Beverly R King. 21st Century Psychology: A Reference Handbook. Editor: Stephen F Davis & William Buskist. Volume 1. Thousand Oaks, CA: Sage Publications, 2008.
In 1885 the first person to experimentally investigate the properties of memory, German psychologist Herman Ebbinghaus (1885/1913), published information on the curve of forgetting, which showed the progression of forgetting with the passage of time. Essentially, this curve demonstrated that (up to a point) as more time elapses, more information is forgotten—at least in the case of rather innocuous material such as nonsense syllables. Ebbinghaus’s observations are still valid, more than 120 years later. In 1932 British psychologist Sir Frederic C. Bartlett concluded that human memory is a reconstructive rather than a reproductive process. That is, our memories are never exact duplicates of the experiences they purport to capture but rather are rebuilt from bits and fragments with each recollection. Both cognitive psychologists and neuroscientists have accumulated research evidence that Bartlett’s observation was accurate (Schacter, 1996). Memories typically are less available over time and may not be recalled with absolute accuracy, but is it possible for a person to have complete amnesia for an entire experience and then retrieve relatively intact memories of that experience after a lengthy period of time has elapsed? This is the question at the heart of one of the most contentious debates in the history of psychology—the recovered memory debate.
Although psychologists had discussed the notion of memory repression at least since Freud, it garnered special attention in the late 1980s and early 1990s with the publication of books such as The Courage to Heal (Bass & Davis, 1988). These books encouraged individuals who knew that they had been sexually abused to seek help and encouraged others to think about whether they might have been sexually abused based on the possession of particular symptoms sometimes presented in the form of a checklist. The implication was that memories such as those of childhood sexual abuse (CSA) might be put completely out of awareness (or “repressed”) due to their traumatic nature but could be subsequently recalled (or “recovered”) with the help of certain techniques and/or an experienced therapist.
In a number of cases, adult women came forward to report memories of CSA recovered in therapy but later recanted their stories (Loftus, 1997) after therapy was discontinued. In addition to the stress undoubtedly engendered by the recovery of “false” memories, some of these women accused family members of the abuse and even brought criminal charges against their alleged abusers before recanting (Ost, 2003). Thus began what Ost and others have referred to as “the memory wars.” The “wars” center on whether it is actually possible to repress and later recover memories of trauma. The two sides of the debate are summarized as follows: On one hand are those, often mental health practitioners, who believe that traumatic events such as CSA may be stored by a different process than “normal” memories such that they are difficult to recall but can be retrieved relatively intact years later with the right cues. This special storage process or amnesia protects victims of CSA from the pain of these traumatic memories. On the other hand are individuals, often academics or research scientists, who claim that there is no empirical evidence for this special process (American Psychological Association [APA], 1995; Clancy, Schacter, McNally, & Pitman, 2000) and that traumatic events typically are not forgotten, although people may try not to think about them for long periods of time (McNally, 2004). They also point out that in many incidences of so-called recovered memories, the memories were “recalled” after particular media exposure (e.g., television shows or self-help books about repressed memories) or following participation in inappropriate, suggestive, and perhaps unethical therapy techniques.
No one, on either side of the debate, disagrees that sexual abuse of children occurs or that forgetting can occur. In both retrospective and prospective studies, researchers have found some adults who had periods in their lives during which they could not recall a traumatic event such as sexual or physical abuse (see Berliner, Hyman, Thomas, & Fitzgerald, 2003, for a review of some of these studies). Some psychiatric disorders even have as part of their diagnostic criteria amnesia for all or part of a traumatic episode (e.g., post traumatic stress disorder [PTSD] and acute stress disorder; American Psychiatric Association, 2000). However, there usually is consensus that most individuals who were sexually abused as children remember all or part of what happened to them although they may not fully understand or disclose it (APA, 1995). An extensive literature based on research with both animals and humans indicates that traumatic memories, especially the central details, are not easily forgotten (Alison, Kebbell, & Lewis, 2006).
The question remains: is it at all possible for a person to have complete amnesia for an entire experience and then retrieve relatively intact memories of that experience after a lengthy period of time has elapsed? Due to ethical constraints, we cannot test this issue directly, and neither clinicians nor researchers would argue that it is impossible for a memory to be repressed and later recovered. After all, as McNally (2004) has pointed out in regard to this issue, “one can never prove the null hypothesis that a phenomenon does not occur” (p. 100). However, several lines of research over the past two decades have focused on the following questions:
- Memory for actual traumatic events: How likely is it that memories of traumatic events will be recalled, both shortly after the events have occurred and years later?
- Memory distortion: How, and under what conditions, might recollections of past events become distorted?
- False memory creation: Is it possible for individuals to construct entirely false memories (or “pseudomemories”) for events that never occurred but in which they strongly believe? Especially of interest has been whether or not techniques employed in psychotherapy can lead to the creation of false memories.
The concept of repression comes from Freudian theory. To Freud (1910), the process of repression involved driving a memory out of the conscious into the unconscious mind even though the repressed material could continue to have influences on behavior. Repression could occur either during or after an event and was one mechanism that could lead to amnesia for traumatic events (Allison et al., 2006). Dissociation was a different mechanism, in which memories of traumatic events were somehow disrupted and which resulted in memory impairment rather than fullblown amnesia (Allison et al., 2006). Currently, theorists use the term “dissociative amnesia” to describe a condition of memory loss greater than ordinary forgetfulness (Porter, Campbell, Birt, & Woodworth, 2003) and “repression” to describe one process by which an event might be forgotten, at least for a while.
Two other theoretical models are of relevance to the repressed and recovered memory issue. Both of these would relate to how false memories might be created. One is a model such as that of Smith (2000), which describes how memories are constructed upon recollection. In Smith’s model,
The raw material of memory is a set of stored fragments … Memory retrieval is an iterative process involving interactions between memory fragments and a coherent jigsaw, called theCurrent Memory, which is assembled from the memory fragments…. In assembling this jigsaw, a measure of goodness-of-fit is computed (Harmony) and this corresponds to the confidence the memory retriever might feel about the authenticity of the retrieved information. (p. 246)
In other words, just as Bartlett observed in 1932, memory is a reconstructive process. When bits and pieces of memory fragments are put together for recollection, the memory retriever might have a great deal of confidence that the memory is being recalled accurately when, in fact, it may be that the “measure of goodness-of-fit” is computed incorrectly due to other forces such as emotions, imagination, or familiarity.
The second relevant model is one that describes how false memories might be created (Loftus & Bernstein, 2005). In this model (Mazzoni, Loftus, & Kirsch, 2001), a three-step process leads to the formation of false memories:
- Individuals must see events to be incorporated into memory as plausible.
- They must come to believe that they had experienced the event.
- They must reinterpret narratives or images about events to produce what they feel are genuine memories.
A large body of empirical evidence exists that indicates not only that the creation of false memories is possible but also that it is surprisingly easy to implant them in the minds of research participants.
Methods and Research Results
Support for the Mazzoni et al. (2001) model and research related to false memory creation is discussed following presentation of research methods and findings relevant to memory for actual traumatic events and memory distortion.
Memory for Actual Traumatic Events
Are memories for traumatic events remembered more vividly than memories for everyday events (the trauma superiority argument), or are memories of traumatic events somehow encoded, stored, or retrieved by different processes such that they can be forgotten for long periods of time (the traumatic memory argument; Alison et al., 2006)? This question is typically addressed by asking individuals who have experienced past trauma to recount their experiences, either directly or while being questioned on another issue. Overwhelmingly, the research supports the trauma superiority argument—memories for stressful experiences are not easily forgotten, especially the central details of the events. For example, in studies of children who have experienced traumatic events such as parental homicide, physical abuse, or holocaust experiences, they remember these experiences very well and can provide detailed, vivid accounts of the events (see McNally, 2003, for a review). Similarly, in both retrospective and prospective studies with adult respondents, traumatic memories for events such as genocide, torture, disasters, and accidents are usually remembered both continuously and vividly, with central details more likely to be remembered than peripheral details (see McNally, 2003, and Porter et al., 2003, for reviews). Certainly forgetting can occur, not only for relatively ordinary events but also for more significant ones (Porter et al., 2003). However, if adult victims of trauma have forgetfulness problems after their experiences, it is usually everyday forgetfulness and not an inability to remember the event(s) (McNally, 2003).
In studies specifically focused on participants who had experienced sexual abuse as children, a minority (percentages vary by study) do state that there was a period of time in their lives when they had no recall of the experience and recovered the memory after a period of forgetting. In a national probability sample of 711 women in the contiguous United States, more than a fourth of those who reported sexual coercion or abuse while growing up indicated that they had forgotten the abuse for some length of time and later remembered it on their own (S. C. Wilsnack, Wonderlich, Kristjanson, Vogeltanz-Hohn, & R. W. Wilsnack, 2002). However, in the majority of historical cases of abuse, individuals bringing the complaints report that they had always been able to recall the incident(s) since its (their) occurrence but simply chose to delay disclosure (Porter et al., 2003). The APA (Alpert, 1996) maintains that most victims of CSA remember at least some of what happened to them continuously from the time of abuse.
There are a number of reasons that individuals may not think about a traumatic experience such as CSA for an extended period of time. For example, they may not have interpreted the experience as traumatic when it happened (they had no schema for the experience in childhood); they may have intentionally suppressed the experience; they may simply choose not to talk about abusive experiences for various reasons; or they may have truly forgotten the experience (McNally, 2003; Porter et al., 2003).
Goodman and colleagues (2003) contacted 170 individuals who were victims of CSA and who had participated in a study of court testimony 13 years earlier. They surveyed these individuals three times: first by phone, then by questionnaire, and finally in a face-to-face interview. Only slightly over 8 percent of these individuals did not report the abuse in any phase of the study, and they were more likely than those who did report to have been under age five at the time of the abuse and to have been victims of less severe abuse.
One of the leading researchers on issues related to repressed and recovered memory is Elizabeth Loftus. Her work on how memory can be distorted began in the early 1970s when she began studying how leading questions can influence memory; later, she studied what has come to be called “the misinformation effect” (Loftus & Hoffman, 1989). Literally hundreds of studies have now been conducted by Loftus and her students, all of which show that when people witness an event and are later exposed to new and misleading information about it, they often “misrecall.” Their recollections become distorted as the result of suggestion (Loftus, 1997). In the real world, this information might come from media sources or interrogators; in the laboratory, the information is fed to participants by the researcher.
In one of her earliest studies, Loftus showed research participants films of traffic accidents and then asked questions in various forms about what they saw (Loftus & Palmer, 1974). Individuals who were asked “How fast were the cars going when they smashed into each other?” gave faster speed estimates than individuals who were asked “How fast were the cars going when they hit each other?” Using the word smashed versus the word hit also led more individuals to report seeing broken glass when there was no broken glass in the original film. Leading questions can distort memory, as can information coming from a source other than the event to be remembered. In another study, participants saw a simulated automobile accident; half of the participants then received misleading information about the accident (the false suggestion that the stop sign they saw in the simulated accident was actually a yield sign). These individuals were much more likely to have inaccurate memories (to incorporate the yield sign into their memories) than the research participants who did not receive misleading information (Loftus, Miller, & Burns, 1978). In addition to recalling yield signs instead of stop signs, people in these types of studies have recalled nonexistent tape recorders, a blue vehicle as white, and Mickey Mouse as Minnie (see Loftus, 2005a, for a review of Loftus’s 30 years of investigations on this topic). Hundreds of studies on the misinformation effect show that misleading information “… can change an individual’s recollection in predictable, and sometimes very powerful, ways” (Loftus, 2005b, p. 1).
False Memory Creation
Psychological research has demonstrated not only that memory can be distorted but also that entirely false whole memories can be created. These rich false memories involve “the subjective feeling that one is experiencing a genuine recollection, replete with sensory details, and even expressed with confidence and emotion, even though the event never happened” (Loftus & Bernstein, 2005, p. 101). Loftus and Pickrell (1995) conducted the first research study in which an attempt was made to create an entire memory for something that never happened. The procedure they used to do this has been called the familiar informant false-narrative procedure (Lindsay, Hagen, Read, Wade, & Garry, 2004) or, more simply, the lost-in-the-mall technique (Loftus, 2003b). In this study, participants read short narratives—all of which they thought were true stories supplied by family members; however, one was a false event. The false event was a plausible shopping trip to a mall during which the participant was ostensibly lost for an extended period of time but eventually rescued and comforted by an elderly woman and returned to the family. After participants read each story, the researchers instructed them to write what they remembered about the events described. In two follow-up interviews, participants provided as much detail about the memories as they could and described how their memories compared with those of the relative who provided the information. Parts of the narratives were provided as retrieval cues. A majority (68 percent) of participants recalled something about the true events; 29 percent remembered, either completely or in part, the false event after the initial reading; and 25 percent continued to assert that they remembered the false event in the follow-up interviews.
The Loftus and Pickrell (1995) study was followed by numerous studies that showed that both strong and relatively subtle suggestions could lead to people believing “that they had experiences in childhood that they almost certainly did not have” (Loftus, 2003a, p. 108). For example, in several studies by Hyman, Husband, and Billings (1995), college students were asked to recall events from their childhood. Some of the events were true events as reported by parents and some were false events, such as a hospitalization, a birthday party, spilling punch at a wedding, or evacuating a grocery store when an overhead sprinkler system went awry. No student recalled the false event in a first interview, but in a second interview 18 to 20 percent recalled something about the false event. A long list of subsequent studies further documents the types of false memories that research participants can be lead to believe: being attacked by an animal (Porter, Yuille, & Lehman, 1999); falling off a bicycle and receiving stitches in their leg (Heaps & Nash, 1999); and being saved from drowning (Heaps & Nash, 2001), to name a few.
These types of studies have been criticized on two fronts:
- Perhaps instead of implanted false memories, participants were simply recalling memories of events that had really happened to them. Maybe several participants in the lost-in-the-mall study really had been lost at some point in childhood and were simply elaborating on an existing memory.
- More pertinent to the repressed/recovered memory debate, all of these events were relatively common or at least entirely plausible; that is, they could have happened. Wouldn’t it be much less likely that someone could, under suggestion, recall a traumatic event such as sexual abuse that never actually happened?
Researchers have devised several techniques to address both of these criticisms. In each technique, researchers have made an attempt to implant memories for events that could not have happened or that are highly improbable. Pezdek and colleagues (e.g., Pezdek, Finger, & Hodge, 1997; Pezdek & Hodge, 1999) found that less plausible events (e.g., having had a rectal enema as a child) were less successfully implanted than more plausible events. However, some researchers have been successful at implanting quite implausible memories. Mazzoni et al. (2001) used the not just implausible but bizarre target event of seeing, as a child, someone possessed by demons. All participants in this study rated a list of events in terms of the plausibility that they could have happened to other people and the confidence that they had not been experienced by the participants. Then, some of the participants read fake articles that described demonic possession as common; these same participants then took a fear survey and had it interpreted for them, indicating evidence that they had witnessed demonic possession in childhood. Finally, all participants repeated the plausibility and confidence ratings. Participants who had been exposed to the fake articles and survey results thought it was more plausible and more likely that they had indeed witnessed demonic possession as a child. The manipulation was effective even with people who entered the study believing that demonic possession was rather implausible.
This study by Mazzoni et al. (2001) demonstrates the stages in the process by which false memories may be created. Research participants came to see demonic possession as plausible though the use of fake articles that portrayed demonic possession as common and came to believe that they had experienced the event by virtue of survey results that they thought indicated a history of witnessing demonic possession. What might lead people to come to believe that they have experienced an event even when they have not? Other studies indicate that belief may be created through a process of “imagination inflation” (imaging an event can increase confidence that one has actually experienced the event; Garry, Manning, Loftus, & Sherman, 1996). Garry et al. (1996) asked research participants to indicate the likelihood that each of 40 events had happened to them during childhood. Two weeks later, they were instructed to imagine that some of these events had happened to them (e.g., breaking a window with their hand) and, later again, were asked again to indicate the likelihood that each of the 40 events had happened to them. For the broken window scenario, 24 percent of those who imagined the event reported increased confidence compared to 12 percent of those not asked to imagine the event.
Thomas and Loftus (2002) investigated imagination inflation for more unusual experiences. Participants were asked to perform or imagine some ordinary actions and some more unusual ones (e.g., sitting on dice). Later, they imagined some of the actions zero to five times. Imagination inflation occurred even for the unusual actions; the more participants imagined an action, the more inflation occurred. Why might the act of imagining an event increase a person’s belief that the event happened to them in the past? Loftus (2005a) posits two possible reasons: (a) imagining an event might remind some participants of a true experience from their past or, more likely, (b) imagining made an event seem more familiar to a participant, and the familiarity was mistaken as childhood memories.
Another interesting set of studies further illustrates how repeated exposure can raise the likelihood of false memory formation perhaps by increasing familiarity. In these studies, researchers asked participants to evaluate advertising copy for a Disney resort. Some of the participants saw a fake Disney ad containing a picture of Bugs Bunny and later were asked about any trips to Disney they had taken as children. Of individuals in one study (Braun, Ellis, & Loftus, 2002), 16 percent who were exposed to the fake ad claimed to have met Bugs Bunny when they visited Disney (an impossibility, as Bugs Bunny is a Warner Bros. cartoon character). Repeated exposure to the fake ad raised the false memory rate to 25 percent in one study and 36 percent in another (Grinley, 2002). In one of these studies, participants who falsely recalled having met Bugs at Disney had quite detailed recollections; some remembered shaking his hand; others recalled hugging him; and still others had recollections of touching his ear, touching his tail, or hearing him say his familiar line, “What’s up, Doc?”
False memory rates can be boosted even higher when photographs are used as memory cues. Wade, Garry, Read, and Lindsay (2002) gave research participants photos of themselves as young children and asked them to recall the events depicted in the photos. Most of the photos were genuine but one for each participant was faked by digitally altering a photo of a hot-air balloon ride to include a real childhood image. Of participants, 50 percent recalled some memory of the hot-air balloon and often provided detailed descriptions of the ride despite the assurance of family informants that this event had never happened. In response to the criticism that studies using “doctored” photographs lack ecological validity, Lindsay, Hagen, Read, Wade, and Garry (2004) used a real photograph to help implant a false memory. Over two interviews, they asked participants to remember three school experiences—two were real but the earliest was false. The false event involved putting toy slime in a teacher’s desk. All participants received a narrative of the experience; half also received a class picture. Less than half of the narrative-only participants reported slime images or images with memories; about two-thirds of the photo participants did. The authors speculated that three different mechanisms might explain the high incidence of false memories in their study: (a) The photo may have added additional authoritativeness to the narrative; (b) the photo may have lead participants to speculate about details related to the false event; and (c) memories of details in the photo may have been mixed with imagination, leading to the creation of vivid images of the false event (a source monitoring error—see Comparisons section).
Given that the most heated arguments regarding whether memories can be repressed and then recovered are in cases of CSA, the memory research described is most applicable to judges, juries, and legal experts, who must assess the credibility of recovered memories of CSA, and to therapists, who might inadvertently lead their clients to develop false memories of CSA. Both Alison et al. (2006) and Porter et al. (2003) present recommendations and guidelines for judges, juries, and experts to help in evaluating allegations of CSA based on recovered memories. Neither set of authors argues that every case of alleged recovered memory of CSA is false; no doubt some cases of alleged historical abuse are accurate. How can the criminal justice system distinguish between the two? There is often no corroborating evidence in alleged cases of historical abuse, and there is always the fear of false convictions. Alison et al. (2006) suggest that each case should be evaluated on its own merits alongside research-based information. These authors advocate an approach in which supporting evidence is required to establish credibility in recovered memory cases.
Porter et al. (2003) provide four guidelines for deciding on the veracity of historical allegations that judges, juries, expert witnesses, lawyers, and police investigators could use. The first guideline deals with the context in which memories are recalled. When recovery of memories is accomplished by use of suggestive techniques, there is a greater likelihood of illusory memories being recovered than if memories are recalled spontaneously. The second and third guidelines deal with differences in the content of recalled memories or among persons who recover mistaken memories (both of which are described more fully in the Comparison section). Although more research is needed, researchers have found certain content differences between real and mistaken memories. Individuals with a greater susceptibility to suggestion and a tendency toward dissociation (and perhaps those who are more introverted and imaginative) tend to recall more mistaken memories than those who are low in these characteristics. Finally, in line with the recommendations by Alison et al. (2006), Porter et al. (2003) suggest that corroboration of any alleged event involving historical CSA should be sought. Corroboration adds credibility to a claim; lack of it raises doubts.
Probably more than any other application, attention has been directed toward whether therapeutic techniques can foster false memory recollections and what advice should be given to therapists related to this issue. Recovered memories that emerge in the process of psychotherapy have engendered one of the most heated aspects of the repressed memory controversy, not only because of the difficulty in distinguishing true from created memories but also because of ethical considerations—the lives of clients and their families can be significantly altered if CSA is suspected where none actually existed.
Researchers have applied the term “memory work” to psychotherapy techniques used to help individuals retrieve ostensibly repressed memories of childhood sexual abuse (Loftus, 1993). Some of the techniques in this category are guided imagery, suggesting false memories, hypnosis, searching for early memories, age regression, hypnotic age regression, past-life regression, symptom interpretation, bogus personality interpretation, dream interpretation, physical symptom interpretation, and bibliotherapy (Lynn, Lock, Loftus, Krackow, & Lilienfeld, 2003). How might psychotherapeutic techniques encourage the development of false memories? First of all, demand characteristics may be operating in a therapeutic setting, just as they frequently do in research settings. In research, demand characteristics refer to the cues that may influence the behavior of participants, perhaps by suggesting an experimental hypothesis. In psychotherapy, the term may refer to the cues that convey the therapist’s expectations to clients and influence their behavior (Kanter, Kohlenberg, & Loftus, 2002). A client may feel he or she needs to produce memories consistent with suggestions from a therapist simply because the therapist is seen as a trusted authority figure (Alison et al., 2006). In fact, the effects of suggestive techniques may be more pronounced in a clinical than in a laboratory setting because the influence of a regularly seen and trusted therapist on a vulnerable patient may be much greater than the influence of an experimenter on a research participant seen once or only a few times and who is participating for money or course credit (Lynn et al., 2003).
Secondly, all the research discussed above indicates that these techniques can not only help a person retrieve memories but also help create memories. (See Lynn et al., 2003, for a full review of these techniques and how they may be used to help create memories.) Loftus and Pickrell’s (1995) lost-in-the-mall studies as well as others using the same methodology have demonstrated that, when scenarios are suggested to individuals and they are asked to either imagine or otherwise think about the events, they may create detailed memories that are obviously false but which they come to believe are true parts of their history. These suggestions may come in a variety of forms such as symptom interpretation or guided imagery (Lynn et al., 2003). Checklists of symptoms that therapists or authors may suggest are indicative of a history of abuse (but which in reality could apply to almost anyone) may lead some people to create a history to match the symptoms. Even more likely to lead to the formation of false memories would be a therapist’s suggestion that a person has certain personality characteristics indicative of abuse. The APA has been clear in its assertion that there is no one set of symptoms that indicates that someone has been abused. A therapist or book author who suggests otherwise is not basing his or her assertion on empirical evidence (APA, 1995).
Another psychotherapeutic technique that can be problematic in terms of leading to false memory creation is guided imagery. In guided imagery, clients “are instructed to relax, close their eyes, and imagine various scenarios described by the therapist” (Lynn et al., 2003, p. 207). Often used successfully in systematic desensitization (where imagery is combined with relaxation techniques to treat a phobia), the technique is more controversial when used to help a client try to recall repressed traumatic memories. The studies of imagination inflation discussed above show that imagining events makes them seem more familiar and increases confidence in their occurrence. This familiarity and confidence can be misattributed to true childhood memories.
Another pertinent question is: How many clinicians actually use directive therapies designed to help recover repressed memories? Polusny and Follette (1996) surveyed a random sample of clinical and counseling psychologists and found that 25 percent of them reported using guided imagery, dream interpretation, bibliotherapy regarding sexual abuse, and free association of childhood memories with clients who had no specific memories of CSA. In another survey of British and American doctoral-level psychotherapists and counselors conducted by Poole, Lindsay, Memon, and Bull (1995), 25 percent of respondents thought that memory recovery was an important part of therapy and 71 percent of therapists reported using suggestive memory-recovery techniques (including hypnosis, dream interpretation, age regression, and guided imagery) when they suspected CSA. They also reported a wide variety of behavioral symptoms they thought could indicate child sexual abuse including sexual dysfunction, relationships problems, low self-esteem, and depression. However, 90 percent of these therapists were aware that clients could have mistaken memories; more recent surveys also indicate widespread awareness of the possibility of false beliefs concerning CSA (Gore-Felton et al., 2000). This finding, coupled with the report in a recent national probability sample of women in the United States that only 1.8 percent had recovered memories of sexual abuse with the help of a therapist (S. C. Wilsnack et al., 2002), indicates that the problem of false recovered memories in therapy may be less prevalent than in previous decades.
Researchers and theorists have written many articles with recommendations for therapists related to suspected CSA and potentially repressed memories. Some of the recommendations are as follows:
- Therapists should not suggest an explanation of abuse but rather stick with information reported by the client (APA, 1995).
- Clinicians should be aware of how much they can influence their clients’ recollections and exercise restraint in using imagery to help recover assumed memories (Loftus, 1997).
- Mental health professionals should consider all alternative explanations for a recovered memory (e.g., that the person simply hasn’t thought about the experience in a long time) and consider the plausibility of the memory (Taylor, 2004). For example, the average age of a first autobiographical memory is 3 1/2 years; a memory of something that happened substantially earlier than this is highly unusual.
The APA (1995) also recommends that individuals seeking a therapist be cautious if it appears that a large number of the therapist’s clients have recovered memories of childhood sexual abuse while in treatment.
Comparisons have been made between persons in terms of their propensity to distort memories and between recollections of true versus false memories. Research comparing individuals with suspected repressed memories with those with continuous memories and controls has been conducted by McNally and colleagues (e.g., Clancy, McNally, Schacter, Lenzenweger, & Pitman, 2002; Clancy et al., 2000) using variants of a research design called the Deese/Roediger-McDermott (DRM) paradigm. In this procedure, the researcher presents participants with lists comprising words related to a theme word; however, the theme word is not present in the list. For example, a list containing the words sour, candy, sugar, and bitter would not contain the primary theme word of sugar. After presenting the lists, the researcher gives the participants a recall test and then a recognition test (in which they select list words from among presented words, nonpresented theme words, and nonstudied words). From these tests, researchers identify false recall and false recognition (both of which involve individuals indicating that they studied a nonpresented theme word).
People who report recovered memories of highly unlikely events show a tendency for false memory formation in the DRM paradigm. Clancy et al. (2002) investigated memory distortion in people claiming to have recovered memories of abduction by space aliens. This group was compared to two other groups: (a) people who believed that they had been abducted but had no memories of the event (repressed group); and (b) people who said they had no history of abduction by aliens (control group). The recovered-memory group was more prone to false recognition than was the repressed-memory group, and both of these groups were more prone to false recognition than was the control group. The authors identify false recognition in the DRM paradigm as a type of source-monitoring error. Source monitoring is “remembering how, when, and where a memory is acquired” (Clancy et al., 2002, p. 460). Not remembering the source of information on alien abductions may be related to the creation of false memories of alien abduction (for example, incorrectly remembering an alien abduction as having been a personal experience rather than something watched in a movie years before).
Clancy et al. (2000) compared four groups of women using the DRM paradigm: a group reporting recovered memories of CSA; a group who believed that they had been sexually abused as children but had no memory of the abuse (repressed-memory group); women who had been sexually abused as children and had always remembered it (continuous-memory group); and finally a control group of women with no histories of sexual abuse. Women reporting recovered memories of CSA had a higher rate of false recognition of nonpresented theme words than did women who had never forgotten their abuse; that is, they were more likely to mistakenly report having seen a word that captured the gist of words they had seen. Again, as a type of source-monitoring error, these women may have generated an image of the nonpresented theme word while processing semantically related words and then mistakenly “remembered” having seen the theme word. McNally, Clancy, Barrett, and Parker (2005) refer to this as a deficit in reality monitoring—a form of source monitoring in which one must distinguish memories of perceived events from memories of imagined events. Clancy et al. (2000) recommend using caution when extrapolating their findings to clinical settings but point out that the results are “consistent with the hypothesis that women who report recovered memories of sexual abuse are more prone than others to develop certain types of illusory memories” (p. 30). Are there other characteristics of individuals that make them more susceptible to implanted memories? There is some evidence that more susceptible individuals are more dissociative, introverted, and imaginative than those not as susceptible (e.g., Hyman & Billings, 1998; Porter et al., 1999).
Other comparisons have been made between reports of memories known to be true and those known to be false. Researchers have found statistical differences in some studies but not in others (Loftus, 2004), and much more work remains to be done before it is possible to definitively determine the difference between the two. Studies that have demonstrated distinctions between true and false memories have shown the following: True memories are held with higher confidence than are false memories (Loftus & Pickrell, 1995; Smith et al., 2003); people use more words to describe the true memories (Loftus & Pickrell, 1995); false memories are less coherent than real memories (Porter et al., 1999); and people report false memories as less clear and vivid than are true ones (Loftus & Pickrell, 1995). In one study, there was a trend for the two types of memories to differ in perspective: Participants were more likely to describe implanted memories from a “participant” perspective and to describe true reports from an “observer” perspective (Porter et al., 1999).
Repressed memories are those that, in Freudian terms, are blocked from consciousness because of their threatening or anxiety-provoking nature. Freud theorized and wrote about repressed memory in the early part of the 20th century. In the last two decades of the 20th century, media sources such as self-help books introduced the notion of repressed memory to the general public. Subsequently, the topic was sensationalized by the media, and many people became convinced that this phenomenon was common (APA, 1995). By the late 1990s and early 2000s, prestigious organizations (such as the American Psychiatric Association) were warning about the problems of false memories being created in therapy, and individuals who had been led to recover false memories in therapy were successfully suing their therapists; public attitude began to change toward recovered memory therapy (Loftus, 2003a).
Before this attitude shift began, however, the notion of repressed and recovered memories in cases of childhood sexual abuse led to one of the most heated controversies in the history of psychology. The debate was not over the existence of CSA; all concerned agreed that CSA occurs all too frequently. (The percentages of adult survivors of CSA in the United States have been estimated at 20 percent of women and 5 to 10 percent of men; Finkelhor & Dzuiba-Leatherman, 1994). The debate was over whether or not traumatic memories could, in fact, be repressed and then recovered years later with no recall in between—that is, could traumatic memories differ so dramatically from nontraumatic memories that they did not conform to the typical forgetting curve? There also was the concern that memories of CSA could be created through suggestion (either in therapy or in the form of self-help books and checklists), and that if false memories were created and then recanted, real survivors of CSA might not be believed (Madill & Holch, 2004). Efforts at helping individuals recover ostensibly repressed memories of CSA led, in some cases, to false-memory reports and subsequent accusations (and sometimes criminal proceedings) against alleged abusers. Controversy raged not only over potentially unethical therapeutic practices but also over whether “[u]ncritical acceptance of every trauma memory report [could] harm the false victims and, also sadly, trivialize the experiences of the true victims [of CSA]” (Loftus, 2003b, p. 871).
Although it is impossible to conduct experimental research on memories of CSA due to ethical constraints, psychological methods have contributed greatly to our understanding of how people remember, how they forget, and how they may come to believe things that never happened. There is no empirical evidence from survey research or experiments on non-CSA memories that traumatic memories differ in any substantial way from nontraumatic memories or that memories can be repressed (Porter et al., 2003). There are case reports indicating that it is possible that people can sometimes recall things in adulthood that had not been thought of for years (Lynn et al., 2003), and these memories are not necessarily false. However, suggestion can lead to false memories and these memories can be expressed with a great deal of confidence, detail, and emotion (Loftus, 2003b)—that is, rich false memories can seem and feel as real as true memories (Loftus & Bernstein, 2005).
Human memory is a reconstructive process; it is malleable and sometimes misleading (Loftus & Bernstein, 2005). Psychologists who have conducted applied memory research have an obligation to dispel common myths of memory and to help reduce social problems associated with the misleading aspects of human memory. Loftus and Bernstein advocate education of three groups of individuals in order to accomplish these goals: education of the general public so that people are less susceptible to the creation of false memories, education of police and other investigators about the power they have to influence memory, and education of clinicians that the use of techniques such as guided imagery to assist clients in getting in touch with so-called repressed memories may be helping them to create, rather than remember, a past (Loftus & Bernstein, 2005).