Psilocybin promoted the formation and maturation of neuronal connections in mice while revealing that the cellular pathway driving these structural changes can be separated from a behavioral proxy for hallucinations, according to a study published October 15, 2026, in Nature Neuroscience. Researchers at the University of California, Santa Cruz used repeated live-brain imaging and cell-specific receptor manipulations to examine how the psychedelic compound reorganizes synapses, including in mice exposed to chronic stress.
Psilocybin changed how mouse neurons formed and retained connections
Psilocybin is a naturally occurring psychedelic compound found in certain mushroom species. Its effects involve serotonin receptors, especially the serotonin 2A receptor (5-HT2AR), which is abundant in parts of the cerebral cortex. Researchers have become interested in whether changes in neuronal connectivity help explain the lasting effects reported in some controlled studies of psychedelic treatment.
The new study, led by Jacob Baker, Emily Kogan and colleagues in Yi Zuo’s laboratory, focused on dendritic spines. These small protrusions extend from neuronal dendrites and are the receiving sites of many excitatory synapses. A newly formed spine can develop, persist or disappear. Measuring these events provides a structural view of how a neuronal circuit changes, although the presence of a spine alone does not establish that a functional synapse has formed.
Using two-photon microscopy, the researchers repeatedly imaged neurons in the living mouse cortex. After psilocybin administration, they observed greater formation and maturation of dendritic spines. The team also tracked changes in the removal of pre-existing spines. This is more nuanced than simply saying that the drug increased the number of brain connections: formation, stabilization and elimination are distinct processes.
New spine formation appeared rapidly, but longer-term remodeling was more complex
The authors report that increased spine formation was detectable within 24 hours of treatment. Tracking neurons over subsequent days allowed the investigators to assess which new spines stabilized and what happened to established connections. Some newly formed spines matured into larger shapes associated with more stable synaptic structures, while some older spines were eliminated.
These observations fit a broader body of experimental work on neuronal plasticity, the ability of neural circuits to adapt their structure or function. Earlier mouse research found persistent dendritic-spine changes after psilocybin. The October 2026 work advances that question by identifying the cortical neuronal population and receptor signaling needed for structural remodeling.
Such cellular measurements are not interchangeable with improved cognition, memory or mood. They describe physical changes in experimental neural circuits. Establishing whether those changes produce durable benefits for patients requires separate clinical evidence.
Chronic stress changed the balance of spine formation and loss
The investigators also examined mice subjected to unpredictable mild stress. This experimental paradigm is used to study biological and behavioral responses to repeated stress exposure, but it does not reproduce the full complexity of human major depressive disorder.
According to the researchers’ October 22 account, psilocybin produced a different pattern in stressed animals than in unstressed controls. In mice exposed to chronic stress, the drug enhanced formation of new dendritic spines and reduced the excessive loss of existing structures. In control mice, increased formation was accompanied by the subsequent elimination of some older spines, with structural changes evolving over approximately three weeks.
The contrast matters because restoring a balanced pattern of spine turnover is different from indiscriminately increasing spine density. It also suggests that the structural effects of an intervention may depend on the starting condition of the circuit. The study did not establish that individual damaged connections were identified and repaired, or that the observed remodeling reversed a human psychiatric illness.
A specific cortical receptor population controlled plasticity without controlling head twitches
The most decisive experiment addressed whether the same neurons are responsible for structural plasticity and the drug’s hallucinogenic-like behavioral response. The researchers focused on layer 5 pyramidal neurons, a population of cortical projection neurons. They manipulated expression of 5-HT2ARs in these cells and separately measured spine changes and the head-twitch response, a stereotyped movement in mice used as an indirect proxy for psychedelic-like activity.
When receptor expression was restored selectively in layer 5 pyramidal neurons, psilocybin induced new spine formation and stabilization even though the animals did not show the characteristic head-twitch response. Conversely, removing the receptors specifically from those neurons prevented the structural plasticity response while leaving the head-twitch response intact.
This combination of gain- and loss-of-function experiments supports the paper’s central conclusion: 5-HT2AR signaling in this neuronal population is necessary and sufficient for the observed structural remodeling in the tested mouse system, but is dispensable for head twitching. It does not prove that a person can receive psilocybin’s antidepressant effects without altered perception. Head twitching is an animal proxy, not a direct measurement of subjective hallucinations or clinical efficacy.
The results inform psychiatric drug design but do not establish a new treatment
One long-standing question in psychedelic research is whether intense subjective experiences and long-term therapeutic effects must occur together. Lead researcher Baker described the motivation in the university announcement: “One of the central questions in psychedelic research is whether the psychedelic experience itself is necessary for these lasting changes in brain connectivity.”
By identifying cell populations that contribute differently to structural plasticity and head-twitch behavior, the research offers a starting point for testing future compounds designed to engage selected neuronal pathways. Those compounds would still require extensive testing of behavior, safety, dosage and clinical outcomes before any claim of therapeutic benefit.
Other recent Bites of Bio reports have examined how Piezo1 affects myelin sheath length, how researchers identified a brain circuit implicated in obsessive-compulsive disorder, and why Epstein-Barr virus reactivation may precede multiple sclerosis relapse. All address different biological aspects of nervous-system function, and none provides direct clinical validation for the new psilocybin mechanism.
What the evidence cannot yet answer
This was a mechanistic study in mice, not a randomized trial in people with depression. Its imaging sampled selected cortical neurons and dendritic branches, rather than mapping whole-brain connectivity or directly measuring synaptic communication across the entire brain. The chronic-stress model cannot capture the diversity of human depressive illness. Differences between a rodent head-twitch response and human psychedelic experience also limit how strongly the pathway separation can be generalized.
The publicly accessible journal abstract and figures support the cell-specific receptor conclusion, but the complete paper is subscription-restricted. The detailed description of the chronic-stress experiments and 24-hour time course comes from the researchers’ institutional news release; those details should be read in that context rather than presented as independently inspected full-text methods. The article also does not establish the ideal human dose, duration of clinical benefit, safety of repeated exposure or whether a non-hallucinogenic treatment using this mechanism would work.
The study was supported by grants to senior author Yi Zuo from the National Institute of Mental Health and the National Institute on Aging. Psilocybin was supplied through the National Institute on Drug Abuse Drug Supply Program. The authors declared no competing interests in the journal.
Separating neuronal remodeling from psychedelic-like behavior is the next research question
The investigators’ most important advance is not simply that new dendritic spines appeared after psilocybin. It is the identification of a cortical receptor-dependent process that could be experimentally separated from head twitching. Senior author Zuo said the findings challenge the idea that psilocybin merely increases neural connectivity and suggest that effects depend on the brain’s initial state.
Future work will need to test whether selectively engaging that process can improve meaningful behavioral outcomes and, eventually, whether the mechanism translates to safe and effective treatment in humans. For now, the findings explain part of how psilocybin changes mouse cortical circuits, not whether psilocybin use should be recommended to patients.
References
- Baker JJ, Kogan E, Ma S, Lu J, Zuo Y. 5-HT2A receptors in layer 5 pyramidal neurons mediate the neuroplastic but not the hallucinogenic effects of psilocybin. Nature Neuroscience. 2026. DOI: 10.1038/s41593-026-02490-y.
- University of California, Santa Cruz. How psilocybin reshapes brain circuits. October 22, 2026. University news release.