Purpose. Little research exists on the influence of emotion in forensic settings. To start filling this gap, we used a hypothetical interrogation scenario to examine the effects of emotional state on judgement, decision making, and information-processing style across two separate experiments. Methods. The participants were induced a specific emotion. Then, they read a scenario where a suspect was arrested and rated (1) the suspect's guilt, and (2) the extent to which they would use a number of tactics to interview the suspect. Based on the feelings-asinformation theory and cognitive-appraisal theories of emotion, we predicted that relative to angry or happy participants, sad participants would be less inclined to judge the suspect as guilty (judgement), would show a stronger tendency to select benevolent interrogation tactics and a weaker tendency to select hostile interrogation tactics (decision making), and would be more likely to use an analytic (rather than a heuristic) processing style. Results. In Experiment 1 (conducted with college students), the judgement hypothesis was supported. In Experiment 2 (with mTurkers), the decision-making hypothesis was supported. A meta-analysis of the two experiments revealed that participants were more willing to select benevolent than hostile interrogation tactics and that, as predicted, sad participants were more willing than angry or happy participants to select benevolent tactics. However, emotion did not affect the participants' tendency to select hostile tactics. Conclusion. We tested emotion theories in an interrogation scenario. The significant results were consistent with the feelings-as-information and cognitive-appraisal theories of emotion and have practical relevance.