Objectives We completed a systematic review and meta-analysis of the available empirical literature assessing the influence of accusatorial and information-gathering methods of interrogation in eliciting true and false confessions. Methods We conducted two separate meta-analyses. The first meta-analysis focused on observational field studies that assessed the association between certain interrogation methods and elicitation of a confession statement. The second meta-analysis focused on experimental, laboratory-based studies in which ground truth was known (i.e., a confession is factually true or false). We located 5 field studies and 12 experimental studies eligible for the meta-analyses. We coded outcomes from both study types and report mean effect sizes with 95 % confidence intervals. A random effects model was used for analysis of effect sizes. Moderator analyses were conducted when appropriate.Results Field studies revealed that both information-gathering and accusatorial approaches were more likely to elicit a confession when compared with direct questioning methods. However, experimental studies revealed that the information-gathering approach preserved, and in some cases increased, the likelihood of true confessions, while simultaneously reducing the likelihood of false confessions. In contrast, the accusatorial approach increased both true and false confessions when compared with a direct questioning method. Conclusions The available data support the effectiveness of an information-gathering style of interviewing suspects. Caution is warranted, however, due to the small number of independent samples available for the analysis of both field and experimental studies. Additional research, including the use of quasi-experimental field studies, appears warranted.