In 2012, a small team of researchers at Imperial College London placed volunteers inside an MRI scanner and gave them an intravenous dose of psilocybin. The participants expected the drug to increase activity across the brain — to light it up, as the cultural shorthand suggested. The data showed almost the opposite. Specific regions, deeply interconnected and metabolically expensive, went quieter. The most striking decreases clustered in a network that had only been formally named a decade earlier: the Default Mode Network.
That paper, and the dozens that followed, did not prove that psilocybin “expands consciousness.” It did not validate any spiritual claim. What it did was give neuroscience a tractable, measurable place to look. For the first time, the subjective reports of psychedelic experiences — ego dissolution, timelessness, unity — had a candidate neural correlate. The story that emerged is more nuanced than the headlines, and the limits of the evidence matter as much as the findings.
This article explains what the Default Mode Network is, what psilocybin appears to do to it, and what the current research can and cannot tell us. We end with a section on what to read next if you want to understand this literature beyond the press releases.
What Is the Default Mode Network?
The Default Mode Network — usually shortened to DMN — is not a single brain region. It is a coordinated set of regions that became famous because of a counterintuitive observation. In the late 1990s, neuroimaging researchers studying focused tasks noticed that certain brain areas reliably decreased their activity whenever a participant started concentrating on something external. When the task ended and the person was simply lying in the scanner with nothing to do, those same regions came back online.
By 2001, neuroscientist Marcus Raichle had named this pattern the brain’s “default mode.” The regions involved — the medial prefrontal cortex, the posterior cingulate cortex, the precuneus, and parts of the inferior parietal lobe — were always doing something even when the rest of the brain was at rest. They were not idle. They were busy with a different kind of activity entirely.
Subsequent research has linked DMN activity to a cluster of mental processes that share one quality: they involve self-referential thinking. Mind-wandering, autobiographical memory, mental time travel, theory of mind (thinking about what others think), moral reasoning, and the construction of a continuous narrative self all show DMN engagement. When you are not focused on a specific task, your brain runs a kind of identity simulation in the background, and the DMN is the hardware that simulation runs on.
This matters because much of what we call ordinary consciousness is built on top of that self-referential narrative. The DMN is, in some practical sense, where the feeling of being a continuous “I” gets generated.
How Psilocybin Changes the Picture
Psilocybin is the prodrug of psilocin, a serotonin 2A receptor agonist. Receptor pharmacology, however, only tells you what the drug binds to. It does not tell you what happens to coordinated activity at the network level. That requires neuroimaging — and the imaging studies are where the DMN story begins.
The first major paper, Carhart-Harris et al. (2012) in PNAS, found that intravenous psilocybin produced significant decreases in cerebral blood flow in regions of the DMN, particularly the medial prefrontal cortex and posterior cingulate cortex. The decrease in DMN activity correlated with the intensity of subjective effects participants reported. Larger decreases meant larger reported changes in consciousness.
Later studies refined this picture. Psilocybin appears to do two things at once. First, it reduces the integration within the DMN — the regions that normally fire together become less coordinated with each other. Second, it increases communication between networks that normally stay relatively separate. The brain becomes, briefly, less segregated. The technical phrase researchers use is “increased global functional connectivity.”
This dual effect — local disintegration, global integration — is now considered a candidate mechanism for the so-called “ego dissolution” that participants frequently report. If the DMN supports the construction of the self, and psilocybin reduces its coherence, the felt sense of being a bounded individual may temporarily loosen. Activity that would normally have been routed through the self-narrative gets dispersed across regions that do not normally talk to each other. The result, subjectively, is described by participants as a profound but temporary change in the sense of self.
What the Evidence Can Support — and What It Cannot
It is tempting to read those findings and conclude that we have a complete neurobiological account of mystical experience. We do not. The evidence supports several careful claims and rules out very few alternative interpretations.
What the evidence supports:
- Psilocybin produces measurable, reproducible decreases in DMN activity in human subjects across multiple independent studies.
- The magnitude of those decreases tracks, statistically, with the magnitude of subjective effects on the Mystical Experience Questionnaire and similar instruments.
- The brain shifts toward a more globally integrated state under psilocybin, then returns to baseline as the drug clears.
- These effects appear in healthy volunteers and in clinical populations alike.
What the evidence does not support — at least not yet:
- That decreased DMN activity causes the subjective experience, rather than accompanying it. Correlation is not causation, and fMRI cannot establish causal direction.
- That a single mechanism explains the wide variation in subjective reports. Two people with similar DMN patterns can describe wildly different experiences.
- That the DMN changes explain the lasting clinical or psychological effects that some people report weeks or months later. The acute neural signature and the long-term outcomes are studied separately, and the connection between them is still hypothetical.
- That what is seen in 30-person fMRI studies will generalize cleanly to the general population, or to specific clinical conditions.
A useful rule of thumb when reading any neuroscience paper: the strength of the claim about brain activity is usually well-supported. The strength of the claim about what that activity means subjectively is often weaker than the press release suggests.
The REBUS Hypothesis: A Theoretical Frame
Karl Friston and Robin Carhart-Harris published an influential theoretical paper in 2019 proposing a framework called REBUS — “Relaxed Beliefs Under Psychedelics.” The idea is rooted in the predictive coding model of the brain.
In predictive coding theory, the brain constantly generates predictions about incoming sensory data. When predictions match, you experience an ordinary, expected world. When they mismatch, the brain updates the model. Over a lifetime, certain predictions — about self, identity, social roles, emotional reactions — become extremely entrenched. They are what predictive coding researchers call “priors.” Priors save energy. They also constrain what experiences are possible.
The REBUS hypothesis proposes that psychedelics like psilocybin temporarily relax those high-level priors. Decreased DMN activity, in this frame, corresponds to weakened top-down constraints on perception, emotion, and self-modeling. With looser priors, bottom-up sensory and emotional information can be processed in unfamiliar ways. New associations form. Old patterns of self-narrative become temporarily revisable.
REBUS is a hypothesis, not a confirmed mechanism. It fits a wide range of subjective reports and clinical findings, but fitting evidence is different from being uniquely supported by evidence. Other models exist. The thalamic filter model, for example, emphasizes changes in how the thalamus gates sensory information. The neuroplasticity model focuses on the structural changes psilocybin may induce in dendritic spines. None of these models are mutually exclusive. They may be describing different parts of the same elephant.
The honest position is that we have multiple plausible frameworks, each supported by different lines of evidence, and no consensus on which is most complete.
Clinical Implications: Cautious Optimism
The DMN findings have generated significant interest in clinical research, particularly for treatment-resistant depression. The reasoning is straightforward. Many psychiatric conditions involve patterns of rigid, self-focused negative cognition — exactly the kind of activity the DMN seems to support. Major depression, in particular, has been associated with DMN hyperactivity and increased connectivity in functional imaging studies.
If psilocybin temporarily disrupts DMN coherence, perhaps it offers patients a brief window in which entrenched negative self-narratives become more malleable. Combined with psychotherapy in that window, the hypothesis goes, lasting cognitive shifts might be possible.
Several Phase 2 clinical trials have produced encouraging results in treatment-resistant depression, with reductions in depressive symptoms that persist weeks beyond a single dosing session. However, these trials are small, often unblinded (because the drug effects are obvious), and rarely include active comparators that would mimic the subjective intensity of psilocybin. They are not yet sufficient to recommend psilocybin as a general treatment, and the regulatory pathway in most countries remains uncertain.
We cover the specific depression trial literature in a separate article. For the purposes of this piece, the point is narrow: the DMN hypothesis is part of why psilocybin is being studied clinically, but the hypothesis is not the same thing as a proven mechanism of clinical benefit.
What Critics Get Right
Not everyone in the field is persuaded by the DMN story. Several critiques deserve a fair hearing.
The first is methodological. Most psilocybin neuroimaging studies have small sample sizes — often fewer than 20 participants. Effect sizes can be inflated in small samples. Replication has been better than for many areas of neuroscience, but it is still limited. Several null and partially null findings exist that the public-facing narrative tends to underweight.
Second, fMRI itself is an indirect measure. It does not measure neural activity. It measures the blood-oxygen-level-dependent (BOLD) signal, which is a proxy for metabolism, which is a proxy for activity. Conclusions about specific neural events from BOLD data require careful interpretation.
Third, the concept of a “Default Mode Network” is itself debated. Some researchers argue that what we call the DMN is actually multiple overlapping subnetworks with different functions. Lumping them together may obscure more than it reveals.
Fourth, the language of “ego dissolution” is freighted with cultural and even religious meanings that may not map cleanly onto what the brain is doing. Reducing self-referential processing for 90 minutes is not the same thing as a unitive mystical experience, even if participants sometimes describe the latter when given the former.
None of these critiques invalidate the research. They contextualize it. The DMN story is one of the better-supported lines of psychedelic neuroscience, but “better-supported” is a relative claim within a young field.
What This Does Not Prove
A few claims you might encounter elsewhere are not actually supported by current evidence:
- Psilocybin does not “turn off the ego.” The ego, neurologically speaking, is not a single switch.
- The DMN is not the location of consciousness. It is one of several large-scale networks involved in self-related processing.
- Decreased DMN activity is not unique to psychedelics. Deep meditation, certain forms of breathwork, and even some sleep stages show overlapping changes.
- Subjective intensity does not always correlate with clinical benefit. In trials, some participants who report mild experiences improve significantly, and some who report intense experiences do not.
These distinctions matter because they affect what we can responsibly claim about the drug, the experience, and the science.
How to Read the Literature Yourself
If you want to engage with this research directly, three habits will help.
First, prefer original research papers over journalism summaries. The Carhart-Harris group at Imperial, the Johns Hopkins group, the NYU group, and several European labs publish openly. Most of the foundational papers are accessible through PubMed.
Second, pay attention to sample sizes, study designs, and whether the imaging was conducted under the acute effects of the drug, after the experience, or both. The implications differ.
Third, look for the limitations sections. Researchers in this field have been notably honest about what their data cannot show. Those sections are often more informative than the abstracts.
We maintain a curated index of foundational papers in our research library, with confidence ratings and plain-English summaries of what each study can and cannot support.
The Larger Frame
Psilocybin and the Default Mode Network is one of the most popular narratives in contemporary neuroscience, partly because it offers a satisfying bridge between the molecular and the experiential. We can point to a receptor, a network, a connectivity pattern, and a subjective report, and tell a story that connects them. That bridge is real, but it is narrower than the popular discussion implies.
What we know is that psilocybin produces measurable, reproducible changes in large-scale brain network organization, and that those changes track with the intensity of subjective effects in a statistically meaningful way. What we are still learning is what those changes mean, how they relate to lasting outcomes, and whether the DMN frame will hold up as more sophisticated tools become available.
The story is unfinished. That is not a flaw. It is the condition of any active scientific field, and the honest position is to follow the evidence wherever it goes — including, sometimes, away from the most compelling narratives.
This article is part of the Magic Mushroom Institute’s research literacy series. We summarize peer-reviewed neuroscience for general readers without overstating findings. Last reviewed May 2026.