UC Santa Cruz researchers put the synapse growth and the head-twitching on separate circuits, and found the structural effect was far larger in mice that had been chronically stressed.

The clinical interest in psilocybin rests on an awkward package. The compound appears to produce lasting changes in brain structure, which is the part researchers want, and it produces hours of profound perceptual disturbance, which is the part that makes administering it expensive, logistically difficult and unsuitable for many patients.

Whether those two things can be separated is the central question in the field. A study in Nature Neuroscience from the University of California, Santa Cruz reports evidence in mice that they can.

Yi Zuo, a professor of molecular, cell and developmental biology at UC Santa Cruz and the study’s senior author, says the work shows “that therapeutic circuit remodeling and hallucinogenic effects operate through distinct cellular pathways” — which would change what a drug designer is aiming at.

Watching single synapses for weeks

The measurements were made on dendritic spines — the small protrusions on a neuron’s branches where synapses form. Using two-photon microscopy, the team tracked individual spines in living mice over several weeks, which is what makes the result more than a snapshot.

Psilocybin rapidly increased the formation of new spines, with the increase visible within 24 hours. It also accelerated their maturation, so a larger share of the new spines became the large, stable forms that persist rather than the short-lived immature ones that mostly disappear. And it promoted the elimination of some pre-existing connections.

That third effect is the one that revises the standard account. “These findings challenge the prevailing view that psilocybin simply increases neural connectivity,” Zuo said.

Pruning alongside growth is what remodelling looks like as opposed to mere addition, and it is a more plausible substrate for a therapeutic change. A brain that has added connections everywhere has not reorganised anything; one that has added some and removed others has.

Stressed mice responded differently

In healthy control mice the boost in new connections was temporary, returning toward baseline over three weeks.

In mice exposed to chronic unpredictable mild stress — a standard laboratory model used to produce depression-like behaviour — psilocybin produced a significantly larger surge in new spines, and halted the excessive loss of existing connections that the stress had been causing.

The asymmetry is the most clinically suggestive finding in the paper. A drug that does little in a healthy brain and a great deal in a stressed one is behaving the way a treatment should, and it offers an explanation for why psychedelics show effects in depression trials that are hard to reproduce in healthy volunteers. Whether the mechanism is restoration of something stress removed, or something else that merely coincides, the study does not establish.

Separating the two effects

The dissociation was demonstrated by manipulating one receptor in one cell population. The effects were traced to serotonin 2A receptors in Layer 5 pyramidal neurons — the large output cells of the cortex.

The experiment then ran in both directions. Restoring those receptors only in those neurons restored synapse formation and stabilisation, while the mice showed no head-twitch response — the involuntary rapid head movement used as a behavioural proxy for hallucinogenic activity in rodents. Deleting the receptors only from those neurons eliminated psilocybin-induced synapse growth, while the head-twitch behaviour remained intact.

Those two results are the argument. One shows the structural effect occurring without the hallucinogenic proxy; the other shows the proxy occurring without the structural effect. Each alone would be suggestive; together they establish that the pathways are separable, at least in this animal and by this measure.

The caveat is in that last clause. The head-twitch response is a proxy and an imperfect one: it correlates well with hallucinogenic potency across compounds, which is why it is used, but a mouse twitching its head is not a mouse reporting an altered state. Whether a compound engineered on this principle would produce the structural changes in a human without the subjective experience is exactly what the proxy cannot tell you.

Mice, and what that limits

This is a mouse study throughout, and the usual distance applies. Dendritic spine dynamics in mouse cortex are a reasonable model for synaptic plasticity in general; depression-like behaviour induced by unpredictable stress is a much looser model for human depression, and the field has a long record of interventions that corrected the mouse phenotype and failed in people.

There is also a question the study design raises about the therapeutic logic. If the subjective experience turns out to be part of how psilocybin works clinically — as several trial designs assume, pairing the drug with psychotherapy during the session — then separating the structural effect from the perceptual one would remove something load-bearing rather than something unwanted. The mouse data cannot speak to that at all.

Jacob Baker, a PhD student in Zuo’s lab and the study’s lead author, frames the appeal of the compounds in terms that acknowledge it. “Psychedelics are fascinating compounds because they have the potential to profoundly transform who we are,” he said.

Zuo’s own assessment is about mechanism rather than application: “I consider this work a major advance, particularly in our understanding of how psilocybin reshapes neural circuits,” she said.

Psilocybin remains a controlled substance in the United States, where the Controlled Substances Act was passed in 1970, though the Food and Drug Administration has granted Breakthrough Therapy designations for psilocybin and MDMA and held a public hearing on the subject in September.

The study’s co-authors are Emily Kogan, Shaorong Ma and Ju Lu, all molecular and cell biologists at UC Santa Cruz. The work was funded by the National Institute of Mental Health and the National Institute on Aging, with psilocybin supplied through the National Institute on Drug Abuse Drug Supply Program.

---

Notes

1. Baker, J., et al. Nature Neuroscience, 2026. DOI: 10.1038/s41593-026-02490-y.
2. University of California – Santa Cruz. “How psilocybin reshapes brain circuits.” EurekAlert!, 22 October 2026. https://www.eurekalert.org/news-releases/1148940.

Photo by Pixabay on Pexels.com