Does psilocybin rewire the brain? What imaging and plasticity studies show

Headlines say psilocybin “rewires” the brain. The word is a metaphor, and the evidence behind it comes in three very different kinds: studies of neurons in dishes and animals, brain scans of people, and clinical studies that track symptoms. They answer different questions. None of them yet shows lasting rewiring in people.

This guide separates the three, rates each one, and says what is not known. It is about research. It is not about use, and it makes no treatment claims.

So, does psilocybin rewire the brain?

Not in any way that has been shown to last in people. Psilocybin changes the brain’s activity and connectivity while it is active. In animals it can change the structure of neurons. Whether a human brain is rewired for weeks, months or years, and whether that would matter, is still open.

What does “rewire” mean, and what does it not mean?

Neuroscientists use “neuroplasticity” for an induced change in brain function or structure (Carhart-Harris et al., 2026). That is a low bar. Learning something new changes the brain too, an idea that goes back to the late 19th century (Carhart-Harris et al., 2026). “Rewire” sounds like a repair job. The term measures only that something changed.

Researchers use the word loosely too. One mouse paper describes its result as synaptic rewiring (Shao et al., 2021). A systematic review says effects of this kind are proposed to rewire pathological cortical networks, then adds that it is hard to say how much benefit comes from changes in cells and how much from changes in networks (Gattuso et al., 2023). So “rewire” is a hypothesis, not a finding.

A change is not automatically a good change. A 2026 paper by researchers in the field argues that many common plasticity markers reflect learning processes that can also lock in unhelpful habits, and it cautions against leaping from a plasticity marker to better mental health (Carhart-Harris et al., 2026).

The word usually bundles four different things together:

1. Structural change: more, fewer or larger spines (small protrusions on neurons where connections form) or more branching.
2. Functional change: neurons responding or firing differently.
3. Network change: how the activity of whole regions is coordinated on a scan.
4. A lasting outcome: a change in a symptom, behaviour or test score.

The studies below measure the first three at most, mostly in different species. The fourth is a clinical question, and a scan cannot answer it.

What do animal and cell studies show about psilocybin and plasticity?

Evidence: preliminary. This is the strongest source of the “rewiring” idea, and it is also the furthest from people.

Neurons in dishes

A 2018 study treated cultured rat cortical neurons with many psychedelics. Nearly all of those tested promoted growth of neuron branches, including psilocin (see what is psilocybin), DMT, LSD and MDMA. The spine experiments in cells used DOI, DMT and LSD (Ly et al., 2018).

The live-animal part of that paper did not test psilocybin. It gave DMT to rats and LSD to fruit fly larvae and saw growth there too (Ly et al., 2018). So this work supports a general idea about psychedelics in cells and animals. It is not a test of psilocybin in a living mammal.

A living mouse brain

The study most often cited came from a Yale team led by Alex Kwan and was published in Neuron in 2021. Researchers imaged the same dendritic spines in the frontal cortex of living mice over time with a two-photon microscope. After a single dose of psilocybin, spine density and size rose by about 10%, driven by more new spines forming. The change appeared within 24 hours (Shao et al., 2021).

The details matter. The live imaging covered 12 mice, six male and six female, and only four were imaged again at 34 days. About a third of the new spines were still there then. The effect on spine formation was larger in females than in males (Shao et al., 2021). The authors also write that it is not yet clear whether the results will extrapolate to humans, partly because the serotonin receptor involved binds psychedelics differently across species (Shao et al., 2021).

A rat study that found something different

A 2026 study in rats looked months after a single dose of psilocybin. It found lasting changes in how medial prefrontal cortex neurons behave electrically, but no difference in synaptic density or spine types between treated and control animals. Its authors concluded that the enduring effects came from functional rather than structural plasticity (Kramer et al., 2026).

Another 2026 mouse study measured protein markers rather than spines. It reported changes in the prefrontal cortex but not in the amygdala, which suggests that effects can differ by brain region (Maltby et al., 2026). Different species, time points and methods give different pictures, and that is why we rate this evidence preliminary.

What do human brain scans show?

What fMRI measures, in plain terms

Functional MRI (fMRI) does not watch neurons fire. It tracks signals linked to blood flow and blood oxygen, which stand in indirectly for activity (Carhart-Harris et al., 2012). “Functional connectivity” describes how closely the signals from two regions rise and fall together. The default mode network (DMN) is a set of connected regions that has been tied to self-reference, mind wandering and autobiographical memory (Gattuso et al., 2023).

During the experience

Evidence: established for the acute state. In 2012, a London team scanned healthy volunteers during an intravenous psilocybin infusion: 15 people for blood flow and a separate 15 for the oxygen-level signal. They expected increases. Surprisingly, they saw only decreases, largest in hub regions such as the thalamus and the cingulate cortex. A bigger drop in one frontal region went with stronger subjective effects, and the coupling between two DMN regions weakened (Carhart-Harris et al., 2012). The volunteers had used hallucinogens before.

A systematic review of 28 papers (17 on psilocybin) found consistent acute disruption of connectivity within the DMN and more connectivity between networks (Gattuso et al., 2023). A 2024 Nature study tracked seven healthy adults with about 18 scans each. Psilocybin changed connectivity more than three times as much as a stimulant used as a comparison, with the biggest changes in the DMN, and differences between people tracked their subjective experience (Siegel et al., 2024).

These are changes in a drug state. They tell us the brain responds. They do not tell us that anything is left rewired afterwards.

Weeks later

Evidence: preliminary. In the 2024 study, connectivity between the anterior hippocampus and the DMN stayed lower for about three weeks after a dose. It had returned to its starting level at a repeat visit 6 to 12 months later, but only four people came back, too few to detect small changes. The authors suggest the reduced connectivity may be a correlate of plasticity. They also say they studied non-depressed volunteers, and that confirming any link to antidepressant effects needs studies in patients (Siegel et al., 2024).

A 2022 Nature Medicine paper scanned patients in two depression trials. One was open-label, meaning everyone knew what was given. The other compared psilocybin with a daily antidepressant. In both, improvements in symptom scores went along with lower network “modularity” (a measure of how separate the brain’s networks are), which the authors read as more integration across networks. The comparison arm showed no such change (Daws et al., 2022). This is a correlation. The randomized trial was the same one whose main symptom measure showed no significant difference between the two arms (Carhart-Harris et al., 2021).

Synapses in living humans

Evidence: preliminary, and so far not supportive. A 2026 Translational Psychiatry study gave fifteen healthy volunteers a single dose and used PET imaging of SV2A, a protein found at synapses that serves as an indirect measure of synapse density. It found no statistically significant overall increase a week later. The authors report a larger rise in the group dosed in a therapeutic-like room than in the group dosed inside an MRI scanner, and suggest context may matter (Johansen et al., 2026). With fifteen people split in two, that is a hint to test, not a result to rely on.

Do brain scans show that psilocybin helps?

No. A scan can show that activity changed. It cannot show that a person is better off. Five things get in the way:

• Small samples. In the systematic review, the average psilocybin study had about 26 participants, and only 2 of the 28 papers had 60 or more, the size a power analysis suggested was needed (Gattuso et al., 2023).
• Not a unique signature. Alcohol also lowers connectivity within the DMN, and other drugs shift network connectivity too, so a changed network does not point to a particular outcome (Gattuso et al., 2023).
• Correlation, not cause. It is unclear whether DMN changes are a by-product of the drug or play a part in its psychological effects, and the review advises researchers to be tentative about causal conclusions (Gattuso et al., 2023).
• Healthy volunteers are not patients. The most detailed human scan study used non-depressed volunteers and says patient studies are needed to test the proposed mechanism (Siegel et al., 2024).
• Change is not benefit. More plasticity can serve learning that helps or learning that hurts (Carhart-Harris et al., 2026).

Where clinical outcome studies fit

Clinical trials measure something else: symptom rating scales. A 2021 trial of 59 patients with depression found no significant difference between psilocybin and a daily antidepressant on its main scale, and the secondary results were not corrected for multiple comparisons (Carhart-Harris et al., 2021). A 2022 phase 2 trial of 233 people reported lower scores at three weeks in the highest-dose group than in the lowest-dose control group. A sustained response at 12 weeks did not clearly follow, and adverse events occurred in 77% of participants (Goodwin et al., 2022).

Those are early, mixed and condition-specific results, covered in our guide to the depression trials. They are one reason brain-change stories should not be read as outcome stories. Whether any of this could become a treatment is a matter for regulators and clinicians, and nothing here is medical advice.

Is the change permanent?

Unresolved. The longest looks we found are short next to a human life:

• Mice: new spines seen within 24 hours, and a fraction still present at about a month in four animals (Shao et al., 2021).
• Rats: functional changes in neurons months later, without a structural difference (Kramer et al., 2026).
• Humans, synapses: one PET scan a week after the dose, with no significant overall change (Johansen et al., 2026).
• Humans, networks: lower hippocampus-DMN connectivity for about three weeks in a few people, back to baseline by 6 to 12 months in a small, underpowered follow-up (Siegel et al., 2024).

We did not find a study that follows structural or network change in people for months and ties it to outcomes. Until one exists, “permanent rewiring” is a claim without evidence behind it.

How should you read a “psilocybin changes the brain” headline?

A few questions sort most claims quickly:

• Which species? Cells, mice, rats and people are not interchangeable.
• What was measured? Structure, electrical function, network connectivity or symptoms?
• How many? A dozen animals or volunteers is a pilot, not a verdict.
• Healthy people or patients? Results can differ, and benefit can only be studied in patients.
• Was there a control group, and was it blinded?
• How long after the dose? Acute effects fade. Lasting ones need follow-up.
• Does it say correlated or caused? Most imaging results are the first.

What we still don’t know

• Whether psilocybin increases synapse number or spine density in people (unresolved; the one PET study found no significant overall increase).
• Whether the mouse spine increase lasts beyond a month, matters for behaviour, or holds up against the rat result that found no structural change (unresolved).
• Whether network changes on scans cause any outcome or just accompany it (unresolved).
• How long human network changes last, beyond early data of a few weeks in a handful of people (preliminary).
• Whether findings in healthy volunteers apply to people with a condition (unresolved).
• Whether the setting of a dose changes plasticity, which rests on one small, exploratory finding (preliminary).
• Whether more plasticity is good in every case (unresolved).

So the honest answer to the headline is “it changes the brain, and we do not yet know what, how much or for how long in a person.” For the basics, read what psilocybin is and how it differs from LSD and DMT. Check the date on this page, because this field moves quickly.

Shroomverse sells mushroom products. This article is general information about brain research. It is not medical, legal or purchasing advice, and it does not recommend any product.