The Psychedelic Experience: Mystical States, Set and Setting, Expectancy and Blinding
Psychedelic-Assisted Medicine: Science and Practice
Learning objectives for this lesson:
- Contrast pharmacological and experiential accounts of psychedelic therapy, and describe mystical-type experiences, emotional breakthrough, psychological insight and psychological flexibility, naming the questionnaire used to measure each.
- Evaluate the evidence that features of the acute experience predict later therapeutic benefit, distinguishing correlation, prediction and mediation.
- Define set and setting and trace their history from the Saskatchewan researchers and Hollywood Hospital to the guidelines used in modern trials.
- Summarize the evidence that personality, intentions, music, the physical environment and the therapeutic alliance shape psychedelic experiences and their outcomes, and explain why trials need to report set and setting consistently.
- Explain placebo response, expectancy and functional unblinding, and how unblinding can bias the results of a psychedelic trial.
- Calculate and interpret the share of correct allocation guesses and the Bang blinding index for each arm of a trial, and describe how trials and regulators try to reduce the bias that unblinding and expectancy create.
- Weigh the arguments and evidence on whether the subjective experience is necessary for therapeutic benefit.
- Describe how challenging experiences are measured, how often they occur in different samples, how people make meaning of them and what extended difficulties some people report.
- Apply the lesson's measures, set and setting standards, blinding checks and safeguards for challenging experiences to plan a trial of psilocybin-assisted therapy.
This course was developed by Dr. Kiffer G. Card.
Experience and Outcome: Mystical-Type Experiences, Emotional Breakthrough, Insight and Psychological Flexibility
Introduction
Accounts of psychedelic-assisted therapy rarely stop at the drug. Participants describe a sense of unity, grief that finally broke through, or a new view of a relationship or of their own habits. Many researchers argue that the quality of this acute experience helps determine who benefits, alongside the dose and the person's condition.
Lesson 1 introduced the subjective effects of psychedelics and the scales used to record them, and Lesson 4 described the accompanying brain changes. Three narrower questions organize the evidence here. Which features of the experience go with later improvement? How strong is the association? What kind of study would show that the experience causes the improvement? Section 2 then asks what shapes the experience, Section 3 asks how expectations complicate trials, and Section 4 turns to experiences that are frightening or painful.
Learning Objectives
- 1.1 Contrast the pharmacological and experiential accounts of psychedelic therapy, and describe the causal chain that the experiential model proposes.
- 1.2 Describe the mystical-type experience and the MEQ30 used to measure it, including the criterion for a complete mystical experience.
- 1.3 Describe emotional breakthrough and psychological insight, and name the questionnaires used to measure each.
- 1.4 Explain psychological flexibility and why researchers propose it as a process linking the acute experience to later change.
- 2.1 Distinguish correlation, prediction and mediation, and explain why each falls short of showing that an experience causes improvement.
- 2.2 Summarize the evidence that features of the acute experience predict later benefit, including the design and size of the main studies.
1.1 The Experiential Model of Psychedelic Therapy
Two broad accounts of how psychedelics might help a person with depression or addiction shape how trials are designed and read. In a pharmacological account, the drug changes the brain directly, for example by activating serotonin 2A (5-HT2A) receptors and promoting the growth of new connections between neurons (Lessons 3 and 4), and the subjective effects accompany that action without carrying the benefit themselves. In an experiential account, the subjective experience is part of the active treatment: what a person feels, perceives and understands during the session sets off psychological changes that continue after the drug has left the body.
Roseman, Nutt and Carhart-Harris (2018) describe the experiential account as a basic principle of the psychedelic treatment model, in which the quality of the acute experience mediates long-term improvements in mental health. The two accounts can both be partly true. A drug may open a period of greater neural plasticity while the content of the session, and the therapy around it, shapes what is learned during that period. Figure 5.1 sets out the experiential model as a chain of proposed links and adds the alternative paths that any test of the model has to rule out.
The solid arrows in Figure 5.1 are the claims a study would need to support, and the dashed paths explain why that is difficult. Expectancy can raise both the intensity of the experience a person reports and the improvement they report afterward. A strong alliance with the therapists can improve both the session and the recovery. A higher dose, or greater sensitivity to the drug, can produce both a more intense experience and a larger pharmacological effect. Each can create a correlation between experience and outcome without the experience causing the change. The following parts describe the features of experience that researchers measure and then ask how far the evidence supports the solid arrows.
1.2 Mystical-Type Experiences
The feature studied longest is the mystical-type experience, a state that resembles the experiences described in religious and contemplative traditions. The term "mystical-type" signals that researchers are describing the form of an experience by its reported features, without making any claim about its spiritual truth.
1.2.1 Origins of the Construct
The modern measure grew out of the comparative study of religion. Barrett, Johnson and Griffiths (2015) trace its dimensions to William James (1902) and Walter Stace (1960), who described recurring features of mystical experience across traditions: a profound sense of unity, a felt sense of sacredness, a sense of encountering truth or reality at a fundamental level (the "noetic" quality), deeply felt positive mood, transcendence of ordinary time and space, and difficulty putting the experience into words. In the 1960s, Walter Pahnke and William Richards built a questionnaire on Stace's scheme (Roseman et al., 2018). Its most widely used version had 43 items, each rated on a six-point scale from 0 ("none; not at all") to 5 ("extreme").
The revised 30-item Mystical Experience Questionnaire (MEQ30) was derived from retrospective accounts of experiences with psilocybin mushrooms and validated with data from five laboratory studies in which 184 participants received at least 20 mg of psilocybin per 70 kg (Barrett et al., 2015). The flip cards below describe its four factors. A response pattern counts as a complete mystical experience when the person scores at least 60 percent of the maximum possible score on each of the four factors (Barrett et al., 2015). In the same validation, MEQ30 scores predicted the persisting changes in attitudes, behaviour and well-being that participants attributed to their psilocybin session, after controlling for how intense they rated the drug effects overall.
1.2.2 The Johns Hopkins Studies of Healthy Volunteers
The first controlled modern study of the mystical-type experience used healthy volunteers and remains the reference point for later clinical work. Case 5.1 summarizes it, including the difficult experiences reported alongside the positive ones.
Thirty-six adults who had never used a psychedelic and who took part regularly in religious or spiritual activities received a high oral dose of psilocybin (30 mg per 70 kg) in one session and methylphenidate (40 mg per 70 kg), a stimulant used as the comparison drug, in another, under double-blind conditions in a comfortable, supportive room. Volunteers were encouraged to close their eyes and direct their attention inward (Griffiths et al., 2006).
At 14 months, 58 percent rated the psilocybin session as among the five most personally meaningful experiences of their lives and 67 percent rated it as among the five most spiritually significant. Sixty-four percent said it had increased their well-being or life satisfaction, and 58 percent met the criteria for a complete mystical experience. In correlation and regression analyses, the mystical experience measured on the session day was the strongest predictor of the ratings of meaning and significance at follow-up (Griffiths et al., 2008).
The same report noted that 11 of the 36 volunteers had a period of significant fear during the psilocybin session and that 6 had transient ideas of reference or paranoia, none of which persisted after the session (Griffiths et al., 2008). The study shows that a high dose given with careful preparation can occasion experiences that people value highly. It does not show a clinical benefit, because the participants were healthy and the outcomes were their own ratings of meaning.
Two limits carry forward to the clinical studies. The ratings come from people who know they have taken a powerful drug, which exposes them to the expectancy effects discussed in Section 3, and a rating of personal meaning differs from a clinician's measure of depressive symptoms. Later studies therefore asked whether the experience predicts clinical outcomes, and whether features other than the mystical one matter.
1.3 Emotional Breakthrough and Psychological Insight
Olson (2021) notes that many patients without full mystical experiences still find psychedelic treatment beneficial. Researchers at Imperial College London and Johns Hopkins University therefore developed measures of two other features of the session: emotional release and new understanding of oneself.
1.3.1 Emotional Breakthrough
An emotional breakthrough is an experience of confronting difficult emotions or memories that are usually avoided and of feeling relief or release afterward. Roseman and colleagues (2019) asked 379 people who were planning to take a psychedelic on their own initiative to complete online surveys before and after the experience. Factor analysis identified six items forming the Emotional Breakthrough Inventory (EBI), with high internal consistency (Cronbach's alpha of 0.932). Scores rose with dose and were higher when the psychedelic was taken with therapeutic planning and intent. In 75 participants with low well-being at baseline, the EBI, the MEQ and the Challenging Experience Questionnaire together predicted later changes in well-being (r = 0.45). Emotional breakthrough and mystical-type scores predicted larger increases, and challenging experience scores predicted smaller increases (Roseman et al., 2019).
1.3.2 Psychological Insight
Psychological insight refers to new understanding of one's own patterns, motives or history, for example recognizing how avoidance has kept a problem in place. Two scales measure it. The Psychological Insight Questionnaire, developed in an online survey of 1,661 people who had used psilocybin or LSD, has 23 items in two subscales: insights about avoidance and maladaptive patterns, and insights about goals and adaptive patterns. Its scores predicted retrospectively reported gains in psychological flexibility and well-being beyond what mystical and challenging experience scores predicted (Davis et al., 2021). The Psychological Insight Scale, a six-item measure with a seventh item on behaviour change, was validated in a prospective study of 886 participants; insight measured this way statistically mediated the relationship between emotional breakthrough and long-term well-being (Peill et al., 2022).
Table 5.1 brings the measures of this section together. It also includes two measures discussed elsewhere in the lesson, so that the full set used in the studies below is in one place.
Table 5.1. Questionnaires used to measure the acute psychedelic experience and the psychological changes proposed to follow it.
| Measure | What it records | Structure | Development or validation |
|---|---|---|---|
| Mystical Experience Questionnaire (MEQ30) | Mystical-type features of the session | 30 items; four factors; "complete" if at least 60 percent on every factor | Barrett et al., 2015 (184 participants in five laboratory studies) |
| Five-Dimensional Altered States of Consciousness Rating Scale (5D-ASC) | Broad changes in consciousness, including oceanic boundlessness and dread of ego dissolution | 94 items; five dimensions (also visionary restructuralization, auditory alterations and vigilance reduction), or 11 lower-order scales (Lesson 1) | Studerus et al., 2010; used by Roseman et al., 2018 in a depression trial |
| Emotional Breakthrough Inventory (EBI) | Facing and releasing difficult emotions | 6 items | Roseman et al., 2019 (379 participants) |
| Psychological Insight Questionnaire (PIQ) | Insight during the session | 23 items; two subscales | Davis et al., 2021 (1,661 survey respondents) |
| Psychological Insight Scale (PIS) | Insight in the days and weeks after the session | 6 items plus a behaviour item | Peill et al., 2022 (886 participants) |
| Acceptance and Action Questionnaire II (AAQ-II) | Psychological inflexibility, so lower scores mean more flexibility | 7 items | Used by Davis et al., 2020 |
| Challenging Experience Questionnaire (CEQ) | Fear, grief, paranoia and other difficult features | 26 items; seven factors | Barrett et al., 2016 (Section 4) |
Every measure in Table 5.1 is a self-report completed after the experience. That is unavoidable for a private mental state, and it means the measures share any bias in how a person remembers and describes the session. The next part moves from the session to the change in psychological habits proposed to carry its effects forward.
1.4 Psychological Flexibility as a Process of Change
If an experience is to improve depression months later, something about the person must change and stay changed. One candidate comes from acceptance and commitment therapy, a behavioural therapy that Lesson 7 returns to when it describes models of psychedelic-assisted therapy.
Psychological flexibility is described as a set of processes that help people manage stressors and act in line with their values: being open to and aware of difficult thoughts and feelings, adapting to what a situation demands, and committing to behaviour consistent with what matters to them (Davis et al., 2020). Its opposite, inflexibility, includes avoiding painful internal experiences and being caught in rigid patterns such as rumination. The idea fits the experiential model well, since participants often describe facing material they normally avoid.
2.1 Survey and Trial Evidence on Flexibility and Mediation
Two studies with different designs show the current state of the evidence. Davis, Barrett and Griffiths (2020) surveyed 985 people who reported having used a psychedelic (71.6 percent male, 84.1 percent white, mean age 32.2). Respondents rated their psychological flexibility on the AAQ-II as it had been before and as it was after a memorable psychedelic experience. Mystical and insightful effects were associated with reported decreases in depression and anxiety, and in a path analysis the increase in psychological flexibility fully mediated that association. The authors called for prospective experimental studies, because both the "before" and "after" ratings were recalled at the same time.
Sloshower and colleagues (2024) then measured flexibility prospectively in people with moderate to severe major depressive disorder. In a fixed-order design, participants received placebo first (19 participants) and psilocybin (0.3 mg per kg) four weeks later (15 participants), within a manualized psychotherapy that incorporated principles of acceptance and commitment therapy. Psychological flexibility, several facets of mindfulness and values-congruent living improved after psilocybin and stayed improved to week 16, and improvements in flexibility were strongly associated with reductions in depression. Because every participant received placebo before psilocybin, the design cannot separate the effect of psilocybin from the effect of time and continuing therapy, and the sample was small. The box below explains why the word "mediates" in studies like these has a specific meaning.
Correlation, prediction and mediation
A correlation says that two measures move together, for example that higher MEQ30 scores go with lower depression scores. A prediction adds time order: the experience is measured before the outcome, so the experience score helps forecast the outcome. Mediation is a causal claim: the treatment changes the mediator, and the change in the mediator produces the change in the outcome.
Statistical mediation analysis can be consistent with that causal story, but it cannot establish it alone. A convincing mediation claim needs the treatment to be randomized, the mediator to be measured before the outcome, and no unmeasured factor that drives both the mediator and the outcome. A retrospective survey meets none of these conditions, and a single-arm trial meets only the second. Expectancy, discussed in Section 3, is the most obvious unmeasured factor in psychedelic studies.
With these distinctions in hand, the final part of the section weighs the clinical evidence that the acute experience predicts outcomes.
2.2 How Strong Is the Link Between Experience and Outcome?
The claim that experience predicts outcome rests on a growing number of mostly small studies. The evidence is consistent in direction, and the designs leave open other explanations.
2.2.1 Evidence That the Experience Predicts Later Benefit
In an open-label trial of psilocybin for treatment-resistant depression, 20 patients received 10 mg and then 25 mg of psilocybin a week apart. Higher oceanic boundlessness (a dimension resembling the mystical-type experience) and lower dread of ego dissolution (a dimension resembling anxiety) during the 25 mg session together predicted 54 percent of the variance in clinical change at five weeks. Oceanic boundlessness was a better predictor than the visual and auditory effects of the drug (Roseman et al., 2018).
Later analyses within randomized trials point the same way. In the psilocybin arm (30 participants) of a trial comparing psilocybin with escitalopram, a stronger therapeutic alliance predicted greater emotional breakthrough and mystical-type experience, and these in turn predicted lower depression scores six weeks after treatment (Murphy et al., 2022). In a waiting-list-controlled trial of 24 adults with major depressive disorder, peak ratings of mystical experience (r = −0.45) and psychological insight (r = −0.75) were correlated with depression scores at four weeks (Levin et al., 2024). A systematic review of 12 studies of psilocybin, ayahuasca or ketamine found that 10 reported a significant association between mystical experience and symptom reduction, while noting that the samples were small and not diverse and that 6 of the 12 studies were open-label (Ko et al., 2022).
2.2.2 Reasons for Caution
None of these studies manipulated the experience while holding the drug and dose constant, which a direct test of the solid arrows in Figure 5.1 would require. Participants rate both the experience and their outcomes, so a general tendency to report positively can link the two, and people who expect a strong drug effect to help may report more improvement after experiencing one. Olson (2021) argues that mystical-type scores may work like a marker of how strongly 5-HT2A receptors were activated, as a rise in blood pressure tracks the dose, so that the correlation reflects a dose-response relationship. Yaden and Griffiths (2021) reply that the associations generally persist after statistically controlling for the overall intensity of drug effects. Section 3 returns to this debate, because the question of whether the experience is necessary is also a question about how trials can be blinded.
Interactive 5.1 asks you to classify the studies in this section by design, to show that the finding "the experience goes with improvement" carries different weight depending on how it was obtained.
What kind of evidence is it?
Roseman and colleagues (2018) found that two dimensions of the acute experience predicted 54 percent of the variance in clinical change at five weeks in an open-label trial of 20 patients. Write down three explanations for this result other than "the experience caused the improvement", using Figure 5.1 and the box on correlation, prediction and mediation in 2.1 Survey and Trial Evidence on Flexibility and Mediation as prompts. Then check your list against these: expectancy raised both the experience and the reported improvement; patients who were more sensitive to psilocybin had both a stronger experience and a larger drug effect; patients with a stronger bond with their therapists had both a better session and better recovery; and a small open-label sample can produce large correlations that shrink in larger studies.
Summary
The experiential model holds that the acute experience is part of the treatment. Researchers measure mystical-type experiences with the MEQ30, emotional breakthrough with the EBI, and insight with the PIQ and PIS, and they propose psychological flexibility as a process that carries the effects of a session forward. Across healthy-volunteer studies, open-label trials, surveys and analyses within randomized trials, these features predict later benefit in most studies, while anxious, dreadful experiences predict less benefit. The studies are small, rely on self-report and have not manipulated the experience independently of the drug, so they show association and prediction more clearly than causation. If the experience matters, so do the conditions that shape it, the set and setting of Section 2.
Key Takeaways
- 1.1 In a pharmacological account the drug changes the brain directly and the subjective effects only accompany that action; in an experiential account the experience itself is part of the treatment and sets off psychological changes that outlast the drug. The two accounts can both be partly true. The experiential model proposes a chain from drug and setting to acute experience, process change and outcome, and expectancy, the therapeutic alliance, the intensity of drug effects and plasticity offer alternative paths that any test of the model has to rule out.
- 1.2 The mystical-type experience is measured with the 30-item MEQ30, whose four factors are mystical, positive mood, transcendence of time and space, and ineffability. A complete mystical experience requires at least 60 percent of the maximum score on every factor. In healthy volunteers given a high dose of psilocybin, the mystical experience was the strongest predictor of ratings of personal meaning 14 months later, although 11 of 36 volunteers had a period of significant fear.
- 1.3 Emotional breakthrough, confronting avoided emotions or memories and feeling release afterward, is measured with the six-item Emotional Breakthrough Inventory. Psychological insight is measured during the session with the Psychological Insight Questionnaire and in the days and weeks after with the Psychological Insight Scale. Like the MEQ30, all of these measures are self-reports completed after the experience.
- 1.4 Psychological flexibility, a concept from acceptance and commitment therapy, combines openness to difficult thoughts and feelings, adaptation to what a situation demands and commitment to values-consistent behaviour, and the AAQ-II measures its opposite. Researchers propose it as the lasting change that carries the effect of a session forward, because participants often describe facing material they normally avoid.
- 2.1 A correlation says that two measures move together, a prediction adds time order, and mediation is a causal claim that the treatment changes the mediator and the mediator changes the outcome. A convincing mediation claim needs randomization, a mediator measured before the outcome and no unmeasured factor driving both. The retrospective survey of 985 people meets none of these conditions and the small fixed-order trial meets only the second, and expectancy is the most obvious unmeasured factor.
- 2.2 Mystical-type experience, emotional breakthrough and insight predict later benefit in most studies that have tested them, from an open-label trial of 20 patients to analyses within randomized trials, while anxious, dreadful experiences predict less benefit. A systematic review found an association in 10 of 12 studies, but the samples were small and not diverse and half the studies were open-label. No study has manipulated the experience while holding the drug constant, so the evidence shows prediction more clearly than causation.
Reflection
A hypothetical news story describes an open-label trial in which 20 patients with treatment-resistant depression received psilocybin. Two dimensions of the dosing session, high "oceanic boundlessness" and low "dread of ego dissolution", together predicted 54 percent of the variance in the change in depression five weeks later. The story concludes that "the mystical experience is what heals". In one or two paragraphs, explain what this finding shows and what it does not show. Address (a) the difference between prediction and mediation, (b) at least two explanations other than the experience causing the improvement, and (c) the kind of study that would come closest to testing whether the experience causes improvement.
The finding shows prediction: the experience was measured during the session, before the five-week outcome, and people with more oceanic boundlessness and less dread improved more. It does not show mediation, which is a causal claim that the drug changed the experience and the experience then changed depression. Mediation would need a randomized comparison, a mediator measured before the outcome and no unmeasured factor driving both, and an open-label study of 20 people meets only the second condition.
Two other explanations fit the data. Everyone knew they had received psilocybin, so patients who expected a strong effect to help may have reported both a more intense experience and more improvement. Patients who were more sensitive to the drug may also have had both a stronger experience and a larger pharmacological effect, so the experience could be a marker of dose response, as Olson (2021) argues. A strong bond with the therapists could also improve both the session and recovery. The closest test would manipulate the experience while holding the drug constant, for example by giving the same dose with and without a drug that blocks memory of the session, in a randomized and blinded design, or by comparing a psychedelic with a non-hallucinogenic analogue that has the same plasticity effects. Small studies of this kind, such as the midazolam study by Nicholas and colleagues (2024), are only a beginning.
Minimum 20 characters required.
Question 1: In the revised Mystical Experience Questionnaire (MEQ30), what counts as a "complete mystical experience"?
Question 2: Davis, Barrett and Griffiths (2020) reported that psychological flexibility fully mediated the link between mystical and insightful effects and reduced depression and anxiety. Which feature of the study most limits a causal reading?
Question 3: In the open-label trial analyzed by Roseman, Nutt and Carhart-Harris (2018), which pattern during the 25 mg session predicted the largest improvement in depression at five weeks?
Question 4: Which statement best summarizes the systematic review of mystical experience and outcomes by Ko and colleagues (2022)?
Set and Setting: From Hollywood Hospital to Modern Trials
Introduction
Two people can take the same dose of the same drug and have very different experiences. Researchers in the 1950s noticed this quickly, and the phrase "set and setting" became the shorthand for the influences that lie outside the pharmacology. Section 1 showed that the quality of the experience predicts later benefit. If that is so, the conditions that shape the experience are part of the treatment and need the same scrutiny as the dose.
The history of set and setting has a strong Canadian thread, running through Saskatchewan, Toronto and New Westminster, British Columbia. Lesson 2 tells that history in full and Lesson 7 describes how preparation and dosing sessions are conducted in practice. The focus here is on what the concepts mean, how they developed and what the research evidence says about them.
Learning Objectives
- 3.1 Define set and setting and distinguish the physical, social and cultural layers of setting.
- 3.2 Trace the history of set and setting from the Saskatchewan researchers and Hollywood Hospital to the guidelines used in modern trials.
- 4.1 Summarize the evidence that personality traits, intentions and mood before a session shape the acute experience.
- 4.2 Summarize the evidence that music, the physical environment, the therapeutic alliance and preparation shape the experience and its outcomes.
- 4.3 Explain why inconsistent reporting of set and setting limits comparison between trials, and describe the 2025 consensus reporting guidelines.
3.1 Defining Set and Setting
Precise definitions matter here, because the two terms are often used loosely. Set refers to what the person brings to the experience: expectations, intentions, mood on the day, personality traits, previous experience with psychedelics and the meanings the person attaches to the drug. Setting refers to the environment in which the experience takes place. It has a physical layer (the room, the light, the music, the presence or absence of medical equipment), a social layer (who is present, how they behave and the relationship with them) and a cultural layer (the beliefs, rituals and laws that surround the use of the drug). Hartogsohn (2017) groups these influences under the label "extra-pharmacological" factors.
Figure 5.2 arranges the layers as rings around the drug. The diagram is a teaching device: in practice the layers interact, for example when a trusted therapist (social setting) reduces a person's anxiety (set) on the morning of a session.
With the terms defined, the next part traces how researchers came to recognize these influences and build them into the design of sessions.
3.2 A Short History of Set and Setting
The idea that context shapes a drug experience was established in clinical practice before it had a name. The modern trial setting, with its comfortable room, music and trained monitors, draws on choices made in Canada in the 1950s.
3.2.1 Before the Term: Hubbard, Saskatchewan and Hollywood Hospital
Al Hubbard, an American promoter of LSD whose early career is poorly documented (Hartogsohn, 2017, and other accounts describe him as a former officer of the United States Office of Strategic Services), ran LSD sessions for people with alcohol problems in which he used music and religious imagery. Hubbard persuaded Humphry Osmond and Abram Hoffer in Saskatchewan to adopt these setting-based methods, and Hartogsohn notes that the Handbook of the Therapeutic Use of LSD written by Blewett and Chwelos of the Saskatchewan group in 1959 acknowledges its debt to Hubbard's work. Case 5.2 describes a private hospital in British Columbia that applied these methods for over a decade.
Hollywood Hospital in New Westminster, British Columbia, had treated people with alcohol problems before J. Ross MacLean, who became its medical director, bought it in 1956. Under MacLean, and with Hubbard's involvement, the hospital used LSD between 1957 and 1968 to treat alcohol use and other problems, including depression and anxiety. In their history of the hospital, Donaldson and Dyck describe a treatment that attended closely to patient comfort through the design of the room, lighting, artwork, flowers and music. Patients completed questionnaires before their sessions and wrote about their experiences afterward during recovery (Marquis, 2023, reviewing Donaldson and Dyck, 2022).
More than 500 patient files from the hospital went to the provincial archives (Marquis, 2023). The written accounts that patients produced are an early example of the self-report measures in Table 5.1, and the arranged room is an early example of a deliberately designed setting.
3.2.2 Leary Names the Idea
The phrase itself came from the United States. Hartogsohn (2017) notes that the phrase is commonly credited to Leary and, citing earlier historical work, dates the first published use of "set and setting" to a paper that Timothy Leary presented to the American Psychological Association on 9 September 1961, and notes that at a 1959 conference Frank Fremont-Smith had already observed that the response to a drug depends as much on the situation as on the drug. Leary turned the observation into a slogan and studied psilocybin under a range of sets and settings. Walter Pahnke's 1962 Good Friday study, in which seminary students received psilocybin or placebo in a basement chapel while a Good Friday service was broadcast from the main sanctuary (Johnson et al., 2008), is the best-known early attempt to design a setting for a particular experience; Lesson 2 discusses it.
3.2.3 When the Setting Was Stripped Away
An instructive comparison came from Toronto. Researchers at the Alcoholism and Drug Addiction Research Foundation published a controlled study of LSD for alcohol use disorder in 1966 (CBC News, 2012). Hartogsohn (2017) describes how this group tried to replicate the Saskatchewan results in a deliberately more "objective" setting, in which observers were told not to interact with patients, and found lower improvement rates than the Saskatchewan group, while other researchers who used supportive settings reported results closer to those from Saskatchewan. The comparison is suggestive, but the studies may also have differed in patients, doses and outcome measures, so it falls short of an experiment on setting.
3.2.4 Ceremony as Setting
Indigenous and ceremonial use of psilocybin mushrooms, peyote, ayahuasca and iboga took place within structured settings long before clinical research began, and Lesson 2 describes these traditions and their colonial suppression. The box below summarizes a concern that Indigenous scholars and practitioners have raised about how the psychedelic field treats those settings.
Setting, relationality and appropriation
An Indigenous-led, globally represented group of practitioners, scholars, lawyers and human rights defenders set out eight ethical principles for Western psychedelic research and practice: reverence, respect, responsibility, relevance, regulation, reparation, restoration and reconciliation (Celidwen et al., 2023). The authors describe Indigenous healing traditions as grounded in relationality, the understanding that human lives depend on ethical relations with other people, ancestors, plants, animals and the natural world.
They argue that Western psychedelic research has separated traditional medicines from the context of their original environments, and that many research projects, training programs and facilitated sessions take up Indigenous ceremonies, principles and ideas without the free, prior and informed consent of the communities concerned (Celidwen et al., 2023). Ceremonial settings are part of living cultural practices, and borrowing their elements for a clinical session raises questions of consent and benefit that Lesson 12 examines.
3.2.5 The Modern Trial Setting
Modern trials codified the lessons of the first research era. The safety guidelines published by Johnson, Richards and Griffiths (2008) recommend a living-room-like session room with comfortable furniture, because an overly clinical environment with white walls, extraneous medical equipment and staff in white coats may increase anxious reactions. They recommend at least two monitors so that the participant is never alone, at least eight contact hours over at least four preparatory meetings to build rapport and trust, and, for therapeutic studies, eyeshades and headphones playing supportive music. As of July 2026, guidance from the United States Food and Drug Administration (FDA) expects two monitors for the whole session, led by a clinician with graduate-level training in psychotherapy (FDA, 2026). Lesson 7 describes how these elements are used in practice.
These guidelines rest partly on clinical experience. The next two parts of the section ask what controlled and observational studies show about set and about setting.
4.1 Evidence on Set
Studies of set ask whether characteristics of the person, measured before the session, predict the kind of experience they have. The two largest studies used very different samples, one from laboratories and one from people using psychedelics on their own initiative, and they agree on several points.
Studerus and colleagues (2012) pooled data from 23 controlled experimental studies conducted at the University Hospital of Psychiatry in Zurich between 1992 and 2011, covering 409 psilocybin administrations to 261 healthy volunteers, and tested 24 possible predictors. Dose was the most important predictor of every response measured. Among the non-drug variables, high trait absorption (a tendency to become deeply immersed in sensory and imaginative experiences), being in an emotionally excitable and active state just before taking the drug, and having had few psychological problems in recent weeks were most strongly associated with pleasant and mystical-type experiences. High emotional excitability as a personality trait (as distinct from the excitable state measured just before dosing), younger age and an experimental setting involving positron emission tomography (a type of brain scan) most strongly predicted unpleasant or anxious reactions.
Haijen and colleagues (2018) followed people who planned to take a psychedelic through online surveys at five time points, from 654 participants at baseline to 212 at the last follow-up. Having clear intentions for the experience was associated with mystical-type experiences. A positive set, and intentions related to recreation, were associated with a lower likelihood of a challenging experience. Trait absorption and higher doses increased every aspect of the acute experience, including mystical-type experiences, challenging experiences and visual effects. Baseline traits explained more of the change in well-being than any other group of variables.
Table 5.2 summarizes these studies with the evidence on setting described in the next part. Reading across the rows shows that most of the evidence is observational and that the experimental studies are small.
Table 5.2. Selected studies of set and setting, with their designs, samples and main findings.
| Study | Design and sample | Main finding | Main limitation |
|---|---|---|---|
| Studerus et al., 2012 | Pooled analysis of 23 laboratory studies; 261 healthy volunteers | Dose mattered most; absorption and mood before dosing predicted pleasant experiences; a scanner setting predicted anxious ones | Healthy volunteers in research settings |
| Haijen et al., 2018 | Prospective online survey; 654 at baseline, 212 at final follow-up | Clear intentions predicted mystical-type experiences; positive set predicted fewer challenging ones | Self-selected sample with heavy attrition; unverified drugs and doses |
| Kaelen et al., 2018 | Interviews and ratings; 19 patients with treatment-resistant depression | The experience of the music predicted reduced depression at one week; general drug intensity did not | Small open-label sample; music was rated after the session |
| Olson et al., 2020 | Single-arm placebo study in a party-like setting; 33 students | 61 percent reported some drug effect after placebo | No comparison group; deception was part of the design |
| Murphy et al., 2022 | Path analysis within the psilocybin arm of a randomized trial; 30 patients | Alliance predicted emotional breakthrough and mystical-type experience, which predicted outcome | Small sample; alliance was not randomized |
| Levin et al., 2024 | Waiting-list-controlled trial; 24 adults with major depressive disorder | Alliance before dosing correlated with mystical experience, insight and depression at 4 weeks, 6 months and 12 months (not significantly at 3 months) | Small sample; correlations only |
The findings on set point toward practical steps that preparation sessions already take, such as helping a person clarify intentions and arrive in a calm state. The evidence on setting adds the environment and the people in it.
4.2 Evidence on Setting: Music, Environment, Alliance and Preparation
Setting has been studied in four main ways: through the music played in sessions, through the physical environment, through the relationship between participant and therapists, and through the preparation that comes before the session. The tabs below summarize the evidence on each.
Kaelen and colleagues (2018) interviewed 19 patients who received psilocybin for treatment-resistant depression about the music played during their sessions and rated how much each patient expressed liking, resonance (the music feeling in harmony with their emotional state) and openness to the music-evoked experience. Patients described welcome influences, such as meaningful emotion, guidance and safety, and unwelcome ones, such as unpleasant imagery. The experience of the music was associated with mystical experiences and insightfulness and predicted reductions in depression one week later, while general drug intensity did not. The authors called music a "hidden therapist"; the sample was small and the ratings came from later interviews.
Johnson and colleagues (2008) recommend a living-room-like room because a clinical look may increase anxious reactions. The pooled Zurich data support the concern about clinical environments: sessions involving positron emission tomography predicted unpleasant or anxious reactions, and the authors suggested that the clinical atmosphere of the scanning centre, with technical equipment, white walls and staff in white lab coats, may partly explain the association (Studerus et al., 2012). The environment can also produce effects with no drug at all, as Case 5.3 shows. No randomized trial has compared session rooms directly, so most of the evidence on the physical environment comes from comparisons across studies and from clinical experience.
The therapeutic alliance is the working relationship between a client and therapist, including their agreement on goals and the bond between them. In the psilocybin arm of a trial against escitalopram, a stronger alliance predicted rapport before the session, greater emotional breakthrough and mystical-type experience, and lower final depression scores; emotional breakthrough in the first session in turn predicted a stronger alliance before the second (Murphy et al., 2022). In a Johns Hopkins trial for major depressive disorder, a stronger alliance at the final preparation session predicted lower depression scores at four weeks (r = −0.65), six months (r = −0.47) and twelve months (r = −0.54), and was correlated with higher peak mystical experience (r = 0.49) and insight (r = 0.52) (Levin et al., 2024).
Preparation is where set and social setting meet. In the Johns Hopkins guidelines, monitors discuss the person's history, relationships and beliefs, explain the range of possible experiences, and encourage the person to surrender to difficult experiences, trusting that their usual state of consciousness will return; the guidelines also cite earlier findings that negative mood before a session predicted anxious experiences during it (Johnson et al., 2008). In the survey by Haijen and colleagues (2018), clear intentions predicted mystical-type experiences. Lesson 7 describes the content of preparation sessions in detail.
The strongest single demonstration that setting and expectation can shape a psychedelic-like experience comes from a study in which no psychedelic was given. Case 5.3 describes it.
Researchers at McGill University in Montreal recruited 33 students to what was presented as a study of how a psychedelic drug affects creativity. All participants received a placebo, described to them as a drug resembling psilocybin. The four-hour session took place in a group room arranged like a psychedelic party, with music, paintings, coloured lights and visual projections. To strengthen expectations, confederates subtly acted out the stated effects of the drug, and participants were led to believe there was no placebo group (Olson et al., 2020).
The placebo effects varied widely. Many participants reported no change, while others reported effects on the Five-Dimensional Altered States of Consciousness Rating Scale of a size usually associated with moderate or high doses of psilocybin. In all, 61 percent verbally reported some effect of the drug; several said the paintings on the walls moved or reshaped themselves (Olson et al., 2020). The study shows how far context and expectation alone can go, and it foreshadows the problem of expectancy in trials that Section 3 takes up.
Together, the tabs and Case 5.3 suggest that the setting is an active ingredient, which raises a practical question: if setting matters this much, trials need to describe their settings well enough for others to compare and repeat them.
4.3 Reporting Set and Setting
A finding about a treatment can only be repeated if the treatment is described fully. Two recent studies show how far psychedelic research has to go in describing its settings, and what a standard description might include.
Estric and colleagues (2025) systematically reviewed clinical trials of psychedelics in people with psychiatric disorders published up to December 2023. Of 4,912 articles screened, they retained 27 articles reporting on 25 studies with 763 participants in total; 13 were randomized and 12 were open-label, and psilocybin accounted for 47 percent of the studies. They assessed 16 domains of set and setting. Participant selection and the creation of a safe environment were consistently reported, but reporting was uneven for monitor training (52 percent of the 25 studies), control of visual distractions (64 percent) and creation of a pleasant environment (68 percent). The authors concluded that differences in set and setting limit the comparability and reproducibility of trials.
In response to gaps of this kind, an international Delphi consensus study gathered 89 experts from 17 countries, including researchers, clinicians and past trial participants. After four rounds, 30 variables reached the predefined consensus level (rated important or very important by at least 70 percent of experts). These form the Reporting of Setting in Psychedelic Clinical Trials (ReSPCT) guidelines, grouped into the physical environment, the dosing session procedure, the therapeutic framework and protocol, and subjective experiences (Pronovost-Morgan et al., 2025). The authors also noted significant ambiguities in how the field currently defines set and setting.
Interactive 5.2 gives you practice in placing influences into the layers of Figure 5.2. Some items fit more than one layer, and the feedback explains the best fit and the evidence behind it.
Set, physical setting, social setting or cultural setting?
Look back at Case 5.2 (Hollywood Hospital) and Case 5.3 (the McGill placebo study). For each case, list two features that belong to the physical setting, one that belongs to the social setting and one that belongs to the set of the participants. Then name one detail that the ReSPCT categories would ask a modern trial to report that neither case description gives you, such as the training of the people present or what happened at the end of the session.
Summary
Set is what the person brings to the experience, and setting is the physical, social and cultural environment in which it occurs. The concepts were developed in practice by Hubbard, the Saskatchewan group and Hollywood Hospital in the 1950s, named by Leary in 1961, and codified in the safety guidelines that modern trials follow. Pooled laboratory data and prospective surveys show that dose matters most, and that absorption, intentions and mood before the session also shape the experience. Studies of music and of the therapeutic alliance link the setting to outcomes, and a placebo given in a party-like setting produced psychedelic-like effects in many participants. Reporting of setting in trials is uneven, and the 2025 ReSPCT guidelines propose a standard. Because expectation alone can produce strong effects, Section 3 asks how trials can separate the effect of a psychedelic from the effects of what participants expect it to do.
Key Takeaways
- 3.1 Set is what the person brings to the experience: expectations, intentions, mood on the day, personality traits, previous experience and the meanings attached to the drug. Setting is the environment, with a physical layer (the room, light, music and equipment), a social layer (the people present and the relationship with them) and a cultural layer (beliefs, rituals and laws). Hartogsohn groups these as extra-pharmacological factors, and in practice the layers interact.
- 3.2 Set and setting were developed in practice before they were named: Al Hubbard, the Saskatchewan group and Hollywood Hospital in New Westminster (1957 to 1968) attended closely to the room, the music and patient comfort. The phrase was first published by Timothy Leary in 1961, and a Toronto study run in a deliberately "objective" setting found lower improvement rates. Indigenous scholars warn that borrowing ceremonial settings raises questions of consent and benefit, and modern trials follow the Johns Hopkins safety guidelines and, as of July 2026, FDA guidance on monitors.
- 4.1 In pooled Zurich laboratory data from 261 healthy volunteers, dose was the most important predictor of every response; trait absorption, an excitable and active state before dosing and few recent psychological problems predicted pleasant and mystical-type experiences, while a scanner setting predicted anxious ones. In a prospective survey, clear intentions predicted mystical-type experiences and a positive set predicted fewer challenging ones. Most of this evidence is observational.
- 4.2 The experience of the music predicted reduced depression one week later in a study of 19 patients, and a clinical environment may increase anxious reactions, although no randomized trial has compared session rooms. A stronger therapeutic alliance predicted stronger experiences and lower depression in two trials, and preparation is where set and social setting meet. In the McGill placebo study, 61 percent of 33 students reported some drug effect after a placebo given in a party-like setting.
- 4.3 A systematic review of 25 trials found that monitor training, control of visual distractions and the creation of a pleasant environment were reported unevenly, which limits the comparability and reproducibility of trials. An international Delphi study of 89 experts produced the ReSPCT guidelines, 30 variables grouped into the physical environment, the dosing session procedure, the therapeutic framework and protocol, and subjective experiences.
Reflection
A hypothetical hospital in British Columbia wants to adapt an ordinary clinic room for psilocybin-assisted therapy sessions. The room has white walls, a hospital bed, a bright examination light, a window onto a busy corridor and medical equipment on display, and staff wear white coats. The team plans to meet each patient once, for 30 minutes, before the dosing day. Using the evidence on set and setting (the pooled Zurich data, in which a scanner setting predicted anxious reactions; the Johns Hopkins safety guidelines; the music study of 19 patients; and the alliance studies of Murphy and colleagues and Levin and colleagues), recommend four changes to the setting and preparation. Give the evidence for each change and state one limit of that evidence.
First, the room should look less clinical: soft lighting, comfortable furniture in place of the hospital bed, and medical equipment kept out of sight but available. The Johns Hopkins guidelines warn that white walls, visible equipment and white coats may increase anxious reactions, and in the pooled Zurich data a scanner setting, which the authors described as clinical and "antiseptic", was one of the strongest predictors of anxious responses (Studerus et al., 2012). Second, the corridor window should be covered and the session protected from interruptions, with eyeshades and headphones offered, as the guidelines recommend for therapeutic studies.
Third, the playlist should be chosen with care and checked with patients, because the experience of the music predicted reductions in depression one week later in the study of 19 patients (Kaelen et al., 2018). Fourth, preparation should be much longer than one 30-minute meeting. The guidelines describe at least eight contact hours over at least four meetings, and stronger alliance before dosing predicted better experiences and lower depression in two trials (Murphy et al., 2022; Levin et al., 2024). The main limit is that most of this evidence is observational or comes from small samples: no randomized trial has compared session rooms, the music ratings came from interviews after the session, and alliance was measured without being randomly assigned, so people who form strong alliances may differ in other ways.
Minimum 20 characters required.
Question 1: According to Hartogsohn (2017), who is credited with the phrase "set and setting", and when was it first published?
Question 2: In the pooled analysis of 23 laboratory studies by Studerus and colleagues (2012), which factor was the most important predictor of every response to psilocybin?
Question 3: In the McGill University placebo study described in Case 5.3, 33 students received only a placebo in a party-like setting. What share verbally reported some effect of the drug?
Question 4: What did the systematic review by Estric and colleagues (2025) conclude about set and setting in psychedelic trials?
Expectancy, Placebo and Blinding
Introduction
In a randomized, double-blind, placebo-controlled trial, randomization makes the groups comparable at the start, and blinding keeps participants, therapists and assessors from knowing who received the drug, so that their expectations cannot explain a difference between groups. Psychedelics strain this design. Once the effects of a high dose of psilocybin or MDMA begin, most participants can tell whether they received it, and so can the therapists sitting with them.
Section 2 ended with a placebo producing psychedelic-like effects in a party-like setting, and the same power of expectation can inflate the apparent benefit of a psychedelic in a trial. Lesson 1 introduced blinding as one item in a checklist for judging claims; it is the hardest item for psychedelic research. Lessons 8 to 10 apply the tools described below to the trials for each indication, and Lesson 9 applies them to the 2024 review of MDMA-assisted therapy by the United States Food and Drug Administration (FDA).
Learning Objectives
- 5.1 Define placebo response, expectancy, blinding and functional unblinding, and explain how unblinding can bias the results of a psychedelic trial.
- 6.1 Describe how trials check whether blinding held and what trials of psychedelics and ketamine have found.
- 6.2 Calculate the share of correct allocation guesses and the Bang blinding index for each arm of a trial, and interpret the results.
- 6.3 Describe the comparators, instructions and rating procedures that trials use to reduce bias, and how well they work.
- 6.4 Explain how expectancy is measured, what the evidence shows about its influence and what the self-blinding microdosing study showed.
- 7.1 Explain the debate over whether the subjective experience is necessary for therapeutic benefit and the evidence gathered since 2021.
5.1 Placebo Response, Expectancy and Blinding
Four terms need precise definitions. A placebo response is the improvement seen in people who receive an inactive treatment, including natural recovery, regression of extreme scores toward the average and the effects of expectation and care. Expectancy is what a person anticipates a treatment will do, and it can change both how they feel and how they report feeling. Blinding, also called masking, keeps participants and staff unaware of who received which treatment. Functional unblinding occurs when participants or staff work out the allocation in practice because the treatment has effects they can recognize, even though the formal blind is kept.
The FDA's final guidance on psychedelic trials, issued in July 2026, states that the intense perceptual changes these drugs produce increase the potential for bias from functional unblinding of patients, therapists, monitors or raters: those who experience or observe the changes may expect benefit, and those who receive a placebo and notice nothing may expect none (FDA, 2026). Figure 5.3 traces how this can widen the difference between groups.
The lower row of Figure 5.3 is visible in the second phase 3 trial of MDMA-assisted therapy for post-traumatic stress disorder. Only 1.9 percent (1 of 53) of participants in the MDMA arm dropped out, compared with 15.7 percent (8 of 51) in the placebo arm, and the authors attributed the difference partly to the less effective treatment in the placebo arm and partly to disappointment from ineffective blinding (Mitchell et al., 2023). A systematic review of psychedelic randomized trials published between 1990 and 2020 found that they had generally reported neither participants' expectancy nor the success of blinding, and argued that their effect sizes are likely overestimated because of unblinding and high response expectancy (Muthukumaraswamy et al., 2021). The first step is to ask participants what they think they received.
6.1 Measuring Whether Blinding Held
Blinding is usually checked with a short questionnaire asking participants, and often therapists and raters, which treatment they believe was received and how confident they are, as the FDA guidance recommends (FDA, 2026). Table 5.3 shows what trials that asked found.
Table 5.3. Reported results of blinding assessments in selected trials of psychedelics and ketamine.
| Trial | Comparison | What participants guessed |
|---|---|---|
| Bogenschutz et al., 2022 (alcohol use disorder) | Psilocybin versus diphenhydramine | 93.6 percent correct after session 1 and 94.7 percent after session 2; therapists 92.4 and 97.4 percent |
| Mitchell et al., 2021 (MAPP1, post-traumatic stress disorder) | MDMA versus inactive placebo | Not formally assessed; at unblinding, at least 7 of 44 placebo and 2 of 46 MDMA participants had believed the opposite |
| Mitchell et al., 2023 (MAPP2, post-traumatic stress disorder) | MDMA versus inactive placebo | MDMA arm 49 of 52 correct; placebo arm 33 of 44 correct |
| Jelovac et al., 2025 (KARMA-Dep 2, depression, Ireland) | Ketamine versus midazolam | Ketamine arm 78.1 percent correct; midazolam arm 46.4 percent after infusion 1 and 60.0 percent at the end |
| Szigeti et al., 2021 (self-blinding microdosing) | Microdoses versus empty capsules | 72 percent of capsule guesses correct, against 62.5 percent by chance |
Participants in trials of high-dose psilocybin or MDMA against an inactive or weak comparator identified their allocation in most cases, and some trials report results for both arms together while others report each arm. The two equations below put the results on a common footing.
6.2 The Share of Correct Guesses and the Bang Blinding Index
For each arm, count the participants who guessed correctly (c), those who guessed incorrectly (i) and those who said they could not tell (u). Equation 5.1 gives the share who guessed correctly; with two equal arms and a forced choice, about half would be right by chance.
The blinding index proposed by Bang and colleagues in 2004 subtracts incorrect from correct guesses and divides by everyone asked (Equation 5.2), so that unsure answers pull the index toward zero (Muñoz Laguna et al., 2025). Calculated separately for each arm of a two-arm trial with equal allocation, it runs from −1 to +1: 0 is what random guessing produces, +1 means everyone in the arm identified their allocation, and −1 means everyone believed they were in the other arm. Values between −0.2 and 0.2 are taken as consistent with blinding under a stringent rule, and values between −0.3 and 0.3 under a less stringent one, so an index of exactly 0.2 fails the stringent rule; these cut-offs are ad hoc (Muñoz Laguna et al., 2025).
Worked example. In the MAPP2 blinding survey, 49 of 52 participants in the MDMA arm thought they had received MDMA, 1 thought placebo and 2 could not tell; in the placebo arm, 33 of 44 thought placebo, 9 thought MDMA and 2 could not tell (Mitchell et al., 2023). For the MDMA arm, p = 49 / 52 = 94.2 percent and BI = (49 − 1) / 52 = 0.92. For the placebo arm, p = 33 / 44 = 75.0 percent and BI = (33 − 9) / 44 = 0.55. Both indices are far above the cut-offs, so most participants in each arm knew their allocation. Use Calculator 5.1 to reproduce these numbers, then compare the ketamine trial that used midazolam as its comparator.
Three cautions apply. First, a correct guess can follow from improvement as well as from side effects, since a participant who feels better may conclude they received the drug. Second, the chance level depends on the design: in the self-blinding microdosing study, three of every four capsules were placebos, so a person guessing at random in that proportion would be right 62.5 percent of the time (0.75 × 0.75 + 0.25 × 0.25), the figure against which the observed 72 percent must be judged (Szigeti et al., 2021). Third, pooling the arms can hide the pattern. In the last preset of Calculator 5.1, everyone believes they received the drug: half the guesses are correct overall, which looks like perfect blinding, while the arm indices of +1 and −1 show that nobody guessed at random.
KARMA-Dep 2 gave up to eight infusions of ketamine or midazolam to adults hospitalized with depression in Dublin (Jelovac et al., 2025). Load its two presets in Calculator 5.1, compare the midazolam arm's blinding index after the first infusion with its value at the end of treatment, and decide whether each meets the stringent cut-off of 0.2. Then write one sentence explaining why a sedating active comparator might protect the blind better at the start of a trial than at the end.
Measuring blinding shows the size of the problem. The next part describes what trial designers have tried.
6.3 Design Responses: Active Placebos, Instructions and Independent Raters
Researchers have responded to functional unblinding in three ways: by choosing a comparator with noticeable effects of its own, by changing what participants are told, and by keeping the people who rate outcomes away from the dosing sessions. None has yet solved the problem for high-dose classic psychedelics.
6.3.1 Active Placebos
An active placebo is a comparator with noticeable effects, so that effects alone do not reveal allocation. The accordion lists comparators used in trials discussed in this lesson. Bogenschutz and colleagues (2022) concluded that diphenhydramine had failed to maintain the blind, as had methylphenidate (used by Griffiths et al., 2006), niacin and low-dose psilocybin in earlier psilocybin trials.
Niacin at 250 mg was the active control in a crossover trial of psilocybin for people with life-threatening cancer; the authors listed its limited blinding among their limitations (Ross et al., 2016).
Griffiths and colleagues (2016) compared a high dose (22 or 30 mg per 70 kg) with a low dose (1 or 3 mg per 70 kg). Carhart-Harris and colleagues (2021) gave the comparison arm two 1 mg doses of psilocybin with daily escitalopram and the psilocybin arm two 25 mg doses with daily placebo capsules, a double-dummy design.
A sedating antihistamine (50 mg, then 50 or 100 mg in the second session) in a trial of psilocybin for alcohol use disorder; guesses were correct after 93.6 and 94.7 percent of sessions (Bogenschutz et al., 2022).
A short-acting sedative (0.045 mg per kg) used as the comparator for ketamine in KARMA-Dep 2, where patients receiving it guessed at about chance after the first infusion and more accurately later (Jelovac et al., 2025).
Aday and colleagues (2025) propose dextromethorphan, an NMDA receptor antagonist found in cough medicines, as an active placebo for classic psychedelics. In a laboratory study of 20 healthy participants that they cite, high-dose dextromethorphan matched psilocybin's time course and peak strength, but psilocybin produced significantly greater insight and mystical-type scores.
In 40 adults with depression who were having routine surgery, ketamine (0.5 mg per kg) or saline was infused during general anaesthesia, so that the acute effects occurred while participants were unconscious. Only 36.8 percent guessed their allocation correctly, with similar guesses in both arms, and ketamine produced no greater acute reduction in depression than placebo; the authors note that the approach is impractical for most trials (Lii et al., 2023).
6.3.2 Instructions That Widen Expectations
A second strategy changes what participants believe the design to be. In their cancer trial, Griffiths and colleagues (2016) told participants and monitors that psilocybin would be given in both sessions at doses that might range from very low to high, and that at least one dose would be moderate to high. Monitors' guesses of the dose, on a 10-centimetre line, averaged 7.0 for high-dose and 1.7 for low-dose sessions, but more than 13 percent of high-dose sessions were rated 4 or less and more than 12 percent of low-dose sessions 4 or more, which the authors took as some protection against monitors' expectations (Griffiths et al., 2016).
6.3.3 Independent Raters and Regulatory Expectations
A third strategy protects the measurement of outcomes. MAPP2 used a central pool of independent assessors who were blinded to the study design, visit number and allocation, and each assessed a participant on the primary outcome scale only once (Mitchell et al., 2023). As of July 2026, the FDA states that sponsors should use central raters blinded to allocation and visit number, that blinding questionnaires for participants and investigators can be helpful, and that expectancy questionnaires before randomization and at the end of treatment should be considered. It suggests pairing a dose-ranging trial without placebo with a placebo-controlled trial, and keeping the monitor in the dosing session out of the later psychotherapy. It also states that the contribution of psychotherapy to any observed efficacy has not been characterized, and that a factorial design could separate the contributions of drug and psychotherapy (FDA, 2026). These measures reduce bias in rating, but participants still know their allocation, which is why expectancy itself also has to be measured.
6.4 Measuring Expectancy
If participants' expectations differ between arms, they can at least be measured so that their influence can be estimated. The evidence on how much expectancy matters in psychedelic trials is mixed.
The Stanford Expectations of Treatment Scale, a six-item questionnaire with positive and negative subscales, predicted 12 to 18 percent of the variance in outcomes of surgical and pain treatments in its development studies (Younger et al., 2012). In the trial comparing psilocybin with escitalopram, Szigeti and colleagues (2024) analyzed data from 55 participants, who expected more from psilocybin than from escitalopram. Expectancy predicted improvement with escitalopram but did not predict response to psilocybin, while trait suggestibility measured at baseline was associated with response to psilocybin only. The authors concluded that psychedelic therapy may be less vulnerable to expectancy bias than previously suspected. This finding conflicts with the warning of Muthukumaraswamy and colleagues (2021), and the question is unresolved: the analysis was small, and expectancy measured before a trial misses the expectations that form once a participant recognizes the drug's effects. Case 5.4 describes the study that showed most clearly how far expectation can explain apparent benefit.
Microdosing means taking small doses of a psychedelic, too low to cause hallucinations, one to three times a week. Researchers at Imperial College London invited people already planning to microdose to add a placebo control, using identical opaque capsules and envelopes labelled with QR codes so that the researchers could later recover which capsules each person had taken. Participants were randomized to four weeks of placebo, four weeks of microdosing, or two weeks of each, and 191 completed the study (Szigeti et al., 2021).
Outcomes improved in the microdose group, but the placebo group also improved, with no significant differences between groups on the accumulated outcomes. Small differences on some daily and weekly measures disappeared when participants' guesses about their capsules were added to the models, except for ratings of drug intensity. Participants guessed their capsules correctly 72 percent of the time, against about 62.5 percent expected from random guessing, and people who reported larger doses were more likely to guess correctly (Szigeti et al., 2021).
The substances and doses were not verified, adherence could not be confirmed, and participants had positive expectations. The authors concluded that the anecdotal benefits of microdosing can be explained by the placebo effect (Szigeti et al., 2021). For trials of full doses, the lesson is that guesses, and the expectations that follow them, can produce differences that look like drug effects.
At full doses, expectancy can be measured but cannot be removed while participants still have the experience, which raises the question with which the lesson began: would the treatment work without it?
7.1 Is the Subjective Experience Necessary?
In its April 2021 issue, the journal ACS Pharmacology and Translational Science published two viewpoints on opposite sides of this question. The debate matters for blinding, because a treatment whose active ingredient is a recognizable experience can never be fully blinded, and for access, because a drug that worked without the experience could be given without long supervised sessions. The tabs summarize both positions and the later evidence.
Yaden and Griffiths (2021) propose that the subjective effects of psychedelics are necessary for their enduring benefits and account for most of the benefit. They point out that particular subjective effects, such as mystical-type experience and insight, predict therapeutic outcomes after the overall intensity of drug effects is statistically controlled, and they regard plasticity and other neurobiological mechanisms as likely necessary and insufficient alone. They proposed a critical test: give a psychedelic to people rendered fully unconscious, for example by deep anaesthesia, and see whether the benefit remains; they wrote that they would not be surprised to see some therapeutic effects, of lower magnitude or more transient.
Olson (2021) argues that the subjective effects may not be necessary, noting that the evidence for their role is largely correlational. He suggests that mystical-type scores may track the strength of 5-HT2A receptor activation, much as a rise in blood pressure does, and cites mostly animal evidence that brief drug exposure starts plasticity lasting days and that compounds engineered to avoid hallucinogen-like behaviour in rodents keep their plasticity-promoting effects. He argues that non-hallucinogenic treatments could reach more people, including those excluded from psychedelic therapy by a family history of psychosis. Olson disclosed that he was president and chief scientific officer of Delix Therapeutics, and the checklist from Lesson 1 asks readers to weigh such disclosures alongside the evidence.
Two studies have since approached the test that Yaden and Griffiths described. In the anaesthesia trial described above, ketamine given while patients were unconscious produced no greater acute improvement in depression than saline (Lii et al., 2023). The result is consistent with ketamine's acute effect depending partly on conscious experience or expectations, but it concerns ketamine over a short follow-up. In a study of 8 healthy volunteers, adding the amnestic sedative midazolam to 25 mg of psilocybin allowed a conscious experience while partly impairing memory of it, and higher midazolam doses and greater memory impairment tended to go with lower ratings of salience, insight and well-being (Nicholas et al., 2024). The authors suggested that memory of the experience contributes. Lesson 4 describes the non-hallucinogenic analogues that would provide the most direct test, and Lesson 12 discusses their commercial context.
The debate is unresolved: the correlational evidence in Section 1 and the small manipulation studies fit the view that the experience matters, and the animal evidence keeps the alternative open. Every trial of a full-dose psychedelic has to be read with functional unblinding in mind.
Summary
Psychedelics produce effects that participants and therapists recognize, so the formal blind is often broken in practice, raising expectations in the drug arm and lowering them in the placebo arm. Blinding can be summarized for each arm by the share of correct guesses and the Bang blinding index; in trials of high-dose psilocybin and MDMA, most participants in both arms identify their allocation. Active placebos, broadened instructions, independent raters and expectancy measures reduce some sources of bias, and the evidence on how much expectancy drives outcomes conflicts. The self-blinding microdosing study showed that guesses can explain apparent benefits, and the debate over whether the experience is necessary remains open. Section 4 turns to challenging experiences.
Key Takeaways
- 5.1 A placebo response includes natural recovery, regression toward the average and the effects of expectation and care, and expectancy is what a person anticipates a treatment will do. Functional unblinding occurs when participants or staff work out the allocation from recognizable drug effects, raising hope and engagement in the drug arm and bringing disappointment and dropout in the placebo arm, as in MAPP2 (1.9 against 15.7 percent dropout). The observed difference between arms then combines the drug effect with a gap in expectations.
- 6.1 Blinding is checked by asking participants, and often therapists and raters, which treatment they believe was received and how confident they are. In trials of high-dose psilocybin or MDMA against inactive or weak comparators, most participants identified their allocation, for example 93.6 and 94.7 percent of guesses in a psilocybin trial with diphenhydramine as the comparator.
- 6.2 For each arm, the share correct is the number who guessed correctly divided by everyone asked (Equation 5.1), and the Bang blinding index subtracts incorrect from correct guesses before dividing (Equation 5.2), so it runs from −1 to +1 with 0 expected from random guessing. In MAPP2 the indices were 0.92 in the MDMA arm and 0.55 in the placebo arm, far above the ad hoc cut-offs of 0.2 and 0.3. The chance level depends on the design, a correct guess can follow from improvement, and combining the arms can hide a strong pattern of guessing.
- 6.3 Active placebos such as niacin, low-dose psilocybin, diphenhydramine and methylphenidate have failed to maintain the blind in psilocybin trials, midazolam protected it only early in a ketamine trial, and masking by general anaesthesia is impractical for most trials. Instructions that widen expectations about doses give some protection against monitors' expectations, and central raters blinded to allocation and visit number, which the FDA's 2026 guidance expects, protect outcome ratings even though participants still know their allocation.
- 6.4 Expectancy can be measured with questionnaires such as the Stanford Expectations of Treatment Scale, and the evidence on its influence conflicts: in 55 participants it predicted improvement with escitalopram but not response to psilocybin, while a review argued that trial effect sizes are likely overestimated. In the self-blinding microdosing study of 191 participants, both groups improved with no significant differences on the accumulated outcomes, and small differences disappeared once participants' guesses were modelled.
- 7.1 Yaden and Griffiths argue that the subjective effects are necessary because they predict outcome after the overall intensity of drug effects is controlled; Olson argues they may not be, citing mostly animal evidence on plasticity and the promise of non-hallucinogenic compounds, and he disclosed a commercial interest. An anaesthesia study of ketamine and a small midazolam study of psilocybin fit the view that conscious experience or memory of it matters, but the debate is unresolved.
Reflection
A hypothetical trial of psilocybin for depression reports this blinding survey at the end of treatment. In the psilocybin arm of 50 participants, 45 guessed psilocybin, 3 guessed placebo and 2 were unsure. In the placebo arm of 50 participants, 30 guessed placebo, 15 guessed psilocybin and 5 were unsure. Calculate the share who guessed correctly and the Bang blinding index for each arm, interpret each index against the cut-offs of 0.2 and 0.3, and explain how these results should affect how a reader interprets the reported difference in depression scores between the arms. Suggest two design changes for the next trial.
In the psilocybin arm, the share correct is 45 / 50 = 90 percent and the blinding index is (45 − 3) / 50 = 0.84. In the placebo arm, the share correct is 30 / 50 = 60 percent and the index is (30 − 15) / 50 = 0.30. The psilocybin arm is clearly unblinded. The placebo arm's index of 0.30 equals the less stringent cut-off, and both rules require a value below the cut-off, so this arm meets neither rule, although its participants were more mixed in their beliefs than those in the drug arm.
Because most participants in the psilocybin arm knew their allocation, part of the difference in depression scores could reflect higher expectations in that arm and disappointment in the placebo arm, in addition to any effect of the drug. A reader should treat the reported effect as an upper estimate, check whether outcomes were rated by blinded central raters, and look for analyses that adjust for participants' guesses or for expectancy measured before randomization. For the next trial, I would use an active comparator, such as a low dose of psilocybin, together with instructions that widen expectations about the doses, as Griffiths and colleagues (2016) did, and I would add an expectancy questionnaire before randomization and at the end of treatment, keep the in-session monitors out of later psychotherapy, and use central raters blinded to allocation and visit number, as the FDA's 2026 guidance recommends or asks sponsors to consider.
Minimum 20 characters required.
Question 1: Which statement best defines functional unblinding?
Question 2: In the MAPP2 placebo arm, 33 of 44 participants guessed placebo, 9 guessed MDMA and 2 were unsure. What is the Bang blinding index for that arm?
Question 3: What happened to patients' guesses in KARMA-Dep 2, which compared ketamine with midazolam?
Question 4: Which statement about the self-blinding microdosing study (Szigeti et al., 2021) is correct?
Challenging Experiences: Frequency, Measurement, Meaning and Extended Difficulties
Content note
This section discusses frightening and distressing experiences after psychedelic use, including reports of suicidal thoughts and suicide attempts. Support is available at any time from the 9-8-8 Suicide Crisis Helpline (call or text 9-8-8, 24 hours a day) and, for Indigenous people, from the Hope for Wellness Help Line (1-855-242-3310).
Introduction
Not every psychedelic experience is pleasant or meaningful. People who use psychedelics often describe frightening sessions as "bad trips", and researchers and clinicians use the term challenging experience for acute experiences marked by fear, grief, confusion, paranoia, physical distress or a sense of losing one's mind. The course uses the clinical term throughout and keeps "bad trip" for people's own descriptions. The choice of term reflects a finding discussed below: many people come to regard such experiences as valuable, while a minority experience lasting harm.
The study of challenging experiences connects the earlier sections of the lesson. Section 1 showed that anxious, dreadful experiences predict less benefit in trials, Section 2 showed that set and setting shape the likelihood of a difficult session, and Section 3 explained why the drug's recognizable effects complicate trials. Lesson 6 covers the clinical side: psychiatric risks such as psychosis and hallucinogen persisting perception disorder, screening, the management of difficult sessions and the support services that exist outside clinics. The focus here is on what challenging experiences are, how often they occur, how people make sense of them and what happens when difficulties persist.
Learning Objectives
- 8.1 Define a challenging experience and describe the seven factors of the Challenging Experience Questionnaire.
- 8.2 Compare estimates of how often challenging experiences occur in supervised studies, in a national population sample and in self-selected surveys, and explain why they differ.
- 8.3 Explain how people make meaning of difficult experiences and why a difficult session can be associated with both benefit and harm.
- 8.4 Describe the extended difficulties that some people report after psychedelic use, their duration and their predictors, and state what self-selected surveys can and cannot show.
- 9.1 Apply the measures, set and setting standards, blinding checks and safeguards from the lesson to plan a trial of psilocybin-assisted therapy for end-of-life distress.
8.1 What Is a Challenging Experience?
A challenging experience is defined by its content during the drug's acute effects, and researchers have developed a questionnaire to describe that content systematically. Its development shows how the field moved from anecdotes to measurement.
Barrett and colleagues (2016) took items from three existing questionnaires, the Hallucinogen Rating Scale, the States of Consciousness Questionnaire and the Five-Dimensional Altered States of Consciousness questionnaire, and analyzed responses from an online survey in which people described their most challenging experience with psilocybin mushrooms. The result was the 26-item Challenging Experience Questionnaire (CEQ), validated as a stand-alone instrument in a second sample. Its seven factors, listed in Table 5.4, provide a profile of the difficult aspects of an experience. Factor scores were associated with how difficult the experience was and also with its meaningfulness, spiritual significance and the change in well-being attributed to it, and the factor structure did not differ by gender or by a previous struggle with anxiety or depression (Barrett et al., 2016).
Table 5.4. The seven factors of the Challenging Experience Questionnaire, with plain-language descriptions.
| Factor | Plain-language description |
|---|---|
| Fear | Anxiety, panic or feeling frightened during the session |
| Grief | Sadness, sorrow or emotional pain, for example about a loss |
| Physical distress | Unpleasant bodily sensations, such as nausea or physical discomfort |
| Insanity | A fear of going mad or of losing one's mind permanently |
| Isolation | Feeling alone or cut off from other people |
| Death | Thoughts or feelings about dying or one's own death |
| Paranoia | Suspicion or fear that other people intend harm |
The CEQ is not the only measure of difficulty. The Altered States of Consciousness questionnaire includes the dimension of dread of ego dissolution, which Section 1 showed predicted less improvement in a depression trial (Roseman et al., 2018). In the prospective survey that validated the Emotional Breakthrough Inventory, higher CEQ scores predicted smaller gains in well-being (Roseman et al., 2019). Yet the CEQ factors were also associated with meaningfulness and improved well-being in the original survey (Barrett et al., 2016). These findings point in different directions, and 8.3 Making Meaning of Difficult Experiences returns to the apparent contradiction. First, the question of how often such experiences occur.
8.2 How Often Do Challenging Experiences Occur?
Estimates of the frequency of challenging experiences range widely, and most of the range comes from differences in who was asked. Figure 5.4 places three kinds of evidence side by side, and the paragraphs that follow describe each.
8.2.1 In Supervised Studies
Challenging experiences occur even under the best conditions that research can provide. Barrett and colleagues (2016) summarized three highly controlled Johns Hopkins studies in which 69 participants received a high dose of psilocybin (0.429 mg per kg, equivalent to 30 mg per 70 kg) over 204 sessions: about 30 percent of participants had marked periods of anxiety or fear, and between 17 and 39 percent experienced paranoia. Case 5.1 gave the figures for one of these studies, in which 11 of 36 volunteers had a period of significant fear (Griffiths et al., 2008). In a pooled analysis of eight double-blind, placebo-controlled studies with 110 healthy participants, acute adverse reactions marked by strong dysphoria or anxiety and panic occurred only at the two highest doses and in a relatively small proportion of participants, and all were managed with interpersonal support without medication. Follow-up questionnaires found no lasting impairment of functioning in any participant (Studerus et al., 2011). These figures come from healthy volunteers. A 2024 meta-analysis of studies that gave classic psychedelics in clinical or research settings found serious adverse events, including worsening depression, suicidal behaviour, psychosis and convulsive episodes, in about 4 percent of participants with existing neuropsychiatric disorders and in none of the healthy participants (Hinkle et al., 2024). Lesson 6 examines these risks.
8.2.2 In the General Population
Population surveys give a broader picture because they include people who used psychedelics without screening or support. Simonsson and colleagues (2023) analyzed a nationally representative sample of 2,822 adults in the United States, of whom 613 reported having used a classic psychedelic at least once. Most of these 613 (59.1 percent) had never had a challenging, difficult or distressing experience, so 40.9 percent had had at least one. Of the 613, 8.9 percent reported functional impairment lasting longer than a day after such an experience, and 2.6 percent sought medical, psychiatric or psychological help in the days or weeks after their most difficult one. Six set and setting variables were associated with how difficult the experience had been: no preparation, a disagreeable physical environment, a negative mindset, no psychological support, a dose that was too large and a major life event before the experience. Taking lithium or other mood stabilizers at the same time was also associated with greater difficulty, an interaction that Lesson 3 describes.
8.2.3 In Surveys of Difficult Experiences
The most detailed descriptions come from people who chose to report a difficult experience. Case 5.5 describes the largest such survey, which also provided the data from which the CEQ was built.
Researchers at Johns Hopkins University invited people to complete an online survey about their single most psychologically difficult or challenging experience after taking psilocybin mushrooms. Of the 1,993 respondents (mean age 30, 78 percent male), 39 percent rated that experience as among the five most challenging of their lives. Eleven percent said they had put themselves or others at risk of physical harm during it, which was more likely with a higher estimated dose, a longer and more difficult experience, and the absence of physical comfort and social support. In all, 2.6 percent had behaved in a physically aggressive or violent way and 2.7 percent had received medical help.
Among respondents whose experience had happened more than a year earlier, 7.6 percent had sought treatment for enduring psychological symptoms. Three cases appeared to be associated with the onset of enduring psychotic symptoms and three with attempted suicide. Despite these difficulties, 84 percent said they had benefited from the experience. The authors noted that the incidence of risky behaviour or enduring distress is extremely low when psilocybin is given in laboratory studies to screened, prepared and supported participants (Carbonaro et al., 2016).
The box below explains why the three kinds of evidence in Figure 5.4 should not be compared as if they measured the same thing.
Same question, different denominators
A percentage is only as meaningful as its denominator. The supervised studies counted participants who were screened to exclude people at higher psychiatric risk, prepared over several meetings and supported throughout, so their figures describe difficulty under protective conditions. The national survey counted everyone in a representative sample who had ever used a classic psychedelic, so it estimates how common difficult experiences are among people who use psychedelics, although it relies on memory over many years. The worst-experience survey counted only people who chose to respond to a study about difficult experiences, and it asked each about the worst one, so its figures describe the range and severity of difficult experiences and cannot estimate how often they happen.
None of the three can show what fraction of people with a particular diagnosis, given a psychedelic in a clinic, will have a challenging experience. That question needs data from the clinical trials that Lessons 8 to 10 examine and the adverse event reviews that Lesson 6 discusses.
Frequency is only part of the picture. The next question is what difficult experiences mean to the people who have them.
8.3 Making Meaning of Difficult Experiences
People interpret their experiences, tell stories about them and decide what they mean. Two kinds of evidence describe this process: qualitative interviews about how people talk about difficult sessions, and survey data on how difficulty relates to later well-being.
8.3.1 Narrative Work
Gashi, Sandberg and Pedersen (2021) conducted in-depth interviews with 50 people in Norway who used psychedelics. Almost all had had frightening experiences, and many called them "bad trips". The defining feature was a feeling of losing oneself or going crazy, which the authors linked to ego dissolution. Drawing on narrative theory, the authors described how the stories people told about these experiences did several kinds of work, summarized in the flip cards below. They concluded that "bad trip" narratives may be a potent coping mechanism for people who use psychedelics in uncontrolled environments, and that this narrative work also makes continued use more likely after unpleasant experiences.
8.3.2 Difficulty, Duration and Benefit
The survey evidence helps explain why difficult experiences can be associated with both benefit and harm. In the worst-experience survey, the degree of difficulty was positively associated with enduring increases in well-being, while the duration of the difficulty was negatively associated with them (Carbonaro et al., 2016). In other words, an intense difficulty that resolved during the session tended to go with benefit, while a long one tended to go with less. This pattern fits the clinical advice in the Johns Hopkins guidelines, which encourage participants to surrender to frightening imagery or feelings, trusting that their usual state of consciousness will return (Johnson et al., 2008), and it fits the finding that dread of ego dissolution, a resistant, anxious response, predicted worse outcomes in a depression trial (Roseman et al., 2018).
Two cautions apply. People who decided that a difficult experience was valuable may be more willing to complete a survey about it, so samples of this kind may underrepresent those who were harmed. The stories people tell can also make a harmful experience easier to live with without making it less harmful. The integration practices that Lesson 7 describes are, in part, structured support for this meaning-making, and they depend on honest discussion of experiences that did not resolve.
In the worst-experience survey, 84 percent of respondents said they had benefited from their most challenging experience (Carbonaro et al., 2016). Write two sentences. In the first, explain how this figure supports the view that challenging experiences can be valuable. In the second, explain how the way the sample was recruited might make the figure higher than it would be among everyone who has had a challenging experience. Then check whether your second sentence uses the reasoning in the box on denominators (8.2 How Often Do Challenging Experiences Occur?).
8.4 Extended Difficulties
For some people, problems continue after the drug's effects have ended. Researchers call these extended difficulties, a broader term than the diagnosis of hallucinogen persisting perception disorder that Lesson 6 describes, because many of the reported problems are emotional or existential and do not involve perception.
Evans and colleagues (2023) recruited people through social media, a newsletter, student email lists and a newspaper advertisement, asking for anyone who had experienced difficulties after using a psychedelic that negatively affected their functioning for more than a day. Between October 2022 and January 2023, 608 people completed the survey, including a written description of their difficulties. The most common difficulties were anxiety and fear, existential struggle, social disconnection, and depersonalization and derealization (feeling detached from oneself or that the world is unreal). For about one-third of participants the problems lasted more than a year, and for about one-sixth they lasted more than three years. Suicidal thoughts or thoughts of self-harm were reported by 6 percent. A more challenging acute experience predicted a longer duration of difficulties, taking the drug in a guided setting predicted a narrower range of difficulties, and reported durations were longer after ayahuasca, DMT and LSD than after psilocybin and MDMA (Evans et al., 2023).
Interactive 5.3 lets you explore the published figures from this survey and from the two surveys in Figure 5.4. As you switch between them, read the note on who was asked before interpreting the bars.
Explore published figures on challenging experiences and extended difficulties
The survey by Evans and colleagues describes the forms that extended difficulties take and suggests factors to study further, but its design limits what it can show. Every participant had difficulties, so the survey cannot estimate how common extended difficulties are among people who use psychedelics. The drugs and doses were self-reported and unverified, reports were retrospective, and the predictors explained only a small share of the variation in duration (6 percent in the regression model). The survey does show that a minority of people who use psychedelics experience problems that last months or years, that these problems are often emotional, existential and social, and that guided settings may narrow them. Those findings support the course's view that honest discussion of difficult experiences belongs in consent, preparation and follow-up.
9.1 Applying This Lesson
The four sections of the lesson fit together when they are applied to a single real situation. On 4 August 2020, Canada's federal Minister of Health granted exemptions under section 56 of the Controlled Drugs and Substances Act to four people with incurable cancer, allowing them to use psilocybin for end-of-life distress. Their applications were supported by TheraPsil, a non-profit organization founded in Victoria, British Columbia, in 2019 (Rafi and Jarvis, 2020; TheraPsil, 2026). Lesson 11 describes this access route. The task below asks you to design the research that would test this kind of care.
Suppose a hypothetical research team at a Canadian university plans a randomized trial of psilocybin-assisted therapy for anxiety and depression in people with incurable cancer, the group that received the first section 56 exemptions in August 2020. Write a one-page plan with four parts, one for each section of the lesson.
First, name the measures of the acute experience and of psychological change you would collect (Table 5.1), say when each would be completed, and state what analysis would show prediction and what would be needed to support mediation (see 2.1 Survey and Trial Evidence on Flexibility and Mediation). Second, describe three features of set and setting you would standardize and report using the ReSPCT categories, drawing on the practices of Hollywood Hospital and the Johns Hopkins guidelines. Third, choose a comparator from Section 3 (6.3 Design Responses: Active Placebos, Instructions and Independent Raters), explain how you would assess blinding in each arm with Equation 5.2 and measure expectancy, and explain how you would keep outcome raters blind. Fourth, describe how you would prepare participants for a possible challenging experience, which questionnaire you would use to measure one, and how you would follow up participants for extended difficulties.
Summary
A challenging experience is an acute experience marked by fear, grief, physical distress, fear of insanity, isolation, confrontation with death or paranoia, the seven factors of the Challenging Experience Questionnaire. About 30 percent of carefully screened and supported participants given a high dose of psilocybin have marked anxiety or fear, about 41 percent of United States adults who have ever used a classic psychedelic report at least one challenging experience, and surveys of people's worst experiences show a minority with risky behaviour, medical help or lasting symptoms. The figures differ because the denominators differ. Many people transform difficult experiences into valuable ones through narrative work, and intense difficulty that resolves is associated with benefit while prolonged difficulty is not. A minority report extended difficulties lasting months or years, often emotional, existential or social. The final assessment that follows draws the four sections of the lesson together.
Key Takeaways
- 8.1 A challenging experience is an acute experience marked by fear, grief, confusion, paranoia, physical distress or a sense of losing one's mind. The 26-item Challenging Experience Questionnaire, built from a survey of difficult experiences with psilocybin mushrooms, profiles seven factors: fear, grief, physical distress, insanity, isolation, death and paranoia. Its scores predicted smaller gains in well-being in one prospective survey, yet were associated with meaningfulness and improved well-being in the original one.
- 8.2 About 30 percent of screened, prepared participants given a high dose of psilocybin had marked anxiety or fear; 40.9 percent of United States adults in a national sample who had used a classic psychedelic reported at least one challenging experience, and 8.9 percent were impaired for more than a day; and in a self-selected survey of 1,993 people describing their worst experience, 11 percent had put themselves or others at risk. The figures differ because the denominators differ, and none estimates the risk for patients given a psychedelic in a clinic.
- 8.3 Interviews with 50 people in Norway showed that "bad trip" narratives set rules for avoiding them, mark insiders, and turn fear into value, which helps people cope and makes continued use more likely. In the worst-experience survey, greater difficulty went with enduring increases in well-being while longer duration went with smaller ones, so intense difficulty that resolves is linked to benefit. Samples of this kind may underrepresent people who were harmed, and a story can make harm easier to live with without making it less harmful.
- 8.4 Among 608 people recruited because they had difficulties lasting more than a day, the most common problems were anxiety and fear, existential struggle, social disconnection, and depersonalization and derealization; for about one-third they lasted more than a year, and 6 percent reported suicidal thoughts or thoughts of self-harm. A more challenging acute experience predicted longer difficulties and a guided setting a narrower range. Because everyone recruited had difficulties, the survey cannot estimate how common they are.
- 9.1 The four sections come together in the design of a trial, such as one of psilocybin-assisted therapy for end-of-life distress in people with incurable cancer, the group that received Canada's first section 56 exemptions in August 2020. Such a plan names measures of the experience and of psychological change with an analysis suited to prediction or mediation, standardizes and reports set and setting with the ReSPCT categories, chooses a comparator and checks blinding in each arm, and prepares and follows up participants for challenging experiences.
Reflection
A hypothetical friend tells you that a survey of 1,993 people found that 84 percent of those who had their worst experience with psilocybin mushrooms said they benefited from it, and concludes that challenging experiences are nothing to worry about. Write a respectful reply of one or two paragraphs. Explain what the 84 percent figure does and does not show, what other figures from that survey and from the national United States sample (613 people who had used classic psychedelics) and the survey of 608 people with extended difficulties indicate about harm, and what these studies suggest about the conditions associated with more or less difficulty. Give no advice about using drugs.
The 84 percent figure is real, and it fits interview research showing that many people make sense of frightening experiences and come to value them. It comes from people who chose to answer a survey about difficult experiences, though, so people who were harmed and did not want to revisit the experience may be underrepresented, and a person's sense of benefit does not rule out harm. The same survey found that 11 percent had put themselves or others at risk of physical harm, 2.7 percent had needed medical help, and 7.6 percent of those whose experience was more than a year old had sought treatment for lasting symptoms; a few cases involved enduring psychotic symptoms or suicide attempts (Carbonaro et al., 2016).
The national sample is better for estimating frequency: about 41 percent of people who had used classic psychedelics reported at least one challenging experience, 8.9 percent had functional impairment for more than a day and 2.6 percent sought help (Simonsson et al., 2023). Among 608 people who reported extended difficulties, about one-third had problems for more than a year (Evans et al., 2023). The studies associate greater difficulty with no preparation, a negative mindset, no psychological support, a disagreeable environment, a dose that was too large and a recent major life event, and they associate guided settings with a narrower range of later difficulties. Challenging experiences are therefore worth taking seriously, and the research describes conditions under which they are more or less likely.
Minimum 20 characters required.
Question 1: Which list contains only factors of the Challenging Experience Questionnaire?
Question 2: In the nationally representative United States sample studied by Simonsson and colleagues (2023), what share of the 613 people who had used a classic psychedelic reported functional impairment lasting longer than a day after a challenging experience?
Question 3: In the worst-experience survey of Carbonaro and colleagues (2016), how were the difficulty and the duration of the challenging experience related to enduring increases in well-being?
Question 4: Why can the survey of 608 people by Evans and colleagues (2023) not estimate how common extended difficulties are among people who use psychedelics?
Final Assessment
Bringing It All Together
This lesson examined the psychedelic experience as a possible part of the treatment. Section 1 described the features researchers measure, the mystical-type experience, emotional breakthrough, psychological insight and psychological flexibility, and showed that they predict later benefit in small studies, while anxious, dreadful experiences predict less. Section 2 traced set and setting from Al Hubbard, the Saskatchewan researchers and Hollywood Hospital to the Johns Hopkins safety guidelines and the 2025 ReSPCT reporting guidelines, and reviewed evidence that dose, absorption, intentions, music and the therapeutic alliance shape the experience.
Section 3 explained why the same recognizable experience makes psychedelic trials hard to blind. Most participants in high-dose trials identify their allocation, and the share of correct guesses and the Bang blinding index make that visible for each arm. Active placebos, broadened instructions, independent raters and expectancy measures reduce some bias, and the question of whether the experience is necessary remains open. Section 4 turned to challenging experiences: how they are measured with the Challenging Experience Questionnaire, why estimates of their frequency differ with the sample, how people make meaning of them, and what extended difficulties a minority report.
Taken together, the sections support the course's view that psychedelic-assisted therapy is a process in which the experience, the setting and the relationship all matter, and that claims about it should be read with attention to design, blinding and the people who were studied.
Key Takeaways from this lesson
- Mystical-type experiences, emotional breakthrough and psychological insight, measured with the MEQ30, EBI, PIQ and PIS, are the most studied features of the acute experience, and the experiential model treats them as part of the treatment. Psychological flexibility is a plausible process linking the acute experience to lasting change, supported by a retrospective survey and a small fixed-order trial.
- These features predict later benefit in most studies that have tested them, but the studies are mostly small, rely on self-report and cannot show that the experience causes improvement. Correlation, prediction and mediation are different claims, and a convincing mediation claim needs randomization, correct time order and no unmeasured factor driving both mediator and outcome.
- Set and setting were developed in practice by Hubbard, the Saskatchewan group and Hollywood Hospital in the 1950s, named by Timothy Leary in 1961 and codified in the safety guidelines that modern trials follow.
- Dose is the strongest predictor of the acute response, and absorption, intentions, mood before the session, music, a non-clinical environment and the therapeutic alliance also shape the experience. Uneven reporting of these features limits comparison between trials, which the 2025 ReSPCT guidelines address.
- Functional unblinding occurs when participants or staff recognize a drug's effects, and in trials of high-dose psilocybin and MDMA most participants in both arms identify their allocation.
- The share of correct guesses and the Bang blinding index should be reported for each arm, because combining the arms can hide a strong pattern of guessing. Active placebos, instructions that widen expectations, blinded central raters and expectancy measures reduce some sources of bias, and none has solved the problem for high-dose classic psychedelics.
- Whether the subjective experience is necessary for benefit is unresolved, with correlational and small experimental studies on one side and mostly animal evidence on the other.
- Challenging experiences are common outside clinics, and about 30 percent of supported research participants given a high dose of psilocybin have marked anxiety or fear; many people find meaning in them, and a minority report difficulties lasting months or years.
- Designing a trial draws the lesson together: it needs measures of the experience and of psychological change, a standardized and reported setting, a comparator with blinding checked in each arm, and preparation and follow-up for challenging experiences.
Core Concepts Reviewed
Section 1: mystical-type experience, complete mystical experience, emotional breakthrough, psychological insight, psychological flexibility, oceanic boundlessness, dread of ego dissolution, and the distinction between correlation, prediction and mediation.
Section 2: set, setting and its physical, social and cultural layers, extra-pharmacological factors, absorption, the therapeutic alliance, the history from Hollywood Hospital to modern guidelines, and the ReSPCT reporting guidelines.
Section 3: placebo response, expectancy, blinding, functional unblinding, active placebos, the share of correct guesses, the Bang blinding index, the self-blinding microdosing study and the debate over whether the experience is necessary.
Section 4: challenging experiences and the seven factors of the Challenging Experience Questionnaire, frequency estimates in supervised, population and self-selected samples, narrative work and meaning-making, and extended difficulties.
The final reflection asks you to bring the four sections together by answering the questions a person might ask before joining a trial of psilocybin-assisted therapy.
Reflection
On 4 August 2020, Canada's federal Minister of Health granted section 56 exemptions to four people with incurable cancer to use psilocybin for end-of-life distress. Imagine a hypothetical person with incurable cancer who is considering joining a randomized trial of psilocybin-assisted therapy for anxiety and depression. Before consenting, the person asks four questions: (1) "Do I need to have a mystical experience for this to work?" (2) "Why do the room, the music and the team matter so much?" (3) "If I get the placebo, will I be able to tell?" (4) "What if the session goes badly?" Write the answers a well-informed clinician might give, in plain language, drawing on all four sections of the lesson and stating honestly what the evidence does and does not show.
(1) "Studies find that people who have strong mystical-type experiences, emotional breakthroughs or new insights during the session tend to improve more, but these studies are small and cannot prove that the experience itself does the healing. Some people improve without a mystical experience, and researchers still disagree about whether the experience is necessary."
(2) "Your mindset and the setting shape what you experience. Research links a calm, comfortable room, carefully chosen music and a strong relationship with your therapists to better sessions and outcomes, and a clinical, unfamiliar environment to more anxiety. That is why we will meet several times before the session and why two trained people will be with you throughout."
(3) "Possibly. In trials with an inactive placebo, most people correctly guess which group they were in, and trials now ask about guesses, use comparison treatments with some effects, and have independent raters who do not know your group, so that expectations can be accounted for."
(4) "About three in ten carefully prepared research participants given a high dose of psilocybin have a period of marked fear or anxiety. The team is trained to support you through it, and these episodes have usually passed during the session. In trials with patients, a small percentage of participants have had serious problems, such as worsening depression, and a minority of people who use psychedelics outside research report problems lasting months, so we will follow up with you closely and you can stop taking part at any time."
Minimum 30 characters required.
Final Knowledge Assessment
Question 1: In a psilocybin trial, MEQ30 scores correlate with improvement within the psilocybin arm. Which explanation connects Section 1 with Section 3 as an alternative to the experience causing improvement?
Question 2: Murphy and colleagues (2022) found that stronger therapeutic alliance predicted greater emotional breakthrough and mystical-type experience, which predicted lower depression. What does this suggest about the social setting?
Question 3: In a hypothetical drug arm of 50 participants, 40 guessed correctly, 8 guessed incorrectly and 2 were unsure. What is the Bang blinding index, and what does it suggest?
Question 4: Why can the combined share of correct guesses across both arms be misleading?
Question 5: What does the McGill placebo study (Olson et al., 2020) imply for randomized trials of psychedelics?
Question 6: Simonsson and colleagues (2023) found that no preparation, a negative mindset, no psychological support, a disagreeable environment and a dose that was too large were associated with more difficult experiences. Which concept from Section 2 do these findings illustrate?
Question 7: Section 1 reported that dread of ego dissolution predicted less improvement, while Section 4 reported that greater difficulty was associated with enduring increases in well-being. Which interpretation reconciles these findings best?
Question 8: Which of the following is a recommendation in the FDA's July 2026 guidance on clinical investigations of psychedelic drugs?
Question 9: In the trial by Lii and colleagues (2023), ketamine or saline was infused during general anaesthesia for surgery. What did the trial find, and why does it matter for the debate in Section 3 (7.1 Is the Subjective Experience Necessary?)?
Question 10: Which description of Hollywood Hospital in New Westminster (1957 to 1968) is supported by the sources in this lesson?
Question 11: Which source in Section 4 gives the best estimate of how common challenging experiences are among people who use classic psychedelics?
Question 12: A hypothetical researcher wants to measure the insight a participant reports in the weeks after a psilocybin session, including behaviour change. Which measure fits best?
Question 13: Bogenschutz and colleagues (2022) used diphenhydramine as the comparator in a psilocybin trial for alcohol use disorder. What does the trial's blinding result show about active placebos?
Question 14: Evans and colleagues (2023) found that taking a psychedelic in a guided setting predicted a narrower range of extended difficulties. How does this finding relate to Section 2?
Question 15: Which finding would most directly support Olson's (2021) position that the subjective experience may not be necessary for benefit?
Glossary: Key Terms, People & Frameworks
📚 Reference page, available throughout the lesson
This glossary defines the terms, measures and people introduced in this lesson on the psychedelic experience, set and setting, expectancy and blinding, and challenging experiences.