The effect of a presumption of guilt on police guilt judgments

ABSTRACT

Police conduct pre-interrogation interviews with suspects whom they presume might be guilty. This research tested whether a presumption of guilt causes police to misclassify innocent suspects as guilty during pre-interrogation interviews by virtue of biasing their interpretation of suspects’ emotional states. In two experiments, college students (n = 33) and police officers (n = 33) each watched eight videos in which an experimenter interviewed a student who was either factually guilty or factually innocent of having cheated on a problem-solving task. After watching each video, participants reported their judgments of the interviewed student’s emotional state and guilt-status. The results indicated that both college students and police officers reported higher guilt judgments when they presumed the interviewed students to be guilty versus innocent. Additionally, participants’ perceptions of the interviewed students’ emotional states mediated this effect. Factual guilt-status did not influence judgments rendered by either college students or police officers. The results suggest that police may be susceptible to misclassifying innocent suspects as guilty when they hold a presumption of guilt at the outset of a pre-interrogation interview.

KEYWORDS:

• pre-interrogation interview
• confirmation bias
• guilt judgment
• police interrogation

Disclosure statement

No potential conflict of interest was reported by the author(s).

Author note

We appreciate the support of the Scholarly and Creative Activities Grant from the University of Nevada, Reno for this research project. Portions of the findings were presented at annual meetings of the American Psychology-Law Society in Portland, Oregon in March 2019 and in New Orleans, Louisiana in March 2020. All data, materials, and analysis codes that support the research findings are openly available in the Center for Open Science at https://osf.io/3wuv4/. We have no known conflict of interest to disclose.