Psilocybin Rewires Brain Circuits in Mice, Separating Structural Repair From Hallucinogenic Response

Within 24 hours of a dose of psilocybin, the brains of mice at UC Santa Cruz had begun sprouting new connections between neurons at a rate the researchers described as substantial. The finding comes from a study published October 15 in Nature Neuroscience, and it goes beyond the headline that magic mushrooms change the brain. It describes how, where, and in which kind of brain.

The team, led in the lab of Yi Zuo, a professor of molecular, cell, and developmental biology, used two-photon microscopy to watch individual connections on neurons in living mice over several weeks. They focused on dendritic spines, the tiny projections that receive signals from neighboring nerve cells and help form synapses. Synapses underpin learning, memory, and emotional regulation, so changes in their number and strength can alter how information moves through the brain.

The drug did more than prompt new spines to appear. New spines often start small and "stubby," and many vanish before becoming real connections. After psilocybin, a much larger share matured into the mushroom-shaped form, which is generally more stable and carries more receptors for chemical signals. Lead author Jacob Baker, a Ph.D. student in Zuo's lab, said these new connections can integrate into existing circuitry and persist, leaving a lasting structural trace of the drug.

The effect was not the same in every brain. In healthy mice, psilocybin first spurred new spines, but over the next three weeks the total drifted back toward its earlier level, partly because the brain eliminated some connections that existed before treatment. The healthy brain seemed to reorganize rather than simply accumulate synapses.

Mice exposed to chronic stress responded differently. The researchers used an unpredictable mild stress model, which exposes animals to varied low-intensity stressors and produces changes resembling some features of human depression. Prolonged stress is known to cause substantial synapse loss. When these mice received psilocybin, new spine formation rose especially strongly, and the drug prevented the unusually high rate of connection loss that stress produces. Zuo said the results challenge the prevailing view that psilocybin simply increases connectivity, and suggest its effects depend on the brain's underlying condition.

The second major finding concerned hallucinations, or at least a proxy for them. The team examined serotonin 2A receptors, central to the action of classic psychedelics, in Layer 5 pyramidal neurons of the cerebral cortex. These cells carry information from the cortex to other brain regions and are rich in the receptors. Using genetically modified mice, the researchers restored the receptors only in those neurons, and psilocybin still stimulated and stabilized new synapses. Yet those mice did not show the head twitch response, the behavior researchers commonly use to indicate psychedelic activity in rodents. In the reverse experiment, removing the receptors only from Layer 5 pyramidal neurons erased the structural growth, while the head twitch response remained.

The conclusion is that receptor signaling in these neurons is necessary and sufficient for the structural changes, but not required for that behavioral marker. There is a caveat the researchers acknowledge: mice cannot describe subjective experience, so the head twitch does not directly show whether an animal is hallucinating. The evidence points to distinct cellular mechanisms, not to a proven hallucination-free effect.

The work arrives as regulators take a fresh look at psychedelics. Research slowed sharply after the Controlled Substances Act of 1970. The FDA has since granted Breakthrough Therapy designations to psilocybin and MDMA, and in September 2026 it held a public hearing, "Considerations for Potential Future Therapeutic Use of Psychedelic Drugs," covering practitioner qualifications, Risk Evaluation and Mitigation Strategy programs, and trial design.

If compounds can be found that trigger this rewiring without an intense altered state, treatments for depression, PTSD, and other conditions might need less supervision. That remains a prospect, resting on mouse data; human trials would have to show whether it holds.