What I understand about the research, and where it runs out

Psilocybin acts on the brain's serotonin system, binding especially to the 5-HT2A receptor. Neuroimaging work — much of it out of Imperial College London under Robin Carhart-Harris — has found that psilocybin quiets activity in what's called the default mode network, a set of regions involved in self-referential thought: the inner monologue, the rumination, the story you tell about who you are. At the same time, communication increases between brain regions that don't usually talk to each other much. Carhart-Harris calls this an increase in the brain's "entropy," and his REBUS model (relaxed beliefs under psychedelics) suggests the brain becomes temporarily less locked into its habitual patterns of prediction and more open to revision.

There's also early evidence, mostly from animal studies and a smaller amount of human imaging, that psilocybin promotes neuroplasticity — new dendritic growth, new connections between neurons. It's tempting to treat this as the whole explanation, the mechanistic proof that "it works." I don't think that's right. Neuroplasticity is a window of flexibility, not a guarantee of a particular outcome. A city with new roads being built isn't the same thing as a city with better traffic patterns. What gets built in that window of openness depends enormously on what happens around the person during and after the experience.

Here's where the mechanism and the room meet: none of that neuroscience explains what it's actually like to sit across from someone whose defenses have softened enough that they can finally say the thing they've never said out loud, or weep for a parent they haven't grieved yet, or feel, briefly, unified with something larger than their own suffering. The entropic brain model can describe why the door opens. It has nothing to say about what walks through it, or how it's held once it does. That's the part that isn't in the fMRI data, and it's the part I think matters most.

Small samples. Many of the headline-grabbing trials involve a few dozen participants, sometimes fewer. That's normal for early-phase clinical research, but it means the confidence intervals around these findings are wide, and a handful of dramatic responders or non-responders can swing the results considerably.

Short follow-up. A lot of the "lasting effects" reporting covers weeks or a few months. Some longer follow-ups exist — Ohio State's five-year check-in on an earlier depression trial found many participants remained in remission — but that's a small group of people who were carefully selected and closely supported, followed for one specific window. It's not evidence about what happens at ten years, or in someone whose life circumstances change.

Highly screened participants. Trial volunteers are typically screened out for psychosis risk, certain medications, unstable housing, active substance use disorders, and a long list of other factors. The people who make it into these studies do not look like a general population walking in off the street, and that matters when you try to generalize the results.

You can't really blind a psychedelic trial. If you take a strong dose of psilocybin, you generally know you didn't get the placebo. So do the people running the session, and so, often, do the raters scoring the outcome measures. That unblinding is one of the most serious methodological problems in this entire field, and it's rarely mentioned in press coverage.

Expectancy effects. People who volunteer for a psilocybin depression trial tend to arrive hopeful, often after reading about the drug's reputation for producing breakthroughs. Hope and belief can genuinely influence outcomes, which is worth taking seriously as part of the effect rather than explaining away — but it also means the drug alone, outside of a hopeful, prepared, supported context, has not really been isolated and tested.

Publication bias. Positive, dramatic results get published and covered. Null results and dropouts are underreported. That skews the visible literature toward optimism.

The medicine never travels alone. In every one of these trials, psilocybin is delivered inside hours of preparatory conversation and follow-up integration sessions with trained staff. When a study reports that "psilocybin reduced depression," it's really reporting on psilocybin plus a significant amount of psychological support — and that second part rarely makes it into the headline. I think it's dishonest to talk about the drug's effects as though the room, the relationship, and the aftercare weren't part of what was being measured.

Put plainly: this is promising early clinical evidence for some conditions, gathered under conditions that don't fully resemble real life, and I would not build a promise to you on top of it. If you want a fuller sense of what I mean by that, I've written more about what this work does not do.