Directed forgetting (DF) studies demonstrate that humans can intentionally forget item information. In the presented study, participants learned three lists of words. After studying the first two lists (L0+L1), we cued half of the participants to forget these lists before learning a new list (L2), the other half remembered all three lists. Typically, such a forget instruction impedes recall of previously-studied to-be-forgotten words but enhances memory for subsequent to-be-remembered items. Instead of recalling the words, we asked participants to select the list a word was studied in, assessing how DF affected both item- and list-memory. In line with the context-change hypothesis, list-memory for L1 did not differ between the two groups suggesting that even if recall of to-be-forgotten words is typically impaired, list-memory is still intact. Furthermore, after the forget instruction, participants’ list-memory was enhanced particularly for early L2 words, providing evidence for a reset of encoding or rehearsal processes.