Phoenix Counselling Services
Authentic Abundant Living
Phoenix Counselling Services
Telephone 613-224-7172
Home False Memory/Recovered Memory Debate by Lorraine Jaksic


Pastoral Sciences , 18 (1999), pp.127-150

'False Memory/Recovered Memory' Debate by Lorraine Jaksic


Psychological 'trauma' is understood to originate from outside the individual (Briere, 1992, 1996; Herman, 1992; Horowitz, 1986; Terr, 1991, 1994). For the purposes of this paper, however, trauma is extended to include recognition that trauma is also an introspective event. That is, "the conflict between the will to deny horrible events and the Will to proclaim them aloud is the central dialectic of psychological trauma" (Herman, 1992, p. 1). On the one hand, a person known to have experienced childhood sexual abuse is most likely to exhibit social problems and physical symptoms including "posttraumatic stress, low self-esteem and guilt, anxiety, depression, somatization, dissociation, interpersonal dysfunction, eating disorders, sexual problems, substance abuse and suicidality" as a direct result of their trauma (Briere, 1992, p. 196). On the other hand, a person who has experienced a single traumatic event usually does not experience pervasive low self-esteem, guilt and the resulting interpersonal dysfunction that follows the betrayal of trust and personal boundaries. Many other researchers and practitioners in this field of psychology/psychiatry concur with Briere's list of social problems and physical symptoms that become part of a victim's life-pattern (Cronin, 1994; Finkelhor, 1979, 1988; Herman, 1981, 1992; vari der Kolk, 1994; Williams, 1994; Bass & Davis, 1994; Gale, Thompson, Moran & Sack, 1988). In addi-tion, recent studies indicate that childhood sexual abuse is associated with multiple short and long term psychological difficulties (Briere, 1992, 1996; Herman, 1981, 1992; Lifton, 1988; Terr, 1991, 1994; Williams, 1994; Briere & Zaidi, 1989; Schacter & Moscovitch, 1984; Tong, Oates & McDowell, 1987; Gale, Thompson, Moran & Sack, 1988).

Traumatic memory is a new field of investigation in which there are many unanswered questions. The documented findings of researchers and clinicians who work with traumatized people "show that these memories are formed in an altered state of consciousness induced by terror" (Lawrence, 1993, p.10). The victims 'dissociate' themselves from the traumatic experience as it is happe-ning (Courtois, 1992; Horowitz, 1986; Lifton, 1988; Terr, 1991, 1994; van der Kolk, 1994; Herman & Schatzow, 1987; Schacter & Moscovitch, 1984). Lifton (1988) calls this process "psychic numbing" and claims that it "stops the symbolizing or formative process" which the mind needs to create images (p.8). Therefore, the focus of attention is narrowed and the surrounding context of the experience falls away. Kristiansen (1994) calls this "speechless terror" where the overwhelming emotional impact of trauma interferes with the victim's ability to encode the experience in words or symbols (p.5). Still, certain details of imagery and sensation are deeply etched into the psychobiology of the person. Many people suffer highly elevated physiological responses to neutral stimuli such as sudden noises, or unexpected images and thoughts.

Many studies have confirmed that traumatized individuals respond to such stimuli with significant conditioned autonomic reactions- for example, increases in heart rate, skin conductance, and blood pres-sure. The highly elevated physiological responses accompanying the recall of traumatic experiences that happened years, and sometimes decades, before illustrate the intensity and timelessness with which traumatic memories continue to affect current experience (van der Kolk, 1994, p.254).

These memories seem to be accurate in some aspect. For example, adults (includes men as victims) may give a detailed description of a room in which they were raped, even though they had not seen the room since a very young age (Blume, 1990; Briere, 1996; Courtois, 1992; Herman, 1981, 1992; Terr, 1994; Bass & Davis, 1994). However, some memories associated with the trauma, such as those dealing with time sequence, the context and/or frequency of the experience, may be fragmentary and imprecise (Herman, 1992; Ho-rowitz, 1986; Lifton, 1988; Reber, 1993; Terr, 1991, 1994). In short, flaws in memory recall do not necessarily imply that the remembered events did not occur. Learning is fundamental and precedes memory (Reber, 1993; Thorndike & Rock, 1934); a victim of childhood sexual abuse may implicitly learn to put traumatic memories into an inaccessible memory network which will be recovered only in a 'hypermnesia' state (Erdelyi, 1992). Trauma-related memories further increase arousal (in hypermnesia) and this state can result in the re-release of stress hormones that consolidate the strength of the memory trace, such as norepinephrine (van der Kolk, 1994). Consequently, the traumatic memories return as implicit 'body memories' through the sensorimotor systems, but are devoid of their original context and appear as kinesthetic sensations, olfactory stimuli, or visual images, in the form of flashbacks and nightmares (Kristiansen, 1994; van der Kolk, 1994)

Partial or even complete amnesia for childhood trauma is well documented (Courtois, 1992; Freyd, 1993; Herman, 1981, 1992; Kristiansen, 1994; Terr, 1991, 1994; Williams 1992, 1994; Briere & Conte, 1993; Briere & Zaidi, 1989; Herman & Schatzow, 1987). However, other researchers believe the term 'repression' adequately defines the coping mechanism utilized by many childhood sexual abuse victims (Loftus, 1993; Loftus, Polonsky & Fullilove, 1994). The term 'repression', however, does not adequately address the fact that people have multiple memory systems which include amnesic properties (Schacter & Moscovitch, 1984; Sherry & Schacter, 1987), or account for how the psychophysiological well being of victims is affected (van der Kolk, 1994). Thus, while trauma begins with events outside the child, a number of internal changes in the child occur once the events take place.

Terr (1991) cites four characteristics related to childhood trauma that appear to last for long periods of life, no matter what diagnosis the patient eventually receives. These are: "visualized or otherwise repeatedly perceived memories of the trauma event, repetitive behaviors, trauma-specific fears and changed attitudes about people, life and the future" (Terr, 1991, p.12). Also, Terr (1991) divides childhood trauma into two basic types: Type ] trauma includes full, detailed memories and misperceptions of a single specific event or tragedy; and Type 2 trauma includes denial and numbing, self-hypnosis and dissociation, and rage associated with long-standing repeated ordeals. With the onset of a single sudden traumatic event a child's behavior will differ from a child suffering from Type 2 trauma which is caused by events originating from outside the child that are "marked by prolonged and sickening anticipation" and "[nione is generated solely within the child's own mind" (Terr, 1991, p. 11). The issue of whether or not memories conceming childhood sexual abuse are real or fantasy is still a controversial one.

Further evidence reported by several researchers and clinicians indicates that traditionally psychiatry/psychology avoided therapy models which would expose histories of sexual abuse (Briere, 1996; Cronin, 1994; Courtois, 1992; Herman, 1981, 1992; Bass & David, 1994). Briere and Zaidi (1989) conducted a study with 100 nonpsychotic female patients coming through a hospital's psychiatric emergency room. The first 50 charts written by clinicians who were specifically asked to query an abuse history were exarnined. A control group was formed by randomly selecting 50 charts of women who were not queried. They hypothesized that people who were sexually abused as children were most likely representative of an emergency psychiatric population. The results of the study indicated much higher rates of molestation history in clinical populations than in the general population. Only 6% of those not specifically asked about a sexual abuse history volunteered information pertaining to abuse; yet 70% of the clients who had been directly asked reported sexual abuse. "Further analysis linked molestation history to suicidality, substance abuse, sexual difficulties, multiple psychiatric diagnoses, and axis II traits or disorders - especially borderline personality" (Briere & Zaidi, 1989, p.1602). These results suggest that sexual abuse may place certain adults at a higher risk for later psychological crisis. Therefore, this type of knowledge may encou-rage clinicians to investigate abuse histories of those who come into emergency care.

In a follow-up study of 100 women who had been hospitalized and treated for childhood sexual abuse 17 years earlier, Williams (1992, 1994) suggests that 38% did not recall the experiences that had been reported and carefully documented by hospital staff, despite the child's presence and examination at the hospital. Of the women who had no memory of the reported abuse, 53% reported other childhood victimizations, such as other incidents of molestation or physical abuse. This would suggest that the patients were not embarrassed to talk about such personal matters with the clinicians doing the study survey (Williams, 1992, 1994). This study replicated research conducted several years earlier by Herman and Schatzow (1987). The 1987 study found that 64% of patients did not have full recall of the abuse, and just over one-quarter of the women (28%) reported severe amnesia. Also, it is suggested that the earlier the onset of abuse (preschool), the longer its duration, and the degree of severity of violence involved, are significant indicators of the degree of amnesia. Furthermore, in response to hearing others in the group talk about their stories, patients reported recovery of additional memories.

Patients in this category suffered some increased anxiety in the process of recovering new memories, but were usually able to integrate them without prolonged distress, and generally reported that the new memories enabled them to form a more realistic picture of their families and a less critical estimate of themselves (Herman & Schatzow, 1987, p.7)

The groups were structured around defining and achieving a goal related to the sexual abuse. The most commonly chosen goals were: (a) disclosure of the abuse to a family member, (b) recovery of memories, and (c) perpetrator confrontation. Most clients defined a goal that included the potential for gathering corroborating evidence outside of their own memories. The majority of women (74%) were able to confirm their memories of childhood sexual abuse with concrete evidence (verbal acknowledgments, diaries, and pictures) and 34% discovered that other children, usually siblings, had been abused by the same perpetrator. Many clinicians report similar findings when working with their client populations (Cronin, 1994; Herman, 1992; Bass & Davis, 1994). "The presumption that most patient's reports of childhood sexual abuse can be ascribed to fantasy no longer appears tenable" (Herman & Schatzow, 1987, p. 11). However, this statement is challenged.

Elizabeth Loftus, a cognitive psychologist and a member of the Advisory Board to the False Memory Syndrome Foundation, recognizes that "actual childhood sexual abuse is tragically common", yet does not believe it is common to repress memories of childhood sexual abuse. Despite the growing body of literature with regard to the acquirement of implicit and tacit knowledge, she questions "how common is it to repress memories of childhood sexual abuse?" (1993, p.521) without any reference to research conceming the psychobiological effects of Type 2 trauma as described and differentiated by researchers such as Briere (1991, 1996),Herman(1992) and Terr(1991, 1994). Additional evidence has been obtained which supports the concept of recovered memories.

Loftus (1993) reviewed the study conducted by Briere and Conte (1993) and disputes the findings that suggest the possibility of 'recovered memories'. She claims that the interviewers may have unintentionally guided the subjects to disclose abuse which never happened. At the core of the Briere and Conte (1993) study, where 450 adult clinical clients who had reported sexual abuse histories were asked if they had experienced amnesia for the abuse, is an affirmative response of 59.3% for the sample. This is not surprising considering the outcome of the study (Briere and Zaidi, 1989), where 70% of the clinical population queried for sexual abuse reported abuse. However, further ques-tions need to be asked concerning 'amnesic' episodes for the remaining 30%. For example, Williams (1992, 1994) noted that 38% of the women in this longitudinal study were either amnesic for the abuse or had chosen not to report the abuse to the interviewers 17 years after having documented the abuse. As well, Herman and Schatzow (1987) reported that 74% of the women in their study who had recovered memories were able to validate them with corroborating evidence. These two studies validate recovered memories, and suggest that "retrospective studies which rely on self-reports of childhood experiences of sexual victimization are likely to result in an underestimation of the true prevalence of such abuse" (Williams, 1992, p.20).

Loftus, Polonsky, and Fullilove (1994), question Herman and Schatzow's criteria for corroboration. The 1987 study documents that 74% of the women obtained corroborating evidence. This factor is broken down into two groups with 40% getting information directly from other family members, and/or from pictures, or diaries; the other 34% discovering that the perpetrator had sexually abused other children, usually a sibling. Of the remaining 26% in the study, 9% reported that family members had indicated a strong likelihood that they had been abused but would not confirm direct questions, 11% of the women made no attempt to obtain corroborating evidence, and 6% stated that they had tried to obtain such evidence but were unsuccessful. The degree of amnesia was independent of corroboration of evidence obtained. Given that 64% of the sample reported some degree of amnesia (with 28% severely amnesic) and that 74% of the study found corroborating evidence to support the recovered memories, some of the individuals in the 'severe memory' (amnesia) category were able to obtain corroboration.

Loftus et al. (1994) also studied women recruited from a substance abuse rehabilitation program; all had previously used drugs. The results show that, of 105 women, 54% had experienced some form of childhood sexual abuse. Of this group, 69% claimed they had always remembered the abuse, 12% had at least partial amnesia and 19% claimed total amnesia for a period of time but later regained the memories. These figures indicate a total of 31% had amnesic memories with later recall and not the 19% figure quoted in the results section of the article (see Kristiansen, 1994, as this statement concurs with her analysis). However, this research failed to examine three specific aspects which other trauma researchers investigate: (a) the possibility that some of the sample had not yet recovered memories of childhood sexual abuse at the time of the study and, therefore, the results may underestimate the 'severely amnesic' group (Williams, 1992, 1994); (b) the more destructive and maladaptive coping strategies such as substance abuse which people may practice when they cannot forget their abuse (Briere, 1996; Herman, 1992; and Herman and Schatzow, 1987); and (c) the distinguishing features which differentiate simple posttraumatic stress disorder (Type 1) and the complex model (Type 2) which involves denial, psychic numbing and/or dissociation which is all too common in victims of chronic child abuse (Briere, 1992, 1996; Herman, 1992; Horowitz, 1986; Terr, 1991; van der Kolk, 1994; Briere & Conte, 1993; Herman & Schatzow, 1987). The question of whether or not recovered memories are possible has been addressed:

By saying the figures range from 19 to 59%, on the basis of these data it seems that most studies suggest that 50 to 60% of survivors forget their abuse at some point, a figure that is actually consistent with the estimate of forgetting cited in popular recovery books (Kristiansen, 1994, p.12).

Loffus et al. (1994) subsequently dispute the findings of Williams (1992), questioning the validity of the self-reports due to infant amnesia. In response, Williams (1994) argues that the group of women who had no recall of their abuse did not remember or forget that abuse because of their age at the time of the abuse (the children ranged in age upwards starting from 18 months). Moreover, Terr (1991) argues that even infants and toddlers are able to "lay down, store or retrieve full verbal memories of their traumas" through their play, art and visual images (p.12). This position is further substantiated by work done at a biological level by van der Kolk (1994) and Sherry and Schacter (1987). That is, recollections may not appear in semantic form but the implicit and tacit knowledge of the traumatic experiences are in the sensorimotor memory networks. Many personal stories show the ability to re-see or occasionally to re-feel a terrible series of events when tactile, positional or smell memories are triggered (Briere, 1996; Herman, 1981, 1992; Reber, 1993; Bass & Davis, 1994)

Loftus (1993) also argues that if people assume that 'repression' is common, then the individuals in the Briere and Conte (1993) study will infer that repression is likely to have happened to them. However, the consensus of most trauma researchers today is that false complaints by children for child abuse are rare, approximately 2-8% of reported cases (Briere, 1996; Salter, 1989). Also, Loftus ignores research by Finkelhor (1979) which identifies four dynamics unique to sexual abuse. Other traumas of childhood, such as parental divorce, do not generate these results in dynamics. They are: (a) traumatic sexualization, (b) betrayal, (c) stigmatization, and (d) powerlessness (p.357). These four dynamics are ongoing processes which interact with the structure of the whole environment (familial and cultural) of the victim and create an aura of secrecy. Children are not likely to report abuse and undergo the necessary interrogation to document the abuse if it did not happen (Finkelhor, 1979, 1988; Herman, 1981; Tedesco & Schnell, 1987). Furthermore, it is argued that most adults are not going to claim 'repressed' memories for childhood sexual abuse when it is not true because of the social and cultural backlash (Cronin, 1994; Faludi, 1991; Freyd, 1993). In fact, many researchers and clinicians report that false retractions are far more common, especially when the victim is insufficiently protected after disclosure and therefore succumbs to intimidation by the perpetrator or other family members who feel that they must preserve secrecy (Cronin, 1994; Herman, 1981, 1992; Lawrence, 1993; Bass & Davis, 1994).

Loftus (1993) argues that it is possible to implant incorrect memories of childhood trauma in older people. The claim comes from speculation that people who are in therapy for some other type of problem (substance abuse, depression or anger management) come to believe that they were abused as children because of suggestions from the therapist or in 'self-help' books. This assumption is based on the fact that a young teenager (aged 14) could be convinced by an older brother that he had been lost in a shopping mall when he was very young and had experienced trauma.

Loftus' (1993) research is problematic since many young people can have feelings of being lost; it is questionable that such an experience would entail posttraumatic stress disorder symptomatology or the pschophysical changes that researchers hypothesize affect traumatic memory recall (van der Kolk, 1994) or that of hypermnesia (Erdelyi, 1992). The four personal dynamics characteristic of having been sexually abused (Finkelhor, 1979) and the elements of shame, secrecy, and fear of disclosure (Herman, 1981, 1992; Terr, 1994) that typically coincide with sexual abuse, are absent from the research findings. Therefore, is Loftus (1993) testing Type 2 trauma? Secondly, it is possible that these individuals were describing actual events that happened to them, or events happening within other contexts like being lost at a fair ground or on a picnic. Those who work with Type 2 trauma victims would suggest that it is the sensation of 'lost' that may be remembered and retrieved without necessarily remembering the context of the event (Terr, 1991). Undoubtedly, such an event as being lost at a mall may be uncomfortable and stressful but, in the absence of other abnormal psychophysical and psychosocial problems, is hardly Type 2 trauma.

However, the research done by the 'false memory syndrome' side of this debate has raised some solid viewpoints that require attention and conscious awareness. The concept that memories could be implanted within memory reconstruction during therapy is a worthwhile, healthy, therapeutic stance to be mindful of. Recent research shows there is a growing consensus as to how a therapist may potentially create false memories of abuse with the client. Research conducted by Loftus and Ketcham (1994) and Spanos (1994) suggests that there is a consistent model containing particular factors by which some clients can be lead to believe memories for abuse that did not specifically occur. They are: (a) therapist-driven suggestions of abuse, (b) the therapist's own beliefs about abuse (i.e., over identification or erroneous beliefs), (c) a confirmatory bias wherein evidence for abuse is weighted on the affirmative side and non-substantial evidence is discarded, (d) the focused, and perhaps pressured, search for abuse related memories, (e) particular memory recovery methodologies that may confuse or superimpose imagination and memory, and (f) therapy procedures that indirectly suggest to the client procedures that will fill in their memory blanks for them. It is suggested that these elements interact with the therapist's expectations for the therapy regime and that the client is persuaded to agree that abuse may have occurred (Brown, Scheflin & Hammond, 1998). "[W]hile the above arguments offer a lot of [sic] useful hypotheses about suggestive influences in psychotherapy, there is as yet no consistent theory about how and under what conditions therapy may be sugges-tive" (Brown, Scheflin & Hammond, 1998, pg. 34).

The difficulties in obtaining a consensus among those who study 'trauma' and 'memory' may be due to the terminology used to discuss the issues in this debate. Therefore, it is appropriate to discuss the terminology used in 'traumatic memory' discourse. The consequences of not having a consensus are far-reaching and will be discussed in following sections of this paper.


Copyright 2007, Phoenix Counselling Services. All Rights Reserved.