Objective: False confessions are prevalent in wrongful convictions, and much research has examined investigation factors and interrogation methods that can contribute to false confessions. However, not all these factors are under the control of the legal system, and improving the effectiveness of interrogation methods has a limited effect on evaluating the veracity of confessions. We suggest incorporating an important but often-neglected factor in interrogations: suspects' prior probability of guilt ("the prior," a Bayesian term meaning suspects' likelihood of being guilty before police conduct an interrogation). Method: By connecting interrogation practices to probability concepts, we discuss a gap in the literature between questions traditionally answered by lab research and a distinct question faced by the legal system. Results: On the basis of our analysis, we argue that police should increase priors by collecting additional evidence to satisfy an evidence-based suspicion of guilt before interrogating suspects. Conclusions: Implementing the evidence-based suspicion practice can help police reduce false confessions, reallocate investigation time and resources, and assist prosecutors in building strong cases for trial. Likewise, researchers should expand the empirical and legal questions they ask and incorporate priors into their interrogation experiments to improve the generalizability of findings to the criminal justice system. Public Significance StatementFalse confessions contribute to wrongful convictions. This research shows the importance of police achieving evidence-based suspicion of guilt before interrogating suspects, meaning that there is enough evidence suggesting a suspect's guilt to justify an interrogation. Such a practice could help the legal system prevent false confessions and protect innocent suspects.