The issue of repressed memory is surprisingly heated, and comes loaded with the weight of several fundamentally different conceptions of the human mind. The real issue, hidden away in the word “repressed,” is whether a memory once repressed can be unleashed, or “recovered.” Thus we find ourselves in the awkward scientific predicament of having, for each paper published presenting evidence of repressed and recovered memories, at least a few papers published expressly to disclaim it. This article will attempt to make some sense of the available literature, drawing in alternative paradigms in addition to empirical studies.
Examples & Evidence
The issue of anecdotal versus experimental evidence becomes extraordinarily important in discussions of recovered memories, as we shall see, primarily because experimentalists have not managed to invent a way to replicate the phenomenon under controlled conditions. This creates a kind of paradigmatic crisis for experimental psychologists, for whom the temptation to discount uncontrollable phenomenon is a normal hazard of the workplace, so to speak. However, we may do well to consider, as Terence McKenna pointed out in an unrelated discussion, the etymology of “anecdote”, which stems from the Greek “anekdota”, meaning “unpublished items.” An anecdote then, can be thought of as a casually observed phenomenon; once that phenomenon is subjected to scrutiny, corroborated, and–specifically–distributed in a peer-reviewed publication, it ceases to be anecdotal evidence by definition and can instead be considered “medical history” (McKenna, 1998)
Probably the best source of this type of medical history of repressed-and-then-recovered memories is Dr. Ross E. Cheit’s Recovered Memory Project, which maintains an internet database of individual cases of recovered memory. “The cases are annotated and all have corroboration, including medical evidence, confessions, multiple victims, or even eyewitness testimony” (Cheit, in press). The database includes cases in three categories: those extracted from legal proceedings, in which the facts surrounding the repressed memories were subjected to extreme scrutiny and were in most cases corroborated; clinical accounts printed in scientific and academic journals; and other cases, drawn from sources such as legal proceedings that were not allowed to continue for various reasons, or from meticulous journalistic investigations.
One excellent example to be found in the Recovered Memory Project’s archive of clinical accounts is the case of “Claudia”, published in the highly respected journal Science News. Claudia had enrolled herself in an intensive inpatient weight-loss program to battle the severe obesity she was suffering from at that time. After losing more than one hundred pounds in the program, she began to experience flashbacks of sexual abuse at the hands of her older brother. While still an inpatient in the weight loss program, Claudia joined a therapy group for incest survivors at the hospital. During meetings, she remembered and revealed to the group that “from the time she was 4 years old to her brother’s enlistment in the Army three years later, he had regularly handcuffed her, burned her with cigarettes, and forced her to submit to a variety of sexual acts” (Bower, 1993). Upon returning home from the hospital, Claudia was able to examine her brother’s old room and belongings, which had remained untouched in the fifteen years since he had died in Vietnam. “Inside a closet she found a large pornography collection, handcuffs, and a diary in which her brother had extensively planned and recorded what he called sexual ‘experiments’ with his sister.” (again, Bower, 1993). Dozens of powerfully corroborated cases like this exist and are referenced within the archive.
There is a great deal of additional published work in which the recovered memory phenomenon is found to be not only very real, but also relatively common. Psychologists conducting a broad survey of British Psychological Society practitioners, for example, found that “memory recovery appears to be a robust and frequent phenomenon” in cases of total amnesia related to childhood sexual abuse as well as cases relating to other types of traumatic events (Andrews et al, 1995). A 1993 survey of 450 adult clinical subjects reporting histories of sexual abuse found that more than half of the subjects could identify a period of life before the age of eighteen when they had possessed no recall of the event (Briere & Conte, 1993). Another study (Feldman-Summers & Pope, 1994) interviewed a national sample of psychologists; of the psychologists from this sample who reported childhood abuse of any kind, about 40% reported a period during which they had forgotten some or all of the abuse. Of these, nearly half reported some form of corroboration for the occurrence of the abuse. Also significant was the finding that age and gender were not related to the occurrence of amnesia, but the severity of abuse was a strongly correlated factor. Finally, a particularly interesting study interviewed women with previously documented histories of sexual abuse, asking detailed questions about each of their abuse histories. “A large proportion of the women (38%) did not recall the abuse that had been reported 17 years earlier” (Williams, 1994).
Objections & Rebuttals
The major objection to reports of recovered memories is the difficulty in distinguishing them from false memories (Reisner, 1996). While this objection is certainly valid, and the potential for inducing false memories is a serious one for the therapist (Ofshe & Watters, 1996; Loftus, 1994), the preponderance of evidence in support of the existence of legitimate instances of repressed-and-then-recovered memory precludes any serious debate as to the general existence of recovered memories as a phenomenon. The recovered memory phenomenon has been sufficiently observed in a naturalistic environment that it is now fair to assume that any shortcomings in the way of controlled modeling or reliable detection of repressed information can be seen as problems of experimental science and therapeutic technique, rather than as invalidations of the existence of the phenomenon itself.
Furthermore, claims that individuals who report recovered memories are more likely to present symptoms of false memory syndrome (FMS) have been found by some studies to be grossly inaccurate. Hovdestat and Kristiansen (1996) found that FMS indicators were much less common in their sample than the hype surrounding FMS would suggest–no more common, in fact, for subjects reporting recovered memory than for those reporting that their memories of traumatic events had been continuous. In “the first direct investigation of suggestibility among patients who report recovered memory,” Rush Medical College’s Frank Leavitt used the Gudjonsson Suggestibility Scale to compare the overall suggestibility of subjects reporting recovered memories of childhood sexual abuse with that of subjects reporting no history of sexual trauma. “Results indicate that patients who recover memories were remarkably less suggestible than the clinical field has been led to believe by advocates of false memory…. Paradoxically, patients without a history of sex abuse were more at risk for altering memory to suggestive prompts” (Leavitt, 1997).
Models & Theories
Neuroscience, unfortunately, has little to offer us in the way of explaining this phenomenon thus far. While it is able to tell us that extraordinarily stressful conditions can lead to malfunctions of the hippocampus which prevent explicit memories of the surrounding events from being stored, and that implicit memory systems remain functional under these conditions, it is unable to explain how an explicit, conscious memory could be derived from an implicit, unconscious-emotional one. It would seem, in fact, that the implicit memories stored under such circumstances lack sufficient detail to reasonably allow for such a reconstruction to take place with any degree of accuracy at a later time (LeDoux, 1996). Therefore, the current research in this area leads us to assume–for the time being–that proper explicit memories are, in fact, being formed during the traumatic episodes in question and then repressed by way of an unknown mechanism. However, the research is incomplete. It is entirely possible that there is indeed enough information stored by emotional systems to reconstruct traumatic events in detail, and that neuroscientists have simply not yet discovered a way to reliably elicit such reconstruction. It is also possible that there are entire yet-unexplored memory systems at work.
The available research and case studies do, however, offer us significant clues as to the structure and function of memory repression. For example, the Feldman-Summers & Pope finding (1994) that personal indicators and demographics are generally unrelated to the occurrence of amnesia for traumatic events, and that this occurrence is instead reliant only on the severity of abuse, seems to support the hippocampal-interference model of memory formation under severe stress, suggesting that perhaps retrieval of and inference from implicit memory–or the storage of such memories in general–may be underestimated. Another study, however, indicates that the duration of the abuse is also a factor in the extent to which memory repression occurs in victims of childhood sexual abuse (Herman & Schatzow, 1987), which would seem to support the opposite hypothesis that explicit memories are in fact being stored (in the early stages of abuse at minimum), and then actively repressed at the psychological level.
Surprisingly, the most widely accepted model for this phenomenon may still be the Freudian one. It does appear that memory repression is a system of self-defense “by which an individual defends himself or herself against the conscious recollection of a traumatic memory and its associated consequences” (Golding, Sanchez, & Sego, 1996). Jennifer Freyd (1994) builds on this model by speculating that “victims may need to remain unaware of the trauma not to reduce suffering but rather to promote survival. Amnesia enables the child to maintain an attachment with a figure vital to survival, development, and thriving.” Individual cases like that of “Claudia”, in which memories were recovered during the course of extreme weight-loss (Bower, 1993), seem to point us in the direction of exploring externalized methods that individuals might use to assist them in this repression, such as abnormal weight gain. These could serve as a kind of emotional red herring, a reason for the emotions to exist without the necessity of remembering the events that actually created them.
Competing theories include the dissociative model, in which the memories are passively excluded from conscious awareness (Smith, 2000), and Philip T. Smith’s jigsaw-puzzle model: He provides a fairly eloquent description in a paper published in Memory in 2000:
According to the jigsaw model, unwanted memories can be rendered inaccessible in two different ways: by assigning a low importance value to their constituent fragments, or if the importance value is not sufficiently within a person’s control, by trying to ensure that few fragments are simultaneously encoded with the fragment to be forgotten. The former process shares features with repression, the latter process shares features with dissociation. What is novel about the jigsaw approach is that repression-like and dissociation-like phenomena could be seen to derive from initial encoding processes: there is no need to invoke active coping strategies in later storage and retrieval (Smith, 2000).
While this theory is compelling, it is clear that a great deal more research–particularly neuropsychological research–will be required to validate any model of memory recovery.
Practical Recovery
Finally, and perhaps most controversially, is the issue of inducing the recovery of lost memories. One study found that short-term group therapy “proved to be a powerful stimulus for recovery of previously repressed traumatic memories” (Herman & Schatzow, 1987), while another found that, in reality, “psychotherapy was the least commonly reported trigger” for such recall (Elliott, 1997). This study reported that most repressed events are actually recalled while watching television. In cases where recovered memories do arise during the course of psychotherapy, it would appear they do so “more typically during periods of positive rather than negative feeling toward the therapist, and they were more likely to be held with confidence by the abuse victim” (Dalenberg, 1996). At the present time, there simply does not appear to be a reliable way to elicit memory recovery, nor does there appear to be a reliable way to discern whether a “recovered” memory has any basis in truth, short of establishing external corroboration (LeDoux, 1996).
Conclusion
While it is clear that memories of traumatic events are in some cases lost or inaccessible for indefinite periods of time and then recovered, it is unclear what underlying neurological or psychological mechanisms might be responsible. A biological basis has been found for traumatic amnesia on its own, but this explanation is unable to account for–and would seem to preclude the possibility of–the potential of recovery. The major psychological models used to account for the phenomenon are repression and dissociation, although neither has much more experimental support than the other. What’s more, psychotherapy seems to be hit-or-miss at best when it comes to recovering lost traumatic memories, and the potential for the creation of false memories is profound. A great deal more research will be necessary, but a great deal more theoretical clarity will be necessary before research in this area is likely to be beneficial.
