The administration of psilocybin to healthy, hallucinogen-experienced volunteers in a mock-functional magnetic resonance imaging environment: a preliminary investigation of tolerability

Psilocybin is a tryptamine hallucinogen and pro-drug of psilocin that acts as a partial agonist at 5-HT2A receptors and is the active constituent of psilocybe mushrooms. After decades of restricted research following widespread recreational use and associated adverse events, clinical investigation of classical hallucinogens has resumed in recent years. Functional magnetic resonance imaging (fMRI) is a powerful tool for studying brain function, but psilocybin had not previously been administered in the MR environment and human hallucinogen research guidelines advise caution about exposing intoxicated participants to potentially anxiogenic settings. Carhart-Harris and colleagues therefore designed a preliminary tolerability study to determine whether intravenously administered psilocybin could be safely given to healthy, hallucinogen-experienced volunteers in a mock-fMRI environment. Intravenous delivery was selected because prior work suggested a rapid onset and brief duration of effects, features that are convenient for controlled fMRI experiments; the study aimed to establish whether doses up to 2 mg produced acceptable physiological and psychological responses under these conditions.

This was an open-label pilot tolerability study conducted at the Bristol Royal Infirmary clinical research unit. Nine healthy, hallucinogen-experienced volunteers (seven males, two females; mean age 35.8 years, SD 4.9, range 28–43) were recruited by word of mouth and provided written informed consent. Eligibility required prior use of a hallucinogen without an adverse reaction and the absence of personal or family history of psychiatric disorder; exclusion criteria included age under 27, pregnancy, substance dependence, cardiovascular disease, claustrophobia, blood/needle phobia, or a significant adverse response to hallucinogens. Screening included medical history, physical examination, ECG, routine blood tests, urine drug/pregnancy screens, psychiatric assessment, and baseline questionnaires (Beck Depression Inventory, BDI, and State Trait Anxiety Inventory, STAI). Sessions were performed in a consulting room containing a wooden mock-MR scanner (a board and wooden arch approximating the dimensions of an MR bore) to simulate the scanner environment; in the final three sessions an auditory recording of scanner noise was also played. Three participants initially received 1.5 mg psilocybin intravenously; after tolerability was established ethics permission was obtained to dose the remaining six participants at 2 mg. A commercially procured, purity-certified psilocybin vial was reconstituted and sterile-filtered to produce a 1 mg/ml solution, and either 1.5 ml or 2 ml was made up to 10 ml with saline and infused over 60 s, with the end of infusion taken as time zero. Subjective state ratings for 'happy', 'sleepy', 'relaxed', 'anxious', and 'confused' (0–10 scale) were collected pre-drug, and subjects rated overall intensity of drug effects (0–10) at 1-minute intervals for 20 minutes. Cardiovascular data (heart rate and blood pressure) were recorded for 3 minutes before and 20 minutes after the bolus. Approximately 60–90 minutes after dosing participants completed the 94-item 5-Dimensions of Altered States of Consciousness questionnaire (5D-ASC), which yields scores on oceanic boundlessness, anxious ego dissolution, visionary restructuralization (visual alterations), auditory alterations, and reduced vigilance; a global score combines several dimensions. Participants received a phone call the evening of dosing and an informal check 14 days later, when BDI and STAI were re-administered. The analysis reported included repeated measures ANOVA for time and dose effects, related t-tests for within-subject comparisons versus baseline, and independent t-tests for dose comparisons.

Nine volunteers completed the protocol; all had prior psilocybin use (mean 17.9 uses, SD 19.1, range 1–50), and baseline BDI and STAI scores were low and clinically non-significant (BDI mean 1.3, SD 1.2; STAI mean 28.4, SD 5.5). Three participants received 1.5 mg and six received 2 mg intravenously. At 1.5 mg psilocybin produced mild-to-moderate subjective effects with a rapid onset (first noticed ~60 s after infusion end), peaking at around 5 minutes and subsiding thereafter. Peak single-item 'drug effects' ratings in these three subjects were 5/10, 3/10 and 8/10. Heart rate and blood pressure increases at this dose were transient and did not exceed roughly 20 bpm or 20 mmHg. No acute or subacute adverse phenomena were reported and participants described the experience as pleasant and relatively mild. At 2 mg all six participants reported marked effects beginning at the end of the 60 s bolus, with peak intensity at about 4 minutes and effects sustained for ~20 minutes. Peak intensity ratings ranged from 5 to 8.5/10. Typical subjective phenomena included warmth, proprioceptive changes (e.g. floating), visual distortions (apparent 'breathing' of surfaces), altered time perception, and in one case synaesthesia. Transient cardiovascular responses were observed (heart rate increases around 15 bpm and systolic blood pressure increases around 20 mmHg). No participants showed signs of distress during the acute experience, and those exposed to simulated scanner noise did not react adversely. Statistical analysis of the minute-by-minute intensity ratings (repeated measures ANOVA) showed a significant effect of time (F = 15.2, d.f. = 21, p < 0.001), a significant dose-by-time interaction (F = 2, d.f. = 21, p = 0.01), and a significant main effect of dose (F = 10.3, d.f. = 1, p = 0.005). Retrospective peak-period ratings (0–10) indicated generally positive mood: 'happy' mean was 8.0 after 1.5 mg and 8.3 after 2 mg; 'relaxed' was 8.7 versus 7.3; 'anxiety' was 2.7 versus 3.8; and 'confusion' was 0 versus 1.7. None of these retrospective ratings showed significant change relative to pre-drug baseline on related t-tests at α = 0.05, although two individual participants reported transient higher anxiety (7/10 and 8/10 at peak) without subsequent distress. On the 5D-ASC the global and subscale scores tended to be lower after 1.5 mg than 2 mg, but differences were not statistically significant; the authors note that the 1.5 mg mean global score approximated previously reported low oral psilocybin doses (~8 mg oral equivalent) and the 2 mg mean global score was roughly consistent with medium oral doses (~15 mg oral equivalent) as reported in earlier work. At 14-day follow-up there were no significant changes in BDI or STAI compared with screening, no persistent adverse events were reported, and all participants indicated they expected to tolerate the drug in a real MR scanner.

Carhart-Harris and colleagues interpret their findings as evidence that intravenous psilocybin, at doses up to 2 mg administered as a 60 s bolus, is both psychologically and physiologically well tolerated in healthy, hallucinogen-experienced volunteers when given in a mock-MR environment with appropriate screening and supportive care. The rapid onset and short duration observed with intravenous delivery were viewed as advantageous for designing future fMRI studies, and the mock-scanner setting—made comfortable and aesthetically managed by the study team—appeared compatible with the acute psychedelic experience. The investigators emphasise the likely importance of 'set and setting'—that is, participant preparation and the supportive environment—in shaping tolerability, and suggest that future research might explicitly manipulate environmental factors as an independent variable. They also highlight a steep dose-response relationship for intravenous psilocybin: a relatively small increase in dose produced substantially greater subjective effects, and previous reports of adverse reactions at 3 mg indicate increasing risk with higher doses. For doses above 2 mg the authors recommend slower infusion rates to mitigate acute adverse effects. The extracted text does not present an extended formal limitations section beyond discussion of dose-related risk and the pilot nature of the work. The authors conclude that, with careful screening and subject care, controlled fMRI investigations of psilocybin in humans are feasible, while noting that dose escalation requires caution due to potential for transient but intense adverse responses.

Intravenous psilocybin administered up to 2 mg in healthy, hallucinogen-experienced volunteers was psychologically and physiologically well tolerated in a mock-fMRI setting. No significant acute or persistent adverse phenomena were reported, and participants generally rated the experience as pleasant or interesting. These results support the feasibility of conducting carefully screened, closely supervised psilocybin studies in an fMRI environment, while cautioning that higher intravenous doses carry greater risk and may require slower infusion protocols.