A central issue in the recovered memory debate is whether it is possible to "remember" a highly emotional incident which never occurred. The present study provided an in-depth investigation of real, implanted, and fabricated (deceptive) memories for stressful childhood events. We examined whether false memories for emotional events could be implanted and, if so, whether real, implanted, and fabricated memories had distinctive features. A questionnaire was sent to participants' parents asking about six highly emotional, stressful events (e.g., serious animal attack) which the participant may have experienced in childhood. Next, across three sessions, interviewers encouraged participants (N = 77) to "recover" a memory for a false event using guided imagery and repeated retrieval attempts. In the first interview, they were asked about one real and one false event, both introduced as true according to their parents. In two subsequent interviews, they were reinterviewed about the false event. Finally, after the third inquiry about the false event, participants were asked to fabricate a memory report. Results indicated that 26% of participants "recovered" a complete memory for the false experience and another 30% recalled aspects of the false experience. Real, implanted, and fabricated memories differed on several dimensions (e.g., confidence, vividness, details, repeated details, coherence, stress). These findings have important implications for the debate over recovered and false memories.