How psilocybin works, and where the explanation runs out
The receptor pharmacology is settled. The imaging findings are small and unreplicated. The step from either of those to a person being better six months later is the part nobody has demonstrated, and it is usually the part told most confidently.
The short answer
Psilocybin is a prodrug, converted in the body to psilocin, which acts primarily at the serotonin 5-HT2A receptor. Imaging studies report changes in brain connectivity and structure that outlast the acute drug effects.[1] The mechanism by which any of this produces a durable clinical benefit is not established. What is known is where the molecule binds. What is not known is why that would leave someone less depressed a month later — and the confident neuroplasticity narrative in circulation is an interpretation, not a finding.
- Primary receptor target
- 5-HT2A
- Active metabolite
- Psilocin
- Imaging study size
- n=28
- Mechanism of benefit
- Not established
Three claims, three confidence levels
What is actually established
These are usually run together into one sentence. They are not equally certain, and separating them is most of what this page does.
Psilocybin is a prodrug converted to psilocin
Psilocybin itself is largely inactive. It is dephosphorylated to psilocin, which is the compound that acts on the brain. This is basic, uncontested pharmacology.
Settled
Psilocin acts primarily at the serotonin 5-HT2A receptor
The subjective effects are attributable to agonism at 5-HT2A; blocking that receptor blocks the experience, which is also why antipsychotics blunt it. Primarily does not mean exclusively — psilocin has affinity at other serotonin receptors, and their contribution is less well characterized.
Settled
Something about the brain changes for longer than the drug is present
Imaging studies report changes in connectivity and structure that outlast the acute effects. The observation is real. Its size, its durability, its clinical meaning and whether it replicates are all open.
Exploratory
A fourth claim — that these changes are what produce clinical improvement — is routinely presented alongside the first three and belongs in a different category entirely. Nobody has shown it.
Pharmacokinetics
How long it stays in you
Pooled from the human studies in a 2025 systematic review. Ranges rather than single values, because that is what the evidence supports.
- Time to peak psilocin, oral
- 1.8–4 hours
- Elimination half-life, oral
- 1.5–4 hours
- Psilocin bioavailability
- ≈53%
- Duration of subjective effects
- 4–6 hours
Wide, and it is why supervised sessions allow a long onset window rather than a fixed one.
Intravenously it is about 74 minutes, which tells you the spread above is mostly absorption rather than clearance.
52.7 ± 20% in the single study reporting it. The uncertainty on that figure is as important as the figure.
Longer than the plasma half-life implies, which is ordinary for a drug acting on a receptor rather than in proportion to blood level.
Source: Meshkat S, et al. Pharmacokinetics of Psilocybin: A Systematic Review. Pharmaceutics. 2025;17(4):411. Open access under CC BY. We publish the ranges it reports rather than picking a value from inside them.
The ledger
The imaging study everyone is citing
Reported the way every trial on this site is reported: design, sample size, comparator and limitation next to the result rather than after it.
| Trial | Design | Participants | Primary endpoint | Result and limitation |
|---|---|---|---|---|
| Brain changes after first psilocybin use2026 · Healthy volunteersSource ↗ | Placebo-controlled imaging study, fixed order ControlledSingle-blind Compared against: Placebo | 28 | No primary endpoint | Measurable changes in brain structure persisting beyond the acute drug effect. Read it with this: Twenty-eight healthy volunteers, exploratory, not pre-registered, and fixed-order rather than randomized. The authors explicitly call for replication. |
This single study is doing an enormous amount of work in public discussion of how psilocybin works, so it is worth being precise about what it can carry.[1]
- Twenty-eight participants. Small enough that individual variation dominates, and small enough that a finding of this kind is expected to shrink on replication even if it is real.
- Healthy volunteers. Not patients. Whatever was observed was observed in brains that were not depressed to begin with, so it cannot describe the mechanism of a clinical improvement that did not occur.
- Exploratory and not pre-registered. Without a pre-specified analysis plan there is no protection against the analysis being shaped by the data, which is the standard route from noise to a publishable pattern.
- Fixed order rather than randomized. Every participant received the conditions in the same sequence, so order effects and drug effects are not separable.
- The authors explicitly call for replication. They are more cautious about this result than most of the coverage of it. That is worth noticing.
None of that makes the study bad. Exploratory imaging work in healthy volunteers is how a field generates hypotheses, and this is a reasonable example of it. It is simply not the thing it is being used as, which is proof that psilocybin rewires the brain in a way that treats illness.
Why mechanistic plausibility is not evidence of benefit
This is the part of the page worth reading twice, because it is the error that makes every other error on this subject possible.
A mechanism is a story about why something should work. A clinical result is a measurement of whether it did. These are different kinds of claim and they are established by different kinds of study — the first by pharmacology and imaging, the second by randomized controlled trials measuring outcomes in patients. A compelling story about neuroplasticity is not a clinical result and cannot be upgraded into one by being told well.
Psychiatry in particular should have learned this. The field has an extensive history of mechanisms that were broadly correct and treatments built on them that failed to outperform placebo, and an equally long history of treatments that worked while the accepted explanation for why turned out to be wrong. The relationship between “we know what this molecule does” and “this helps patients” has been weak enough, often enough, that plausibility should count for very little on its own.
The practical test is simple. When a mechanism story is doing the persuading, ask what the controlled outcome data say. For psilocybin the answer is: modest and real in depression, strongest in alcohol use disorder, and entirely absent in PTSD — a condition it is marketed for on exactly the strength of the mechanism story. That is the failure mode in its purest form.
Does the experience itself have to happen?
Whether the “mystical-type experience” is necessary for benefit is one of the genuinely open questions in this field, and it is unresolved rather than quietly settled. It is worth setting out because the answer has large consequences: if the subjective experience is doing the work, psilocybin therapy will always require a supervised day and cannot be reduced to a pill. If it is not, a compound with the same receptor action and none of the experience would be cheaper, safer and far easier to deliver.
The case that it matters. Scores on mystical-type experience measures have repeatedly been reported to correlate with clinical outcomes. That is a real observation and it is why most of the field designs around the experience rather than trying to strip it out.
The case that this is not enough. A correlation within a treated group cannot establish that the experience causes the benefit. People who respond well to a treatment may rate the experience more highly because they responded, and both may follow from a third thing — expectancy, dose exposure, or how the session went in ways nobody measured. This is the same blinding problem that runs through the whole literature, arriving in a different place.
One controlled data point. Two weeks of escitalopram pretreatment reduced anxiety and bad drug effects during a psilocybin session while leaving positive mood and mystical-type experience unchanged.[2] That shows the components of the experience can be pulled apart pharmacologically — that they are not one indivisible thing. It does not tell you which component, if any, carries the clinical benefit, because that study was not designed to measure clinical outcomes.
We are not going to resolve this. Anyone who tells you it is settled in either direction is ahead of the evidence.
Duration and pharmacology, at the level we will state it
Psilocin’s acute effects last long enough that every legal program in the United States is built around a supervised session of roughly six to eight hours with a facilitator present throughout.[3] That is the operationally relevant fact, and it is the level of detail this page will go to.
We do not publish doses, dose-response figures, or duration-by-dose tables. Those are decisions for a prescriber or a licensed facilitator who knows an individual history, and a page that cannot know that history has no business supplying the number. The rule and the reasoning are on our editorial standards page. What we will cover is on the safety and risks page, including the interactions that make the pharmacology matter in practice.
Common questions
Questions about mechanism
How does psilocybin work in the brain?
Does psilocybin permanently change your brain?
Is the mystical experience necessary for psilocybin to work?
How long does psilocybin last?
Does a plausible mechanism mean psilocybin works?
References
- 1.Lyons T, Spriggs M, et al. Human brain changes after first psilocybin use. Nat Commun. 2026;17:3977.n=28, placebo-controlled, fixed order, exploratory and not pre-registered. Suggestive of durable brain changes; explicitly requires replication.
- 2.Becker AM, et al. Acute Effects of Psilocybin After Escitalopram Pretreatment. Clin Pharmacol Ther. 2022;111(4):886–95.Two weeks of escitalopram reduced anxiety and bad drug effects but left positive mood and mystical-type experience unchanged. No serotonin syndrome.
- 3.Oregon Health Authority. Oregon Psilocybin Services.Program rules for Oregon's regulated psilocybin services, including the supervised administration session.