Using a novel autobiographical think/no-think procedure (ATNT; a modified version of the think/no-think task), 2 studies explored the extent to which we possess executive control over autobiographical memory. In Study 1, 30 never-depressed participants generated 12 positive and 12 negative autobiographical memories. Memories associated with cue-personal word pairings were learned to criterion. Participants were then asked to recall the memory associated with some of the cue-personal word pairs (i.e., think condition) or to avoid saying or thinking about the memory associated with others (i.e., no-think condition). In a subsequent test of recall, systematic forgetting effects emerged for no-think autobiographical memories compared to baseline that received neither no-think nor think instructions. These findings were extended and replicated in a second ATNT study (using a further 30 never-depressed participants), which showed that the forgetting of autobiographical memories in the no-think condition was unlikely to be a function of thought substitution or demand characteristics. (Contains 6 tables and 7 figures.)