Recovering Memories of Trauma: A View From the Laboratory


Richard J. McNally1

Department of Psychology, Harvard University, Cambridge, Massachusetts




The controversy over the validity of repressed and recovered memories of childhood sexual abuse (CSA) has been extraordinarily bitter. Yet data on cognitive functioning in people reporting repressed and recovered memories of trauma have been strikingly scarce. Recent laboratory studies have been designed to test hypotheses about cognitive mechanisms that ought to be operative if people can repress and recover memories of trauma or if they can form false memories of trauma. Contrary to clinical lore, these studies have shown that people reporting CSA histories are not characterized by a superior ability to forget trauma-related material. Other studies have shown that individuals reporting recovered memories of either CSA or abduction by space aliens are characterized by heightened proneness to form false memories in certain laboratory tasks. Although cognitive psychology methods cannot distinguish true memories from false ones, these methods can illuminate mechanisms for remembering and forgetting among people reporting histories of trauma.



recovered memories; trauma; repression; sexual abuse; dissociation


            How victims remember trauma is among the most explosive issues facing psychology today. Most experts agree that combat, rape, and other horrific experiences are unforgettably engraved on the mind (Pope, Oliva, & Hudson, 1999). But some also believe that the mind can defend itself by banishing traumatic memories from awareness, making it difficult for victims to remember them until many years later (Brown, Scheflin, & Hammond, 1998).

            This controversy has spilled out of the clinics and cognitive psychology laboratories, fracturing families, triggering legislative change, and determining outcomes in civil suits and criminal trials. Most contentious has been the claim that victims of childhood sexual abuse (CSA) often repress and then recover memories of their trauma in adulthood.2 Some psychologists believe that at least some of these memories may be false--inadvertently created by risky therapeutic methods (e.g., hypnosis, guided imagery; Ceci & Loftus, 1994).

            One striking aspect of this controversy has been the paucity of data on cognitive functioning in people reporting repressed and recovered memories of CSA. Accordingly, my colleagues and I have been conducting studies designed to test hypotheses about mechanisms that might enable people either to repress and recover memories of trauma or to develop false memories of trauma.

            For several of our studies, we recruited four groups of women from the community. Subjects in the repressed-memory group suspected they had been sexually abused as children, but they had no explicit memories of abuse. Rather, they inferred their hidden abuse history from diverse indicators, such as depressed mood, interpersonal problems with men, dreams, and brief, recurrent visual images (e.g., of a penis), which they interpreted as "flashbacks" of early trauma. Subjects in the recovered-memory group reported having remembered their abuse after long periods of not having thought about it.3 Unable to corroborate their reports, we cannot say whether the memories were true or false. Lack of corroboration, of course, does not mean that a memory is false. Subjects in the continuous-memory group said that they had never forgotten their abuse, and subjects in the control group reported never having been sexually abused.


            To characterize our subjects in terms of personality traits and psychiatric symptoms, we asked them to complete a battery of questionnaires measuring normal personality variation (e.g., differences in absorption, which includes the tendency to fantasize and to become emotionally engaged in movies and literature), depressive symptoms, posttraumatic stress disorder (PTSD) symptoms, and dissociative symptoms (alterations in consciousness, such as memory lapses, feeling disconnected with one's body, or episodes of "spacing out"; McNally, Clancy, Schacter, & Pitman, 2000b).

            There were striking similarities and differences among the groups in terms of personality profiles and psychiatric symptoms. Subjects who had always remembered their abuse were indistinguishable from those who said they had never been abused on all personality measures. Moreover, the continuous-memory and control groups did not differ in their symptoms of depression, posttraumatic stress, or dissociation. However, on the measure of negative affectivity--proneness to experience sadness, anxiety, anger, and guilt--the repressed-memory group scored higher than did either the continuous-memory or the control group, whereas the recovered-memory group scored midway between the repressed-memory group on the one hand and the continuous-memory and control groups on the other.

            The repressed-memory subjects reported more depressive, dissociative, and PTSD symptoms than did continuous-memory and control subjects. Repressed-memory subjects also reported more depressive and PTSD symptoms than did recovered-memory subjects, who, in turn, reported more dissociative and PTSD symptoms than did control subjects. Finally, the repressed- and recovered-memory groups scored higher than the control group on the measure of fantasy proneness, and the repressed-memory group scored higher than the continuous-memory group on this measure.

            This psychometric study shows that people who believe they harbor repressed memories of sexual abuse are more psychologically distressed than those who say they have never forgotten their abuse.


            Some clinical theorists believe that sexually molested children learn to disengage their attention during episodes of abuse and allocate it elsewhere (e.g., Terr, 1991). If CSA survivors possess a heightened ability to disengage attention from threatening cues, impairing their subsequent memory for them, then this ability ought to be evident in the laboratory. In our first experiment, we used directed-forgetting methods to test this hypothesis (McNally, Metzger, Lasko, Clancy, & Pitman, 1998). Our subjects were three groups of adult females: CSA survivors with PTSD, psychiatrically healthy CSA survivors, and nonabused control subjects. Each subject was shown, on a computer screen, a series of words that were either trauma-related (e.g., molested), positive (e.g., charming), or neutral (e.g., mailbox). Immediately after each word was presented, the subject received instructions telling her either to remember the word or to forget it. After this encoding phase, she was asked to write down all the words she could remember, irrespective of the original instructions that followed each word.

If CSA survivors, especially those with PTSD, are characterized by heightened ability to disengage attention from threat cues, thereby attenuating memory for them, then the CSA survivors with PTSD in this experiment should have recalled few trauma words, especially those they had been told to forget. Contrary to this hypothesis, this group exhibited memory deficits for positive and neutral words they had been told to remember, while demonstrating excellent memory for trauma words, including those they had been told to forget. Healthy CSA survivors and control subjects recalled remember-words more often than forget-words regardless of the type of word. Rather than possessing a superior ability to forget trauma-related material, the most distressed survivors exhibited difficulty banishing this material from awareness.

            In our next experiment, we used this directed-forgetting approach to test whether repressed- and recovered-memory subjects, relative to nonabused control subjects, would exhibit the hypothesized superior ability to forget material related to trauma (McNally, Clancy, & Schacter, 2001). If anyone possesses this ability, it ought to be such individuals. However, the memory performance of the repressed- and recovered-memory groups was entirely normal: They recalled remember-words better than forget-words, regardless of whether the words were positive, neutral, or trauma-related.


            The hallmark of PTSD is involuntary, intrusive recollection of traumatic experiences. Clinicians have typically relied on introspective self-reports as confirming the presence of this symptom. The emotional Stroop color-naming task provides a quantitative, nonintrospective measure of intrusive cognition. In this paradigm, subjects are shown words varying in emotional significance, and are asked to name the colors the words are printed in while ignoring the meanings of the words. When the meaning of a word intrusively captures the subject's attention despite the subject's efforts to attend to its color, Stroop interference--delay in color naming--occurs. Trauma survivors with PTSD take longer to name the colors of words related to trauma than do survivors without the disorder, and also take longer to name the colors of trauma words than to name the colors of positive and neutral words or negative words unrelated to their trauma (for a review, see McNally, 1998).

            Using the emotional Stroop task, we tested whether subjects reporting either continuous, repressed, or recovered memories of CSA, relative to nonabused control subjects, would exhibit interference for trauma words (McNally, Clancy, Schacter, & Pitman, 2000a). If severity of trauma motivates repression of traumatic memories, then subjects who cannot recall their presumably repressed memories may nevertheless exhibit interference for trauma words. We presented a series of trauma-related, positive, and neutral words on a computer screen, and subjects named the colors of the words as quickly as possible. Unlike patients with PTSD, including children with documented abuse histories (Dubner & Motta, 1999), none of the groups exhibited delayed color naming of trauma words relative to neutral or positive ones.


            Some psychotherapists who believe their patients suffer from repressed memories of abuse will ask them to visualize hypothetical abuse scenarios, hoping that this guided-imagery technique will unblock the presumably repressed memories. Unfortunately, this procedure may foster false memories.

            Using Garry, Manning, Loftus, and Sherman's (1996) methods, we tested whether subjects who have recovered memories of abuse are more susceptible than control subjects to this kind of memory distortion (Clancy, McNally, & Schacter, 1999). During an early visit to the laboratory, subjects rated their confidence regarding whether they had experienced a series of unusual, but nontraumatic, childhood events (e.g., getting stuck in a tree). During a later visit, they performed a guided-imagery task requiring them to visualize certain of these events, but not others. They later rerated their confidence that they had experienced each of the childhood events. Nonsignificant trends revealed an inflation in confidence for imagined versus nonimagined events. But the magnitude of this memory distortion was more than twice as large in the control group as in the recovered-memory group, contrary to the hypothesis that people who have recovered memories of CSA would be especially vulnerable to the memory-distorting effects of guided imagery.

            To use a less-transparent paradigm for assessing proneness to develop false memories, we adapted the procedure of Roediger and McDermott (1995). During the encoding phase in this paradigm, subjects hear word lists, each consisting of semantically related items (e.g., sour, bitter, candy, sugar) that converge on a nonpresented word--the false target--that captures the gist of the list (e.g., sweet). On a subsequent recognition test, subjects are given a list of words and asked to indicate which ones they heard during the previous phase. The false memory effect occurs when subjects "remember" having heard the false target. We found that recovered-memory subjects exhibited greater proneness for exhibiting this false memory effect than did subjects reporting either repressed memories of CSA, continuous memories of CSA, or no abuse (Clancy, Schacter, McNally, & Pitman, 2000). None of the lists was trauma-related, and so we cannot say whether the effect would have been more or less pronounced for words directly related to sexual abuse.

            In our next experiment, we tested people whose memories were probably false: individuals reporting having been abducted by space aliens (Clancy, McNally, Schacter, Lenzenweger, & Pitman, 2002). In addition to testing these individuals (and control subjects who denied having been abducted by aliens), we tested individuals who believed they had been abducted, but who had no memories of encountering aliens. Like the repressed-memory subjects in our previous studies, they inferred their histories of trauma from various "indicators" (e.g., a passion for reading science fiction, unexplained marks on their bodies). Like subjects with recovered memories of CSA, those reporting recovered memories of alien abduction exhibited pronounced false memory effects in the laboratory. Subjects who only believed they had been abducted likewise exhibited robust false memory effects.


            The aforementioned experiments illustrate one way of approaching the recovered-memory controversy. Cognitive psychology methods cannot ascertain whether the memories reported by our subjects were true or false, but these methods can enable testing of hypotheses about mechanisms that ought to be operative if people can repress and recover memories of trauma or if they can develop false memories of trauma.

            Pressing issues remained unresolved. For example, experimentalists assume that directed forgetting and other laboratory methods engage the same cognitive mechanisms that generate the signs and symptoms of emotional disorder in the real world. Some therapists question the validity of this assumption. Surely, they claim, remembering or forgetting the word incest in a laboratory task fails to capture the sensory and narrative complexity of autobiographical memories of abuse. On the one hand, the differences between remembering the word incest in a directed-forgetting experiment, for example, and recollecting an episode of molestation do, indeed, seem to outweigh the similarities. On the other hand, laboratory studies may underestimate clinical relevance. For example, if someone cannot expel the word incest from awareness during a directed-forgetting experiment, then it seems unlikely that this person would be able to banish autobiographical memories of trauma from consciousness. This intuition notwithstanding, an important empirical issue concerns whether these tasks do, indeed, engage the same mechanisms that figure in the cognitive processing of traumatic memories outside the laboratory.

            A second issue concerns attempts to distinguish subjects with genuine memories of abuse from those with false memories of abuse. Our group is currently exploring whether this might be done by classifying trauma narratives in terms of how subjects describe their memory recovery experience. For example, some of the subjects in our current research describe their recovered memories of abuse by saying, "I had forgotten about that. I hadn't thought about the abuse in years until I was reminded of it recently." The narratives of other recovered-memory subjects differ in their experiential quality. These subjects, as they describe it, suddenly realize that they are abuse survivors, sometimes attributing current life difficulties to these long-repressed memories. That is, they do not say that they have remembered forgotten events they once knew, but rather indicate that they have learned (e.g., through hypnosis) the abuse occurred. It will be important to determine whether these two groups of recovered-memory subjects differ cognitively. For example, are subjects exemplifying the second type of recovered-memory experience more prone to develop false memories in the laboratory than are subjects exemplifying the first type of experience?


Recommended Reading

Lindsay, D.S., & Read, J.D. (1994). Psychotherapy and memories of childhood sexual abuse: A cognitive perspective. Applied Cognitive Psychology, 8, 281-338.

McNally, R.J. (2001). The cognitive psychology of repressed and recovered memories of childhood sexual abuse: Clinical implications. Psychiatric Annals, 31, 509-514.

McNally, R.J. (2003). Remembering trauma. Cambridge, MA: Belknap Press/Harvard University Press.

McNally, R.J. (2003). Progress and controversy in the study of posttraumatic stress disorder. Annual Review of Psychology, 54, XXX-XXX.

Piper, A., Jr., Pope, H.G., Jr., & Borowiecki, J.J., III. (2000). Custer's last stand: Brown, Scheflin, and Whitfield's latest attempt to salvage "dissociative amnesia." Journal of Psychiatry and Law, 28, 149-213.


Acknowledgments-- Preparation of this article was supported in part by National Institute of Mental Health Grant MH61268.



            1. Address correspondence to Richard J. McNally, Department of Psychology, Harvard University, 1230 William James Hall, 33 Kirkland St., Cambridge, MA 02138; e-mail:

             2. Some authors prefer the term dissociation (or dissociative amnesia) to repression. Although these terms signify different proposed mechanisms, for practical purposes these variations make little difference in the recovered-memory debate. Each term implies a defensive process that blocks access to disturbing memories.

            3. However, not thinking about a disturbing experience for a long period of time must not be equated with an inability to remember it. Amnesia denotes an inability to recall information that has been encoded.



Brown, D., Scheflin, A.W., & Hammond, D.C. (1998). Memory, trauma treatment, and the law. New York: Norton.

Ceci, S.J., & Loftus, E.F. (1994). 'Memory work': A royal road to false memories? Applied Cognitive Psychology, 8, 351-364.

Clancy, S.A., McNally, R.J., & Schacter, D.L. (1999). Effects of guided imagery on memory distortion in women reporting recovered memories of childhood sexual abuse. Journal of Traumatic Stress, 12, 559-569.

Clancy, S.A., McNally, R.J., Schacter, D.L., Lenzenweger, M.F., & Pitman, R.K. (2002). Memory distortion in people reporting abduction by aliens. Journal of Abnormal Psychology, 111, 455-461.

Clancy, S.A., Schacter, D.L., McNally, R.J., & Pitman, R.K. (2000). False recognition in women reporting recovered memories of sexual abuse. Psychological Science, 11, 26-31.

Dubner, A.E., & Motta, R.W. (1999). Sexually and physically abused foster care children and posttraumatic stress disorder. Journal of Consulting and Clinical Psychology, 67, 367-373.

Garry, M., Manning, C.G., Loftus, E.F., & Sherman, S.J. (1996). Imagination inflation: Imagining a childhood event inflates confidence that it occurred. Psychonomic Bulletin & Review, 3, 208-214.

McNally, R.J. (1998). Experimental approaches to cognitive abnormality in posttraumatic stress disorder. Clinical Psychology Review, 18, 971-982.

McNally, R.J., Clancy, S.A., & Schacter, D.L. (2001). Directed forgetting of trauma cues in adults reporting repressed or recovered memories of childhood sexual abuse. Journal of Abnormal Psychology, 110, 151-156.

McNally, R.J., Clancy, S.A., Schacter, D.L., & Pitman, R.K. (2000a). Cognitive processing of trauma cues in adults reporting repressed, recovered, or continuous memories of childhood sexual abuse. Journal of Abnormal Psychology, 109, 355-359.

McNally, R.J., Clancy, S.A., Schacter, D.L., & Pitman, R.K. (2000b). Personality profiles, dissociation, and absorption in women reporting repressed, recovered, or continuous memories of childhood sexual abuse. Journal of Consulting and Clinical Psychology, 68, 1033-1037.

McNally, R.J., Metzger, L.J., Lasko, N.B., Clancy, S.A., & Pitman, R.K. (1998). Directed forgetting of trauma cues in adult survivors of childhood sexual abuse with and without posttraumatic stress disorder. Journal of Abnormal Psychology, 107, 596-601.

Pope, H.G., Jr., Oliva, P.S., & Hudson, J.I. (1999). Repressed memories: The scientific status. In D.L. Faigman, D.H. Kaye, M.J. Saks, & J. Sanders (Eds.), Modern scientific evidence: The law and science of expert testimony (Vol. 1, pocket part, pp. 115-155). St. Paul, MN: West Publishing.

Roediger, H.L., III, & McDermott, K.B. (1995). Creating false memories: Remembering words not presented in lists. Journal of Experimental Psychology: Learning, Memory, and Cognition, 21, 803-814.

Terr, L.C. (1991). Childhood traumas: An outline and overview. American Journal of Psychiatry, 148, 10-20.