Ketamine vs Psilocybin: A Neuroscience Comparison

For most of the twentieth century, the pharmacology of depression was a single story told in one language: serotonin, and the drugs that nudged it. Then two very old molecules broke the plot open. Ketamine, a battlefield anesthetic from the 1960s, turned out to lift severe depression in a matter of hours. Psilocybin, the active compound in “magic mushrooms,” turned out to loosen entrenched depression for weeks after a single guided session. This is how they actually compare — mechanism, tempo, evidence, risk, and access. This article is education, not medical advice.

They are now the two most-discussed rapid-acting antidepressants in neuroscience — and they could hardly be more different in how they touch the brain. This is not a contest to crown a winner. It is a study in contrast: two doorways into the same wounded system, opening onto the same downstream repair. Understanding where they diverge, where they quietly converge, and what each one actually asks of a patient is the most useful thing a comparison can offer.

(Educational overview only — not medical advice. If you are in crisis, contact a local emergency line or the 988 Suicide and Crisis Lifeline in the US.)

The core contrast: two receptors, one repair

Start with the molecules, because everything else flows from there.

Ketamine is an NMDA-receptor antagonist. It blocks a specific glutamate receptor — and counterintuitively, blocking it increases glutamate signaling. The prevailing “disinhibition hypothesis” holds that ketamine preferentially silences NMDA receptors on inhibitory GABA interneurons, taking the brakes off nearby excitatory neurons. The result is a brief, sharp surge of glutamate in the prefrontal cortex. That surge activates a second class of glutamate receptor, AMPA, which kicks off a signaling cascade — BDNF release, mTOR activation — that drives the rapid formation of new dendritic spines and synapses. Block the AMPA step and the antidepressant effect disappears, which is strong evidence that the plasticity, not the dissociation, is doing the therapeutic work (Kim & Monteggia, Neuropsychopharmacology 2023).

Psilocybin arrives from the opposite direction. Once metabolized to psilocin, it is a 5-HT2A serotonin-receptor agonist. Activating 5-HT2A receptors — densely packed on layer-V cortical pyramidal neurons — increases cortical excitability and, at the whole-brain scale, desynchronizes activity. Neuroimaging describes a rise in neural “entropy” and a loosening of the default mode network, the tightly coupled circuit associated with self-referential rumination (Siegel et al., Nature 2024). Where ketamine feels like a targeted electrical jolt, psilocybin feels like the brain’s habitual grooves being temporarily flattened so new patterns can form.

And here is the twist that keeps the two drugs bound together in the literature: they converge downstream. Both promote rapid, lasting growth of dendritic spines. Both funnel into BDNF–TrkB–mTOR signaling — the brain’s core plasticity machinery (Shao et al., Neuron 2021). A 2023 study even found that psychedelics bind the BDNF receptor TrkB directly, with roughly a thousand-fold higher affinity than classical antidepressants (Moliner et al., Nature Neuroscience 2023). A 2025 analysis framed the two families as sharing a mechanism ceiling: rapid neuroplasticity by different keys into the same lock (Molecular Psychiatry 2025).

The honest caveat

The “plasticity restores mood” story is a compelling and well-supported hypothesis, not settled fact. Much of the synaptogenesis data is from animals, the causal chain in humans is inferred, and no one can yet say precisely how a molecular spine-growth event becomes a lifted mood. Treat the mechanism as a strong working model, not gospel.

Onset and duration: the tortoise, the hare, and the maintenance problem

The most practically important difference between these two drugs is tempo.

Ketamine is the hare. In the landmark 2000 trial by Berman and colleagues, a single 0.5 mg/kg intravenous infusion improved depression within hours (Berman et al., 2000). Zarate’s 2006 replication in treatment-resistant patients found significant improvement within about 110 minutes, sustained for roughly a week (Zarate et al., 2006). That speed is genuinely without precedent in psychiatry — it is why ketamine draws attention for acute suicidality, where waiting six weeks for an SSRI is not an option.

But the hare tires quickly. A single ketamine dose typically fades within one to two weeks. This is ketamine’s defining clinical burden: it usually requires repeated dosing. Repeated-infusion studies show responders relapsing a median of around 18 days after a series ends (Murrough et al., 2013), which is why real-world protocols involve induction series followed by ongoing maintenance — an open-ended commitment with cost, logistics, and cumulative-exposure implications.

Psilocybin is the tortoise. The antidepressant effect does not announce itself in two hours; it unfolds over the days following a dosing session as the acute experience integrates. But when it holds, it holds. Trials show meaningful benefit at three to six weeks from one or two sessions, with subsets of patients maintaining response at six and twelve months (12-month follow-up, J Psychopharmacol 2022). The dosing model is closer to a discrete intervention — a session, or two spaced a few weeks apart — than a standing prescription.

Setting amplifies the divergence. Ketamine’s therapeutic ritual can be minimal: an infusion chair, monitoring, a dissociative interval. Psilocybin trials are built around a full-day supervised session with a prepared “set and setting,” eyeshades, curated music, and — crucially — psychological support before and after. In the psilocybin paradigm, the drug is arguably the smaller half of the treatment.

The clinical scoreboard, honestly kept

If mechanism is where they converge and tempo is where they diverge, evidence is where they are least comparable — because they sit at different stages of the regulatory pipeline.

Ketamine and esketamine: the established players

Esketamine, the S-enantiomer delivered as a nasal spray (brand name Spravato), was FDA-approved in March 2019 for treatment-resistant depression as an add-on to an oral antidepressant, and in 2020 for depressive symptoms with acute suicidal ideation. In January 2025, the FDA approved Spravato as a standalone monotherapy for treatment-resistant depression — the first such approval, based on a placebo-controlled trial showing rapid MADRS improvement by week 4 (J&J press release, 2025). Racemic (generic) ketamine, meanwhile, is used off-label by IV — not FDA-approved for depression, but widely administered.

The honest asterisk: esketamine’s approval does not mean its benefit is dramatic. A 2024 systematic review and meta-analysis found a modest pooled effect size (roughly 0.15–0.23), comparable to atypical-antipsychotic augmentation, with several individual short-term trials returning negative primary results and persistent debate about “functional unblinding” — patients can tell they got the active drug because they feel it (Am J Psychiatry 2024 meta-analysis). Established is not the same as overwhelming.

Psilocybin: the ascending player

The modern evidence base runs from Carhart-Harris’s 2016 open-label feasibility study (Lancet Psychiatry 2016), through Johns Hopkins’ 2021 randomized trial reporting large between-group effect sizes (Davis et al., JAMA Psychiatry 2021 — note a published erratum corrected some figures), to the pivotal work. COMPASS Pathways’ Phase 2b trial in 233 treatment-resistant patients found a single 25 mg dose beat a 1 mg control on MADRS at three weeks, with response around 37% and remission near 29% — alongside reported serious adverse events including suicidal ideation (Goodwin et al., NEJM 2022). Usona Institute’s 2023 trial found a single 25 mg dose improved MADRS by 12.3 points more than placebo at six weeks (Raison et al., JAMA 2023).

The decisive news is recent. In 2025, COMPASS reported that both Phase 3 trials (COMP005 and COMP006) met their primary endpoints — COMP005 the first Phase 3 efficacy readout for any classic psychedelic — and the company is pursuing an FDA submission under Breakthrough Therapy designation, with longer-term 26-week COMP006 data expected in the second half of 2026 (COMPASS Pathways, 2025). Psilocybin also has arguably its strongest signal in end-of-life distress: Griffiths’ 2016 cancer trial reported clinician-rated response rates near 80% at six months (Griffiths et al., 2016), echoed by Ross’s NYU trial the same year (Ross et al., 2016).

So which is more established for what? Ketamine, unambiguously, for approved, available, acute treatment of resistant depression and suicidal crisis. Psilocybin has deeper durability signals and a uniquely strong existential-distress dataset — but remains investigational until the FDA rules.

The experience, and the very different hazards

Subjectively, these are not variations on one theme. Ketamine produces dissociation — a detachment from body and environment, at higher doses the disorienting “K-hole.” Psilocybin produces the classic psychedelic journey — perceptual shifts, emotional intensity, and in the deepest sessions a sense of ego dissolution. One is a numbing distance; the other is a flooding immersion.

Their safety profiles differ in kind, and it is a disservice to file one as “safe” and the other as “dangerous.” They are dangerous in different directions.

Ketamine: risks of the drug itself and chronic exposure

Ketamine carries real abuse and dependence liability — it is a Schedule III controlled substance and a recreational drug — which matters acutely now that at-home telehealth prescribing has expanded access dramatically. Heavy, sustained use causes a well-documented bladder toxicity (ketamine-induced ulcerative cystitis), with urinary frequency, pain, and in severe cases irreversible bladder damage (safety review, 2024). It transiently raises blood pressure and heart rate, requiring caution in cardiovascular disease. The dissociation itself, while usually brief, can be frightening without support.

Psilocybin: risks that are psychological and acute

The signature hazard is the “bad trip” — overwhelming anxiety, fear, or panic during a session — which in controlled settings is transient and manageable with trained support, but can be genuinely destabilizing without it. HPPD (persistent visual disturbances) appears rare and is essentially unreported in modern controlled trials. The firm contraindications are personal or family history of psychosis or bipolar disorder, where psychedelics can precipitate prolonged psychotic or manic episodes; a survey of bipolar users found roughly a third reported new or worsened symptoms (bipolar survey, 2022). Psilocybin also transiently raises blood pressure and heart rate. A more theoretical concern is 5-HT2B receptor agonism, the mechanism behind serotonergic valvular heart disease — a real worry for hypothetical chronic dosing or frequent microdosing, but not established as a risk for occasional supervised therapeutic sessions (5-HT2B review, 2023).

The cleanest summary: ketamine’s tail risks accumulate with repeated exposure and misuse; psilocybin’s sharpest risks concentrate in the acute session and in vulnerable populations. Different hazard curves, not a hierarchy.

Access and legality in 2025–2026

This is where the practical asymmetry is starkest, and where facts change fast.

Ketamine is legal medicine. As a Schedule III substance, it is available through IV infusion clinics, in the approved esketamine (Spravato) form, and — increasingly — through telehealth. In January 2026, the DEA and HHS extended telehealth prescribing flexibilities for controlled substances through the end of 2026, keeping the door open for remote ketamine prescribing without an in-person visit. That access is a double-edged sword: it has widened reach while raising legitimate concerns about screening, monitoring, and misuse in an under-regulated online market.

Psilocybin remains Schedule I federally — the most restrictive classification, formally reserved for substances with no accepted medical use — and no federal rescheduling has occurred; any change is contingent on FDA approval of a specific product. But state-level supervised pathways exist. Oregon operates licensed psilocybin service centers for adults 21+ without a prescription. Colorado opened its first licensed healing center in April 2025 and had dozens more approved by early 2026. New Mexico passed a Medical Psilocybin Act in April 2025 — the first via legislation rather than ballot — with a program still under construction and initial patients targeted for late 2026 (state law summaries and trackers, 2026). These are supervised-use frameworks, not retail legalization, and they do not change federal law.

Which one, for whom?

Resist the urge to crown a winner — the useful question is fit.

If the need is speed — acute, severe, or suicidal depression where days matter — ketamine’s hours-fast onset and legal availability make it the pragmatic option today. If the goal is durability from a discrete, integrative intervention, psilocybin’s weeks-to-months profile is the more natural shape, once and where it is legally accessible.

Comorbidities tilt the choice. A history of substance use disorder weighs against ketamine’s dependence liability; a history of psychosis or bipolar disorder weighs heavily against psilocybin. Willingness and capacity to engage in an intense, hours-long psychological experience favors psilocybin; a preference for a briefer, more clinical procedure favors ketamine. And setting is not incidental — psilocybin’s benefit appears bound up with preparation and integration in a way ketamine’s is not.

In the end the most honest framing is the one the neuroscience keeps pointing to: two different keys, opening onto the same room of neuroplastic repair. The interesting future is less “which wins” than what each teaches us about how a depressed brain can be coaxed to change — and, perhaps, how they might one day be used in sequence rather than opposition.

This article is education, not medical advice.