Summary: Study reveals people are only 53% likely to identify whether someone else in recounting and accurate or false memory of a crime.
People are no better than chance at identifying when someone else is recounting a false or real memory of a crime, according to a new UCL study.
The findings, published in Frontiers in Psychology, build on a previous study that was the first to successfully implant false memories of committing a crime – involving either assault or assault with a weapon that resulted in police contact.
Study author Dr Julia Shaw (UCL Psychology & Language Sciences) said: “Everyone thinks that they couldn’t be tricked into believing they have done something they never did, and that if someone were telling them about a false memory, they would be able to spot it. But we found that actually, people tend to be quite susceptible to having false memories, and they sound just like real memories.”
For the previous study, published in 2015, Dr Shaw and a colleague invited young adults into a study about emotional memories, and also spoke with a member of their family to learn about events from the participants’ early adolescence, in as much detail as possible.
The researchers spoke to the participants about their past, and used leading questions and suggestive tactics, as well as visualisation techniques to convince the participants that they were helping them recover a forgotten memory – while in fact they were implanting a false memory that the participant had committed a crime when they were young, such as theft or assault.
“We were essentially doing exactly ‘what not to do’ when conducting a police interview,” explained Dr Shaw.
The 2015 study reported that the majority of participants developed a false memory of committing a crime, and the participants consistently reported that the false memories felt incredibly real.
The current research involves two studies that used videos from the 2015 study, of the study participants recounting their false memories of a crime, which they believed to be real. The new participants watched those videos and were asked if the person was describing an event that actually happened or not.
Participants were only 53% accurate (no better than chance) at identifying false memories of committing a crime. These results were replicated in the second study. Even when participants were explicitly told that one of the memories they watched was false, their judgment was still no better than tossing a coin.
In addition to incorrectly believing false memories to be true, participants were just as likely to watch someone recount a genuine memory, and then misidentify it as false, once the research team told them that some of the videos would feature false memories. In other words, many true memories looked like false memories.
“Legal professionals and police officers need to realise how easy it is to manipulate someone’s memories. Judges in particular should never assume that they can tell when someone has a false memory, and should consider the entire process to see if there was any risk of contamination of a defendant or witness’ memories,” Dr Shaw said.
“The findings really highlight how important it is to ensure that criminal proceedings are done right. The questioning process should be evidence-based, to reduce the risk of implanting false memories in people being questioned by the police.”
Do False Memories Look Real? Evidence That People Struggle to Identify Rich False Memories of Committing Crime and Other Emotional Events
Two studies examined whether people could identify rich false memories. Each participant in both studies was presented with two videos, one of a person recalling a true emotional memory, and one of the same person recalling a false memory. These videos were filmed during a study which involved implanting rich false memories (Shaw and Porter, 2015). The false memories in the videos either involved committing a crime (assault, or assault with a weapon) or other highly emotional events (animal attack, or losing a large sum of money) during adolescence. In study 1, participants (n = 124) were no better than chance at accurately classifying false memories (61.29% accurate), or false memories of committing crime (53.33% accurate). In study 2, participants (n = 82) were randomly assigned to one of three conditions, where they only had access to the (i) audio account of the memory with no video, (ii) video account with no audio, or (iii) the full audio-visual accounts. False memories were classified correctly by 32.14% of the audio-only group, 45.45% of the video-only group, and 53.13% of the audio-visual group. This research provides evidence that naïve judges are not able to reliably identify false memories of emotional or criminal events, or differentiate true from false memories. These findings are likely to be of particular interest to those working in legal and criminal justice settings.