The Good Friday Experiment: what Walter Pahnke’s 1962 Marsh Chapel study measured, and what it left out

The Good Friday Experiment asked whether psilocybin, taken by religious people in a religious setting, could produce something like the experiences described by mystics. It ran on Good Friday 1962 in a basement chapel at Boston University, while the main service went on upstairs. It is still cited as one of the most influential psychedelic studies, and as a warning about how such studies get reported (Doblin, 1991).

This guide covers what the record shows: the design, what was measured, what the results were, what a follow-up decades later added, and how the study shaped later research. Our separate guides cover how researchers score mystical experiences today and where the idea of set and setting came from.

What was the Good Friday Experiment?

It was the doctoral research of Walter Pahnke, who was both a physician and a minister. He ran it for his Ph.D. in Religion and Society at Harvard University, with Timothy Leary as his principal academic adviser (Doblin, 1991). A 2025 commentary dates it to 20 April 1962, Good Friday that year, and names the participants’ school as Andover Newton Theological School (Simonsson and Goldberg, 2025).

The setting was chosen on purpose. Pahnke believed the best conditions would be a group of believers taking part in a familiar ceremony meant to stir religious feeling. He secured permission from Rev. Howard Thurman, Boston University’s chaplain, to use Marsh Chapel. Participants sat in a smaller basement chapel with an altar, pews and stained glass, and the main service above was broadcast to them (Doblin, 1991; Seale-Feldman, 2026).

The study grew out of the Harvard Psilocybin Project, the wider Leary-era research program that has its own article in our journal and is not covered here.

How was the Good Friday Experiment designed?

On paper, it was a strong design for its time. Doblin describes it as a randomized, controlled, matched-group, double-blind experiment using an active placebo. The volunteers were 20 white male Protestant students from one Boston-area theological school, none of whom had taken psilocybin or anything similar before. They were tested in advance and sorted into 10 matched pairs by religious background, past religious experience and psychological makeup. On the morning itself, a helper who knew none of them flipped a coin to decide which member of each pair got psilocybin (Doblin, 1991).

The control capsule held nicotinic acid, a B vitamin that causes a warm flush and tingling. Pahnke chose it so that the controls, who knew psilocybin had physical effects, would feel something too; in his thesis he wrote that it was meant to “potentiate suggestion” in them (Doblin, 1991). This idea of an active placebo, a control that produces noticeable effects, is still debated in psychedelic trials today (Aday et al., 2022).

Support was built in. Ten research assistants acted as group leaders, two to each group of four students, and the groups met for two hours beforehand to build trust. Leaders were told not to describe mystical experiences, to avoid steering the students toward the scale, and the participants Doblin interviewed later agreed they had not been coached. Over Pahnke’s objection, and at Leary’s and one leader’s insistence, the leaders also took a capsule, one in each pair receiving psilocybin; their reactions were never recorded (Doblin, 1991).

What did Pahnke measure?

Pahnke built his own measure, because no questionnaire for mystical experience existed then. Drawing on scholars of mysticism such as William James and W. T. Stace, he defined eight categories: a sense of unity, transcendence of time and space, a sense of sacredness, a sense of objective reality, deeply felt positive mood, ineffability (feeling beyond words), paradoxicality and transiency (Doblin, 1991). Unity was scored in two parts, inner and outer, which is why some later accounts count nine categories (Barrett and Griffiths, 2018).

The scores came from three directions. Blind raters scored the written accounts and the transcripts of three interviews, held right after the service, a few days later and six months later. The students also filled in a 147-item questionnaire a day or two after the service and a 100-item questionnaire at six months. The six-month form also asked about lasting changes in attitudes and behavior, such as joy, reverence, vocational commitment and relationships (Doblin, 1991).

Pahnke set his own bar for a “complete” mystical experience: at least 60 to 70 percent of the maximum score, both overall and in each category. Doblin calls that cut-off arbitrary (Doblin, 1991).

What did the 1962 results show?

By Pahnke’s cut-off, four of the ten students given psilocybin had a complete mystical experience and none of the controls did. The psilocybin group’s average was higher in every category and every question, and Doblin reports the differences as statistically significant across all categories (Doblin, 1991; Mosurinjohn et al., 2023). At six months, the psilocybin group also reported more lasting positive changes than the controls.

The controls did not feel nothing. Their highest scores were in sacredness, positive mood and a sense of reality, the categories Doblin notes a moving service is most likely to produce. Pahnke described one control who rated the day as very meaningful but said he had had stronger experiences before (Doblin, 1991).

What did Rick Doblin’s 1991 follow-up find?

Pahnke died in 1971, before he could follow up further, and the raw data, including the list linking names to groups, had been lost. Doblin spent four years tracing the participants and found 19 of the 20. Between November 1986 and October 1989 he interviewed 16 of them, seven from the psilocybin group and nine controls, and had them fill in Pahnke’s six-month questionnaire again. Of the three psilocybin participants missing, one had died, one could not be identified and one declined (Doblin, 1991).

The scores barely moved over 24 to 27 years. The psilocybin group’s average across the mystical categories went from 60.8% at six months to 66.8%, and the controls’ from 11.8% to 12.2%, with the gap still significant in every category (Doblin, 1991). Doblin warns that the long-term average may overstate the psilocybin group somewhat, because one of the missing participants was known to have scored lower.

In interviews, the psilocybin participants all described their Good Friday as having genuinely mystical elements and as a high point of their spiritual lives. They linked it to a deeper sense of joy, more tolerance of other religions, steadier responses to crises and, for some, involvement in the civil rights movement. Most controls could barely remember the service. Two controls later sought out a psychedelic experience of their own, while others said what they saw that day put them off for good (Doblin, 1991).

The interviews also brought out what the published reports had played down. Of the seven psilocybin participants Doblin interviewed, only two described a wholly positive day. The rest recalled moments when they feared they were going mad, dying, or too weak for what was happening; one felt trapped in the chapel and spent much of the service outside. They said these struggles resolved and taught them something, but Doblin concludes they were significantly underemphasized in Pahnke’s thesis and later papers (Doblin, 1991).

The adverse event that went unreported

The most serious omission concerned one participant. Pahnke’s thesis mentions that injectable Thorazine (chlorpromazine, a tranquilizer) was on hand for emergencies, and in one technical section that a participant later called his experience “a psychotic episode”. What it never reports, Doblin found, is that a participant was actually given a Thorazine injection during the experiment. Several participants and group leaders remembered it independently (Doblin, 1991).

Doblin’s account is second-hand, because that participant was the one psilocybin subject who declined to be interviewed. According to others, he was deeply moved by the sermon and went outside; there was a struggle when group leaders, worried for his safety, tried to bring him back, and Pahnke then sedated him. At six months he rated his fear experience as “slightly harmful”. A 2025 commentary adds that he reportedly believed he had been chosen to announce the dawn of a messianic age (Doblin, 1991; Simonsson and Goldberg, 2025).

Doblin suggests Pahnke stayed silent so as not to hand ammunition to critics of the research. His verdict is short: perhaps good politics, but bad science, and with no justification. The episode now appears in the safety guidelines that Johns Hopkins researchers published for hallucinogen studies, as an example of why researchers must stop disoriented volunteers from leaving the study site. That paper describes the man as believing God had chosen him to announce an age of peace (Johnson et al., 2008).

Why the blind broke, and other criticisms

The active placebo worked only briefly. Nicotinic acid acts faster than psilocybin, and early in the service several controls believed they had received the drug. As the psilocybin took hold and the flush faded, the blind collapsed: by the end of the day every participant correctly knew which group he was in (Doblin, 1991). A 2022 review of trial methods calls this the experiment that “infamously demonstrated” how hard psychedelic trials are to mask (Aday et al., 2022).

That matters for interpretation. Once people know they got the drug, expectation and the group’s mood can add to the effect, so the study cannot separate psilocybin from suggestion. Doblin argues the damage is smaller than it looks, because Pahnke deliberately tested psilocybin together with a supportive religious setting rather than the drug alone; Barrett and Griffiths say the broken blind likely contributed to the group difference (Doblin, 1991; Barrett and Griffiths, 2018).

• The sample: 20 white Protestant men from one seminary, all already religious, run as a single group session (Doblin, 1991).
• Self-report only: no reports from family or friends, and a few questions about losing one’s sense of self could score high for reasons unrelated to mysticism (Doblin, 1991).
• The definition of mystical: theologian R. C. Zaehner objected that the scale had no specifically Christian category, and a 2023 review argues such scales carry a “specifically Christian bias” (Mosurinjohn et al., 2023).
• Group leaders on psilocybin: Pahnke himself feared this would invite charges of experimenter bias (Doblin, 1991).

Record versus retelling

Retellings have stretched the result from early on. Doblin quotes a 1966 Time magazine report that all the students who took the drug had a mystical consciousness like that of saints; Pahnke’s own count was four of ten by his cut-off (Doblin, 1991). In the other direction, Doblin shows that Zaehner misread one of Pahnke’s later papers as repudiating the study, when the passage about absent mystical experiences referred to the controls.

Details of place also drift. Some modern papers say the experiment took place at Harvard or used Harvard Divinity School students (Mosurinjohn et al., 2023). The chapel belonged to Boston University; Pahnke was the Harvard student, while the volunteers came from a Boston-area seminary, which a 2025 commentary names as Andover Newton (Simonsson and Goldberg, 2025). Marsh Chapel is nondenominational, though founded by a Methodist leader (Mosurinjohn et al., 2023).

How it shaped later mystical-experience research

The questionnaire outlived the experiment. A 2023 review calls Pahnke’s Mystical Consciousness Typology the first psychometric measure of mystical-type effects and the starting point for today’s Mystical Experience Questionnaire. William Richards, who worked with Pahnke in the 1960s, later helped develop the 43-item version at Johns Hopkins (Mosurinjohn et al., 2023). Pahnke and Richards also wrote together on experimental mysticism in 1966 (Pahnke and Richards, 1966; we read only its bibliographic record). How the later 30-item version is scored is covered in our guide to DMT and religious experience research.

The 2006 Johns Hopkins study was designed as a replication and extension that fixed known weaknesses. It tested 36 hallucinogen-naive adults individually rather than in a group, used methylphenidate, a stimulant, as a stronger comparison drug in a crossover design, added validated mysticism scales, and asked family and friends to rate changes (Barrett and Griffiths, 2018; Griffiths et al., 2006). At 14 months, 58% met the criteria for a complete mystical experience, and the authors cited Doblin’s follow-up as consistent with their findings (Griffiths et al., 2008).

The longer story of the Johns Hopkins psilocybin program is told in its own article and is not repeated here.

Two lessons from 1962 still frame the field. Masking remains a central problem: a 2022 review recommends measuring how well blinding holds and choosing active placebos with care (Aday et al., 2022). And adverse events have to be reported, which is why a 2025 commentary on psilocybin and religious leaders brings up the Thorazine episode when calling for closer attention to rare harms (Simonsson and Goldberg, 2025).

In Canada, research of this kind now goes through formal trials. Health Canada says the production, sale and possession of magic mushrooms are illegal, that no therapeutic product containing psilocybin is approved, and that clinical trials, which must be approved by a research ethics board, are the most appropriate way to advance research (Health Canada). For trials in Ontario, see psilocybin clinical trials in Toronto.

What we could not check

• Pahnke’s 1963 thesis and his later papers. Every quotation from them here comes through Doblin’s 1991 paper.
• The Thorazine episode at first hand. Doblin’s account comes from other participants and leaders, and the participant concerned was never interviewed.
• The exact date and the seminary’s name. We found both in one 2025 commentary; Doblin gives only Good Friday 1962 and “the same theological school in the Boston area”.
• What the sedated participant believed. A 2025 commentary says a messianic age and the 2008 safety guidelines say an age of peace, both citing sources we did not read; Doblin’s paper does not say.

Shroomverse sells mushroom products. This article is general information about the history of religious and psychedelic research. It is not medical, legal, spiritual or purchasing advice, it does not encourage anyone to use drugs, and it does not recommend any product.