Mixing Memory

One of the criticisms of most false memory research is that it lacks ecological validity. For example, in the Deese-Roediger-McDermott (DRM) paradigm, a common method for inducing false memories in the lab, involves giving participants a bunch of words (e.g., bed, rest, nap, snore, etc.) that are all associated with another word that's not presented (e.g., sleep). During recall, if you ask participants if they saw the word "sleep" after seeing a list of its associates, they're pretty likely to tell you that they did. But it's difficult to know how to generalize the DRM to real-world situations. So it's always exciting to see researchers attempting to create more ecologically valid paradigms to study false memories. And the current issue of the Psychonomic Bulletin & Review just happens to have a paper on just such an attempt.

The two experiments described in the paper, written by Alan Brown and Elizabeth Marsh(1) are really pretty simple. Brown and Marsh asked participants to find a cross in each of a series of photos (216 per participant). There were four types of target photos: mundane (that is, common) locations from the participants' campus, mundane scenes from an unfamiliar campus, scenes unique to the participants' campus, or scenes unique to the unfamiliar campus. Examples of each can be seen in the figure below (Brown and Marsh' figure 1, p. 187):

Each of the target photos a participant saw was presented either once or twice over the series of photographs. One (in experiment one) or three (in experiment two) weeks after the cross-finding task, participants came back and were given a series of photos of campus locations, and asked to indicate whether they'd been there before or not. Each picture was presented for 500 ms, and all of the photos from the first session were presented, along with some new photos from both the familiar and unfamiliar campuses. As you'd expect, in the second session, participants indicated that they'd visited locations on their own campus significantly more than than locations on the unfamiliar campus. However, after both 1 and 3 week delays, almost all participants indicated having visited mundane locations from the unfamiliar campus that they'd seen in the first session, and most participants indicated having visited one of the unique locations from the unfamiliar location from the first session. They were significantly more likely to say that they'd visited locations from the unfamiliar campus if it had been presented in the first session than if it hand't. In fact, for 10% of the mundane locations (5% of the unique locations) from the unfamiliar campus that were presented in the first session, participants said that they had "definitely visited" them. And the false memories for unfamiliar locations occurred regardless of whether the scenes had been presented once or twice in the first session.

Obviously, this study has limited ecological validity as well, but it's a significant step in the right direction. Simply by showing people scenes, Brown and Marsh were able to induce doubt about whether they'd visited locations on a campus half a continent away, and in some cases, they were actually able to make the participants feel certain that they'd visited them. In other words, they've developed a paradigm that actually causes false autobiographical memories. It will be interesting to see where researchers are able to take it from here.

¹Brown, A.S., & Marsh, E.J. (2008). Evoking false beliefs about autobiographical experience. Psychonomic Bulletin & Review, 15(1), 186-190.