Previous research has shown that simulation training using avatars with repeated feedback improves child sexual abuse interview quality. The present study added a hypothesis-testing intervention and examined if the combination of two interventions, feedback and hypothesis-testing, would improve interview quality compared to no intervention and to either intervention alone. Eighty-one Chinese university students were randomly assigned to a control, feedback, hypothesis-testing, or the combination of feedback and hypothesis-testing group and conducted five simulated child sexual abuse interviews online. Depending on the assigned group, feedback on the outcome of the cases and question types used in the interview were provided after each interview, and/or the participants built hypotheses based on preliminary case information before each interview. The combined interventions group and feedback group showed a higher proportion of recommended questions and correct details from the 3rd interview onward compared to the hypothesis-building and control groups. The difference between the number of correct conclusions was not significant. hypothesis-testing alone exacerbated the use of non-recommended questions over time. The results show that hypothesis-testing may impact question types used negatively but not when combined with feedback. The potential reasons for hypothesis-testing alone not being effective and the differences between the present and previous studies were discussed.