Do You Remember Proposing Marriage to the Pepsi Machine? False Recollections from a Campus Walk

Article excerpt

During a campus walk, participants were given familiar or bizarre action statements (e.g., "Check the Pepsi machine for change" vs. "Propose marriage to the Pepsi machine") with instructions either to perform the actions or imagine performing the actions (Group 1) or to watch the experimenter perform the actions or imagine the experimenter performing the actions (Group 2). One day later, some actions were repeated, along with new actions, on a second walk. Two weeks later, the participants took a recognition test for actions presented during the first walk, and they specified whether a recognized action was imagined or performed. Imagining themselves or the experimenter performing familiar or bizarre actions just once led to false recollections of performance for both types of actions. This study extends previous research on imagination inflation by demonstrating that these false performance recollections can occur in a natural, real-life setting following just one imagining.

Are there limits on fictitious actions that we subsequently remember to be real? Considerable memory research, spurred initially by cases of recovered memories of abuse, has indicated that false memory for actions can be reliably demonstrated. For example, research has shown that when a plausible but false childhood event, such as getting lost in a mall, is embedded in a series of true childhood events, and those events are attributed to a trusted family source, adults can subsequently accept that false event as real (Loftus & Pickrell, 1995). Still other studies have shown that more unusual fictitious childhood events, such as running around at a wedding and knocking a punchbowl onto the laps of the bride's parents, can also be falsely remembered by adults when the fictitious action is embedded in a set of real childhood events and attributed to the participants' parents (Hyman & Pentland, 1996). Similarly, when childhood portraits are inserted into photographs to portray an event that never took place, such as a hot air balloon ride with a family member, adults can generate false recollections based on the altered photo (Wade, Garry, Read, & Lindsay, 2002). Typically, such false memories occur when a fictitious event is presented in the context of real events from the distant past, the events are attributed to a trusted source, and, to aid in remembering, the participant is given time and instructions to imagine them repeatedly (for a review, see Brainerd & Reyna, 2005).

Other studies have bypassed the potential problem that a false childhood event might have actually occurred in the past by presenting participants with a series of actions to perform or imagine in a laboratory setting, and later testing the participants' recent memory for both the actions and the source of those actions. Goff and Roediger (1998), for example, presented participants with common objects as they heard simple action statements such as "break the toothpick," heard the statements and performed those actions, or heard the statements and imagined performing the actions. One day later, the participants were given some of the old objects and actions along with some new ones, and they were asked to imagine each action one, three, or five times. Two weeks later, the participants were given a list of actions taken from the first and second sessions, together with new actions, and asked to indicate which actions were presented in Session 1 and, if presented, whether those actions were imagined, performed, or only heard. These researchers found that increasing the number of imaginings of a new action in the second session increasingly led participants to misremember that action as having been performed in the first session, thereby demonstrating false recollection by imagination inflation (see also Sharman, Carry, & Beuke, 2004). Subsequently, Thomas and Loftus (2002) and Thomas, Bulevich, and Loftus (2003) extended these false performance results with a similar procedure and showed that not only would participants misremember imagined common actions (e. …