We examined the qualitative characteristics of genuine, imagined, and deceptive accounts of positive and negative childhood events. We investigated whether trained raters could discriminate between these accounts using the Memory Characteristics Questionnaire (MCQ; Johnson, Foley, Suengas, & Raye, 1988) and the Aberdeen Report Judgment Scales (ARJS; S. L. Sporer, paper presented at the biennial meeting of the American Psychology-Law Society in Redondo Beach, California, March 1998). Participants generated three accounts. The first account was of an event that participants genuinely experienced in childhood. The second account was of an event that participants did not experience, but merely imagined happened in childhood. The third account was of an event that participants did not experience, but wrote a deceptive account to convince someone else that the event really happened in childhood. Half our participants wrote about positive events and half wrote about negative events. Ratings made by two trained judges indicated that genuine, imagined, and deceptive accounts were qualitatively different on both the MCQ and ARJS. Moreover, based on the MCQ and ARJS scores raters could discriminate whether the events had been genuinely experienced. We discuss the theoretical and practical implications of these findings.