This study aims to advance the field of child forensic interviewing by assessing the impact of different refresher training modalities on police officers’ abilities to adhere to the steps of an interview protocol and on the types of questions used. Previously trained police officers (N = 46) were randomly assigned to one of three experimental conditions: (1) supervision with an expert, (2) peer group supervision, and (3) computer-assisted exercises on children’s investigative interview techniques. Comparison of interviews conducted before (n = 136) and after (n = 124) the refresher modalities revealed an improvement in performance across time for almost all steps of the protocol. There were more effects associated with time than with modality of refresher training with regard to question types used during episodic memory training and the substantive phase of the interview. Although there were some differences between the three conditions, no method emerged as clearly superior. Results suggest that all modalities could be useful in refreshing adherence to the steps of an interview protocol and use of best practice questioning approaches. The discussion highlights that the time devoted to the three modalities was likely not enough and that further studies are needed to determine the most optimal delivery of refresher training.