Psilocybin and the brain: what the depression research shows

Louisa Nicola is a neurophysiologist and host of the podcast TNE (The Neuro Experience). In this solo, scripted episode she walks through the neuroscience of psilocybin as a candidate treatment for depression and addiction, moving from brain imaging to cellular biology to clinical trial results and finally to policy. Her stance is enthusiastic but not uncritical: she gives a full section to contraindications and adverse events, and argues that the therapeutic setting, not the molecule alone, produces the outcomes. Its claims are best read as her reading of a field that is still developing.

The default mode network as the starting point

Nicola builds the episode on one structure: the default mode network, the midline regions that become active between tasks rather than during them. She describes it as the seat of self-referential thought, and says that in a healthy brain it switches on and off fluidly. In depression, she argues, that switching breaks down and the network becomes rigid and hyperactive, producing rumination she frames as a network problem rather than a personality trait.

She then contrasts two mechanisms. SSRIs, on her account, change the chemistry at the synapse and turn the volume down on the loop without disturbing the loop itself. Psilocybin, she says, takes the network offline. She cites a 2024 Nature paper by Siegel and colleagues using precision functional mapping, which reported what the authors called massive desynchronization, followed by reorganization, with regions that do not normally communicate beginning to fire together. A 2022 Nature Medicine study found the change still detectable at three weeks, and a 2026 Nature Communications study, she notes, replicated the finding in psilocybin-naive participants, addressing the concern that prior users may already have different brains.

Dendritic spines and the plasticity argument

The second section moves from networks to cells. Nicola explains dendritic spines as the small protrusions through which neurons make contact, and treats spine density as a measure of how richly connected a circuit is. Chronic stress and depression, she says, are associated with spine loss.

Her key animal study is a 2021 Neuron paper (Shao and colleagues) that used two-photon microscopy in mice after a single psilocybin dose and reported roughly a 10 percent rise in spine density in frontal cortex within 24 hours, with about 60 percent of the new spines still present at a month. She attributes the mechanism to serotonin 2A receptor activation and downstream release of BDNF, citing a 2018 Cell Reports paper (Ly and colleagues) on psychedelics and structural neuroplasticity. She also repeats a figure from a 2024 review: that psilocin, the active metabolite, binds the TrkB receptor with roughly 300 times the affinity of conventional antidepressants. That comparison rests on a small and contested literature, and the spine work is animal work, which does not on its own establish a human mechanism.

What the trials actually showed

Section three covers the 2021 New England Journal of Medicine trial led by Robin Carhart-Harris, which compared psilocybin against the SSRI escitalopram in 59 patients with moderate to severe depression. Nicola is precise about the awkward part: on the primary outcome the difference between groups was not statistically significant, and the trial was not powered to detect a small difference there. She argues the secondary outcomes were more interesting: a 70 percent response rate against 48 percent, and a 57 percent remission rate against 28 percent.

The addiction section reports the Johns Hopkins pilot on psilocybin-assisted smoking cessation: 15 long-term smokers, 80 percent biologically confirmed abstinent at six months and 67 percent at twelve, alongside a 2022 JAMA Psychiatry trial in alcohol use disorder that found fewer heavy drinking days at eight months. These are small samples, and every protocol paired the dose with structured therapy. Her framing of addiction as an overlearned pattern the disruption loosens is offered as a hypothesis.

Risks, and why the container matters

The fifth section is the most sober. Nicola lists the most common adverse event as a challenging experience: intense anxiety, confusion, or distress across a six to eight hour session, manageable with trained support and a different proposition without it. Personal or family history of psychosis, schizophrenia, or bipolar I disorder was an exclusion criterion in every trial she discusses, and cardiovascular conditions require screening. She notes a 2025 review that found no serotonin syndrome at standard doses alongside antidepressants, though SSRIs appeared to blunt the effect, and mentions HPPD as rare, real, and poorly understood. The section maps onto standard contraindications used in clinical screening.

She closes on the mystical experience questionnaire developed at Johns Hopkins, and on the observation that higher scores tend to correlate with better clinical outcomes across independent studies. If the effect were purely pharmacological, she argues, outcomes should not vary with the subjective quality of the session. That leads her to treat set and setting as possible active ingredients rather than supportive detail, and to note that the Hopkins and Imperial protocols standardize music, room design, and therapist training.

"The drug alone is not the treatment. The drug plus the therapeutic container is the treatment," Nicola says.

Key takeaways

• ●Nicola frames depression partly as a rigidity problem in the default mode network, and psilocybin as something that takes that network offline rather than modulating it.
• ●Dendritic spine growth within 24 hours of a single dose is the cellular basis of her plasticity argument, though it remains animal work.
• ●The trial against escitalopram missed significance on its primary outcome; the favorable numbers are secondary outcomes from a 59-patient study.
• ●Smoking and alcohol results come from small pilots, and always in combination with therapy.
• ●Psychosis and bipolar I history were hard exclusions in every trial discussed, and cardiovascular screening was standard.
• ●Nicola reports that Australia reclassified psilocybin for supervised prescribing in 2023, and describes United States policy as moving quickly.
• ●Her central caution is that the therapeutic container is not optional: unsupervised use carries a different risk profile from a clinical session.