Psychedelics as edit mode: a psychiatrist's cautious reading of the evidence

Dr. Alok Kanojia (known as Dr. K) is a psychiatrist trained at Harvard Medical School who spent seven years in monastic study before medicine and now runs the HealthyGamerGG channel. In this long, discursive talk he sets out what he thinks a person should understand before deciding whether to use a psychedelic. His stance is not advocacy. He accepts that the clinical signal in treatment resistant depression, PTSD and some addictions is real, then argues that the popular conversation skips the details that decide whether an experience helps or harms. Asked whether a person should use psychedelics, he answers probably not, unless they can assemble the circumstances and guidance that research settings provide.

Not one drug, and not one effect

Kanojia opens by objecting to the word "psychedelics" being used as if it named a single treatment. LSD, psilocybin, ayahuasca and its active DMT, peyote and iboga are grouped together, he says, because they share a mechanism at the serotonin 2A receptor, not because they do the same thing. As a physician, he says, the first questions are which compound, at what dose, for which condition, and with what durability of effect, and he describes hearing little of that precision in podcasts and press coverage.

That receptor, on his account, does two kinds of work. It modulates mood and anxiety in the way antidepressants do, and it also raises markers of neuroplasticity and neurogenesis, including brain derived neurotrophic factor, alongside changes in functional connectivity between brain regions. His framing of this is the video's organising metaphor: psychedelics put the brain into edit mode.

Edit mode cuts both ways

The metaphor is deliberately neutral. Kanojia says he has seen people come out of guided psychedelic work with lasting improvements, and has also seen people traumatise themselves. In the second group, he describes a nervous system that gets locked into a highly aroused state and outcomes that can look like PTSD or panic disorder. His reasoning is that if the brain is editable during the experience, then the content of the experience is what gets written. Positive material tends to be consolidated, and so does frightening material. This is a claim from his clinical practice rather than from a controlled study, and it is worth reading as such.

He contrasts that with what a trial looks like: three or four weeks of preparatory therapy, a dosing session with two trained therapists present for around eight hours, several weeks of integration afterwards, sometimes further dosings, and closing sessions that translate insight into changed behaviour. He notes that traditional contexts, such as extended ayahuasca work with experienced guides, carry comparable structure. His point is that published effect sizes belong to that whole package, not to the molecule alone, which is why set and setting do so much of the work.

Reading the numbers, and reading their limits

Working through a meta analysis of psychedelics for mental disorders, Kanojia makes several points that cut against enthusiasm. Effect sizes shrink once placebo controls are added. Substance and condition interact: on the data he shows, psilocybin looks stronger for major depressive disorder than ayahuasca does, results in alcohol use disorder often cross the line of statistical significance, and MDMA for PTSD looks comparatively promising because so few pharmacological options exist there.

On safety he is more pointed. Adverse events in the reviewed trials are mostly transient (headache, nausea, dizziness, brief anxiety, short lived blood pressure rises), which does not match what he describes seeing clinically. His explanation is that early trial data understates real world risk. He cites benzodiazepines and opioids, whose addictive potential was not visible in early trials, and antidepressants, where withdrawal effects surfaced only after wide use. He also notes that trial participants receive a measured laboratory dose, unlike material grown at home or bought online.

"There is a huge difference between the side effects that we see in a trial and the side effects that we see in real life," Kanojia says.

Microdosing and ego dissolution

On microdosing he is openly sceptical and says so as opinion. He notes there is no agreed definition of the dose, and that the studies he cites found reported improvements that did not match measured outcomes, with one 2024 LSD study finding no alteration of mood or cognition on the measures used. He offers a mechanism for the mismatch: fast acting interventions feel more effective than slow ones, illustrated with stimulant medication versus a year of psychotherapy for ADHD. He acknowledges the research base is thin and that many people will disagree with him.

He closes on the finding he finds most interesting, that therapeutic benefit correlates with ego dissolution rather than with visual intensity. He connects this to where improvements cluster, arguing that depression, PTSD and addiction all involve rigid statements of identity, and notes the parallel with meditative practices that soften the same sense of self. The evidence he cites for this is a review reporting moderate correlations, so it is best treated as a promising direction rather than a settled account.

Key takeaways

• ●Kanojia treats "psychedelics" as a class of different compounds, and says compound, dose, and condition all change the answer.
• ●He frames the shared mechanism as an increase in plasticity, an edit mode in which the content of the experience is what gets consolidated.
• ●On his clinical account, the same mechanism explains both durable improvement and lasting harm.
• ●Published effect sizes come from a package of preparation, supervised dosing, and integration, not from the substance alone.
• ●He argues effect sizes shrink under placebo control and that trial safety profiles tend to understate real world risk.
• ●He views microdosing as offering more subjective relief than measured change, while conceding the research is limited.
• ●Trips involving ego dissolution correlate with benefit more than visually rich trips do, on the review he cites.
• ●His conclusion is conditional: consider it only with appropriate medical or traditional guidance, and only where it is legal.