Robin Carhart-Harris's first psilocybin brain-imaging experiment

Robin Carhart-Harris’s first psilocybin brain-imaging experiment at Imperial College London produced an unexpected result: rather than simply driving the brain into greater activity, psilocybin reduced activity in a centrally connected network, and the strongest reductions accompanied volunteers’ reported loss of a bounded sense of self. The finding shifted the scientific question from what the drug appeared to intensify toward what it might quiet, linking brain measurements to experiences of ego dissolution and helping shape the renewed study of psychedelic consciousness.

From a research ambition to an experiment

Carhart-Harris first approached David Nutt in 2005, hoping to study psychedelics and dreaming at Bristol. He raised the possibility of funding from Amanda Feilding, but Nutt initially regarded the proposal as unrealistic, in part because Carhart-Harris lacked neuroscience experience. Nutt offered him doctoral study on a more straightforward subject, the effects of MDMA on the serotonin system, with the possibility that psychedelic research might follow. That later opportunity arrived in 2009: after completing his PhD, Carhart-Harris was working in Nutt’s lab with Feilding’s funding and obtained approval from the National Health Service and the Home Office to study psilocybin’s effects on the brain.

Carhart-Harris volunteered to take part first, reasoning that researchers who intended to give the drug to volunteers and scan them should experience the procedure themselves. He also told Nutt that he was anxious and might not be in the best psychological condition for participation. Nutt discouraged him for that reason and because taking part could compromise his objectivity. The first volunteer was instead a colleague, who received psilocybin by injection and then entered an fMRI scanner. The episode established an important distinction between the researcher’s willingness to undergo the procedure and the safeguards needed to conduct the study.

The result that reversed the prediction

Carhart-Harris expected the scans to show increased activity, particularly in areas associated with emotion. He imagined a pattern resembling the dreaming brain. That expectation was consistent with earlier work by Franz Vollenweider, whose different scanning method had indicated that psychedelics stimulated activity, especially in the frontal lobes, associated with executive and other higher cognitive functions.

The first measurements did not fit that prediction. They showed decreases in blood flow, one proxy for brain activity measured by fMRI. The unexpected result raised the possibility of a technical mistake, but a second measure—changes in oxygen consumption, used to identify areas of elevated activity—corroborated the blood-flow data. The convergence mattered: the finding was not based on a single indicator. The team concluded that psilocybin reduced brain activity, with the decline concentrated in the default mode network, a system Carhart-Harris had not previously known much about.

That network had been described by neurologist Marcus Raichle in 2001. It links parts of the cerebral cortex with deeper structures involved in memory and emotion, and is active when a person is not occupied by a specific mental task. Its activity is associated with mind-wandering: daydreaming, reflecting on the self, revisiting the past, imagining the future, and worrying. The scan result thus pointed toward a network involved not only in general cognition but also in the ongoing mental activity through which a person maintains a sense of personal identity.

A measurable correlate of ego dissolution

The most striking finding was the relationship between changes in the default mode network and volunteers’ subjective reports. The steepest drops in network activity correlated with reports of ego dissolution: some participants described no longer experiencing themselves as a distinct person, while others reported that the boundary between self and surroundings had disappeared. The sharper the fall in blood flow and oxygen consumption in the network, the more likely a volunteer was to report losing a clear sense of self.

The study did not establish that a scan could fully explain a psychedelic experience. It did, however, provide a measurable correlate for a change participants described from the inside. The result also gave researchers a way to investigate how a temporary disruption in activity associated with self-reflection might accompany the disappearance of the ordinary subject–world boundary. In accounts of mystical experience, that boundary’s dissolution is often described as a merging with a larger whole; the imaging findings offered a neuroscientific question to pursue without, by themselves, settling what such an experience means.

Connections, later work, and limits

After Carhart-Harris published the results in 2012, Judson Brewer noticed that scans of experienced meditators resembled the psychedelic findings. In his meditation research, reported transcendence of the self also appeared alongside reduced activity in the default mode network. The comparison suggested that distinct practices and states might share a neural correlate, while leaving open whether their subjective qualities or causes were the same. It strengthened the case for studying the network as part of the relationship between selfhood and altered states of consciousness.

The experiment also belonged to a longer history in which researchers had debated how to interpret psychedelic sessions. Humphry Osmond and Abram Hoffer’s work with alcohol-dependent volunteers in Weyburn had led them to question whether experiences that resembled psychosis might instead involve transcendence, and to recognize the importance of an engaged, empathetic therapist and the conditions surrounding a session. Later, clinicians at the Addiction Research Foundation in Toronto attempted to replicate Saskatchewan results in neutral rooms, minimizing interaction and sometimes restraining or blindfolding volunteers. Those trials failed to reproduce the earlier outcomes and included frightening experiences. The contrast underscored that a drug’s measurable effects do not occur apart from the circumstances of its administration; it also cautioned against treating different protocols as interchangeable tests of one simple drug effect.

Subsequent clinical studies extended the revival of psychedelic research but did not turn the brain-imaging result into proof of treatment efficacy. In December 2016, reports on the New York University and Johns Hopkins cancer trials described clinically significant reductions in anxiety and depression for about 80 percent of participants, lasting at least six months; the intensity of reported mystical experience closely tracked symptom improvement. The trials included only eighty people altogether and required larger replication. A separate Imperial College London feasibility study gave psilocybin to twelve people with treatment-resistant depression and had no control group: all initially improved, two-thirds were depression-free after a week, and some benefits persisted, though later follow-up showed that some participants relapsed. These studies supplied encouraging clinical signals, not definitive evidence that the brain-imaging experiment had explained or validated a cure.

The original experiment’s lasting contribution was narrower and more consequential than a claim that psilocybin simply switches the brain on or off. It showed that an unexpected decrease in a network associated with self-focused thought could be measured, independently corroborated, and related to volunteers’ accounts of ego dissolution. By bringing subjective reports and brain imaging into the same inquiry, it made the changing sense of self a testable research question while leaving its deeper interpretation open.