Dr. Chu Replies

Given the level of polarized controversy about the issue of recovered memory of childhood abuse, it is not surprising that there have been numerous commentaries on our research. Our study was motivated by the need to further investigate the circumstances of patients’ reports of having recovered memories of childhood abuse. Contrary to the implications of at least one of the critics of our report, we attempted to find some middle ground between those who essentially reject any evidence of traumatic amnesia and recovered memory and those who uncritically accept and validate all patient accounts of childhood abuse. In our investigation we simply sought to replicate previous findings and to test the hypothesis that chronic abuse beginning in early childhood is related to higher levels of dissociative symptoms in adulthood, including amnesia. As in our previous review of this complex subject (1), we endeavored to present a balanced discussion of multiple viewpoints concerning traumatic amnesia.

We made no assertion that any of our research participants corroborated all of their abuse memories—a formidable task indeed, as patients in our programs frequently report long-standing childhood abuse that occurred over years (2, 3). We certainly do not claim to have “proved” that the remembered abuse occurred. This level of confirmation was not only beyond the scope of our investigation but in many cases would have been impossible in that intrafamilial abuse commonly occurs behind closed doors. Given these difficulties, we found it striking that such a high percentage of those who tried to obtain corroboration were able to do so for some of their abuse experiences. We did not rely primarily on scars as evidence for corroboration. In fact, the most common form of corroboration was verbal validation. We do believe that our criteria were relatively stringent, requiring that other individuals report that they knew (rather than believed) that the remembered abuse had occurred. The verbal corroboration was surprisingly high: 13 of 14 cases for physical abuse and 17 of 19 cases for sexual abuse.

Several of the letters regarding our article address the methodological difficulties of the use of participant self-reports in our research (which is one of the primary criticisms of Pope and Hudson [4] in their review of clinical research in this area). In our article we acknowledged the limitations of our study, including the reliance on patients’ self-reports concerning possible abuse and corroboration and the difficulties of determining whether subtle suggestion had been a part of the patients’ psychotherapy. However, the results of our study and similar studies should not be dismissed out of hand for methodological reasons. After all, self-report is a routine and accepted methodology for clinical research in which patients are asked describe a wide variety of variables such as mood and other psychiatric symptoms, perceptions, and life events. Although it is true that patients’ self-reports can be influenced by errors in recall, suggestion, study design, and contagion in treatment settings, we doubt that the cumulative clinical research in this area can be completely misguided and mistaken. As noted by Scheflin and Brown (5), who reviewed 25 studies of traumatic amnesia, “Partial or full amnesia was found across studies regardless of whether the sample was clinical, nonclinical, random or nonrandom, or whether the study was retrospective or prospective. Every known study has found amnesia for childhood sexual abuse in at least a portion of the sampled individuals” (pp. 178–179).

We were puzzled by one detractor of our study who contends that a more reasonable explanation for total amnesia for whole periods of childhood would be no recall because of no abuse and an unremarkable childhood. Such an explanation presupposes that all the reports of abuse were untrue and that it is normal for individuals to forget all the events for these periods. Although very few individuals have detailed memories of childhood events, we find it significant when patients report no memory of such important experiences as school, birthdays, holidays, and special occasions. We also do not agree that nonreporting of abuse was a factor in this study unless many of the participants, when asked directly, deliberately withheld information and misrepresented their previous inability to recall abuse experiences.

Our article reported not only that few of our participants were in therapy sessions when they first recovered memories of abuse but also that approximately half of the participants were not currently in any kind of mental health treatment when they first recovered memories, making suggestion unlikely in these cases. Numerous participants reported that they recovered memories of abuse before treatment and that these memories were the reason for beginning therapy.

Dr. Merckelbach’s letter raises some interesting and pertinent issues concerning individuals who have elevated scores on the Dissociative Experiences Scale. As he notes, some recent studies have also demonstrated that such individuals are more fantasy prone and suggestible. These findings are entirely consistent with research that demonstrates that the innate capacity to dissociate varies considerably (6), and it may well be the case that those with a high ability to dissociate have a heightened ability to use fantasy and imagination. Our results imply that individuals with a high dissociative capacity maintain a high level of dissociative symptoms if they are subjected to chronic traumatization. If these individuals are also prone to the development of pseudomemories, it reinforces our cautions that “clinicians must be open to the possibility of real abuse but must allow patients to reconstruct—without suggestion—a credible personal history that is consistent with past and current symptoms” (our article, p. 754).

Response bias and the suggestion inherent in the questions we asked participants may indeed have been a factor in our results, as argued by Drs. Good and Merckelbach. However, we feel strongly that these factors do not diminish our findings. There is no evidence to suggest that a brief series of direct questions about the possibility of abuse can lead to the immediate creation of complex pseudomemories of such abuse. The format of our reporting did not permit inclusion of the richness of the participants’ responses. For example, their description of confirmation by others frequently included accounts of the abuse being directly observed by others or admitted by the perpetrators, which left little doubt as to the validity of their memories.

As clinicians and clinical researchers, we are involved in the complex issues of trying to determine the etiology of the reports of child abuse that our patients present. There are certainly instances in which such reports stem from grossly inappropriate clinical practices, contamination or contagion, hysterical embellishment, or even malingering. Some such false positive reports may well have been included in our study, as we only recorded our participants’ responses. However, in both our study and our clinical practice, many reports of abuse and recovered memories appear to be authentic, credible, and internally consistent with patients’ past histories and current symptoms. In this context, our study adds some balance to the public and professional debate that sometimes seems to emphasize false memory more than true memory of childhood trauma.

It is striking that one letter expresses incredulity that a specialized trauma treatment unit might customarily house many chronically and multiply traumatized patients. Although it may not be commonplace, it is far from rare for some children from disrupted and chaotic families to have been assaulted and/or molested dozens or even hundreds of times. At a time when more than 1.5 million American children are documented to have been moderately or severely damaged by abuse and neglect each year (7), we feel strongly that research into the prevalence and effects of childhood abuse (including traumatic amnesia) is necessary and warranted. We hope that our study is only a preliminary step to further research on severe childhood trauma and the treatment of its sequelae. Although our study cannot be described as conclusive or definitive, it does underscore the presence and aftereffects of the still underreported and often-denied reality of child abuse in American society.