Psychedelics and The Mind: what the experts say

The Big Picture

Three Huberman Lab conversations approach psychedelics from the clinic and the lab rather than the festival. Dr. Matthew Johnson explains how psychedelic therapy works in clinical trials, why it helps depression and addiction, and what its real risks and limits are [1]. Stanford's Dr. Nolan Williams places psychedelics alongside TMS and ketamine as therapies that recalibrate mood circuits, and argues they belong under medical care [2]. Psychiatrist Karl Deisseroth, who studies these drugs in his lab under strict DEA oversight, supports exploring them carefully while warning that they can be addictive or cause unwanted lasting changes [3].

A shared thread runs through Johnson's and Deisseroth's explanations: the brain constantly predicts and filters, and psychedelics loosen that filter. Johnson calls us prediction machines whose perception is largely top-down [1]; Deisseroth calls the cortex "a hypothesis generation and testing machine" whose unlikely models are usually screened out before they reach awareness [3]. The three conversations also put weight on different things: Johnson on mechanism and its limits [1], Williams on the evidence for specific compounds and circuit-based treatments [2], and Deisseroth on models, risk and learning [3].

Where the Experts Agree

Psychedelics change the brain's models, not just its inputs. Johnson prefers to describe psychedelics as profoundly altering models: a child learns that a thrown ball falls, and psychedelics can loosen such learned predictions so that violations are no longer filtered out [1]. Deisseroth's hypothesis is similar: psychedelics lower the threshold at which unlikely models reach awareness [3].
That loosening could help depression. Deisseroth describes depression as a state where the future seems to run out "like a river just running out into a desert and drying up," and suggests loosening the filter may let new paths forward reappear, something his lab is testing by recording cells that represent future steps [3]. Johnson's conversation covers why psychedelic therapy helps depression and addiction [1], and Williams discusses evidence for psilocybin in depression [2].
Treatment should be structured and supervised. Williams argues these therapies belong under medical care [2]. Deisseroth supports low doses used alongside therapy when done rigorously [3].
The risks are real. Johnson's conversation addresses real risks and limits, including rare but credible cases of people who believed they could fly and a research volunteer who appeared to try to dive through a painting on the wall [1]. Deisseroth notes these drugs can be addictive or cause unwanted lasting changes [3].

Where They Differ

How confident to be about mechanism. Johnson is frank that answers on mechanism are limited. Serotonin itself obviously isn't psychedelic, or eating a bagel would make everyone trip; classic psychedelics activate the 2A receptor, which increases glutamate transmission downstream, possibly a commonality with ketamine [1]. Williams's conversation adds a separate finding on ketamine, discussing a study showing its antidepressant effect depends on opioid receptors, not the "trip" [2].
How to weigh risk. Deisseroth stresses that the risks may not be much greater than what medicine tolerates elsewhere [3], while Johnson's examples of people who believed they could fly or tried to dive into a painting highlight how the loosening of predictions can go wrong [1].
Which disorders and drugs come first. Williams's conversation covers MDMA for PTSD, psilocybin for depression, the role of "letting go," and ibogaine, 5-MeO-DMT and ayahuasca [2]. Deisseroth focuses on MDMA, which releases large amounts of dopamine and serotonin together, and thinks its lasting benefit for trauma comes from learning [3]. Johnson's conversation emphasizes depression and addiction [1].
What counts as a hallucination. Johnson draws a line between classic psychedelics and anticholinergics like atropine and scopolamine, which produce true hallucinations such as conversations with people who aren't there, something the classic psychedelics rarely do [1]. He also notes that alcohol and benzodiazepines scramble space-time orientation, so that effect alone doesn't define a psychedelic [1]. Deisseroth's lens is different: in schizophrenia, faulty models escape the filter and become paranoid delusions, which is the risky side of the same mechanism he hopes can help depression [3].
Where psychedelics fit among other treatments. For Williams, psychedelics are one option alongside TMS and ketamine, and his conversation ends with an accelerated, individually targeted TMS protocol that can bring remission in days [2].

Practical Takeaways

Treat psychedelic therapy as a medical intervention: Williams argues it belongs under medical care, and Deisseroth supports low doses and use alongside therapy when done rigorously [2][3].
Pay attention to what is learned, not only to the drug: Deisseroth thinks MDMA's lasting benefit for trauma comes from learning [3].
Take the safety concerns seriously: these drugs can be addictive or cause unwanted lasting changes [3], and loosened predictions have led to rare but credible cases of dangerous beliefs [1].
Don't equate psychedelics with simply raising serotonin: Johnson notes serotonin itself isn't psychedelic, or eating a bagel would make everyone trip [1].
Know that the mechanism is still unsettled, from 2A receptors and glutamate [1] to an opioid-dependent effect for ketamine [2], so be wary of anyone claiming certainty.
If depression is the concern, psychedelics are not the only new route: Williams describes an accelerated TMS protocol that can bring remission in days [2].
Watch for the results of careful studies, such as Deisseroth's lab work on cells that represent future steps, before drawing firm conclusions [3].