point significantly predicted perspective-taking custody perceptions (R²m = .39; R²c = .77), with an intraclass correlation coefficient of .85.
Conclusions:
Most people do not feel free to leave an interrogation room, even before confrontational interrogation techniques begin. This study informed the "reasonable person" standard by showing that even psychologically healthy young adults typically perceive they are in police custody, even under minimally restrictive circumstances. (PsycInfo Database Record (c) 2026 APA, all rights reserved).
Related Concept Videos
The Stanford Prison Experiment
Surveys
Factors Affecting Perception
An illustrative example of a perceptual set is the scenario where an airline pilot told...
Stereotype Content Model
Deindividuation
Eyewitness Memory
One such error is memory distortion, which occurs because human memory does not function...

