Can psychedelic drugs retrieve traumatic memories?
Participants in clinical trials and therapy patients often report the spontaneous recall of childhood trauma, especially sexual abuse. Recalling these memories may engender a sense of relief that years of suffering and strange relationships may at last be explained.
Yet many of these memories may be false. The literature linking psychedelics to recovered memory is weak. A recentscoping review of over fifty sources from the 1950s onwards found that none even defined “repressed memory”. Nearly ninety per cent failed to raise the question of whether the recovered material might be false.
At its core, we are dealing with vivid and emotionally saturated images, which may be variously interpreted. In the absence of additional evidence, psychedelic memories may be assumed to be broadly false.
With the acceleration in psychedelic therapy expedited by President Trump, America and its satellites should face the prospect of many hundreds, if not thousands, of vulnerable patients forming false memories. The result may be deaths, litigation, torn-apart families, and perhaps the delegitimation of the field for another generation.
Hypnosis and Amytal
Our history begins with Pierre Janet and Sigmund Freud.
Both posited that intolerable experience can be walled off from consciousness, continuing to generate symptoms until retrieved and “abreacted” in therapy. Under hypnosis, Freud’s patients appeared to uncover early incest memories. He oscillated between believing that Vienna was riddled with abuse and concluding that the scenes were fantasies. Freud eventually settled on the latter and built the Oedipal theory around that choice. From the beginning, then, the same clinical material supported both a realist and a sceptical reading, and there was no internal way to decide between them.
Parallel to hypnosis, early twentieth-century psychiatry developed narcosynthesis: abreactive interviews conducted under sodium amytal and other barbiturates. By the late 1940s, narcosynthesis was hailed as a promising technique for shellshock, seemingly capable of unlocking buried trauma by making patients more talkative and emotionally labile. But closer scrutiny revealed that the memories were heavily coloured by suggestion. Nor did abreaction itself reliably predict improvement in clinical outcomes. Much of the drug’s apparent power came from subjects’ belief that sodium pentothal is a truth serum. The notion has since entered popular consciousness through media depictions, including in Kill Bill, Johnny English, True Lies and The Manchurian Candidate, whose original 1959 book even recalled the MK-ULTRA project. That profile is uncannily close to the risk profile of modern psychedelic therapy, whose adherents ascribe almost magical “healing powers” to psychedelic drugs.
When LSD entered psychiatry in the 1950s, it did so largely within a psychoanalytic frame. Psychiatrists like Ronald Sandison treated LSD as an agent for lifting repression. Drawing on cybernetics, the MK-ULTRA programme would appraise various tools to ‘hijack’ the human mind and implant instructions within – in a method analogous to the autonomous memories reportedly ‘stored’ after traumatic events.
A decisive change occurred with Stanislav Grof’s “cartography” of the psyche, which located biographical, perinatal, and transpersonal levels of experience. COEX systems (“systems of condensed experience”) held clusters of emotionally related memories; at the perinatal level, LSD sessions were interpreted as revisiting birth trauma across four Basic Perinatal Matrices; at the transpersonal level, participants reported ancestral, past-life, phylogenetic, and collective recollections.
Grof’s “cartography” has had a pernicious effect. His system deals with memories that are biologically impossible, like birth and past lives. Grof then places them alongside more plausible content such as childhood abuse. There is no principled boundary between these categories. Both are authenticated using the same criteria, namely vividness and emotional force, as well as ”somatic” conviction.
Grof introduced the “inner healing intelligence”. The doctrine professes that the unconscious possesses wisdom and a teleological impulse toward healing. This means that the material surfacing in sessions can be trusted as meaningful. This belief, inherited more or less intact by MAPS-style MDMA therapy, primes participants to regard whatever arises as reliable testimony.
Real-Life Disasters
The dilemma posed by recalled memories is not especially rare. In a mixed-methods study of over six hundred people with extended post-psychedelic difficulties, over two-fifths said those difficulties were linked to childhood trauma. That subgroup was older, more often female and likely to have used psychedelics in guided settings. In the interview sample of 18 people, two-fifths reported direct trauma re-experiencing. Another study of psilocybin in depression found that about six per cent of participants reported resurfaced memories or visions of childhood abuse.
The most consequential public case involves Amy Griffin, who said in her 2025 memoir, The Tell, that underground MDMA therapy helped her recover previously inaccessible memories of sexual abuse by a middle-school teacher. The book became a bestseller and drew high-profile endorsements, including from Oprah Winfrey. There were signs that the story was not as it seemed. The initial memory emerged in an MDMA session and was followed by months of destabilisation. After her husband gave a $1m donation to MAPS, she was connected to an unnamed expert psychedelic guide who helped her uncover further memories. “I knew that these memories were real. My body knew what had happened to me”, Griffin said.
In March 2026, a woman sued Griffin as Jane Doe, alleging that The Tell recounts abuse that the plaintiff, not Griffin, actually suffered in Texas. In June 2026 Griffin counter-sued for defamation, identifying the plaintiff as Joleene Altum and contending that every element of her account is false. A former classmate had earlier told The New York Times that she had been attacked in some of the locations Griffin described by a different teacher.
Whatever the litigation concludes, it shows how muddied the waters have become.
The Memory Wars
While psychedelic research went underground after the regulatory clampdowns of the late 1960s, the recovered-memory idea made its way into mainstream therapy. Books like The Courage to Heal and a generation of therapists in the 1980s and 1990s appeared on The Oprah Winfrey Show and encouraged clients to look for signs of childhood sexual abuse in their present difficulties – often using suggestive prompts and hypnosis. This wave produced both powerful narratives of survivors breaking silence and a series of high-profile controversies: satanic ritual abuse panics and families torn apart by allegations. Some patients judged that the therapy was abusive, and formed the False Memory Syndrome Foundation.
Experimental work by Elizabeth Loftus demonstrated how easily false autobiographical memories could be implanted. Loftus showed participants filmed accidents and changed the question wording, such as “smashed” versus “hit,” to see how that altered later recall. She found that leading questions could change remembered speed estimates and even increase reports of nonexistent details like broken glass. In the false-memory studies, she and colleagues asked people about real childhood events plus one invented event, which some would uncritically accept.
Richard McNally’s Remembering Trauma synthesised much of this work, arguing that “repression” in the strong, unconscious sense had poor empirical support and that trauma is often over-remembered rather than forgotten. It seemed that the sceptics had won the ‘memory wars’. But a paper by Otgaar and colleagues in 2019 showed that belief in repressed memory persists at high rates among clinicians - around 58% - and may have increased since the 1990s, often reframed under the more respectable label of “dissociative amnesia”.
Not all delayed recalls are false. Studies by Geraerts and others compared continuously remembered abuse, spontaneously recovered memories outside therapy, and memories recovered in therapy under suggestive conditions. They found that spontaneously recovered memories were corroborated about as often as continuous memories, whereas therapy-recovered memories had essentially zero independent corroboration. The most important distinction, then, is between memories that emerge unprompted and those produced in contexts steeped with expectation. This distinction becomes crucial once psychedelics are added to the mix.
According to source-monitoring theory, we do not recover a separate record of a memory’s origin along with its content. Instead, we infer the source from features of the memory, like its vividness and emotional register. Anything that enhances the vividness of internally generated imagery – while degrading contextual information about when and how it was generated – will bias the subject toward thinking “this really happened”.
The notion of total repression is also weakly supported. People can fail to talk about distressing experiences for years and misremember when they previously recalled them. Having recalled them again in therapy, these well-known events may be felt as if they were being discovered anew. But clinical discourse often conflates these phenomena with “repression”, which posits an inaccessible store of memory capable of being pried open.
Psychedelics as Machines for Source Confusion
Against this backdrop, what do we know about the effects of psychedelics on memory?
On the whole, it seems that classic psychedelics impair recollection while enhancing familiarity, and can thereby increase certain kinds of false recognition. In double-blind, placebo-controlled work with psilocybin and 2C-B, Doss and colleagues found that both drugs produced emotional memories that felt more familiar, but were recalled with more inaccuracy. Doss characterises this as a “fluency exaggeration” effect: psychedelics turn up the gain on feelings of familiarity and insight, which are then misattributed to whatever is nearby.
The memory-systems limb of the integrated FIBUS model (“False Insights and Beliefs Under Psychedelics”), proposed by McGovern and colleagues, reinforces this picture. Psychedelics are thought to relax high-level predictive priors in the brain, generating more prediction errors and more “aha” moments; dopaminergic mechanisms inflate the precision of these insight signals, regardless of objective correctness. During the trip, people experience more ideas that feel revelatory, but their capacity to encode contextual detail is impaired. Afterwards, they retain the noetic sense – the feeling that something is “deeply true” – while losing much of the detail that could have been checked.
Not all psychedelics show the same pattern. Observational work with experienced ayahuasca users has found enhancements in hippocampally dependent episodic memory without increased false-memory susceptibility. This suggests that different factors beyond a mere psychedelic modulate the risks, including which compound, dose, tolerance, ritual structure, and memory stage, namely encoding versus retrieval.
Other Drugs
Psychedelics are not unique. Many psychoactive drugs distort memory, and some produce more straightforward suggestibility than LSD or psilocybin. Cannabis, for instance, increases false memory and induces a general “yes bias” in recognition tasks. Reddit is chock-full of accounts of recalled sexual abuse while users were stoned. In one study, cannabis-intoxicated participants were more susceptible to misleading questions about virtual-reality mock crimes than controls. MDMA increased some false recognition but did not heighten misinformation susceptibility to the same degree. Alcohol at high blood alcohol concentrations elevates suggestibility, and barbiturates like amytal famously generated highly confident but often inaccurate narratives.
Compared to psychedelics, there is little flourishing discourse about cannabis- or alcohol-assisted recovery of childhood memories. It is the set and setting: the therapeutic notion that psychedelic drugs will reveal a secret past.
Past Lives and Aliens
If we believe that recalled childhood abuse is credible, what are we to do with “memories” of past lives, alien abductions, ancestral trauma, and phylogenetic experiences? All are reported with the same vividness and somatic conviction associated with other revelatory states.
Spanos’s work showed that subjects could be led to construct detailed alternate identities from other eras. The credibility that subjects assigned to these identities was not linked with how liable they were to hypnosis, though. Instead, it was their prior belief in reincarnation and expectations conveyed by the experimenters.
When told that their past-life identities would be questioned about their childhoods, subjects duly produced childhood abuse. Replications in different cultures found that people’s recollections are culturally dependent. In Korean samples, people recalled past lives as animals; in Canada, subjects’ recollections were exclusively human.
Or consider the reports of alien abductees. Claimants sincerely believe their experiences and show physiological responses – heart rate, skin conductance, muscle tension – comparable to PTSD patients reliving real trauma. Claimants also display higher rates of absorption and fantasy proneness, and a vulnerability to false recall. Some reported “abductions” occurred during sleep paralysis, which were elaborated by therapists who believe in UFOs. The body, in other words, responds intensely to false memories, and does not “keep the score” as an objective referee. Unfortunately, practitioners may invoke bodily reactions as proof that psychedelic memories must be true.
Claims about generational and ancestral trauma have likewise blurred distinct phenomena. There is strong evidence that trauma risk and stress phenotypes can be transmitted across generations via behavioural and perhaps epigenetic pathways. There is no credible mechanism, however, for transmitting phenomenology across generations. When participants report vividly “remembering” a grandparent’s experience under psychedelics, or becoming their DNA, the more parsimonious description is a powerful imaginative engagement with known historical narratives.
The West is Obsessed with Memory
If psychedelic drugs encouraged false memories of themselves, we would expect to see long histories of such phenomena in a variety of use contexts. But Indigenous Amazonian frameworks for ayahuasca do not concern themselves with autobiography in the manner of the West. Their rituals are more diagnostic and cosmological, with visions perceived as signs of sorcery, spirit intrusion, or routes to curing physical disease and discovering lost objects. Healers will extract pathogenic darts and consult with spirits.
Anthropologist David Dupuis’s concept of the “socialisation of hallucinations” describes how participants in Amazonian centres gradually learn to see visions that confirm the local cosmology. Immersing oneself in the rhythms of a retreat centre causes one’s attention to become redirected: certain visions are expected and interpreted, creating feedback loops that match the explanatory schemes of the group. The psychedelic visions then feel like direct verification of doctrines they have learned. Once you change the underlying doctrine – be it a plant teacher or a tool for recalling child abuse – the visions change accordingly.
Since the Western preoccupation with the self and memory is fuelled by therapy culture, we must consider the effect of the digital environments in which such dialogues circulate. Jay Olson’s research has linked smartphone addiction with hypnotisability. Tallying hours of screen time per day places the user in a shallow state of absorption, their focus ever prodded by narrowed algorithms. If a user is caught in a loop rich in “therapeutic” content, often of mixed quality, they are bound by similar suggestion dynamics to those experienced in hierarchical therapeutic contexts. It is possible that psychedelic “afterglows” render users more vulnerable to digital exposures. Because they tend to agree with their users’ prompts and be obsequious, the historian Benjamin Breen has theorised that chatbots could prove a dangerous facilitator of false memories.
It would be misleading to reduce all psychedelic re-experiencing of adverse events to Western trauma discourse, however. Surveys of ayahuasca retreat participants show that re-experiencing known adverse life events is common and can be helpful. Women, for instance, report high rates of re-experiencing sexual assault, and veterans report combat trauma. These are not always artefacts of suggestion and culture.
Exploring Memory Ethically
How, if at all, can we assess the veracity of memories surfacing under psychedelics?
We can use external corroboration: documentation and matched testimonies. We may compare our recollections to known probabilities. Childhood sexual abuse is unfortunately common. But the likelihood of a specific recovered narrative may be far lower. Indeed, since we already know that traumatic memories are unlikely to be “blocked”, to recall decades of abuse by multiple perpetrators, wholly forget it, and assemble them in stages with a pushy therapist is almost certainly false. We may suppose that memories that were actively sought and map onto a guide’s favoured theory should be treated with particular caution.
Depending on one’s metaphysics, we may assign such phenomena as birth trauma and memories of ancestors as bearing zero likelihood. It would be odd, all things considered, for such dramatic scientific findings to occur through totally unreliable technology.
It is sometimes suggested that memories can be viewed as “metaphorical”, a kind of symbolic token deposited by the unconscious mind. We may attempt to tread a balance in treating memories as real but confabulated. Perhaps the memories have exactly the mistiness you’d expect with those from infancy and early childhood, especially if they weren’t quite understood at the time.
Even then, the neat trichotomy between veridical, metaphorical, and confabulated memories is likely too clean. Consider the case of a depressed patient who recalled being suffocated with a pillow by his mother during an Imperial College, London trial. He was counselled to view the episode as symbolic of the psychic suffocation that fuelled his mental disorder. Yet the patient struggled to let go. Was his mother guilty of attempted murder? The answer could spell the difference between healthier boundaries and fully reporting them to the police. If we hold to a Grofian account of the “inner healing intelligence”, it would be mysterious that such a specific event was merely symbolic. It would require a kind of theodicy to explain why the unconscious would toy so with the mood and relationships of its subject. The answer may be a begrudged uncertainty, which the patient could have benefited from never engaging.
One interviewee in a recent study expressed this explicitly: “I can’t trust it. I don’t know if that was real… The challenging thing is, like, that’s a horrible thing to not know if it’s real or not.”
A mature ethic for psychedelic work should cultivate this sort of self-aware uncertainty.
Clinicians and facilitators should document participants’ beliefs and prior disclosures before dosing, which makes evaluation possible. Informed consent materials should include memory effects. Participants should clearly understand that psychedelics might degrade recall while inflating feelings of certainty.
Participants should be prepared for the emergence of material – not specifically “forgotten trauma” – and explicitly told that what appears in sessions is not always evidence in the forensic sense. Facilitators should neither immediately validate nor deny emerging narratives, instead adopting a stance of “gentle agnostic respect” that honours the person while refraining from judgement. No irreversible action, like accusations and legal complaints. should be taken solely on the basis of psychedelic memories.
Any memory recovered under or soon after psychedelic intoxication should be treated as contaminated, much as hypnotically refreshed testimony is.