• How downplaying or exaggerating crime severity in a confession affects perceived guilt

      Holt, G. A.; Palmer, M. A.; University of Chester; University of Tasmania
      This study investigated how judgments of guilt are influenced by factual errors in confessions that either amplified or downplayed the severity of the crime. Participants read a confession statement and a police report. Information in the confession statement either was consistent with the facts of the crime in the police report, the suspect admitted to a worse crime than described in the police report, or the suspect admitted to a lesser crime than described in the police report. Mediation analyses showed that, compared to consistent confessions, both types of directional errors reduced judgments of guilt. Inconsistencies that made the suspect look better—but not those that made the suspect look worse—also increased judgments of guilt via a direct effect. Confessions that contain errors that appear to exaggerate the severity of the crime prompt no higher judgments of suspect guilt than confessions that are consistent with the facts of the crime. However, errors in confessions that are perceived to downplay the severity of the crime can prompt an increased perception of suspect guilt, when compared to a consistent confession.
    • The variable influence of confession inconsistencies: How factual errors (but not contradictions) reduce belief in suspect guilt

      Holt, Glenys A.; Palmer, Matthew A.; University of Chester; University of Tasmania
      Wrongful conviction statistics suggest that jurors pay little heed to the quality of confession evidence when making verdict decisions. However, recent research indicates that confession inconsistencies may sometimes reduce perception of suspect guilt. Drawing on theoretical frameworks of attribution theory, correspondence bias, and the story model of juror decision-making, we investigated how judgments about likely guilt are affected by different types of inconsistencies: self-contradictions (Experiment 1) and factual errors (Experiment 2). Crucially, judgments of likely guilt of the suspect were reduced by factual errors in confession evidence, but not by contradictions. Mediation analyses suggest that this effect of factual errors on judgments of guilt is underpinned by the extent to which mock-jurors generated a plausible, alternative explanation for why the suspect confessed. These results indicate that not all confession inconsistencies are treated equally; factual errors might cause suspicion about the veracity of the confession, but contradictions do not.