Numerous eyewitness testimony studies have shown that people can falsely remember parts of an event after being exposed to misleading suggestion about it (Loftus, Miller, & Burns, 1978); however, few researchers have examined whether people falsely remember parts of an event when there is no such suggestion. Across two studies, we show that people systematically develop false memories for unseen aspects of an event. In Experiment 1, subjects saw a movie of a woman making a sandwich; some actions were missing. In a memory test, subjects confidently but falsely remembered 17% of unseen information from the event. In Experiment 2, subjects saw the same event, but the missing actions were either crucial or not crucial. Subjects were more likely to falsely remember the missing noncrucial than missing crucial information. Theoretically, our results fit with a source monitoring account of false memories. Practically, our results suggest a means by which we can predict what aspects of an event are likely to be falsely remembered. Copyright © 2006 John Wiley & Sons, Ltd.