Contextual factors predict self-reported confession decision-making: A field study of suspects’ actual police interrogation experiences.

Abstract

This study examined incarcerated persons’ self-reported interrogation experiences and confession decision-making by investigating which sociodemographic, criminological, and contextual factors were associated with their decisions to deny the allegations, partially admit wrongdoing, or fully confess to the crime. We expected that respondents in this field study would report a wide
\nrange of interrogation experiences. Given mixed prior findings, we did not formulate hypotheses for
\nsociodemographic or criminological factors, but based on contextual variable research, we predicted
\nthat suspects who perceived the evidence against them as strong and who had already decided to confess
\nprior to their interrogation would be more likely to confess. Participants were 249 individuals
\n(86% male; M age = 34.8 years; 49% Black, 41% White, 10% other racial identities) incarcerated in
\nlocal jails in the United States who completed a questionnaire about their most recent interrogation.
\nRespondents described their interrogation experiences (e.g., location, duration, custody), perceptions of
\npolice evidence against them, and thoughts about confession prior to the interrogation. We examined
\ngroup differences according to confession decision and used multinomial logistic regression to examine
\nhow sociodemographic, criminological, and contextual factors relate to suspects’ self-reported confession decisions. Results: Suspects’ interrogation experiences varied considerably, as did their perceptions
\nof custody, beliefs about incriminating evidence, and preinterrogation intent to confess or deny.
\nSociodemographic characteristics and criminological factors were unrelated to self-reported confession
\ndecision-making, but several contextual factors predicted confession outcome. Signing away one’s
\nMiranda rights and already planning to confess predicted suspects’ self-reported confessions, whereas
\nbeing physically restrained, believing that police had no evidence of one’s guilt, and intending in
\nadvance to deny the allegations predicted suspects’ self-reported denials. Suspects who were undecided
\nabout confession prior to interrogation were about as likely to eventually confess as deny. 
\nMost suspects followed through with their initial intention to confess or deny, and suspects’ perceptions
\nabout evidence predicted their self-reported confession. These findings complement existing work
\nfocused on interrogation techniques and inform both police interrogation training and practice.