A new neuroimaging study suggests that the psychedelic compound psilocybin does not simply scramble brain communication into random disorder, but instead reorganizes neural activity to align with a person’s immediate environment. Using brain scans and advanced computational modeling, researchers found that under the drug, brain activity formed structured patterns that tracked what participants were doing and experiencing, especially in those who reported a dissolving sense of self. The findings were published in Nature.
Psilocybin is a psychoactive chemical compound found naturally in certain species of mushrooms. After ingestion, the body converts it into psilocin, which binds mainly to serotonin 2A receptors on brain cells, changing how distinct neural networks interact and altering perception, emotion, and self-awareness. Earlier neuroimaging work characterized psychedelic states as predominantly chaotic, marked by a breakdown of coordinated brain rhythms. An influential 2013 study recorded brain signals after psilocybin administration and found widespread drops in coordinated electrical oscillations, suggesting that the drug plunged neural circuits into an uncoordinated state.
Building on these observations, the authors of a 2018 review led by Robin Carhart-Harris, who is also a co-author of the new study, updated what is known as the entropic brain hypothesis, arguing that psychedelics make brain activity more varied and unpredictable, a property known as entropy, and that this makes the brain unusually sensitive to a person’s mindset and physical setting. Later work began to test this idea directly: a study first posted as a preprint in 2020 and later published in a peer-reviewed journal indicated that sensory inputs like music and video alter the diversity of brain signals under psychedelics, showing that external context helps determine neural activity.
Brain imaging has also shown that psychedelics modify how higher regions of the brain communicate with sensory systems. In a study covered by PsyPost in 2024, psilocybin was found to strengthen signals traveling from higher-order thinking regions down to visual areas, providing a biological basis for vivid closed-eye mental imagery. Despite these insights, scientists still lacked an understanding of whether the brain under psychedelics is truly disorganized or if it follows hidden patterns that track real-world experiences as they unfold over time.
The project was led by Devon Stoliker, a Turner Impact research fellow in the Computational Neuroscience Laboratory at Monash University‘s Turner Institute for Brain and Mental Health, and Adeel Razi, a professor of computational neuroscience and director of the laboratory. Stoliker told PsyPost that the project was motivated by a desire to understand how mindset and setting interact with brain function.
“Psychedelic experiences vary considerably with mindset and setting, but how these differences are reflected in brain activity remains unclear,” Stoliker explained. “Brain-imaging studies often describe disrupted networks or increased disorder, leaving a gap in explaining experiences that people find coherent and deeply meaningful. We wanted to understand how brain activity becomes reorganized during psychedelic experiences, how that reorganization is modulated by what a person is doing and experiencing, and how it relates to subsequent psychological change.”
To carry out the project, the research team recruited 65 healthy adults between the ages of 18 and 55 who had no meaningful prior experience with psychedelic substances. Three participants reported minor or remote exposure without noticeable effects, while the remaining participants had never taken a serotonergic psychedelic. Two individuals left the study before drug administration due to protocol exclusions, leaving 63 participants who received a standardized oral dose of 19 milligrams of psilocybin.
Each volunteer took part in two separate testing visits spaced roughly one week apart: a baseline visit with no drug and an active dosing visit. During each visit, participants completed functional magnetic resonance imaging (fMRI), which measures changes in blood oxygenation across the brain, followed by electroencephalography (EEG), which records electrical brain waves through sensors on the scalp. The imaging sessions began about 80 minutes after taking the capsule, while electrical recordings started roughly 150 minutes post-dose, during the drug’s acute effects.
In both the brain scanner and the EEG session, participants moved through four distinct conditions designed to present varying levels of sensory engagement. These conditions included resting quietly with eyes closed, following a guided audio meditation, listening to an emotionally evocative music playlist with eyes closed, and watching a silent video of drifting clouds with eyes open.
The sequence progressed from low-stimulus to high-stimulus contexts in the scanner, while the movie was presented first in the EEG session. Half of the participants had also completed an eight-week mindfulness meditation course before their scans, but differences between meditators and non-meditators were not statistically significant, leading the authors to combine both groups for the primary analyses.
After applying quality checks to exclude scans affected by excessive head movement, the final analyzed functional imaging dataset included 60 participants during rest, 59 during meditation, 56 during music listening, and 57 during movie watching. A balanced group of 54 participants with complete data across all four conditions was used for machine learning models. The final electrical recording dataset included 58 participants per condition during the psilocybin session.
When the authors analyzed overall communication across the brain, they discovered that psilocybin redistributed functional connections depending on whether participants had their eyes open or closed. During eyes-closed conditions, overall connectivity decreased in sensory regions by up to 39 percent while increasing in associative thinking areas by up to 72 percent. During the eyes-open movie condition, by comparison, connectivity increased across both sensory and associative areas. Overall, the differences in brain connectivity between having eyes open and having eyes closed shrank by 85 percent in the visual system after psilocybin, indicating that eyes-closed states began to resemble eyes-open processing.
The traditional measure of brain modularity, which quantifies how cleanly the brain divides itself into separate functional communities, dropped under psilocybin across all four conditions. Modularity decreased during rest with an effect size of minus 0.63, during meditation at minus 0.60, during music at minus 0.70, and during video viewing at minus 0.91. However, these static, time-averaged modularity scores failed to predict how participants actually felt during the session, and correlations with subjective reports were not statistically significant.
“The striking finding was that reduced separation between the usual brain networks, a common finding known as reduced modularity, coexisted with the organization detectable using machine learning,” Stoliker observed. “Only the machine learning methods, which preserve each individual’s brain activity over time, identified an association with subjective effects.”
To look beyond static averages, the researchers used machine learning to map moment-by-moment brain activity into smooth, lower-dimensional geometric trajectories. An automated classifier could identify which condition a participant was engaged in based on their brain trajectories, and its accuracy rose with the intensity of the experience. In a closer analysis of six participants with strong subjective effects, accuracy climbed from roughly 62 percent at baseline to about 98 percent under psilocybin.
“Uniquely, this method measured each individual separately, rather than an average across the group of participants,” Stoliker remarked. “We think the development of similar measures that preserve each individual’s brain activity over time is an important next step, but whether the current methods can help predict or improve treatment outcomes remains to be tested.”
To capture different facets of this neural reorganization, the scientists used a second machine-learning tool, called TAVRNN, that tracks how connections between brain regions shift over time. “The main analysis captured context-aligned patterns in each person’s brain activity over time, while a complementary machine learning analysis, TAVRNN—developed in-house by co-author Moein Khajehnejad—showed greater cohesion in the connectivity within and across networks,” the researchers pointed out. “Both analyses showed that organization varied with the strength of subjective effects and both showed that the apparent disorder of the psychedelic state contained structure that became visible when we examined brain activity differently.”
This organizational structure tracked the emotional and psychological quality of each participant’s experience. Higher classification accuracy correlated strongly with overall scores on a questionnaire measuring mystical-type experiences (r = 0.64), while correlations with anxiety or impaired thinking were low and not statistically significant. Among specific kinds of experience, accuracy tracked most closely with positively felt ones, such as bliss, unity, and spiritual feelings. Participants who experienced strong feelings of unity showed tightly packed neural patterns aligned with each condition, a pattern the authors call context alignment; they reserve a separate term, embeddedness, for the felt experience that accompanied it.
“Psychedelics may provide a window into how the brain ordinarily maintains the experience of being a separate self within the world,” Stoliker elaborated. “We use the term embeddedness to describe the felt continuity with a broader physical and psychological context that some participants reported. Embeddedness is the phenomenological correlate of context alignment, thus our findings suggest that this change in experience has a measurable counterpart in how brain activity is organized.”
A perturbation analysis indicated that this context-aligned organization depended most heavily on two major brain networks: the default mode network, which typically supports internal self-reflection, and the visual network, which processes external visual inputs. In those six participants, when researchers mathematically replaced the psilocybin-altered signals in either network with that person’s drug-free data, classification accuracy dropped noticeably, with each network contributing roughly 23 percent to the total accuracy gain.
Subjective questionnaires confirmed the profound psychological impact of the session. More than half of the participants rated the experience among the most meaningful of their lives, with 24 placing it in their top five, and most rated the overall intensity at nine or ten out of ten. Participants also completed follow-up assessments one day later, showing positive mindset improvements related to inner peace, patience, and feelings of connection to nature and other people. Mindset change was most strongly associated with feelings of insightfulness during the session (r = 0.65), followed by mystical, blissful, and spiritual experiences.
“The authors acquired neuroimaging data from individuals at baseline and after a dose of psilocybin, the serotonergic psychedelic contained in ‘magic mushrooms’ of the Psilocybe genus,” said Enzo Tagliazucchi, a researcher at the University of Buenos Aires who was not involved in the study. “Instead of obtaining neuroimaging data just during rest, as is most frequent in this type of study, they acquired the data under different conditions, including music and movie viewing, and meditation. There are multiple results included in this paper, but in my opinion the most salient result is the psilocybin-induced enhancement of context sensitivity.”
“In other words, under psilocybin the neuroimaging data allowed for a much clearer distinction between the different experimental conditions, relative to the baseline,” Tagliazucchi told PsyPost. “Moreover, the authors linked this sensitivity to the occurrence of mystical-type experiences under the drug. This experience is interesting from the perspective of basic and clinical neuroscience, as it can predict symptom improvement after psychedelic therapy.”
The findings are in line with a 2024 study covered by PsyPost, which found that psilocybin causes brain networks to shift into temporary states of heightened communication that predict feelings of ego dissolution. The results also align with an international brain imaging analysis covered by PsyPost in 2026, which found that psychedelics consistently bridge the gap between sensory and associative systems, though that study found weak evidence for within-network breakdown while the current work observed drops in within-network connectivity.
In addition, the findings offer a different emphasis from research covered by PsyPost in May 2026, which linked increased brain entropy under psilocybin to later insight and well-being. The new study also found signs of higher entropy, but argues that this apparent disorder contains structured organization aligned with context.
Tagliazucchi also highlighted how the results fit with longstanding concepts in psychedelic science. “The findings in this study align with the context-dependent effects of psychedelics,” he noted. “Today, we are familiar with terms such as ‘set’ and ‘setting,’ coined decades ago to represent the mental and environmental aspects that can influence a psychedelic trip, respectively. From this perspective, the authors found a neural signature of this sensitivity to context; that is, a much easier distinction between the ‘settings’ explored in the study during the psychedelic state compared to the baseline.”
The authors note that earlier psychedelic imaging studies often relied on small samples of experienced users. A second outside expert highlighted the size of the new dataset.
“Psilocybin can produce a rich range of subjective experiences, such as changes in perception to shifts in mood and sense of self,” said Alex Kwan, a professor in the Meinig School of Biomedical Engineering at Cornell University and head of the Kwan Lab, who was not involved in the study. “What makes this study exciting is that it begins to connect these different aspects of the psychedelic experience to distinct patterns of brain activity. One strength of this study is its relatively large sample size, drawing on brain imaging data from more than 60 people. That is a substantial dataset for this type of psychedelic research.”
There are a few things to keep in mind when interpreting these findings. All participants knew they were receiving psilocybin, as the study used an open-label design without an active placebo group. Although an active placebo that convincingly mimics a psychedelic trip is difficult to implement, expectations could have influenced participants’ subjective reports and emotional responses. The four conditions were also presented in a fixed order during imaging to prioritize safety and comfort for first-time participants.
“I think the authors did an excellent job in several aspects, including data collection, participant number, and computational analyses,” Tagliazucchi evaluated. “The latter, especially, are a ‘tour de force’ in neuroimaging data analysis.”
“My only concern relates to the experimental design,” he added. “This study did not include a placebo condition, and the order of the conditions was fixed across all participants. Therefore, the authors’ analysis might be capturing the differential temporal effects of psilocybin on factors such as vigilance, arousal, attention, or even the temporal evolution of the subjective intensity of the effects, which spans several hours.”
For their part, the authors note that context-aligned organization was already strong during rest, the very first condition scanned, before order effects or fatigue could accumulate. In addition, the types of stimuli presented during the scan sessions represent an important consideration for understanding what context meant in this experiment.
“It would be important to distinguish ‘context’ in a broad sense from a more fine grained definition of ‘context’ which implicates the specific contents of consciousness present at any given time,” Tagliazucchi explained. “The authors used visual stimuli composed by slowly drifting clouds, this stimulus is interesting for inducing a certain global ‘setting,’ but it lacks semantic content. The same applies to the auditory stimulation, which comprised slow, ambient music but lacked strong temporal dynamics. A new study would be required to test whether the context-sensitivity also extends to different forms of conscious content, such as faces, objects, or complex situations like those seen in movies.”
Kwan told PsyPost that the physical environment of brain imaging introduces its own unique influences. “Context is indeed important,” Kwan pointed out. “For this study, we should keep in mind that the participants are lying inside a large and loud brain scanner while they are under the influence of a psychedelic drug. That itself is an unusual context that can affect the results and the qualities of the subjective experience.”
In addition, the participants were healthy volunteers who were generally curious and willing to experience an altered state in an imaging scanner, meaning the findings might not generalize directly to clinical populations undergoing psychedelic-assisted therapy for specific psychiatric disorders. Future investigations will need to examine whether similar patterns of brain reorganization occur in individuals being treated for depression, anxiety, or addiction.
The authors also highlighted important caveats regarding how brain scans are interpreted. “‘Organization’ here is inferred from the structure of patterns observed in the machine learning results derived from the fMRI signal. fMRI measures changes in blood oxygenation associated with neural activity, rather than recording neuronal activity directly,” Stoliker cautioned. “Under psilocybin, this context-aligned structure became more detectable, particularly in participants reporting stronger positive subjective effects, but not in those reporting negative subjective effects or just visual effects—the profound changes to self and boundary, a reduced sense of separation, or continuity with the internal and externally oriented environment, seemed to occur with the emergence of the organization we observed.”
“We want to identify the conditions that make this organization more likely, and whether changing those conditions improves outcomes,” Stoliker said. “This means studying individual characteristics, expectations and preparation before dosing, psychological state around administration, and context and support during the session. Behavioral and brain measurements prior to dosing could help establish which factors are modifiable and whether preparation should differ between people. The clinical goal is to determine whether tailoring these conditions can support more reliable and sustained benefit.”
The study, “Psychedelics align brain activity with context,” was authored by Devon Stoliker, Leonardo Novelli, Moein Khajehnejad, Mana Biabani, Matthew D. Greaves, Tamrin Barta, Martin Williams, Sidhant Chopra, Olivier Bazin, Otto Simonsson, Richard Chambers, Frederick S. Barrett, Gustavo Deco, Katrin H. Preller, Robin L. Carhart-Harris, Anil K. Seth, Suresh Sundram, Gary F. Egan, and Adeel Razi.
Leave a comment
You must be logged in to post a comment.