When presented with items that must be generated versus read at encoding, individuals typically remember better those items that they generated versus those that they only read. We examined whether--given the opportunity to experience such differential memorial consequences of generating versus reading--participants might change how they processed future to-be-read information. In a first set of two experiments, participants were able to profit from such an experience to the extent that a generation advantage was eliminated on subsequent memory tests of generated and read items. Two additional experiments demonstrated the critical nature of this experience in leading to improved processing of future to-be-read information and elimination of a generation advantage. We believe that these results relate to the characterization of the learner emerging from recent metacognitive research and have possible implications for how learners might be induced to process information more effectively.