The attempt to forget some recently encoded information can indeed cause later retrieval difficulties. However, such attempts are only effective when new information is learned shortly thereafter. In the present study, we asked whether the new information has to match the format of the to-be-forgotten information for forgetting effects to emerge. Participants studied words or line drawings (L1), and were afterwards instructed to remember or forget these items. Then, a second list (L2) was presented that either matched or mismatched the L1 format. Forgetting effects were only observed when the list formats matched. This result establishes an important boundary condition on intentional forgetting, and can be explained by the context change account (Sahakyan & Kelly, 2002), which assumes that forgetting occurs when retrieval is guided by temporal context only. Salient cues (such as differences in list format) allow for reinstatement of the L1 encoding context, thus eliminating forgetting.