A pattern across my patient panel over the past several months has sharpened a question I have been sitting with since the early days of our ketamine program at Mindbloom: how much of what we call “the therapeutic experience” is actually doing the therapeutic work? Patients who report vivid, emotionally activating medicine sessions sometimes improve less durably than those who describe something quieter and more somatic. Others plateau after meaningful early gains and arrive at follow-up visits asking what, exactly, is supposed to be happening in their brain. I have never had a clean answer. A paper published in Nature Neuroscience this month is the closest the field has come to giving me one.
The finding is specific and consequential: 5-HT2A receptors located in layer 5 pyramidal neurons of the mouse cortex appear to mediate psilocybin’s neuroplastic effects, but those same receptors may not be responsible for its hallucinogenic effects. The pharmacology appears to have been dissociated at the cellular level. Plasticity and perceptual distortion may be separable outputs of the same receptor class, depending on where that receptor lives in cortical architecture. For everyone designing psilocybin trials right now, that sentence should stop the conversation in the room.
What Layer 5 Is Actually Telling Us
When psilocybin activates 5-HT2A receptors on these specific cells, the paper suggests structural synaptic remodeling may be involved, potentially including the kind of synaptic changes associated with antidepressant response. The hallucinations, by contrast, appear to emerge from 5-HT2A activation in different cortical cell populations, according to the paper’s findings.
This matters enormously for how we interpret the receptor pharmacology of competing compounds. LSD, for instance, is reported to bind the 5-HT2A receptor with higher affinity than psilocin (psilocybin’s active form) and to engage a broader receptor profile including dopaminergic and adrenergic targets. The implication, now sharpened by the layer 5 data, is that binding affinity and hallucinogenic potency may be poor proxies for neuroplastic efficacy. A compound engineered to preferentially engage layer 5 5-HT2A receptors could, in principle, drive the synaptic remodeling that produces antidepressant response while substantially attenuating the perceptual side effects that currently require eight hours of dedicated clinical support per session.
COMPASS Pathways has been building its COMP360 program around 25mg psilocybin dosing for treatment-resistant depression, a program that earned FDA Breakthrough Therapy designation and has now extended into PTSD. Every dose-response assumption baked into those trials was built before we knew that neuroplasticity and hallucination were mechanistically separable. That does not invalidate the existing data, but it does mean the field has been optimizing dosing for an outcome bundle without knowing which component was carrying the therapeutic signal.
The Trial Design Problem Nobody Has Solved
In my experience running psychedelic-adjacent trials and treating a large patient population with guided ketamine sessions at Mindbloom, the dominant assumption in the field has been straightforward: the acute subjective experience mediates the therapeutic outcome. Integration of that experience, supported by therapists, produces lasting change. That assumption has driven session-duration protocols, therapist staffing ratios, and the entire support infrastructure the FDA addressed in its July 2026 finalized guidance on psychedelic drug clinical investigations. The layer 5 finding does not disprove that model, but it introduces a serious complication: if the plasticity-driving mechanism operates independently of the hallucinogenic mechanism, then intensity of subjective experience may be a biomarker of 5-HT2A activation in the wrong cortical compartment.
The design implication is uncomfortable. Current trials use subjective experience scales and mystical experience questionnaires as secondary outcome proxies, partly because blinding is impossible and partly because the field has assumed experiential intensity correlates with therapeutic signal. If layer 5 receptor engagement drives plasticity through a pathway that does not generate vivid phenomenology, those proxy measures may actively mislead dose optimization. Sponsors running Phase 2 programs should at minimum be adding objective neuroplasticity markers alongside subjective endpoints: BDNF trajectories, synaptic density measures where feasible, and functional connectivity shifts in the days following a session that may index the plasticity window rather than the perceptual peak.
I want to be careful about what separable actually means here. Showing that plasticity does not require the hallucinogenic pathway is not the same as showing that subjective experience contributes nothing. It may still contribute a great deal. The reading that fits what I see in practice is that biological effect and subjective experience are distinct pathways that interact, not unlike the way pharmacology and psychotherapy interact elsewhere in psychiatry. We understand that interaction poorly. It may turn out to be one of the genuine therapeutic advantages of these compounds rather than a confound to be engineered away.
The populations I work with daily compound this problem. Dual-diagnosis patients with co-occurring SUD and treatment-resistant depression frequently have blunted psychedelic responses due to receptor downregulation from chronic substance exposure, or they carry anxious anticipation of perceptual disruption that keeps them out of trials entirely. If the therapeutic mechanism lives in layer 5 and not in the hallucinogenic response, a substantial population that the field has been quietly excluding from enrollment may be perfectly positioned for the next-generation compounds this finding points toward. The psilocybin-assisted therapy market has been estimated to be in the multi-billion dollar range and is projected to grow steadily; a mechanistically cleaner compound that removes the hallucinogenic barrier to access changes who that market actually reaches.
What This Study Cannot Tell Us
The limits deserve stating plainly, because the headline version of this finding is already traveling faster than the data. This is a mouse study. Inferring anything about subjective experience in a mouse takes a layer of inference that human work does not require, and the head-twitch response, which is what was actually measured, is a behavioral proxy for hallucination rather than a direct measure of experience. A mouse cannot tell us what it underwent. So this work does not settle the role of subjective experience in human therapeutic outcomes, and it should not be read as permission to strip the experiential component out of a protocol.
What I Am Watching Next
The forward question for trial designers is whether any sponsor will move quickly enough to build the layer 5 finding into an active IND. A compound or formulation targeting preferential layer 5 5-HT2A engagement, with objective neuroplasticity readouts as co-primary endpoints alongside traditional symptom scales, would represent the first trial architecture built on this mechanistic dissociation rather than retrofitted to it. I am watching the COMP360 PTSD data closely when it reads out, because the dose-response curve there may show us whether the existing 25mg paradigm is already inadvertently enriching for layer 5 engagement, or whether it has been leaving therapeutic signal on the table all along.
References
- Nature Neuroscience, “5-HT2A receptors in layer 5 pyramidal neurons mediate the neuroplastic but not the hallucinogenic effects of psilocybin”
- COMPASS Pathways, “COMPASS Pathways receives FDA Breakthrough Therapy designation for psilocybin therapy for treatment-resistant depression”
- PubMed, LSD and psilocin 5-HT2A receptor binding affinity and selectivity comparison
- COMPASS Pathways, “Compass Pathways Announces Publication of Results from Phase 2 Study of COMP360 Psilocybin for Post-Traumatic Stress Disorder”
- Baker Donelson, “FDA Finalizes Guidance on Psychedelic Drugs for Clinical Investigations: Key Implications for Hospitals and Trial Sponsors” (July 14, 2026)
- Market Research Future, Psilocybin Assisted Therapy Market Size and Projections (2024,2025)
Dr. Leonardo Vando, MD is Principal Investigator of Psychiatry at Vando Medical Services.
