It’s Crazy How Easy It Is to Make People Falsely Remember Committing a Crime

Photo: Kelly Redinger/Corbis

Memory’s a pretty fluid and complex thing. We don’t always remember specific details of an event well, and what details we do remember can be influenced by stuff that happened after the event itself. This is all pretty standard when it comes to memory research. What the authors of a new paper in Psychological Science just pulled off, though, takes things to a whole new level: They were able to convince study participants they had committed a crime that was completely fabricated.

Julia Shaw and Stephen Porter of the universities of Bedfordshire and British Columbia, respectively, got permission from a group of Canadian undergraduates to send surveys to their primary caregivers in which the caregivers were asked to provide details of negative, emotionally charged events (animal attacks, losing money, and other stuff like that) the students had experienced when they were younger. 

Then, with these details in their hands, the researchers interviewed the students three times, with each interview lasting for about 40 minutes. Some were asked to remember things that had actually happened to them — the animal attack or whatever — while others were asked to remember details of a crime they hadn’t actually committed.

It’s probably best to let the researchers take it from here:

Participants were asked to explain what happened during each of the events in turn, after the interviewer provided some accurate cues from the caregiver questionnaire, including the city that the participant lived in and the name of a friend the participant had at the time of the alleged event (a friend who was supposedly present during the event). The interviewer also provided a number of cues, including the participant’s age at the time of the event, the season when it took place, and an indication that the caregiver was involved after the event occurred; for the true event, these were accurate cues, and for the false event, they were randomly assigned inaccurate cues. As expected, participants successfully provided an account of the true event but were unable to provide an account of the false event in the first interview … When participants had difficulty recalling the false event, the interviewer encouraged them to try to remember it, and (falsely) told them that most people can remember these kinds of memories if they try hard enough. Then, participants were told that the study was an examination of memory-retrieval methods, and they were asked to use context reinstatement and guided imagery to retrieve the memory. They also were told to practice visualization of the false event each night at home. These methods have been shown to effectively generate details that form the foundations of false memories[.]

If you want more details, the study, which is unlocked, has them, but the key takeaway is that by the end of the third interview, after a bunch of carefully crafted nudging to do their best to remember, a full 70 percent of the students said, “Yep, I committed that crime when I was younger,” and they “volunteered … detailed false account[s]” of those crimes.

It’s a pretty stunning example of just how malleable memory is, and how open to suggestion people can be in certain circumstances. And while the researchers point out that it’s unclear whether these findings are directly applicable to police interrogations — there were no negative consequences to copping to the (non-)crimes in question, which makes for a pretty different overall dynamic — it’s hard, after reading this study, not to think about the high-profile false confessions that have caused interrogation procedures to come under a great deal of scrutiny in recent years (there’s a great “This American Life” about this).

Memory: It just keeps getting weirder and fuzzier.