This paper summarizes, analyzes, and critiques the study True Photographs and False Memories published by D. Stephen Lindsay, Lisa Hagen, J. Don Read, Kimberly A. Wade, and Maryanne Garry in March of 2004. Lindsay et al’s study is then compared to Formal Study of Lost-in-a-Mall by Dr. Elizabeth Loftus and JE Pickrell.
D. Stephen Lindsay, Lisa Hagen, J. Don Read, Kimberly A. Wade, and Maryanne Garry worked together to create a study that investigated the formation of false memories. Their research question was, essentially, can a person be compelled to create false memories from their childhood via the use of childhood photographs? How much does the use of photographs affect the likelihood of developing false memories? How detailed and complete are the false memories that people develop with photographs versus without, and how many out of a group develop false memories at all? Their hypothesis was as follows:Order now
… if a person believes that certain kinds of events occurred in his or her childhood, and is motivated to recall such events, childhood photos constitute a source of detailed and vivid perceptual images that may be combined with products of imagination to yield compelling pseudomemories. (p 149, True Photographs and False Memories, Lindsay et al, 2004)
In this study, 45 undergraduate college students (36 female and 9 male) volunteered to participate in exchange for bonus points in an introductory psychology course. Each subject’s parents were asked to describe two school-related events their child experienced, one event from grades 3 or 4 and one from grades 5 or 6. The parents were instructed not to select events that were frequently retold family stories; rather, they were instructed to select events their child wouldn’t be likely to remember. According to parents’ reports, no subject in the study ever put Slime (toy produced by Mattel) in their teacher’s desk in first grade, and so the placing of Slime in the teacher’s desk was used as the pseudoevent. Parents gave the researchers their child’s class photograph from each year for which they had described events.
Interviewers read each event aloud to the subject, beginning with the most recent event and ending with the first grade pseudoevent. A randomly selected half of the subjects were given copies of their class photos corresponding to each event, whereas the other half were not given photos. The interviewers urged the subjects to remember as much as they could about each event by using guided imagery and mental context reinstatement techniques.
The subject then rated (a) the extent to which the memory experience resembled reliving the event (from 1, not at all, to 7, as clearly as if it were happening right now), (b) the extent to which the subject felt he or she was remembering the event (same scale as for the reliving question), and (c) his or her confidence that the event had occurred as described in the narrative (from 1, 0% confident, to 7, 100% confident). The pseudoevent narrative was customized to use the subject’s name and his or her teacher’s name… (p 150, True Photographs and False Memories, Lindsay et al, 2004)
After this initial interviewing session, each subject was told to focus on the oldest event i.e. the pseudoevent. They were heavily encouraged to spend time trying to remember as much as they could on a daily basis, they were provided with printed copies of the pseudoevent narrative (as well as class photographs for half of the subjects), and they were instructed not to discuss the event with anyone else. Four days later, each subject was called by an interviewer and once again heavily encouraged to remember the pseudoevent, and subjected to guided imagery techniques as well as mental context reinstatement techniques.
The subjects then were reinterviewed and once again rated the extent to which they felt they were reliving the event, the extent to which they felt they were remembering the event, and their confidence that the event had occurred as described. Trained blind judges reviewed transcripts of the second interviews and determined whether individual subjects had experienced “(a) no images or memories, (b) images but not memories, or (c) memories of putting Slime in the teacher’s desk” (p 151, True Photographs and False Memories, Lindsay et al, 2004).
Results and Implications
At the point of the first interview, the use of photographs did not make a statistically significant difference in the reports of false memory. 13.6% of the no-photograph group were determined to have pseudomemories of the event, and 31.8% had images but no memories. In the photograph group, results were similar. However, by the second interview, the photograph group took the lead with 65.2% having pseudomemories and 13% having images but no memories, totaling to 78.2%. Clearly, the use of childhood photographs advances the development of pseudomemories when presented alongside coaching, guided imagery, mental context reinstatement techniques, etc.
The introduction of the study suggested that the researchers were interested in mimicking real-world circumstances for false memory development:
…the false-photo procedure suffers an obvious limitation in ecological validity and generalizability: People rarely encounter doctored photos of themselves doing things they have never really done. People do, however, sometimes review old family photo albums. Moreover, some trauma-memory-oriented psychotherapists and self-help books have recommended that adults who think they may have been abused in childhood but do not recall such abuse should review family photo albums (e.g., Dolan, 1991; cf. Poole, Lindsay, Memon, & Bull, 1995). (p 149, True Photographs and False Memories, Lindsay et al, 2004).
However, their methodology is very far from mimicking real-world circumstances. In the case of self-help books and psychotherapists, the subjects already believe or suspect they have been abused in childhood. They would not be seeking trauma-memory-oriented self-help books or psychotherapists otherwise. In the study, the subjects are fed false narratives and given photographs to encourage their initial development of the idea that the false memory exists.
Also, the interviewing of the subjects is quite heavy-handed in terms of encouraging false memories. Except perhaps in the case of a very few misguided psychotherapists and their most vulnerable patients, it is exceptionally rare that a subject in the real world would be told of the false memory alongside real memories, given a printed copy of the false narrative, subjected to guided imagery and mental context reinstatement techniques multiple times, encouraged to force recall multiple times, and interviewed about the false memory multiple times. It is hardly surprising that by the second interview, 78.2% of the subjects had partial or complete false memories. They were meticulously guided to that conclusion.
The study suggests that false memory formation occurs mostly under extraordinary circumstances, and even then, 21.8% — one out of five subjects — do not cave in to false memory formation of any kind. A person who already spontaneously suspects abuse in their background and turns to an overly suggestive self-help book is not equivalent to a subject in this study, in which many steps are taken over the course of days to ensure the development of false memories as thoroughly as possible. The danger of a study such as this one is the potential harm it poses to actual abuse survivors. One could easily imagine the figures of the study being taken out of context by a zealous defense attorney or a parent in denial — 78.2% is an impressive figure. It is clearly possible to develop false memories. Nearly 8 out of 10 people in this study did so. The researchers even go so far as to caution those suspecting childhood abuse:
…there is little reason to doubt that the mechanisms involved in our effect can contribute to other sorts of false memories, and therefore our results warrant concern about the riskiness of encouraging clients to review old photo albums during attempts to “recover” suspected but nonremembered histories of childhood sexual abuse. (p 154, True Photographs and False Memories, Lindsay et al, 2004)
This conclusion is misguided and dangerous. Once again, a person who already suspects abuse and who chooses to peruse childhood photos is in no way equivalent to a subject of the study, for whom the very idea of the pseudoevent was fed to him or her by an interviewer and who was subjected to multiple forms of repeated intensive encouragement to fabricate memories. A more apt conclusion would have been that people often develop a false memory when they are meticulously, repeatedly, almost coercively encouraged to develop that false memory under circumstances that very rarely occur in the real world. Childhood sexual abuse, on the other hand, certainly does occur in the real world frighteningly often.
Another major issue with Lindsay et al’s study is the equation of traumatic memories with non-traumatic memories. How would this study have been different if the pseudoevent had been, for instance, killing one’s pet as opposed to an innocent childhood prank? Perhaps the subjects would have been more resistant to developing that false memory given its upsetting nature. To draw conclusions about trauma, one needs to study trauma, not innocuous happenings.
Lastly, one must call into question the biases the subjects may have had. Almost all the students who participated in the study were female — 80%, or 8 out of 10. Could this disparity have affected the results? Additionally, these were new psychology students being interviewed at their own university in exchange for a reward. Is it possible that the students felt pressure to perform well in the experiment, leading them to agree more readily to false memories? And then there is the role of the parents in providing narratives from each subject’s childhood. Could the students, knowing their parents provided all the narratives (and assuming their parents provided the narrative of the pseudoevent as well), have placed extra weight on the potential veracity of the pseudoevent? After all, their mother and father supposedly remembered it (very unlike cases of supposed “false memory syndrome” in which the child claims abuse and the parents vehemently deny it). The parents asserting that a pseudoevent occurred is markedly different from a therapist or self-help book asserting that a pseudoevent occurred.
Lindsay et al’s study did prove its hypothesis to be true. Using photographs while repeatedly urging a person to remember a pseudoevent appears to contribute to the more frequent and more vivid development of pseudomemories. However, the conclusions drawn from the study about real-world false memory formation are poorly evidenced and potentially harmful. It would be interesting to conduct a study that corrects for some of the issues in the study by Lindsay et al. For example, what if the researchers chose to mimic problematic self-help books by introducing the pseudoevent without simultaneously introducing true events and then providing text that encouraged memory retrieval and false memory formation in lieu of repeatedly coaching the subjects in person and over the phone? The parents would not be involved. Also, perhaps this new study would have a much more diverse sample population of subjects not affiliated with the institution performing the research. In terms of correcting for the disparity between traumatic and non-traumatic events, one has no choice but to study validated traumatic memory in individuals, as it is unethical to induce traumatic pseudomemories.
Dr. Elizabeth Loftus sought to understand if, and how, an entire false memory can be implanted. She wondered as well, how could this be done in an ethical manner, particularly regarding a pseudomemory that is potentially distressing?
In order to conduct this study, Loftus told 24 subjects (3 males and 21 females) that they were participating in a study about “the kinds of things you may be able to remember from your childhood” (p 3, Formal Study of Lost-in-a-Mall, Dr. Elizabeth Loftus et al, 1995). The subjects were then given booklets containing descriptions of four events from their childhood (ascertained by having parents or siblings fill out questionnaires). One of these four events was actually a pseudoevent, that being a description of the child getting lost in a shopping mall at age five. The pseudoevent narratives were tailored to the names and locations of each subject, and the subjects were encouraged to write below each paragraph everything they remembered about what had occurred.
After completing the booklet, each subject attended two interviews. In the first interview, subjects were encouraged to remember as much as they could about each event. The event paragraphs were referenced and quoted to encourage memory retrieval. Subjects rated each memory on clarity and confidence in accurate recall, and as they left, were reminded to continue trying to remember more details.
At the second interview, subjects were asked once again to rate the clarity and confidence of accurate recall for each of the four events described. They were then told that one of the four events was false (19 out of 24 subjects correctly guessed which one was the pseudoevent). The true purpose of the study was revealed and apologies were offered for the deception.
Results and Implications
Of the 72 true events the 24 subjects were asked to reveal, a total of 68% of true events were remembered. 6 out of 24 subjects (25% or one out of 4) claimed to remember the pseudoevent. Subjects on average used 49.9 words to describe pseudomemories and 138.0 words to describe true memories, indicating that people tend to be less descriptive about pseudomemories than they are about true ones.
The study by Loftus contained many flaws as well. For example, most subjects were female, as in Lindsay et al’s study. Subjects may have conflated actual childhood incidents of being lost with the specifics of the pseudoevent. Subjects were heavily encouraged multiple times to remember both true and pseudo events and all events were supposedly provided by family members, which differs greatly from any likely real-world scenarios (again, as in Lindsay et al’s study).
However, Loftus is more cautious in her conclusions:
People can be led to remember their past in different ways, and they can even be led to remember entire events that never actually happened to them. When these sorts of distortions occur, people are sometimes confident in their distorted or false memories, and often go on to describe the pseudomemories in substantial detail. These findings shed light on cases in which false memories are fervently held — as in when people remember things that are biologically or geographically impossible. The findings do not, however, give us the ability to reliably distinguish between real and false memories. Without independent corroboration, such distinctions are generally impossible. (p 10-11, Formal Study of Lost-in-a-Mall, Dr. Elizabeth Loftus et al, 1995)
Unlike Lindsay et al., Loftus does not extrapolate in this study that people who believe they may have been abused should be cautious of certain books or pictures. She merely states that perhaps her research could “shed light” on situations where people report verifiably false memories.
That being said, Loftus has made many public statements since the publication of this study that express support for disproven phenomena such as False Memory Syndrome (in which those coming forward with reports of childhood abuse are accused of having iatrogenically fabricated these memories).
There appears to be a pattern of conclusions being drawn from studies such as the two examined in this review that are biased in favor of recovered childhood sexual abuse memories being false. Interestingly, this research occurred alongside the development of the False Memory Syndrome Foundation and many of the oft-referenced scientists involved (including Elizabeth Loftus, Pamela Freyd, Peter Freyd, John Kihlstrom, and Ralph Underwager) have strong ties to the FMSF. It is worth noting that Peter Freyd and Ralph Underwager were both accused of perpetrating child sexual abuse (and Underwager resigned from his position as board member on the FMSF after it was discovered that he had released pro-pedophilia statements to a Dutch pro-pedophilia publication called Paidika).
This is not to suggest that all those in defense of the existence of false memories are child abusers or child abuse apologists. False memories do occur under specific circumstances, as the research reflects, and this is worth studying further. However, it does call for a thorough review of the motivations and affiliations of all those researchers who used their data to support the FMSF and/or massively discount the stories of survivors coming forward.