Imagination and Memory: Does Imagining Implausible Events Lead to False Autobiographical Memories?

Article excerpt

Previous studies have reported that imagination can induce false autobiographical memories. This finding has been used to suggest that psychotherapists who have clients imagine suspected repressed memories of childhood sexual abuse may, in fact, be inducing false memories for the imagined events. In this study, at Time 1 and then, 2 weeks later, at Time 2, 145 subjects rated each of 20 events on the Life Events Inventory as to whether each had occurred to them in childhood. One week after Time 1, the subjects were told that 2 target events were plausible and 2 were implausible. They were then asked to imagine 1 plausible and 1 implausible target event Plausibility and imagining interacted to affect occurrence ratings; whereas imagining plausible events increased the change in occurrence ratings, imagining implausible events had no effect on occurrence ratings.

Over the past decade, there has been an abundance of research in cognitive psychology on false memories. A PsycINFO search of the empirical publications in cognitive psychology conducted through January 2004, using the subject heading "false memory," produced 198 articles, all published since 1993 (Pezdek & Lam, 2007). Among these studies, one manipulation reported to increase the probability of planting false events in memory is the process of imagining the event. Mazzoni and Memon (2003), for example, reported that after imagining a target event, 40% of participants reported having a memory for the event, in comparison with only 23% of those in the exposure-only condition. Similar findings were reported by Golf and Roediger (1998), Hyman and Pentland (1996), Lampinen, Odegard, and Bullington (2003), and Lindsay, Hagen, Read, Wade, and Carry (2004). The effect of imagination on memory has also been assessed using the imagination inflation procedure (cf. Carry, Manning, Loftus, & Sherman, 1996; Garry & Polaschek, 2000), in which imagining events on the Life Events Inventory (LEI) increases belief that the events occurred in one's childhood.

Pezdek, Finger, and Hodge (1997) have suggested one interpretation of how false events can become planted in memory. According to this interpretation, if a suggested false event is judged to be true, a false memory for the event can be constructed from details of the generic event script, as well as from details of related episodes of the event. Memory for the false event thus becomes constructed from this related information in memory. Imagining the false event then would encourage individuals to activate relevant generic and specific details already in memory and to use this information to construct the memory for the false event.

However, Pezdek et al. (1997) and Pezdek and Hodge (1999) reported that false memories are significantly less likely to be suggestively planted for events that are relatively implausible. The operational definition of an implausible event in this line of research is one that is perceived as having a low probability of occurrence for individuals in the cohort tested. Pezdek et al. (1997, Experiment 2) had 20 confederates read descriptions of one true event and two false events to a younger sibling or close relative. The more plausible false event described the relative's being lost in a mall while shopping; the less plausible false event described the relative's receiving an enema. One week later, only 3 participants recalled a false event, and all three of the recalled events were the more plausible event, being lost in a mall. This finding was replicated in their Experiment 1, in which a suggested description of a Catholic ritual was more likely to be planted in memory for Catholics than for Jews and vice versa.

The interpretation of these results was that if an individual first determines that a suggested event is implausible and not likely to be true, he or she less persistently searches his or her memory for information relevant to the event. Consequently, images constructed for implausible events will contain less information. …