Output monitoring tests reveal false memories of memories that never existed

Three experiments were conducted to assess participants ’ beliefs about potential false memories that might have occurred during free recall tests. An input–output monitoring test was administered that required participants to discriminate between items that were studied and recalled, studied and not recalled, or were entirely new. Critical lures from Roediger and McDermott’s (1995) paradigm were inserted into this test. The results demonstrated that participants believed erroneously recalled items were both studied and recalled. The intriguing finding was that unrecalled items were believed to have been studied approximately 80 % of the time, and half of those were also believed to have been recalled. This result represents a dual false memory effect in which items were believed to have been studied and also to have been recalled. The ramifications of this new procedure are discussed in terms of proposed experi-ments that might clarify the genesis of these false memories. What a person believes to be true about his or her memory is quite important. Because such beliefs determine expectations, confidence, and trust in what one recollects, they are particularly sig-nificant in the context of false memories. In gen-eral, false memories occur when one reports that an event was experienced in a particular situation despite the event never having occurred at all under those circumstances. Although there are several laboratory techniques for studying false memories (see Payne & Blackwell, 1998), none has received as much attention as Roediger an