Psychedelics and Anxiety: What the Johns Hopkins Trials Actually Found

There's a part of your brain that never shuts up. It replays conversations. It rehearses worst-case scenarios. It tells you what you should have said and what people probably think of you. Scientists call it the default mode network, and in people living with serious anxiety, it runs hot. It's like a radio stuck on a station that only plays your fears.

Psilocybin, the compound found in certain mushrooms, does something unusual to that radio. Under carefully controlled conditions, it turns the volume way down. Not permanently, not dangerously, but enough for something interesting to happen. People describe feeling, for the first time in years, like the constant chatter just stopped. The worries didn't vanish because they argued them away. They faded because the part of the brain generating them went quiet for a few hours.

This isn't about recreational use. The research happening at Johns Hopkins and NYU involves a single dose given in a clinical room, with trained guides sitting beside the person for the entire experience. What surprised researchers wasn't just that people felt better during the session. It's that many of them still felt better months later. Something shifted in that window of quiet, and for reasons scientists are still working to understand, the shift stuck.

When someone receives a life-threatening cancer diagnosis, the anxiety that follows isn't ordinary worry. It's existential. It sits in the chest like a stone. It's the kind of dread that doesn't respond well to being told things will be okay, because the person knows they might not be. Standard treatments for this kind of anxiety, medications and talk therapy, help some people. But for many, the fear stays.

In 2016, two teams published results on the same day. Researchers at Johns Hopkins gave cancer patients a single high dose of psilocybin in a supported setting. Researchers at NYU did something very similar. Both found the same thing: people who'd been carrying unbearable anxiety about dying experienced significant relief. Not just during the session, but at follow-up visits five and six months later. Some described it as the most meaningful experience of their lives.

What makes this striking isn't just that it worked. It's that it worked from a single session. Most anxiety treatments require weeks or months of consistent effort. These patients sat in a room for one afternoon with eyeshades on and music playing, accompanied by two trained guides, and many walked out carrying noticeably less fear. That doesn't mean psilocybin is a cure or that it works for everyone. But for people who'd tried everything else, one afternoon changed the weight they were carrying.

It would be easy to read these results and think the answer to anxiety is a mushroom. But the researchers themselves are the first to say: slow down. These studies involved dozens of people, not thousands. The participants were carefully screened. They spent hours in preparation sessions before the actual dose day. And they had trained professionals with them the entire time. What happened in those clinical rooms isn't the same thing as what happens without that structure.

The people guiding these sessions weren't just sitting there. They were trained to hold space for whatever came up, from tears to terror to joy. Participants had been prepared for what the experience might feel like. They had follow-up sessions to make sense of it afterward. Researchers believe all of this matters. The drug opens a window, but what happens during that window depends heavily on who's in the room and how the person was prepared. Take away the preparation and the guides, and the outcomes could look very different.

Right now, psilocybin therapy isn't something you can walk into a clinic and receive for anxiety. MDMA-assisted therapy for PTSD got further along in the FDA process but hit setbacks. The science is moving, and it's genuine. But it's also young. If this research gives you hope, that's a reasonable response to what the data shows. And if it makes you cautious, that's reasonable too. The honest answer is that we're watching something promising unfold, and the full picture isn't in yet.

Your brain has a network that activates whenever you're not focused on anything external. It's called the default mode network, and it handles things like thinking about yourself, remembering the past, and imagining the future. In people with anxiety, this network tends to be overactive. It generates the loop of "what if" thinking, the replaying of awkward moments, the rehearsal of future catastrophes. It's useful for planning, but when it runs unchecked, it becomes a generator of dread.

Psilocybin works by binding to a specific serotonin receptor called 5-HT2A. When it does, something happens that researchers can see on brain imaging: the default mode network's activity drops sharply. The parts of the brain that normally stay separate start communicating in new ways. The rigid patterns that keep anxious thoughts cycling begin to loosen. People describe the experience as feeling like the walls came down between compartments in their mind.

What's important about this isn't just the experience during the session. It's what happens afterward. Researchers noticed that the default mode network didn't just snap back to its old patterns in many participants. The reduction in anxious rumination persisted for weeks, sometimes months. It's as though the brain, given a few hours of relief from its own loops, found a less rigid way to operate going forward. Not a permanent rewiring from one dose, but a genuine shift in the groove the needle had been stuck in.

The anxiety that comes with a cancer diagnosis is different from everyday worry. It's about mortality itself. Standard antidepressants and therapy help some patients manage it, but many continue to carry a weight that doesn't lift. Researchers at Johns Hopkins and NYU independently decided to test whether psilocybin could reach what other approaches couldn't. Both studies used careful randomized designs with crossover controls, meaning each participant eventually received both the real dose and a comparison.

The Hopkins trial, led by Roland Griffiths, gave 51 cancer patients either a high dose of psilocybin or a very low dose (as an active placebo). At five weeks, participants crossed over. The results were striking: about 80% of participants showed clinically significant decreases in anxiety and depression that persisted at the six-month mark. The NYU trial, led by Stephen Ross, found nearly identical results with 29 patients. The convergence between two independent teams using similar designs made the findings hard to dismiss.

One detail that caught researchers' attention was the speed. Antidepressant medications typically take two to six weeks to show effects. Many cancer patients don't have that kind of time, emotionally or literally. In these trials, the shift happened within hours of the session and held. Participants didn't describe feeling numbed or distracted from their diagnosis. Many said they still knew they were dying, but the terror had been replaced by something closer to acceptance. That distinction matters.

The results from Hopkins and NYU are genuine, but the studies were small. Fifty-one people in one trial, twenty-nine in the other. Participants were screened for conditions that might make the experience risky, like a personal or family history of psychosis. They were volunteers who knew they'd be receiving a psychedelic. These aren't limitations that invalidate the findings, but they mean we can't yet say how psilocybin would work across a broader population with different kinds of anxiety.

Researchers in this field consistently emphasize that the drug alone isn't the treatment. The treatment is the entire container: preparatory sessions where participants build trust with their guides, the session day itself with its carefully designed setting (eyeshades, curated music, a comfortable room), and integration sessions afterward where participants process what happened. The guides are trained to be present without directing the experience. Early data suggests that how well the therapeutic relationship is established before the session predicts outcomes just as strongly as the dose itself.

The broader psychedelic therapy field hit a significant moment in 2024 when the FDA declined to approve MDMA-assisted therapy for PTSD, citing concerns about study design despite promising trial data from MAPS. That decision affects the timeline for all psychedelic-assisted treatments. Psilocybin for anxiety is still in earlier phases. If this area interests you, the honest framing is: the mechanism is plausible, the early results are strong, and the path to widespread availability is longer and more uncertain than headlines suggest. Science moves at its own speed, and that's actually part of what makes it trustworthy.

Robin Carhart-Harris and colleagues at Imperial College London published the foundational neuroimaging work on psilocybin's brain effects. Their 2012 PNAS paper was the first to show, using arterial spin labeling and blood-oxygen-level-dependent fMRI, that psilocybin decreased cerebral blood flow and BOLD signal specifically in the default mode network's key hubs: the medial prefrontal cortex and the posterior cingulate cortex. This was counterintuitive. A drug producing a profound subjective experience was reducing, not increasing, neural activity in major brain regions. The magnitude of default mode suppression correlated with the intensity of the subjective experience reported by participants.

Carhart-Harris proposed the "entropic brain hypothesis" to explain what was happening. Under normal conditions, the brain organizes itself in a relatively constrained, hierarchical manner, with the default mode network acting as a kind of conductor. Psilocybin temporarily disrupts that hierarchy. The 2016 PNAS study showed that under psilocybin, functional connectivity between brain networks that normally operate independently increased dramatically. The brain's repertoire of possible states expanded. In computational terms, neural entropy rose. This destabilization of rigid connectivity patterns is what researchers believe gives anxious thought loops a chance to break.

The clinical question was whether these changes persisted. Carhart-Harris' team found that one week after a psilocybin session, resting-state functional connectivity between the default mode network and other regions remained altered compared to baseline. Patients with treatment-resistant depression showed increased connectivity between the default mode network and the ventromedial prefrontal cortex, a region involved in emotional processing and self-referential thought. The shifts weren't permanent. But they were measurable days to weeks after a single dose. Barrett and colleagues at Hopkins similarly found that psilocybin-occasioned changes in default mode functional connectivity correlated with enduring changes in personality traits, specifically increases in openness, measured at twelve months.

The Griffiths 2016 trial was methodologically notable for its crossover design and active placebo. Using a very low dose as the comparator reduced expectancy effects, since participants experienced some perceptible change in both conditions. The primary outcome measures included the Hamilton Anxiety Rating Scale (HAM-A), Beck Depression Inventory, and multiple clinician-rated and self-report measures. The high-dose session produced a mean decrease in HAM-A scores that crossed clinical significance thresholds, and the difference between high and low dose was statistically significant. Five months after the crossover, when all participants had received the high dose, approximately 80% continued to meet criteria for clinically significant anxiolytic and antidepressant response.

Ross and colleagues at NYU took a slightly different approach, using niacin as the active placebo in their crossover design. Their 29-participant trial found that 83% of participants showed clinically significant decreases in anxiety at the 6.5-month assessment, and 60% met criteria for complete remission of anxiety. The convergence between two independent groups using different placebos and slightly different protocols substantially strengthened the evidence. Effect sizes for anxiety reduction in both trials (Cohen's d ranging from 1.0 to 1.3) exceeded those typically observed in SSRI trials for anxiety disorders.

Agin-Liebes and colleagues published the 4.5-year follow-up from the Ross NYU trial in 2020. Of the original participants who could be contacted, approximately 60-80% continued to meet criteria for clinically significant reductions in anxiety and depression. Participants rated the psilocybin experience among the most personally meaningful and spiritually significant of their lives. These are extraordinary numbers for a single-dose intervention. However, the long-term sample was small and uncontrolled. Participants knew they had received psilocybin, and self-selection in follow-up studies introduces bias. The data is genuinely compelling, but it requires larger, more rigorous replication before it can be called definitive.

The mediation analysis from Griffiths' data revealed something pharmacologically unusual. When mystical experience scores were entered as a mediator between psilocybin administration and anxiety reduction, they accounted for a significant portion of the therapeutic effect. Participants who reported feelings of oceanic boundlessness, unity, noetic quality, and deeply felt positive mood during the session improved more. This creates a challenge for standard pharmacological frameworks, which typically attribute drug effects to receptor activity and downstream neurochemistry, not to the subjective quality of the experience the drug produces.

The set and setting concept, borrowed from the broader psychedelic literature, has empirical support beyond anecdote. Set refers to the participant's mindset entering the session, shaped by preparation, expectation, and the therapeutic relationship. Setting is the physical and interpersonal environment. Johnson, Richards, and Griffiths published safety guidelines in 2008 identifying specific environmental and interpersonal factors associated with positive versus adverse outcomes in psilocybin research. A preliminary study by Carhart-Harris' group found that the quality of therapeutic rapport predicted acute subjective experience, which in turn predicted long-term clinical outcomes. The implication is that psilocybin opens a window, but what comes through that window depends on who's holding it open.

The regulatory picture shifted substantially when the FDA issued a Complete Response Letter for MDMA-assisted therapy for PTSD in August 2024, effectively declining approval despite two positive Phase 3 trials. The FDA cited concerns about potential unblinding, functional unblinding, and study conduct. This decision sent ripples through the entire psychedelic therapy field. Compass Pathways and Usona Institute are running psilocybin trials for treatment-resistant depression, not anxiety specifically, and their timelines have been revised. For social anxiety, a small pilot by Goldberg and colleagues showed promising results, but the sample was under twenty. The science is building genuine momentum. But the distance between "exciting Phase 2 data" and "available to your doctor" remains measured in years, not months.

Carhart-Harris, Erritzoe, Williams, Stone, Reed, et al. (2012) published the first human fMRI data on psilocybin's neural effects in PNAS. Using intravenous psilocybin with arterial spin labeling and BOLD imaging in 30 healthy volunteers, they found decreased cerebral blood flow and BOLD signal in default mode network hubs: medial prefrontal cortex, posterior cingulate cortex, and anterior cingulate cortex. The magnitude of suppression correlated with subjective ego dissolution ratings (r = 0.48, p < 0.01). A paradox: a drug producing one of the most intense subjective experiences known was suppressing, not amplifying, activity in the brain's major associative networks.

Carhart-Harris (2014) developed the entropic brain hypothesis to contextualize these findings. The framework proposes that primary consciousness is characterized by elevated entropy in neural dynamics, with 5-HT2A receptors acting as modulators of cortical excitability and constraint. Under normal waking conditions, the default mode network imposes hierarchical constraint on information flow between brain regions. Psilocybin, by agonizing 5-HT2A receptors concentrated in Layer V pyramidal neurons, temporarily destabilizes this hierarchy. The 2016 PNAS study confirmed increased between-module functional connectivity with decreased within-module coherence. Tagliazucchi et al. (2014) showed this state shares computational properties with REM sleep, another condition of reduced default mode constraint and high neural entropy.

Barrett, Johnson, and Griffiths (2015) connected these acute changes to lasting outcomes. In healthy volunteers who received psilocybin, acute mystical-type experience predicted increases in openness to experience at 14-month follow-up, a finding without precedent in personality research where traits are considered stable after age 30. Carhart-Harris' group showed that post-psilocybin reductions in default mode functional connectivity at one week correlated with sustained improvements in depressive symptoms. The mechanism appears to operate through a window of neural flexibility: rigid self-referential patterns are disrupted, allowing new configurations to emerge during a critical period extending beyond the acute pharmacological effect.

Griffiths, Johnson, Carducci, Umbricht, Richards, et al. (2016) published their randomized, double-blind, crossover trial in the Journal of Psychopharmacology. Fifty-one cancer patients with life-threatening diagnoses and DSM-IV anxiety or mood disorder diagnoses received either psilocybin 22 mg/70 kg or 1 mg/70 kg (active placebo) in two sessions separated by five weeks. Primary outcomes included the GRID-Hamilton Depression Rating Scale and HAM-A. At the five-week crossover point, high-dose participants showed significantly greater anxiety reductions (HAM-A: p < 0.001). At six months post-crossover, 83% showed clinically significant response on the Beck Depression Inventory and 79% on state anxiety measures.

Ross, Bossis, Guss, Agin-Liebes, Malone, et al. (2016) published independently in the same journal issue. Their 29-participant crossover trial used niacin 250 mg as an active placebo. The primary anxiety outcome (Spielberger State-Trait Anxiety Inventory) showed a between-group effect size of Cohen's d = 1.32 at seven weeks, qualifying as a large effect. At 6.5-month follow-up, 83% of participants showed clinically significant decreases in anxiety, and 60% met remission criteria. The simultaneous publication of two independent positive trials from separate institutions was strategically coordinated to maximize scientific impact and minimize the likelihood that either set of results would be dismissed as anomalous.

Agin-Liebes, Malone, Yalch, Mennenga, Ponse, et al. (2020) published the 4.5-year follow-up in the Journal of Psychopharmacology. Of 15 participants available for assessment, 60-80% continued showing clinically significant reductions in anxiety and depression. Seventy-one percent rated the session among the top five most personally meaningful experiences of their lives. Sustained effects from a single pharmacological intervention, a profile unlike any existing anxiolytic. The limitations are sample attrition (15 of 29 original participants), absence of a control group at long-term follow-up, and self-selection bias among those willing to be reassessed. The data demands attention because of its effect magnitude, and demands replication because of its methodological constraints.

The mediation analysis from Griffiths et al. (2016) introduced a variable rarely seen in psychopharmacology. The Mystical Experience Questionnaire (MEQ30), measuring unity, noetic quality, sacredness, and transcendence, accounted for a statistically significant portion of variance in therapeutic outcome when entered as a mediator between drug condition and improvement. The pharmacological agent's benefit was partly explained by subjective experience quality. This suggests 5-HT2A agonism isn't sufficient on its own; whatever generates the felt sense of meaning and connection carries independent therapeutic weight. Roseman, Nutt, and Carhart-Harris (2018) replicated this mediating role in their treatment-resistant depression trial.

Johnson, Richards, and Griffiths (2008) published human hallucinogen safety guidelines in Psychopharmacology that codified set and setting as empirical variables rather than folk concepts. They identified specific risk factors (personal or family history of psychotic-spectrum disorders, current lithium use, unsupervised settings) and protective factors (thorough screening, preparation sessions, trained guides, a physically comfortable environment, and post-session integration). Garcia-Romeu and Richards (2018) extended this work, arguing that the therapeutic framework isn't merely a delivery vehicle for the pharmacological agent but an active component of the intervention. Preliminary data from Carhart-Harris' group suggests that therapeutic alliance quality measured before the session predicts both acute experience quality and long-term clinical outcomes, suggesting a three-link chain: relationship predicts experience, experience predicts outcome.

The FDA's Complete Response Letter to Lykos Therapeutics (formerly MAPS PBC) in August 2024 rejected MDMA-assisted therapy for PTSD despite two positive Phase 3 trials. The agency cited concerns about functional unblinding, expectancy bias, and study conduct. Compass Pathways' COMP360 psilocybin program for treatment-resistant depression is in Phase 3 but hasn't reported primary results. Usona Institute runs a parallel psilocybin trial. For anxiety specifically, Goldberg and colleagues' pilot of psilocybin for social anxiety (n = 18) showed promising signal but lacks statistical power. The field stands at a genuine inflection point: mechanism data is plausible, Phase 2 efficacy is strong, but the gap between promising results and regulatory approval has widened. The courage here is patience, and following the science means waiting for the evidence base to mature.