Stop Treating the Trip as a Side Effect

Twenty-seven out of twenty-eight people said it was the strangest state of consciousness they had ever been in. That is not the interesting part. The interesting part is what happened the next morning, when they were sober, sitting in a room, filling out a questionnaire about whether anything about their own life had become clearer.

That answer — the next-day one, the boring clipboard one — turned out to be the number that predicted whether they felt better a month later.

The trip is not the price you pay for the medicine. In the best human data we have this year, the trip is where the medicine happens.

Here is the study. Researchers at UCSF and Imperial College London ran 28 healthy adults who had never used a psychedelic through a placebo-controlled design: first a 1 mg dose of psilocybin that does essentially nothing, then a month later a 25 mg dose that does quite a lot. They scanned people before, during, and a month after.

What they actually measured

Three things, and all three have plain-English versions.

Brain entropy. Forget the physics-class definition. Entropy here just means how varied and unpredictable your brain’s activity is minute to minute — how many different patterns it cycles through instead of running the same groove over and over. Depression, rumination, and craving all look like grooves. Within 60 minutes of the 25 mg dose, EEG showed entropy climbing.

Neural tract density and integrity. The white matter of your brain is cabling — bundles of fiber connecting one region to another. Diffusion imaging tracks how tightly bundled and well-organized that cabling is by watching how water moves through it. One month after a single dose, the cabling looked denser and more orderly. The researchers noted the direction of change was the opposite of what normal aging does to those same tracts.

Psychological insight. Not a mystical term. It is the ordinary experience of suddenly seeing something about your own life you had been working hard not to see — why you keep leaving, why you keep going back, what the drinking was actually for. The study measured it the day after dosing, once the drug was gone.

Then they looked at the order of events. Bigger entropy spike during the session predicted more insight the next day. More insight the next day predicted better well-being at two weeks and at four. Cognitive flexibility — how easily you can drop a rule that stopped working and pick up a new one — was better at one month.

“Psychedelic means ‘psyche-revealing,’ or making the psyche visible,” senior author Robin Carhart-Harris said of the findings. “It suggests that the trip — and its correlates in the brain — is a key component of how psychedelic therapy works.”

Carhart-Harris has been arguing a version of this since 2014, and in June he published an updated defense of the entropic brain hypothesis in Brain, extending it past psychedelics to meditation, REM sleep, near-death states, and early unmedicated psychosis. The through-line: the breadth of what you experience tracks the variability of what your brain is doing, and under psilocybin in a supportive context, that variability predicts what your mental health looks like afterward.

The honest caveats, up front

Twenty-eight people. Healthy people — nobody in this study had a diagnosed mental health condition, which means nobody had a depression score or a craving score to improve. Every participant knew when they had the real dose, because of course they did; that is the functional unblinding problem that has haunted this whole field. The one-month follow-up is a month, not a year. And the entropy-to-insight-to-well-being chain is a statistical relationship, not a proven causal machine.

The animal work is where the causal detail is getting filled in. At Cornell, Alex Kwan’s lab has shown a single psilocybin dose increasing neuronal connections in mice by roughly 10%, weakening negative feedback loops while strengthening sensory input. Mice do not report insight. Humans do not sit still for two-photon microscopy. The two halves of the answer are being assembled in different species, and anyone telling you the picture is finished is selling something.

Why this collides with the money

There is a large, well-funded bet running in the opposite direction: that you can keep the plasticity and throw away the experience. UC Davis chemists reported in May on a compound called D5 that fully activates the 5-HT2A serotonin receptor — the same target classic psychedelics hit — without producing hallucination-like behavior in mice. A pill you take at home, no eight-hour session, no two clinicians in the room, no scheduling nightmare. The commercial logic is obvious and, honestly, the access logic is decent too: people with a personal or family history of psychosis are screened out of nearly every psychedelic trial running today.

But if the insight is doing the work, a trip-free analog is not a cheaper version of the same drug. It is a different drug that happens to share a receptor. That is a testable claim, and nobody has tested it head-to-head in humans.

That is a testable claim, and nobody has tested it head-to-head in humans.

The regulator just walked into the middle of this

On July 14 the FDA finalized its clinical-trial guidance for psychedelics — and conspicuously did not settle the question everyone wanted settled. More than 200 commenters had pushed the agency on whether drug administration can be decoupled from psychotherapy support, and the final document largely sidestepped it, while still specifying that a licensed provider with graduate-level training and clinical experience in psychotherapy should lead session monitoring.

The same day, FDA scheduled a public hearing for September 14 on the therapeutic use of psychedelics in “supervised and supportive settings,” with four topics: provider training and credentialing, patient safety, access, and data standardization. Written comments close October 5. Read that list again — it is entirely about the room, the person in it, and what they are qualified to do. The agency is being asked to regulate set and setting without a framework for measuring it.

This argument is sixty years old

Between 1960 and 1963, a graduate student named Ralph Metzner ran psilocybin sessions at Harvard with Timothy Leary and Richard Alpert — the Harvard Psilocybin Project, whose records now sit in an archive at Purdue. What Metzner contributed was not a molecule. It was the formulation that the content of a psychedelic experience depends on the person’s inner state and the environment around them: set and setting. The project collapsed into scandal, Leary and Alpert were pushed out, and by 1970 the Controlled Substances Act put psilocybin in Schedule I, where it stayed for half a century. The set-and-setting idea survived mostly underground, in living rooms and unlicensed practices, dismissed by serious psychiatry as counterculture residue rather than pharmacology.

Carhart-Harris’s chair at UCSF is named for Metzner. The 2026 imaging data is, functionally, Metzner’s hypothesis with a scanner attached.

If you have been there

You may have done mushrooms at nineteen in a field, or at thirty-four in a basement, or with a guide who charged you $3,000 and called it ceremony. And you may have had the experience of something being genuinely revealed and then nothing changing, because you went home to the same apartment and the same phone and the same people. That gap is not you failing to integrate. That gap is what happens when a mechanism that runs on insight gets delivered without anything built to catch the insight.

What the data says is that the window is real and short. Entropy spikes in an hour. Insight registers the next day. Well-being moves over the following weeks. Every one of those handoffs is a place where a system either shows up or doesn’t.

The field spent fifty years arguing that the experience was the embarrassing part — the thing to be minimized, standardized, or engineered out so the compound could be respectable. The current best guess is that the experience was the active ingredient the whole time, and the pharmacology was the delivery mechanism. If that holds, the hardest problem in psychedelic medicine was never chemistry. It is whether anyone is prepared to pay for a person to sit in a room for eight hours and then call you back on Tuesday.