The Memory Wars and What They Cost
How Dissociative Amnesia Was Erased from Clinical Practice for a Generation Written by Nils Joneborg. Published 2026-06-17 by WONSA Publishing.
Introduction
For a brief period in the late 1980s and early 1990s, mainstream psychiatry began to
catch up with what trauma clinicians and survivors had long known. Child sexual abuse
was not rare. It was not peripheral. It was not, as the field had quietly assumed for most
of a century, primarily an Oedipal fantasy. Survivors began to be heard, and clinicians
began to listen differently. Research on dissociation accelerated. The diagnostic
categories began to expand to make room for what was finally being seen.
That window closed quickly. By the mid-1990s, a counter-narrative had taken hold in the
popular media and in significant parts of academic psychology. The frame was simple:
therapists were implanting false memories of abuse; survivors who reported recovered
memories of childhood sexual abuse were victims of suggestion, not of perpetrators;
and the entire clinical apparatus that had begun to take dissociative disorders seriously
was suspect.
The intellectual figurehead of that counter-narrative was the cognitive psychologist
Elizabeth Loftus. The institutional vehicle was the False Memory Syndrome Foundation
(FMSF). The cultural shorthand was ”false memory syndrome” — a phrase that has
never appeared in any diagnostic manual but that nonetheless reshaped clinical
training, courtroom decisions, public attitudes, and the lives of survivors for three
decades.
This article is not a dismissal of Loftus’s experimental findings, which are real and
important. It is an examination of the gap between what her research actually
demonstrated and what her research was used to claim. It is also an account of why
dissociative amnesia — the loss of access to traumatic autobiographical memory — is
one of the better-documented phenomena in modern psychotraumatology, and why its
erasure from clinical training and public understanding has had measurable costs.
The argument is not that memory is invulnerable to distortion. It is. The argument is that
two things are simultaneously true, and that the field has spent thirty years pretending it
must choose between them: ordinary autobiographical memory is reconstructive and
can be influenced by suggestion under specific conditions; and memory of severe,
prolonged childhood trauma can become inaccessible to conscious recall and can later
return, by mechanisms that are neurobiologically plausible and clinically observed.
Holding both is not a contradiction. It is the precondition for honest practice.
What Loftus Actually Found
Loftus’s most cited contribution to this debate is the 1995 ”Lost in the Mall” study,
published with her research assistant Jacqueline Pickrell (Loftus & Pickrell, 1995). The
original demonstration on which the study was built had been produced earlier, in 1992,
by Loftus’s undergraduate Jim Coan as a class project — a five-credit research
assignment in which Coan persuaded his fourteen-year-old brother that he had once
been lost in a shopping mall. Loftus and Pickrell then formalized the design into a
24-participant study. Each participant was given short narratives describing four
childhood events, three real (verified by their own family members) and one fabricated
— that the participant, as a child, had become lost in a shopping mall and been
rescued by an older stranger. Across multiple interviews, family members were enlisted
to corroborate the false narrative. After repeated suggestion, six of the twenty-four
participants — 25 percent — came to report some form of belief in or memory of the
fabricated event.
The conclusion Loftus drew, and the conclusion that found its way into thousands of
media accounts, courtroom depositions, and undergraduate textbooks, was that false
autobiographical memories can be implanted through suggestion. That conclusion is
defensible at the level of the actual experiment.
What is not defensible — and what shaped the next thirty years of public discourse — is
the leap from ”false autobiographical memories of mundane childhood events can be
cultivated through suggestion delivered by trusted family members across multiple
structured interviews” to ”recovered memories of childhood sexual abuse in adult
psychotherapy patients are probably false memories implanted by therapists.”
The two propositions describe entirely different phenomena. Consider what the Lost in
the Mall paradigm actually requires.
It requires a trusted source. In Loftus’s design, the false narrative is delivered by close
family members — parents or older siblings — who are coached to vouch for an event
that never occurred. The credibility of the source is the central engine of the
manipulation. This is the opposite of what occurs in clinical settings, where a therapist
(a relative stranger, often distrusted by survivors) is alleged to ”implant” memories of
family abuse. The directionality is reversed.
It requires plausibility. Becoming lost in a shopping mall as a child is a common,
unremarkable experience that most participants could believe might have happened to
them and been forgotten. The fabricated event is calibrated to fit ordinary
autobiographical experience. Severe, repeated childhood sexual abuse by a parent or
relative is not.
It requires sustained suggestion. The Lost in the Mall design uses repeated interviews,
structured visualization exercises, and explicit encouragement to ”try to remember.”
This is closer to interrogative pressure than to ordinary clinical conversation.
It uses non-traumatized samples. The participants in Loftus’s classic studies were
typically university students, not survivors of severe childhood trauma. The memory
architecture being probed is the architecture of healthy young adults, not of individuals
whose memory has been organized in childhood around survival under conditions of
inescapable threat.
None of this invalidates the experimental finding. It places it. Loftus showed something
real and clinically useful: confidence in autobiographical memory can be nudged;
partial false beliefs about ordinary childhood events can be cultivated through
trusted-source suggestion combined with structured imagination. This finding has
legitimate implications for eyewitness testimony, for police interrogation procedures,
and for incautious therapeutic technique. It does not, and cannot, address the question
of whether recovered memories of severe childhood trauma in adult psychotherapy
patients are reliable. Loftus did not study that question. The Lost in the Mall paradigm
is not the right instrument for that question.
This was not a private observation among critics. Coan himself, now a tenured
neuroscientist at the University of Virginia, has been clear about it. Asked to testify in
cases involving recovered memories of childhood sexual abuse, he has consistently
refused. He told The Cut in 2021 that “…it took me a while to realize that the study I
was doing was making people who had been sexually abused feel like I was their
enemy.” Of the original Lost in the Mall study, an undergraduate research assignment
for which he received five points of extra credit, he said: ”Five points and decades of
grief” (Heffernan, 2021).
The 2023 replication and the 2024 reanalysis
In 2023, Murphy and colleagues published a preregistered replication of Loftus &
Pickrell in Memory, with 123 participants — five times the original sample size. (Murphy
et al., 2023) After repeated interviews, 35 percent of participants were classified as
showing some form of false belief or memory, partial or full. The result was widely
reported as confirming and even extending Loftus’s original finding.
In December 2024, Bernice Andrews and Chris Brewin reanalyzed the Murphy et
al. transcripts using stricter, prespecified criteria. To count as a false memory under
the Andrews and Brewin scheme, a participant’s response had to contain at least
three specific core details from the planted narrative — such as being lost, crying,
being helped by an older woman, being reunited with family — plus place and time.
Vague associations and general willingness-to-believe were not counted as false
memories (Andrews & Brewin, 2024).
Under this stricter coding, the 35 percent figure collapsed to 4 percent: five participants
out of 123. None of those five reported a complete, sensorially elaborated false
memory. The original finding, on rigorous coding, is that under coordinated, repeated,
family-corroborated suggestion of a plausible mundane event, roughly one in
twenty-five healthy young adults can be brought to produce something that looks, on
the surface, like a partial false memory. The other twenty-four cannot.
This is a more modest finding than the cultural deployment of Lost in the Mall would
suggest. It is also, on its own terms, still scientifically interesting and clinically relevant.
The point is not that Loftus was wrong about everything. The point is that the gap
between what the data show and what the public was told the data show has been
wide enough, for long enough, to do real damage.
A documented misrepresentation: Hoult v. Hoult
The strongest single illustration of how Loftus’s work was deployed is the Hoult v. Hoult
case. In 1995, Loftus published ”Remembering Dangerously” in the Skeptical Inquirer,
in which she presented ”Jennifer H.” as a paradigmatic example of a young woman
who, through suggestive therapy, had developed false recovered memories of paternal
incest and had won a civil judgment against her father on that basis (Loftus, 1995).
The actual court record in Hoult v. Hoult — the case Jennifer Hoult later confirmed
Loftus was describing — tells a substantially different story. Hoult’s account was
supported in court by her mother’s testimony, by a former family babysitter’s testimony,
and by other corroborating witnesses. Loftus’s published account altered Hoult’s age,
occupation, and the timeline of her recovered memories; introduced lurid details (a
bloody towel, a bathroom rape) that did not appear in the trial record; and omitted the
corroborating evidence that had persuaded the jury (Hoult, 2005).
Hoult subsequently filed formal ethics complaints against Loftus, documented the
discrepancies between the published account and the court record, and has continued
to write about the case publicly. The point, for the present article, is not to relitigate the
underlying abuse — the jury verdict stands on its evidence — but to mark a
documented instance in which a leading false-memory researcher published a
substantially altered version of a real adjudicated incest case as evidence for her
position. This is not a small methodological lapse. It speaks to the editorial pattern of
the broader memory wars, in which exceptions were promoted into rules and
corroborated cases were rewritten into cautionary tales.
The Lost in the Mall paradigm is most often cited as evidence that therapy can
implant false memories of abuse. But the paradigm itself, read carefully, points to
a different and more consequential application that has been almost entirely
absent from public discourse.
The same paradigm, applied where it actually operates
Loftus showed that under specific conditions — repeated suggestion, plausible content,
sustained pressure, and most centrally a trusted source whose authority the participant
accepts — autobiographical content can be cultivated that does not correspond to
actual events. Take that finding seriously, and apply it to the conditions in which child
sexual abuse most commonly occurs.
In the typical CSA case, the perpetrator is not a stranger but a caregiver or close family
member: a parent, a step-parent, an older sibling, an uncle, a clergyperson, a coach.
The relationship is one of the most trusted relationships available to the child, and often
the only authority the child has on the meaning of their own experience. The
perpetrator has access, often nightly and across years, to do precisely what the Lost in
the Mall paradigm describes: provide repeated, sustained, authoritative narrative
pressure on the child’s memory of what happened.
The narrative pressure that perpetrators apply is not subtle, and clinicians who work
with CSA survivors hear the same script across cases. Nothing happened. You
imagined it. You’re confused. You wanted it. You’re making it up. This is just how
families are. The naming of these patterns — gaslighting, and more specifically
DARVO (”Deny, Attack, Reverse Victim and Offender,” coined by Jennifer Freyd in this
exact research context) — has become standard in the clinical literature on coercive
abuse and is supported by a growing empirical base (Freyd, 1997; Harsey et al., 2017;
Stark, 2007; Sweet, 2019).
Read this against the Loftus paradigm. The conditions she identified for cultivating
non-veridical autobiographical content — trusted source, repetition, sustained
pressure, plausible counter-narrative — are precisely the conditions that exist inside
the abusive is unreliable.
The directionality of ”false memory” implantation in real-world CSA is therefore the
inverse of the FMSF framing. In the typical case it is not the survivor’s recovered
memory that has been implanted. It is the survivor’s memory of no abuse, or of abuse
that didn’t really mean what it meant — the version they may have lived under for years
or decades — that was the implanted one. The recovery of memory in adult therapy, in
safe relationships, after leaving the perpetrator’s authority, is best understood not as the
introduction of a false belief but as the loosening of a long-standing one. The ”real”
memory was suppressed, fragmented, or counter-narrated under the same conditions
Loftus identified, applied with vastly more force and over vastly longer time periods than
any laboratory paradigm can ethically reproduce.
This inversion is missing from the public discourse on memory not because
it is empirically weaker than the false-memory framing — it is, if anything, better
supported by clinical observation, by betrayal trauma theory, and indirectly by
the Loftus research itself — but because the FMSF strategically owned the term
”false memory” and the narrative built around it. Reclaiming what the term
should describe in this domain is part of the corrective work the field needs to
do.
What Loftus has said about her own memory
There is a further fact about Loftus’s biography, less often discussed in the public
reception of her work, that bears on how the standards she has applied to others apply
to her. In her 1991 memoir Witness for the Defense, Loftus disclosed that she had
herself been sexually molested as a child by a male babysitter. She described a
moment in adulthood when the memory of that abuse, in her own words, ”flew out at
[her], out of the darkness of the past, hitting [her] with full force.” (Loftus & Ketcham,
1991; Hoult, 2023) She wrote of the perpetrator that she hated him, that he had
”betrayed [her] trust, stole [her] innocence, and put an indelible impression, a bad,
black memory into the place where only good, warm, happy memories should be.”
(Loftus & Ketcham, 1991, p. 152) On a separate occasion, in a Psychology Today
interview, she described the same event as ”not that big a deal” (Psychology Today,
1996).
The biography is not raised here as a comment on her credibility as a survivor, which is
not in question. It is raised because Loftus has, over the course of her career,
articulated an evidentiary standard for crediting other women’s accounts of childhood
sexual abuse. In The Myth of Repressed Memory she defined the relevant form of
corroboration as ”medical records indicating venereal disease or obvious scarring of
delicate tissues,” with eyewitnesses, contemporaneous documentation, and similar
independent confirmation in the same register (Loftus & Ketcham, 1994). In other
public commentary she has urged that the accused be interviewed before any
allegation is credited (Psychology Today, 1996; Loftus, 1995).
Her own account meets none of those criteria.
The point of recording this is not to question whether Loftus was abused. It is to mark a
tension at the architectural center of the false-memory position. The most prominent
academic critic of dissociative amnesia is, on her own published account, an instance of
the very pattern she has spent decades casting doubt on: a woman who carries a vivid
memory of childhood sexual assault that surfaced into adult awareness with the somatic
and affective markers of return, who trusts that memory, who has named no perpetrator,
and who has produced none of the corroboration she has demanded of others. The
exception she has made for herself is the rule the false-memory movement has worked
hardest to deny survivors.
Narrative endorsement is not abreaction
A further methodological feature of the false-memory literature is rarely discussed in
summaries of it, and it is the most important one for clinical readers. The Loftus
paradigm, the Murphy et al. replication, the Andrews and Brewin reanalysis, and the
entire family of studies built around them measure a single dependent variable:
whether a participant, in a structured interview, will produce or endorse a narrative
containing elements consistent with the suggested event. The outcome variable is
verbal narrative endorsement under interview conditions. Nothing else is measured
(Loftus & Pickrell, 1995; Murphy et al., 2023; Andrews & Brewin, 2024).
This is not what clinicians observe when a survivor of severe childhood trauma comes
into contact with a previously inaccessible memory. The phenomenology of recovered
traumatic memory in clinical settings is not principally a matter of narrative production.
It is a matter of abreaction: the involuntary return of the somatic, autonomic, and
affective elements of the original experience, often before, and sometimes entirely
without, coherent verbal narrative. Clinicians describe, and survivors report, a
characteristic constellation: autonomic activation (tachycardia, sweating,
hyperventilation, freeze responses), somatic re-experiencing in the body regions
involved in the original event, sensory fragments that arrive in modalities other than
language (visual flashes, tactile sensations, smells, sounds), dissociative state shifts
during which the survivor’s affect, posture, voice, and access to present reality alter in
clinically observable ways, and the involuntary, intrusive quality that distinguishes
traumatic memory from ordinary autobiographical recall. Jennifer Hoult’s trial testimony
in Hoult v. Hoult contains a textbook example: she described a flashback to a childhood
rape as a sensory event of ”static” and a sound rendered as ”plussssh,” not as a
propositional account of who did what to whom (Hoult v. Hoult, 1993).
There is no analogue to this in the false-memory literature. The Loftus paradigm has
never produced, and is not designed to produce, anything resembling abreaction.
Participants in Lost in the Mall who come to endorse the planted narrative do not enter
dissociative states, do not present autonomic crisis, do not arrive in the experimenter’s
office trembling and unable to speak, do not experience involuntary somatic
re-experiencing of being lost in the mall years later. They produce, under structured
interview, a verbal account containing elements of the suggested narrative. That is the
entirety of the phenomenon.
The consequence is that the cultural and forensic deployment of the Loftus literature
against clinically recovered memory of childhood sexual abuse rests on an unstated
equivalence: that what the laboratory paradigm produces (verbal endorsement of a
suggested narrative) is the same kind of phenomenon as what clinicians observe
(involuntary, abreactive, somatically and autonomically marked return of trauma
material). The two are not the same kind of phenomenon. They are not produced
under the same conditions, they do not present with the same features, they do not
respond to the same interventions, and there is now neuroimaging evidence that the
underlying neural states are distinguishable: authentic dissociative states are
associated with measurable psychophysiological and neural signatures that volunteers
cannot voluntarily simulate (Reinders et al., 2012; van der Kolk, 2014; Lanius et al.,
2010). The inference from ”narrative endorsement of mundane false events can be
cultivated in healthy young adults under coordinated suggestion” to ”recovered
traumatic memory in adult survivors of severe childhood abuse is presumptively
unreliable” is not, on close reading, a scientific inference. It is an analogical one, and
the analogy fails on the most basic feature of the clinical phenomenon: the body
remembers in ways that the laboratory paradigm has never reproduced and was never
designed to test.
Dissociative Amnesia is Established Science
While the false-memory framework moved through public discourse and into clinical
training, the science of dissociative amnesia did not stop. It was simply drowned out.
The convergent evidence, across prospective cohort studies, neuroimaging, and
clinical observation, is now strong enough that dissociative amnesia is included as a
formal diagnostic entity in both the DSM-5-TR (300.12) and ICD-11 (6B61). The
category survived multiple revision cycles in which it was scrutinized precisely because
of the false-memory controversy.
Williams (1994): the prospective benchmark
The single most important empirical study on this question remains Linda Meyer
Williams’s 1994 prospective cohort, published in the Journal of Consulting and Clinical
Psychology (Williams, 1994). Williams identified 129 women who, as children in the
1970s, had presented at a hospital emergency department with documented sexual
abuse. The cases had been recorded at the time, with hospital records and police
reports.
Seventeen years later, Williams interviewed these women in adulthood, asking
detailed questions about their abuse histories. Thirty-eight percent of the women — 49
out of 129 — did not report the index abuse that had been documented in their hospital
records. Among the women who did recall it, 16 percent reported having had a prior
period during which they did not remember the abuse.
Two findings about the non-recall group sharpened the picture. Women who had been
youngest at the time of the documented abuse were more likely to have no recall in
adulthood. Women whose abuser had been someone they knew well — a family
member, a close acquaintance — were also more likely to have no recall.
This second finding, that closeness of the abuser predicts forgetting, is the central
prediction of betrayal trauma theory, articulated by Jennifer Freyd in 1994 and
elaborated in subsequent work (Freyd, 1994, 1996). The theory’s core proposal is
straightforward: when the perpetrator is someone the child depends on for survival — a
parent, a primary caregiver — conscious awareness of the abuse threatens the
attachment relationship that the child requires to live. Forgetting becomes adaptive.
Awareness can be selectively walled off so that the dependency relationship can
continue. This is not weakness on the child’s part. It is the only available solution to an
impossible problem.
The Williams cohort has been the focus of sustained methodological attention precisely
because its conclusion — that severe, documented childhood sexual abuse can
become inaccessible to conscious adult recall — is so consequential. Critics have
proposed alternative explanations: that some women may have been reluctant to
disclose to an interviewer; that some may have remembered but not specifically
connected the index event to the questioning; that some may have had encoding
deficits because of age. Each of these alternatives has been examined empirically.
None accounts for the size of the non-recall group, and none addresses the additional
finding of the 16 percent prior-period-of-not-remembering subgroup, which by definition
cannot be explained by simple non-disclosure.
Back and Svedin: verified abuse, fragmented recall
Two Swedish studies by Back and Svedin extended this picture in a particularly clean
way. The cohorts in question consisted of children whose sexual abuse had been
objectively verified through child sexual abuse material seized in police investigations.
The first cohort comprised 10 children; the second, 30. In each case, the abuse was on
video. There was no question of whether it had occurred.
What the Back and Svedin studies showed, across both cohorts, is that even with
photographic and video evidence in hand, many of the children did not spontaneously
disclose the abuse to investigators. Their accounts in early police and clinical interviews
were frequently fragmented, minimal, or clearly incomplete relative to what the seized
material documented. Central sexual events were sometimes omitted, sometimes
denied, sometimes only vaguely described. In some cases the children appeared
genuinely unaware of portions of what had happened to them — a finding particularly
common when abuse had occurred at very young ages, or under conditions of impaired
awareness such as sleep or sedation. Across both cohorts, the children showed
marked trauma-related and dissociative symptoms (Svedin & Back, 1996, 2003, 2011).
These data are particularly hard to dismiss because they remove the central skeptical
move that has been used against retrospective recovered-memory studies. There is no
question of whether the abuse occurred. It was filmed. The question is only whether
the child can narrate it. The answer, repeatedly, is that the child cannot, or can only
partially.
Other prospective and longitudinal evidence
Williams’s findings have been substantially replicated. Catherine Widom’s prospective
cohort followed children with court-substantiated histories of abuse and found, in adult
follow-up, a similar pattern of non-disclosure relative to documented record. (Widom &
Morris, 1997) Fergusson and colleagues’ Christchurch (New Zealand) longitudinal
study found that adult retrospective reports of childhood abuse systematically
under-reported events that had been documented in childhood records — the opposite
of what a false-memory framework would predict (Fergusson et al., 2000).
A 2007 meta-analytic review by Brewin and colleagues, examining the empirical
literature on traumatic memory, concluded that the evidence for delayed recall of trauma
— including childhood sexual abuse — is substantial and is not adequately accounted
for by ordinary forgetting curves, by social-desirability effects, or by retrospective bias
(Brewin, 2007). More recent meta-analytic work by Dalenberg and colleagues has
reinforced this conclusion across a wider clinical and experimental literature (Dalenberg
et al., 2012).
The cumulative pattern, across decades of research and dozens of studies, is that
retrospective accounts of childhood trauma tend to under-report documented events.
Neurobiological substrate
Dissociative amnesia is not a mystical or merely metaphorical phenomenon. It has a
neurobiological substrate consistent with what is known about memory encoding
under extreme stress. Acute and prolonged stress — particularly stress experienced
as inescapable — produces neuroendocrine and autonomic states that interfere with
hippocampally-mediated declarative memory consolidation while preserving, or even
amplifying, amygdala-mediated affective and sensory memory traces. The result is a
memory profile in which narrative coherence is fragmented while affective and somatic
fragments remain accessible, often involuntarily (van der Kolk, 2014; Bremner, 2006).
Functional neuroimaging studies of patients with dissociative identity disorder and
dissociative amnesia have shown reproducible patterns of altered prefrontal-limbic
connectivity, particularly during state switches and during exposure to trauma-related
material (Reinders et al., 2012; Schlumpf et al., 2014). These findings do not ”prove”
that any specific recovered memory is accurate — no neuroimaging finding can do that
— but they establish that dissociative amnesia is associated with measurable, replicable
neural signatures, not with simulation or confabulation.
Convergence
The scientific position on dissociative amnesia is, by 2026, considerably more settled
than the popular discourse would suggest. The DSM-5-TR and ICD-11 entries reflect a
literature in which prospective cohort evidence (Williams, Widom, Back and Svedin),
longitudinal evidence (Fergusson), neurobiological evidence (HPA-axis and
neuroimaging), and theoretical evidence (betrayal trauma) converge.
This is not the picture of a contested or fringe phenomenon. It is the picture of an
established clinical reality whose public reception was, for thirty years, controlled by
counter-narrative built around a small body of memory-implantation experiments that
did not address it.
The False Memory Syndrome Foundation as a
Strategic Operation
The False Memory Syndrome Foundation was incorporated in Philadelphia in March
1992 by Pamela and Peter Freyd, the parents of the developmental psychologist
Jennifer Freyd. Jennifer Freyd had, in the late 1980s, told family members that she had
recovered memories of childhood sexual abuse by her father, Peter Freyd. The FMSF
was founded as a response to that family conflict. Within two years of incorporation, it
had a national membership, a media profile, a scientific advisory board, and an active
program of intervention in courtroom and clinical-policy debates.
Three features of the FMSF are worth noting because they bear on how the
organization shaped the scientific and public discourse it claimed to represent.
”False Memory Syndrome” was never a diagnosis
The phrase ”false memory syndrome” was coined by the FMSF itself. It has never been
included in the DSM, in the ICD, or in any other recognized diagnostic system. There is
no peer-reviewed body of literature establishing it as a clinical entity with reliable
criteria, predictive validity, or characteristic course. As a scientific concept, it does not
exist. As a rhetorical concept, it has been extraordinarily effective.
The clinical effect of the term was to pathologize the act of disclosure. A patient who
came to remember childhood abuse in therapy could, under the FMSF framing, be
characterized as suffering from ”false memory syndrome” — a putative iatrogenic
disorder produced by the therapist. This linguistic move accomplished what the
underlying empirical claims could not: it shifted the burden of proof from the perpetrator
to the survivor, and it pathologized therapeutic settings in which trauma disclosure
occurred.
The advisory board and its conflicts
The FMSF’s scientific advisory board was presented in the foundation’s press materials
as a roster of independent academic memory researchers. The composition was more
complicated than that presentation suggested, in three different ways.
First, the foundation’s own founders had a direct personal stake in the foundation’s
central claims. Pamela and Peter Freyd were the parents of Jennifer Freyd, who had
told family members in the late 1980s of recovered memories of childhood sexual
abuse by her father. The accuser, in this case, was their adult daughter, by then an
established academic psychologist. Peter Freyd’s account of these events appeared in
early FMSF communications under various pseudonyms; Jennifer Freyd’s account, with
the corroborating context her parents disputed, was subsequently published in
peer-reviewed academic work and is the genesis of betrayal trauma theory itself (Freyd,
1996; Salter, 1998). The point for the present article is not to adjudicate the underlying
family dispute but to mark that the founding of an organization positioning itself as the
neutral scientific authority on the credibility of CSA accusers was, at its origin, a
response by the parents of one such accuser. This conflict of interest was rarely
disclosed in press coverage of the foundation.
Second, and substantively more important for the science: several of the most
prominent figures on the FMSF scientific advisory board had professional histories
specifically devoted to the experimental induction, fragmentation, and erasure of
memory under hypnosis and pharmacological manipulation. The institutional authority
of the FMSF rested on its scientific advisors being disinterested experts on the reliability
of human memory. A non-trivial portion of that advisory board had spent decades, in
several cases under direct U.S. government funding, studying how to produce the very
phenomena — fragmented recall, induced amnesia, dissociation — that they would
later, in the FMSF’s name, declare implausible when reported by survivors of childhood
abuse.
This is not contested historical material. It is documented in the U.S. Senate Select
Committee on Intelligence’s 1977 hearings on the CIA’s MKUltra program, in the
surviving project files released through the Freedom of Information Act and now publicly
available through the CIA Reading Room, and in the standard scholarly histories of the
program (U.S. Senate, 1977; Marks, 1979; Kinzer, 2019).
Martin T. Orne, the psychiatrist who co-founded the FMSF in 1992, was one of the
foundation’s most active scientific principals until his death in 2000. Orne’s career-long
research focus, conducted at Harvard and then at the University of Pennsylvania’s
Institute for Experimental Psychiatry, was on hypnosis, hypnotically induced amnesia,
and the experimental manipulation of subjective certainty about memory. A substantial
portion of his early research was funded under MKUltra Subproject 84, documented in
the FOIA-released project files (Marks, 1979; Otterman, 2007; Orne, 1959). Orne
became, in the 1990s, one of the most prolific academic critics of recovered memory
and a frequent expert witness in the United States and abroad against the credibility of
CSA disclosures in adult psychotherapy. The professional path — from CIA-funded
research on the production of experimental hypnotic amnesia to public expert
testimony on the implausibility of trauma-induced amnesia — is, at minimum, a
circumstance that should have been considered in the public reception of his testimony.
It generally was not.
Louis Jolyon West, Chair of Psychiatry at UCLA from 1969 to 1989 and a long-standing
FMSF advisor, was the principal investigator of MKUltra Subproject 43, on hypnosis and
drug-altered states. West later became a public figure in cult-deprogramming work and,
like Orne, a prominent skeptic of recovered-memory testimony in CSA cases (Marks,
1979).
Harold Lief, a University of Pennsylvania psychiatrist who served on the FMSF advisory
board, had professional ties earlier in his career to the Human Ecology Fund —
formally, the Society for the Investigation of Human Ecology — which the Church
Committee identified in 1976 as a CIA front used to fund behavioral research on
coercive persuasion, brainwashing, and dissociative phenomena (U.S. Senate Church
Committee, 1976; Marks, 1979). Margaret Singer, the clinical psychologist who became
a leading academic critic of dissociative-disorder diagnoses and an FMSF affiliate, had
received Human Ecology Fund support in the 1950s for her research on returned
Korean War prisoners of war, work that became foundational to the subsequent
academic literature on coercive persuasion (Marks, 1979; Scheflin & Opton, 1978;
Singer & Lalich, 1995).
The point of naming these professional histories is not that the FMSF’s scientific
positions were wrong by association, nor to suggest that any of these individuals
participated in the documented wrongs committed under MKUltra itself. The point is
more limited and more difficult to evade. An organization that positioned itself as the
public scientific authority denying that memories of severe abuse could be lost and
later recovered was, at its institutional core, advised by individuals whose research had
helped establish that people, under specific circumstances, actually do lose access to
autobiographical memories (U.S. Senate Church Committee, 1976). Whether or not
one regards this as a disqualifying conflict, it is a conflict, and it bears on the framing in
which the foundation’s pronouncements were received. The public — and, more
consequentially, the courts and the clinical training institutions that absorbed the
FMSF’s positions — was not told.
Third, the FMSF’s affiliations included individuals whose public statements on
pedophilia and on the seriousness of childhood sexual abuse were difficult to reconcile
with the foundation’s stated mission of carefully distinguishing genuine cases from false
ones. Ralph Underwager, an advisory board member, gave a 1993 interview to the
Dutch pedophile magazine Paidika in which he described pedophilia as ”an acceptable
expression of God’s will for love” and characterized pedophiles as ”responsible.”
Underwager and his wife Hollida Wakefield, who shared his views and routinely
testified for accused individuals in CSA proceedings, withdrew from the advisory board
after the Paidika interview became public. His role on a board that had positioned itself
as the scientific authority on childhood sexual abuse allegations had been substantial
up to that point (Geraci, 1993; Salter, 1998).
These were not isolated outliers in the FMSF’s roster. They form a pattern that any
responsible historiography of the false-memory movement must account for. The FMSF
was not, in any honest sense, a neutral scientific body. It was an advocacy organization
whose institutional history, founding circumstances, and advisory composition were
systematically aligned with one side of the debate it claimed to mediate.
Cherry-picking and the moral panic
The FMSF’s communications strategy relied heavily on individual cases of
therapeutically-induced false memory — cases that did, and do, exist, and that are
sometimes the result of incompetent or coercive clinical practice. These cases were
promoted as paradigmatic. The far larger population of survivors with corroborated
histories and recovered memories, the survivors whose memories were independently
verified by perpetrator confessions or by other documentary evidence, were not.
This is the structure of moral panic in its classical form: the exception was promoted
into the rule. A small number of bad therapists became, in the FMSF presentation,
evidence that the entire infrastructure of trauma-informed clinical work was suspect.
The FMSF was formally dissolved in 2019, after almost three decades of operation. The
phrase ”false memory syndrome” remained in popular usage, and the cultural and
clinical residue of the foundation’s work — the trained reflex to suspect the survivor
before suspecting the perpetrator — has outlasted the organization itself.
The Clinical and Societal Cost
The cost of the memory wars is not abstract. It is measurable in three domains.
Clinical training and practice
A generation of clinicians was trained to treat the question of childhood abuse with a
defensive caution that exceeds anything justified by the actual data. The fear of being
accused of ”implanting” memories produced a pattern of practice in which clinicians
stopped asking, stopped exploring history, and in some cases actively redirected
patients away from disclosure. This was not because the science required it. It was
because the medico-legal climate, shaped by the FMSF’s strategic operations, made
asking dangerous to the clinician’s practice.
The downstream consequence is delayed and missed diagnosis of post-traumatic and
dissociative conditions. Complex post-traumatic stress disorder (CPTSD) and
dissociative identity disorder (DID) are characterized by, among other features,
chronicity, comorbidity, and substantial healthcare utilization prior to accurate diagnosis.
The mean reported delay from initial symptom presentation to accurate DID diagnosis is
approximately six to seven years across multiple studies (Putnam, 1989; Brand et al.,
2016). This figure should be read as a conservative lower bound rather than as a
complete picture of the problem. The 6–7 year mean is calculated from patients who
eventually received an accurate diagnosis. Patients who remain undiagnosed across
their lifetime, patients who are repeatedly and persistently misdiagnosed, and patients
who disengage from psychiatric care altogether are by construction excluded from
these statistics. The true population-level delay between the onset of dissociative
symptomatology and accurate clinical recognition is almost certainly substantially
longer, and for an unknown but non-trivial fraction of patients is effectively infinite.
Patients in the interval before accurate diagnosis are typically managed under other
diagnostic frames — borderline personality disorder, treatment-resistant depression,
psychosis — that do not address the underlying trauma structure and are
associated with prolonged suffering, repeated hospitalization, and substantial
iatrogenic harm.
Survivor disclosure
The cultural climate produced by the false-memory framework has had a chilling effect
on survivor disclosure independent of clinical practice. Survivors who attempted to
come forward in the 1990s and 2000s frequently encountered family members,
employers, and legal systems that had absorbed the FMSF framing as common sense.
Disclosure was met with skepticism, often hostile skepticism, and the burden of proof
was effectively reversed. The lesson many survivors drew, and still draw, is that
speaking is not safe.
This has implications beyond the individual case. Public-health surveillance of childhood
sexual abuse depends on disclosure. When the cultural cost of disclosure rises,
prevalence estimates fall — not because abuse is decreasing, but because reporting is
suppressed. The result is that policy and resource allocation are made on the basis of
artificially low estimates of a serious public-health problem.
Forensic and legal practice
The false-memory framework was particularly successful in courtroom contexts, where
it was deployed against survivors as expert testimony. The cumulative effect over thirty
years has been a body of legal precedent that treats recovered memory of childhood
sexual abuse with a categorical skepticism that is not warranted by the underlying
science. Civil and criminal cases have been won or lost on the basis of expert testimony
that misrepresents the state of the empirical literature.
The Hoult case, discussed above, is one documented example of the pattern. There are
others. Each represents not only an individual injustice but a contribution to a body of
jurisprudence that future survivors will inherit.
Both / And
The corrective to the memory wars is not a return to uncritical acceptance of every
recovered memory. It is the capacity, missing for thirty years, to hold two propositions at
once.
Memory is reconstructive, and under specific conditions involving trusted-source
suggestion, plausible content, and sustained pressure, it can be influenced. This is real
and clinically relevant. It bears on eyewitness testimony, on interrogation procedure,
and on incautious clinical technique. Loftus’s experimental work documented this in a
useful way and her contribution to the science of memory should be acknowledged.
And: severe, prolonged childhood trauma — particularly trauma inflicted by attachment
figures — can produce dissociative amnesia, a disturbance of declarative memory
access whose neurobiological substrate is increasingly well characterized and whose
clinical course includes the possibility of later, often partial, often involuntary return of
memory in conditions of relative safety. This is also real, and it is established in the
same scientific literature.
These two propositions are not in conflict. They describe different phenomena, in
different populations, under different conditions, mediated by different neural systems.
The thirty years during which the field, the courts, and the public were asked to choose
between them was not a victory for science. It was a victory for a strategic
counter-movement that found the second proposition inconvenient and worked, with
considerable success, to bury it.
The work of the next decade is to recover what was lost: clinical training that takes
dissociation seriously, courtroom practice that represents the science accurately, and a
public discourse in which survivors of childhood sexual abuse are met first with the
attention their accounts deserve and only second with the appropriate caution that any
human memory system warrants.
WONSA’s position is that this is not optional. The clinical conditions that result from
severe childhood trauma — CPTSD, DID, related dissociative presentations — are
common, treatable, and substantially under-diagnosed. The barrier to better outcomes
is not, at this point, the science. The science is reasonably clear. The barrier is the
residue of a cultural and professional climate that was deliberately shaped to make this
material harder to see and harder to address.
That climate can be changed. Restoring scientific accuracy to clinical training, to public
discourse, and to legal practice is a precondition for changing it. This article is part of
that work.
References
Andrews, B., & Brewin, C. R. (2024). Lost in the mall? Interrogating judgements of false memory.
Applied Cognitive Psychology. Advance online publication. https://doi.org/10.1002/acp.70012
Brand, B. L., Sar, V., Stavropoulos, P., Krüger, C., Korzekwa, M., Martínez-Taboas, A., &
Middleton, W. (2016). Separating fact from fiction: An empirical examination of six myths
about dissociative identity disorder. Harvard Review of Psychiatry, 24(4), 257–270.
https://doi.org/10.1097/HRP.0000000000000100
Bremner, J. D. (2006). Traumatic stress: Effects on the brain. Dialogues in Clinical
Neuroscience, 8(4), 445–461. https://doi.org/10.31887/DCNS.2006.8.4/jbremner
Brewin, C. R. (2007). Autobiographical memory for trauma: Update on four controversies.
Memory, 15(3), 227–248. https://doi.org/10.1080/09658210701256423
Center for Inquiry Libraries. (n.d.). False Memory Syndrome Foundation collection. 3965
Rensch Road, P.O. Box 741, Amherst, NY 14228, United States.
https://empireadc.org/search/catalog/nyamcfi_4503
_
Central Intelligence Agency. (n.d.). Reading room — MKULTRA mind-control CIA records.
https://www.cia.gov/readingroom/search/site/mkultra
Dalenberg, C. J., Brand, B. L., Gleaves, D. H., Dorahy, M. J., Loewenstein, R. J., Cardeña, E.,
Frewen, P. A., Carlson, E. B., & Spiegel, D. (2012). Evaluation of the evidence for the
trauma and fantasy models of dissociation. Psychological Bulletin, 138(3), 550–588.
https://doi.org/10.1037/a0027447
Fergusson, D. M., Horwood, L. J., & Woodward, L. J. (2000). The stability of child abuse
reports: A longitudinal study of the reporting behaviour of young adults. Psychological
Medicine, 30(3), 529–544. https://doi.org/10.1017/S0033291799002111
Freud, S. (1962). The aetiology of hysteria. In J. Strachey (Ed. & Trans.), The standard edition
of the complete psychological works of Sigmund Freud (Vol. 3, pp. 191–221). Hogarth
Press. (Original work presented 1896)
Freyd, J. J. (1994). Betrayal trauma: Traumatic amnesia as an adaptive response to childhood
abuse. Ethics & Behavior, 4(4), 307–329. https://doi.org/10.1207/s15327019eb0404
1
_
Freyd, J. J. (1996). Betrayal trauma: The logic of forgetting childhood abuse. Harvard
University Press. ISBN 0674068068, 9780674068063.
Freyd, J. J. (1997). Violations of power, adaptive blindness, and betrayal trauma theory.
Feminism & Psychology, 7(1), 22–32. https://doi.org/10.1177/0959353597071004
Geraci, J. (1993). Interview with Ralph Underwager and Hollida Wakefield. Paidika: The Journal
of Paedophilia, 3(1), 2–12. https://www.scribd.com/document/491974809/Paidika9
Harsey, S. J., Zurbriggen, E. L., & Freyd, J. J. (2017). Perpetrator responses to victim
confrontation: DARVO and victim self-blame. Journal of Aggression, Maltreatment &
Trauma, 26(6), 644–663. https://doi.org/10.1080/10926771.2017.1320777
Heffernan, V. (2021, January 5). The memory war. The Cut.
https://www.thecut.com/article/false-memory-syndrome-controversy.html
Hoult, J. (2005 & 2014).
”Remembering Dangerously” & Hoult v. Hoult: The myth of repressed
memory that Elizabeth Loftus created [Self-published documentation].
https://www.rememberingdangerously.com
Hoult, J. (2023). Using experts’ casework demographics to evaluate expert witness credibility:
An empirical case study of the 1970–2020 legal casework of Elizabeth F. Loftus, Ph.D.
SSRN. https://ssrn.com/abstract=4502988
Hoult v. Hoult, Civ. No. 88-1738-M (D. Mass. 1993). Trial transcript, testimony of Jennifer M.
Hoult, May 1993. Reproduced and discussed in Hoult (2005 & 2014).
Kinzer, S. (2019). Poisoner in chief: Sidney Gottlieb and the CIA search for mind control. Henry
Holt. ISBN 9781250140432.
Lanius, R. A., Vermetten, E., Loewenstein, R. J., Brand, B., Schmahl, C., Bremner, J. D., &
Spiegel, D. (2010). Emotion modulation in PTSD: Clinical and neurobiological evidence for a
dissociative subtype. American Journal of Psychiatry, 167(6), 640–647.
https://doi.org/10.1176/appi.ajp.2009.09081168
Lanius, R. A., Harricharan, S., Kearney, B. E., & Pandev-Girard, B. (2025). Sensory pathways
to healing from trauma. Guilford Press. ISBN 9781462556915.
Loftus, E. F. (1995). Remembering dangerously. Skeptical Inquirer, 19(2), 20–29.
https://skepticalinquirer.org/wp-content/uploads/sites/29/2019/03/Issue-02-17.pdf
Loftus, E. F., & Pickrell, J. E. (1995). The formation of false memories. Psychiatric Annals,
25(12), 720–725. https://doi.org/10.3928/0048-5713-19951201-07
Loftus, E. F. (1998, January 22). Deposition, Vol. I, pp. 105–106, Seignious v. Fair [Court
deposition]. Reproduced in J. Hoult’s catalog of Loftus expert-witness testimony.
https://www.rememberingdangerously.com
Loftus, E. F., & Ketcham, K. (1991). Witness for the defense: The accused, the eyewitness, and
the expert who puts memory on trial. St. Martin’s Press. ISBN 9780312084554.
Loftus, E. F., & Ketcham, K. (1994). The myth of repressed memory: False memories and
allegations of sexual abuse. St. Martin’s Press. ISBN 0312114540.
Marks, J. (1979). The search for the ”Manchurian Candidate”: The CIA and mind control. Times
Books. (Reissued 1991, W. W. Norton.) ISBN 9780393307948.
Murphy, G., Dawson, C. A., Huston, C., Ballantyne, E., Barrett, E., Cowan, C. S. M., Fitzsimons,
C. M., Goorah, J., Holloway, K. C., McGrath, U., O’Quigley, A., Prenderville, S., Walsh, S.
P., & Greene, C. M. (2023). Lost in the mall again: A preregistered replication and extension
of Loftus and Pickrell (1995). Memory, 31(6), 818–828.
https://doi.org/10.1080/09658211.2023.2198327
Orne, M. T. (1959). The nature of hypnosis: Artifact and essence. Journal of Abnormal and
Social Psychology, 58(3), 277–299. https://doi.org/10.1037/h0046128
Otterman, M. (2007). American torture: From the Cold War to Abu Ghraib and beyond. Pluto
Press. ISBN 9780745326702.
Psychology Today. (1996, January/February). The diva of disclosure: A profile of Elizabeth
Loftus (pp. 73, 80).
https://www.psychologytoday.com/us/articles/199601/the-diva-of-disclosure
Putnam, F. W. (1989). Diagnosis and treatment of multiple personality disorder. Guilford. ISBN
9780898621778.
Reinders, A. A. T. S., Willemsen, A. T. M., Vos, H. P. J., den Boer, J. A., & Nijenhuis, E. R. S.
(2012). Fact or factitious? A psychobiological study of authentic and simulated dissociative
identity states. PLOS ONE, 7(6), e39279. https://doi.org/10.1371/journal.pone.0039279
Ross, C. A. (2006). The CIA doctors: Human rights violations by American psychiatrists.
Manitou Communications. ISBN 9780976550808.
Salter, A. C. (1998). Confessions of a whistle-blower: Lessons learned. Ethics & Behavior, 8(2),
115–124. https://doi.org/10.1207/s15327019eb0802
2
_
Scheflin, A. W., & Opton, E. M. (1978). The mind manipulators: A non-fiction account.
Paddington Press. ISBN 9780448229775.
Schlumpf, Y. R., Reinders, A. A. T. S., Nijenhuis, E. R. S., Luechinger, R., van Osch, M. J. P., &
Jäncke, L. (2014). Dissociative part-dependent resting-state activity in dissociative identity
disorder: A controlled fMRI perfusion study. PLOS ONE, 9(6), e98795.
https://doi.org/10.1371/journal.pone.0098795
U.S. Senate, Select Committee on Intelligence and Subcommittee on Health and Scientific
Research. (1977). Project MKULTRA, the CIA’s program of research in behavioral
modification (Joint hearing, 95th Cong., 1st Sess., August 3, 1977). U.S. Government
Printing Office. https://intelligence.senate.gov/sites/default/files/hearings/95mkultra.pdf
U.S. Senate, Select Committee to Study Governmental Operations with Respect to Intelligence
Activities (Church Committee). (1976). Final report, Book I: Foreign and military intelligence.
U.S. Government Printing Office.
https://www.intelligence.senate.gov/wp-content/uploads/2024/08/sites-default-files-94755-i.
Singer, M. T., & Lalich, J. (1995). Cults in our midst. Jossey-Bass. ISBN 9780787900519.
Stark, E. (2007). Coercive control: How men entrap women in personal life. Oxford University
Press. ISBN 9780195154276.
Svedin, C. G., & Back, K. (1996). Children who don’t speak out: About children being used in
child pornography (P. Cracknell, Trans., pp. 45–46, 64). Save the Children Sweden.
Svedin, C. G., & Back, K. (2011). Why didn’t they tell us? On sexual abuse in child
pornography (pp. 27, 53–55). Save the Children Sweden.
https://resourcecentre.savethechildren.net/document/why-didnt-they-tell-us-sexual-abuse-c
hild-pornography
Sweet, P. L. (2019). The sociology of gaslighting. American Sociological Review, 84(5),
851–875. https://doi.org/10.1177/0003122419874843
van der Kolk, B. A. (2014). The body keeps the score: Brain, mind, and body in the healing of
trauma. Viking. ISBN 9780143127741.
Widom, C. S., & Morris, S. (1997). Accuracy of adult recollections of childhood victimization,
Part 2: Childhood sexual abuse. Psychological Assessment, 9(1), 34–46.
https://doi.org/10.1037/1040-3590.9.1.34
Williams, L. M. (1994). Recall of childhood trauma: A prospective study of women’s memories of
child sexual abuse. Journal of Consulting and Clinical Psychology, 62(6), 1167–1176.