Never ever EVER trust a psychologist

Following up on that Scientific American blog that repressed memories don't exist. I was much more interested in the related claim about inducing false memories than the dodgy nature of memory retrieval, and my code is running very slowly, so I did some digging. This is one of those nice cases where the paper is more interesting than the press release.

"Our findings show that false memories of committing crime with police contact can be surprisingly easy to generate, and can have all the same kinds of complex details as real memories," says psychological scientist and lead researcher Julia Shaw of the University of Bedfordshire in the UK. "All participants need to generate a richly detailed false memory is 3 hours in a friendly interview environment, where the interviewer introduces a few wrong details and uses poor memory-retrieval techniques."

Emphasis on "same kinds", as in the paper itself they show that true memories tend to be more vivid and detailed than false ones. How do you go about inducing false memories ? The researchers used a variety of techniques. They don't try and identify which one (if any) is the most important as that's beyond the scope of the study. What they did was as follows, from the press release :

Shaw and co-author Stephen Porter of the University of British Columbia in Canada obtained permission to contact the primary caregivers of university students participating in the study. The caregivers were asked to fill out a questionnaire about specific events the students might have experienced from ages 11 to 14, providing as much detail as possible. The caregivers were instructed not to discuss the questions with the student.

The researchers identified a total of 60 students who had not been involved in any of the crimes designated as false memory targets in the study and who otherwise met the study criteria. These students were brought to the lab for three 40-minute interviews that took place about a week apart. In the first interview, the researcher told the student about two events he or she [the student] had experienced as a teen, only one of which actually happened. [This was either a crime or another emotional event - anxiety seems to be important, though they didn't examine less emotional events]

Importantly, the false event stories included some true details about that time in the student's life, taken from the caregiver questionnaire. Participants were asked to explain what happened in each of the two events. When they had difficulty explaining the false event, the interviewer encouraged them to try anyway, explaining that if they used specific memory strategies they might be able to recall more details.

But there are more subtleties to this, which the paper describes very well. The interviewer (always the same person, and they even tried to make sure the word count of each interview was similar !) was " a senior Ph.D. student who was well trained in police interview tactics and is extraverted — a personality characteristic that has been demonstrated to be related to high success rates for generating false memories." One paragraph from the paper deserves to be quoted in full (omitting only references and splitting this for readability in the G+ format) :

The strategies that were employed throughout all interviews in this study were based on literature regarding factors that facilitate the generation of false confessions. The tactics that were scripted into all three interviews included incontrovertible false evidence (“In the questionnaire, your parents/ caregivers said. . .”), social pressure (“Most people are able to retrieve lost memories if they try hard enough”), and suggestive retrieval techniques (including the scripted guided imagery)... Incorporating true details into the false-memory account — especially the caregiver-provided details regarding the city the participant lived in and the name of a friend the participant had at the time of the alleged event — likely constituted a personalized manipulation in our study.

Other tactics that were consistently applied included building rapport with participants (e.g., asking “How has your semester been?” when they entered the lab), using facilitators (e.g., “Good,” nodding, smiling), using pauses and silence to allow participants to respond (longer pauses seemed to often result in participants providing additional details to cut the silence), and using the open-ended prompt “what else?” when probing for additional memory details. We also used the tactic of presumed additional knowledge if participants asked about the accuracy of details. In other words, participants were told that the interviewer had very detailed information about the event from their caregiver but was able only to vaguely confirm details (e.g., “this sounds like what your parents described,” “I can’t give you more details because they have to come from you”).

Further, when participants reported that they could not recall the false memory, the interviewer seemed disappointed but sympathetic (while saying the scripted line “That’s ok. Many people can’t recall certain events at first because they haven’t thought about them for such a long time.”) and scribbled down a note on her clipboard. Finally, the interview office had a bookshelf intentionally filled with very visible books on memory and memory retrieval to help increase the interviewer’s credibility as a memory researcher.

Also, importantly, the techniques extend beyond the interviews themselves, so that quoted 3 hours is a bit misleading :

The fact that no participants immediately [at the first interview] recalled the false event helped rule out the possibility that participants had actually experienced such an event... 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... 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.

Later on they speculate that false memories are easier to generate if they have "explanatory coherence" :

In other words, imagined memory elements regarding what something could have been like can turn into elements of what it would have been like, which can become elements of what it was like... it is possible that participants increasingly tried to make sense of the introduced false events by spinning explanatory frameworks around what they thought could have happened.

And it's also important for victims participants to use their own imaginations, Inception apparently getting that one right :

Imagination exercises such as this one have been repeatedly associated with the generation of false memories... The relevance of imagination for false memories may be partially explained by the source-monitoring framework, which refers to people’s tendency to confuse imagination with reality. Individuals who are recalling details from a visualization exercise or experimenter misinformation can forget the source of their ideas and may think they are recalling details from a genuine experience.

When analysing the results that students (about 70% success rate !) apparently believe in the induced memories, they make an interesting distinction between false memories and false false memories, i.e. the students might be deceiving themselves (or just the researchers) about the false memories :

It has been argued that false beliefs are qualitatively different from false memories... Participants who provided 10 or more details of the false event but did not claim at the debriefing that they had believed the event actually happened were classified as compliant; they could be seen as having simply acquiesced to the situational demands. Participants who provided fewer than 10 details but claimed at debriefing that they had believed the event actually happened were classified as being accepting of the false memory event. They seemed immune to significant memory generation despite appearing to believe that the event had happened to them.

The paper is aimed at the legal process of getting reliable eye-witness testimony, but I think it surely has much broader implications : political manipulation, the extent to which fiction influences moral beliefs, etc. More of that in tomorrow's concluding part of the Magnesia trilogy. What would have been nice to see in the paper is more discussion on those few individuals who were wholly resistant to this manipulation - do they have unusually good memories, or are they (for instance) just more resistant to forming memories in the first place ? Or were they less trusting for some reason ? Anyway, it's fascinating stuff, or at least I think so. Assuming I haven't misremembered the whole thing, of course.
Shared publiclyView activity