About 45 percent of people who smoke a sufficient dose of DMT report the same strange thing: they are no longer alone. They describe a tunnel, a portal, sometimes a kind of waiting room, and then the beings. Machine elves, insectoid creatures, jesters, deities, entities that seem to manufacture impossibly complex objects in midair as if showing off. Humans, oddly, almost never turn up.
For most of the past century, science has had one answer to all this. Hallucination. Elaborate, sure, but ultimately the brain talking to itself.
A new collaboration between two research outfits, the Trace Institute and the non-profit Noonautics, wants to put that answer to the test rather than simply assume it. The idea sounds almost reckless when you first hear it: build a mathematical framework precise enough to ask whether some of the entities people meet on DMT might, in some sense, be real, then design experiments that could actually falsify the claim. Behind the project are Donald Hoffman, the University of California, Irvine psychologist known for arguing that perception is a user interface rather than a window onto truth, and Andrew Gallimore, a neurobiologist at the Okinawa Institute of Science and Technology who co-invented the technique that makes the whole thing feasible.
The framework they are leaning on is called conscious realism, and it is a properly strange piece of metaphysics. It abandons the assumption that there are mind-independent physical objects at all.
Instead it proposes that reality consists of, and only of, interacting conscious agents, and that space, time, and the objects in them are something like icons on a desktop: useful, but not the underlying code. Each agent perceives only a thin slice of this vast network, the slice evolution tuned it to notice. Everything else is filtered out as noise, or not registered at all. The maths gets dense quickly. An agent’s experience is modelled as transitions through a space of possible states, governed by what the authors call a qualia kernel, and even a modest ten-state version of such a space balloons into a 90-dimensional object with ten billion vertices.
Beyond the basin
In their preprint, Traces of the Other, Gallimore, Hoffman and co-author Niffe Hermansson lay out the central move. Normally we sit inside a stable basin they call the Consensus Reality Space, the shared, fitness-shaped region where most of us spend our waking lives.
Their hypothesis is that DMT is an unusually robust shove. It does not, on this view, simply add a few hallucinated icons to the existing interface; it restructures the interface itself, pushing consciousness into regions of the experience landscape with entirely different dynamics, where the evolved constraints no longer apply. And in those regions, the influence of conscious agents we cannot normally perceive might leave perceptible “traces,” rendered by the altered mind as stable, coherent, meaningful, and apparently autonomous beings. Hence the title. Hence the machine elves. The authors are careful to stress that this is a hypothesis, not a claim, and that the boring hallucination explanation remains the most parsimonious default.
What makes the idea more than armchair speculation is a wrinkle in the phenomenology itself. If DMT visions were just the brain freewheeling, you would expect them to be stocked with the things brains know, namely people and animals, which is exactly what dreams and psychotic hallucinations tend to serve up.
But the DMT state inverts this. Humans appear less than 5 percent of the time, while the worlds fill instead with beings and geometries that have no obvious referent anywhere in human evolutionary history. Users report extra spatial dimensions, non-Euclidean curvature, the unnerving sense of viewing a tesseract turning in 4D.
It is here that the standard account looks, at the very least, incomplete. Why should a brain freed from top-down control so reliably conjure the alien rather than the familiar?
The honest answer is that nobody knows yet, and the authors do not pretend otherwise. There is a complementary thread of work from Andrés Gómez-Emilsson at the Qualia Research Institute, who has argued that DMT drives experience into a regime of such energetic excitation that phenomenal space reorganises into hyperbolic, higher-dimensional forms. Whether that maps onto Hoffman’s conscious agents or simply rhymes with them remains an open question.
Doing science inside the trip
The part that should make even sceptics sit up is that the team thinks this can be tested. The breakthrough enabling it is something Gallimore developed with Rick Strassman: extended-state DMT, or DMTx, a target-controlled intravenous infusion that holds a steady concentration in the brain and stretches the usual five-minute burst into a stable hour or more, with the option to pause and resume by dialling the drip up or down. Suddenly there is time to do science in there. The proposed experiments are ingenious in their simplicity. In one, a computer locked in a separate room flips a screen randomly between blue and yellow; if an entity is genuinely external, the thinking goes, it might be able to report the colour it has no ordinary way of knowing. In another, two isolated volunteers attempt to pass a random word or number to each other through the same entity, a kind of psychedelic dead drop. Null results, the authors note, would count.
There are obvious limitations, and the paper does not hide from them. The colour experiment needs the entity to cooperate, which is hardly guaranteed, and so the team floats recruiting the rare individuals who claim ongoing relationships with the same being across many sessions. Everything ultimately rests on subjective report, expectation effects loom large, and a positive result would demand cautious interpretation rather than triumphant headlines.
Still, the ambition is bracing. “With a theoretical foundation for the highly unusual state of consciousness induced by DMT, we can test these theories experimentally,” says Gallimore. Hoffman frames the work as a way of “exploring the effects of psychoactive substances such as DMT on the structure and function of spacetime,” which is not a sentence you read every day in a press release. The two will discuss it all publicly in Venice Beach in mid-June, with video to follow.
Whether any of this survives contact with a blue-and-yellow screen is anyone’s guess. But the framing is what lingers. Gallimore calls the collaboration “a first step to a mathematics of altered states of consciousness and, ultimately, for engineering our perceptual interface to expand our view of reality,” and even if the entities turn out to be nothing but the brain’s own machinery after all, the question has finally been put in a form that an experiment can answer.
DOI / Source: https://doi.org/10.31234/osf.io/8qvgy_v2
Frequently Asked Questions
Why do people on DMT meet aliens but almost never ordinary humans?
That contrast is exactly what makes researchers uneasy with the simple hallucination story. Dreams and psychotic hallucinations tend to recycle familiar people and animals, because that is what the brain has stored, yet humans show up in fewer than 5 percent of DMT experiences while bizarre non-human entities dominate. Explaining why a freewheeling brain would so reliably produce the alien rather than the familiar is still an open problem.
Is anyone actually claiming DMT entities are real beings?
No, and the researchers are careful about this. Their preprint treats external conscious agents as a hypothesis to be tested, not a conclusion, and explicitly keeps the conventional hallucination explanation as the default. The point is to put the question in a form precise enough that an experiment could settle it either way.
How could you possibly test whether a hallucination is real?
The proposed experiments look for information a purely internal vision should not have access to. One design has a computer in a sealed room randomly showing blue or yellow, to see whether an entity can report the hidden colour; another tries to pass a random word between two isolated volunteers through the same entity. If nothing correlates beyond chance, that counts as evidence against the external-agent idea.
What is extended-state DMT and why does it matter here?
Ordinary DMT trips last only a few minutes, far too short for controlled experiments. Extended-state DMT, or DMTx, uses a continuous intravenous infusion to hold the experience steady for an hour or more, and even lets researchers pause and resume it by adjusting the dose. That sustained, controllable window is what makes structured testing inside the state conceivable at all.
ScienceBlog.com has no paywalls, no sponsored content, and no agenda beyond getting the science right. Every story here is written to inform, not to impress an advertiser or push a point of view.
Good science journalism takes time — reading the papers, checking the claims, finding researchers who can put findings in context. We do that work because we think it matters.
If you find this site useful, consider supporting it with a donation. Even a few dollars a month helps keep the coverage independent and free for everyone.