Synthetic Mysticism: When Technology Can Create Awe, Unity and Revelation

In 2005, a research group led by Pehr Granqvist at Uppsala University published a paper whose title doubles as its conclusion: sensed presence and mystical experiences are predicted by suggestibility, not by the application of transcranial weak complex magnetic fields. For nearly two decades, Michael Persinger’s laboratory had reported that weak, complex magnetic fields applied across the temporal lobes could evoke the felt presence of another being in as many as eight out of ten people. Journalists had already nicknamed the apparatus the “God helmet,” and a consumer version was on sale online. Granqvist’s team built comparable hardware, ran eighty-nine volunteers through a double-blind procedure with a sham-field control, and found no effect of the fields on sensed presence, mystical experience, or the somatosensory oddities the earlier studies had described. What did predict who reported what was the participants’ own suggestibility: absorption, indicators of temporal-lobe lability, and a New Age orientation toward life. Persinger disputed the replication on methodological grounds, and the argument never fully closed. The episode remains the cleanest illustration available of the distance between what a laboratory demonstration establishes and what a headline about it claims.

The question this article takes up is what it would mean if technology could reliably produce awe, a sense of unity, or the feeling that someone is present. Answering it means holding three different kinds of claim apart, because almost every confusion in this debate comes from sliding upward between them.

Demonstrated science: controlled studies have shown that awe has measurable neural and psychological correlates, and that one drug, in one carefully managed setting, can occasion experiences that volunteers describe as mystical and rank among the most significant of their lives.

Plausible engineering: immersive environments, closed-loop stimulation, or pharmacology could credibly be designed to produce components of those states on demand, with different reliability and different risks.

Speculation: that such systems could manufacture revelation, replace religious community, or settle whether a claimed spiritual experience corresponds to anything beyond the brain that produced it.

Popular writing tends to treat the third as though it followed from the first, or the first as though it were irrelevant to the third. Both moves are shortcuts. The interesting work sits in the middle register, where the engineering is real enough to build and the meaning of what it produces is still undecided.

The helmet that promised God

Persinger’s claim was never modest. His laboratory proposed that weak, physiologically patterned magnetic fields applied across the temporal lobes produced transient microseizure-like activity, and that this activity could account for a whole family of anomalous experiences: hauntings, phantom presences, and the felt nearness of a divine other. The reported hit rate of up to 80% traveled from conference talks into newspapers and then into popular memory, where it hardened into a claim nobody had made precisely: that scientists had found the switch for religious experience.

Granqvist’s null result did not show that no technique can induce a sensed presence. It showed something narrower and more useful. Under these conditions, with this equipment, those fields did not produce the effect. The experiences people reported clustered instead around a stable personality disposition toward suggestion and absorption, which is exactly what you would expect if the experimental setting itself — a dark chamber, a reclining subject, a strange helmet, an expectation of something uncanny — supplied most of the effect. The lesson is methodological before it is metaphysical. Any instrument that promises a spiritual state has to survive a blind test against the ordinary power of expectation.

What awe actually is

Keltner and Haidt’s 2003 analysis remains the standard starting point, and its most useful feature is that it refuses to treat awe as one thing. They identify two core appraisals. The first is perceived vastness: something in the world exceeds the scale the observer normally handles. The second is a need for accommodation, which is their term for the mind having to revise its existing categories because they cannot absorb what is happening. Vastness alone is not awe; a huge boring wall does not do it. Vastness plus the failure of one’s mental furniture to fit is awe, and that failure is why awe feels like being taken apart and put back together.

Around those two core appraisals they place five additional ones — threat, beauty, exceptional ability, virtue, and supernatural causation — and these explain why the family of awe experiences runs from terror at a tornado to reverence at a cathedral. Awe induced by an overwhelming threat and awe induced by a Mozart mass share a structure and diverge sharply in what they do to a person afterward.

This taxonomy matters for anyone building the technology. Engineering awe means choosing a branch of the family. A system optimized to maximize the intensity of the accommodation response is not automatically producing something benign, because the same mechanism that opens a person to a new frame also loosens the grip of the old one, and it does so without any guarantee about what gets installed in the gap.

Watching awe in the scanner

The clearest neural finding to date comes from van Elk and colleagues, who showed participants awe-eliciting videos and control material inside an fMRI scanner. During reported feelings of awe, activity dropped in several regions of the default mode network — the frontal pole, the angular gyrus, and the posterior cingulate cortex among them. This network is heavily involved in self-referential thought, in the ongoing narration of “me and my situation.” Participants also described themselves as smaller relative to what they were seeing.

Read this for what it is. It is a correlation between blood-oxygen signals and self-reported experience, elicited by short videos, in a laboratory, in a small sample. It does not isolate awe from related states such as absorption or surprise, and it certainly does not adjudicate what awe means. What it does establish is that the phenomenology of self-diminishment tracks something measurable, which is a real result and a modest one.

The shrinking self and what it does to behavior

The behavioral side is where awe gets interesting for anyone thinking about influence. Piff and colleagues ran five studies with 2,078 participants and found a consistent pattern: people who are more disposed to awe behave more generously in economic games, and induced awe increases ethical decision-making, generosity, and prosocial values relative to control states. In their most striking condition, participants simply stood in a grove of towering eucalyptus trees, and that alone increased helping behavior and reduced entitlement. Statistical mediation suggested that a diminished sense of self carried the effect.

The caveats are ordinary and substantial. Effects measured in a dictator game or a short post-video window are not the same as a life. These were single-session studies. Prosocial intentions are easier to produce than prosocial conduct sustained under pressure. Keep all of that. And notice the asymmetry it implies: if awe reliably shrinks the self and orients people toward something larger, then a technology that can generate awe on demand is a technology that can move moral behavior in directions its operators may not advertise.

What the psilocybin studies actually established

The most cited evidence for chemically occasioned mystical experience comes from Roland Griffiths and colleagues at Johns Hopkins. In their 2006 study, thirty-six volunteers completed a double-blind design in which they received psilocybin at 30 mg per 70 kg of body weight or the active control methylphenidate at 40 mg per 70 kg. The volunteers were hallucinogen-naïve, screened in advance, and recruited because they reported regular participation in religious or spiritual activities. The setting was deliberately supportive: a comfortable room, two monitors present throughout, eyeshades, music, and an instruction to turn attention inward.

At a fourteen-month follow-up, 58% of the volunteers rated the psilocybin session as one of the five most personally meaningful experiences of their lives and 67% placed it among the five most spiritually significant. Sixty-four percent reported an increase in well-being or life satisfaction. The degree of mystical experience measured on the day of the session predicted those ratings more than a year later, which is the part of the finding that resists cynicism: something happened during the session that carried forward.

What the study is: a well-controlled demonstration that a molecule, under conditions designed to make a mystical experience more likely, can occasion one that meets an operational definition drawn from a validated questionnaire, with effects that subjects rate as highly significant and that persist in memory. What it is not: proof that the content of any such experience is true, a demonstration that the state is generally available, or evidence that the transformation lasts. The screening criteria, the preparation, the music, the two seated monitors, and the follow-up interviews are not incidental packaging. They are part of the intervention, and removing them removes the phenomenon with them. That is a claim about dosage, setting, and care, not about metaphysics.

Engineering transcendence in a headset

Virtual reality has now pushed into the same territory from an entirely different direction. Glowacki and colleagues built a distributed multi-person VR experience called Isness-D, in which participants occupied a shared abstract space together, and then measured outcomes with instruments borrowed from psychedelic research: the Mystical Experience Questionnaire, an ego-dissolution inventory, a communitas scale, and a measure of how far participants included their community in their sense of self.

With fifty-eight participants, they reported that scores on some of these factors were statistically indistinguishable from values published in psychedelic studies. The claim is carefully bounded, and it needs to stay bounded. Fifty-eight people is a small sample. The design was citizen-science rather than a randomized clinical trial. Every outcome was self-report. Commentators at the time noted that the study measured experiences rather than consequences: it did not track mental wellness or clinical benefit, and durability was not assessed at all.

Even so, the result is worth pausing over. If a headset and a shared virtual space can move ego-boundary and connectedness measures into the range usually associated with psilocybin, then the phenomenology of self-transcendence is far more reachable through ordinary physics than most people assume. It also means that our best instruments for claiming to have produced it are questionnaires, which is a weak foundation for a strong claim.

A felt presence is not a person

There is a parallel line of evidence on the specific experience of someone being there. In a 2006 report in Nature, Arzy and colleagues described an epilepsy patient undergoing invasive presurgical evaluation whose left temporoparietal junction was stimulated electrically. The stimulation repeatedly produced a felt presence positioned behind the patient, a “shadow person” that mirrored the patient’s own posture and movements.

This is an n-of-1 invasive study, and no one should generalize it to a theory of religion. Its value is that it demonstrates a specific point cleanly: the sense that another being is present has an implementation in the brain and can be triggered by a current. A felt presence is a reportable state with a mechanism. Whether any particular presence that anyone has ever felt corresponds to an actual other being is an entirely separate question, and stimulation studies have nothing to say about it.

The interpretive gap

Here is the argument that the rest of this article rests on. Establishing how an experience arises tells you about the machinery that produced it. It tells you nothing about whether the content of that experience is veridical. A vivid dream has neural correlates; that establishes that the dream had a cause in the dreamer, and it settles nothing about the events the dream described. Applying the same reasoning to awe, unity, or a sensed presence yields the same conclusion. Mechanism and truth are different questions that happen to share a subject matter.

The reflex reply, that such experiences are “just neurons,” fails in the opposite direction. Every experience anyone has ever had depends on neurons, including the most ordinary and most trusted ones. Recognizing that a sunset is processed by a visual system does not demote sunsets. Substrate is the condition of any experience, not a ranking of experiences, and using “it’s biological” as a disqualifier would disqualify everything a human being has ever known.

So the honest position is narrower than either side wants it to be. Demonstrating a mechanism neither proves nor disproves the interpretation a person places on an induced experience. What it changes is the epistemic situation around that experience. An experience that can be produced on demand becomes repeatable, and repeatability makes claims about it testable in ways spontaneous reports never were. An experience that can be produced deliberately also becomes something that can be sold, prescribed, weaponized, or required as a condition of a job, which raises questions no previous generation of contemplatives had to answer.

How traditions already handle claims of experience

The traditions that care most about mystical experience have never treated intensity as the criterion, which is a point frequently lost in these discussions. Christian, Jewish, and Islamic literature developed procedures for testing claimed revelations, and the tests are consistently about consequences and coherence rather than the force of the experience itself. Does the life change in the direction the tradition calls good, over years rather than hours? Does the content cohere with what the tradition already holds, or does it conveniently license what the person wanted? Does a community that knows the person recognize the change? Does the claim connect to anything that can be checked independently?

The reasoning behind that caution is easy to miss. Intensity is precisely what a counterfeit would imitate. If the only test were how overwhelmed you felt, then any sufficiently powerful technique would pass, and the more powerful the technique, the more reliably it would pass. Some traditions went further and grew suspicious of techniques designed to force the state — fasting to the edge of hallucination, sleep deprivation, intoxicants — not because the states were considered false, but because the technique could quietly become the point. This is the framework, and it answers the question “if we could induce it, what then?” more convincingly than most contemporary commentary: the traditions would ask what the induced state did to the person’s conduct, not how convincing it was.

Consenting to a changed self

The design questions that follow are unglamorous and decisive. Who sets the objective function? A system optimizing for a user’s reflective goals is a different moral object from one optimizing for dwell time, purchase rate, or willingness to accept a political claim. Is the state reversible? Does the person return to baseline when the intervention ends, and can they tell whether they have? Is consent given inside the state or outside it? A person in the middle of an awe episode has had their frame of reference loosened by definition, which is the least reliable moment to decide whether to continue. Can the change be reaffirmed later, by the person under ordinary conditions, with the mechanism switched off?

Then there is the asymmetry that should worry anyone building this commercially. Awe is a state in which the mind’s existing categories are being revised, so a claim encountered in that state gets a hearing it would not get otherwise. That is an inference from the structure of accommodation rather than a measured result, but it is the reason the engineering of awe cannot be separated from the ethics of persuasion. The capacity is interesting to contemplatives, to clinicians, and to anyone who wants something from you.

The question worth carrying

The capability is likely to arrive in some form, partial and imperfect, long before anyone settles what it means. The two popular responses are both premature. One says that if awe and unity can be induced, the religious experiences people have reported for millennia are explained away. The other says that no machine could ever produce the real thing. Neither follows from the evidence. Inducing a state and identifying its mechanism leaves the question of that state’s meaning exactly where it was.

What would actually be worth watching is whether engineered experiences start passing the tests the traditions invented for spontaneous ones. Sustained change in conduct over years. Coherence that constrains rather than flatters. Recognition by a community that is not paid to be impressed. Those are far higher bars than a score on a questionnaire, and they are the standards under which most claimed revelation has historically been rejected rather than accepted. A technology that could clear them would be something genuinely new. A technology that produces overwhelming encounters and calls that success would be repeating a very old mistake with better hardware.