Chapter 15
Borjigin, the Pig EEG, and the Limits of a Headline
In late August of 2013, a paper appeared in the Proceedings of the National Academy of Sciences with a title that sounded like it had been written for cable news.1 “Surge of neurophysiological coherence and connectivity in the dying brain.” The lead author was Jimo Borjigin, a young neuroscientist at the University of Michigan. The methods were modest. Nine rats were anesthetized and then given an injection that stopped the heart. Their brains were monitored continuously by EEG — six channels, surface electrodes — from before the heart stopped through cardiac arrest and into the period of biological death. The team was looking for the pattern of activity in those final seconds.
What they found was unexpected. For roughly thirty seconds after the rats’ hearts stopped, before the brains went finally silent, there was a brief surge of high-frequency electrical activity in the gamma band — the frequency range associated, in awake animals, with focused attention and conscious cognition. Coherence across brain regions actually increased during this surge. Cross-frequency coupling tightened. By a number of measures, the dying rat brain looked, electrically speaking, more “awake” in those last seconds than it had in the seconds before cardiac arrest.2
The paper itself was careful. The authors did not claim to have explained NDEs. They proposed that the gamma surge might constitute a candidate neural correlate of conscious experience during the dying process, that this might be relevant to understanding human reports of vivid mental experience around the time of cardiac arrest, and that further work was needed.3 That was about it. Nine rats. Thirty seconds of surface EEG. A modest hypothesis, modestly stated.
The press release was less careful.
Within hours, the headlines began. The Telegraph: “Near-death experiences may be caused by surge of brain activity.” NBC News: “Scientists may have found the cause of near-death experiences.” Wired: “Hyperactive brain may explain near-death experience.” The Guardian ran a piece headlined, “Near-death experiences explained by science.” A Reuters wire article was picked up in hundreds of papers. By the end of the week, the public takeaway was simple. Scientists had figured out NDEs. They were a final fireworks display in a dying brain. Nothing more.4
I was in the middle of dissertation research when this happened. I remember reading the headlines on a Saturday morning and then, an hour later, sitting down with the actual PNAS paper to see what the fuss was about. The gap between the two documents was startling. The headlines said one thing. The paper said something much smaller.
I also remember the conversations that followed in evangelical circles. A friend in seminary, a thoughtful Christian who had been wrestling with conditional immortality and the question of the soul, sent me a link to one of the news stories. “So that’s it, then,” he wrote. “The brain just does this when it dies. The NDE evidence falls apart.” He was not a glib man. He was reading the headline carefully and drawing what looked like the obvious conclusion. The conclusion was wrong, but I did not blame him for drawing it. Anyone reading the headlines would have drawn it. The headlines were doing exactly what headlines do. They were skipping the qualifications, presenting the strong reading, and giving readers a definite answer to the question that — if they knew the actual data — they would not have been able to give.
That gap is the subject of this chapter.
The dying-brain hypothesis — the proposal that NDEs are produced by a final burst of neural activity as the brain shuts down — is, by a comfortable margin, the most widely cited contemporary physicalist response to NDE evidence. When a Christian physicalist in the conditional-immortality conversation tells you they have a naturalistic answer to the kind of evidence in Chapter 10, Chapter 11, Chapter 12, and Chapter 14, the answer they have in mind is almost always Borjigin’s.5 When a secular skeptic in the comments section of an NDE article says, “That’s just the dying brain — it’s been explained,” this is the work they are gesturing at, even if they have not read it. The hypothesis is doing a lot of work. The work is not, in the main, being done by the data.
I want to be careful here. The Borjigin findings are real. They are interesting. They are honest science by a careful researcher who has continued to publish thoughtful follow-up work, including a 2023 paper extending the rat findings to two of four dying human comatose patients.6 The work matters. What does not match the data is the headline that came with it. And that gap — between what the data show and what the headlines claimed they show — is where the dying-brain hypothesis has its center of gravity. The hypothesis is far weaker than its reputation. It explains a small piece of NDE phenomenology. It does not explain the piece that matters most for the dualism-physicalism debate. It does not, on careful examination, even come close.
This chapter walks through the actual experimental record, follows the hypothesis as it has developed, asks what it can and cannot explain, and shows why the strongest piece of evidence covered in earlier chapters — veridical perception during clinical death — is exactly the evidence Borjigin’s work cannot reach. By the end, I hope the reader will have what I had after that Saturday morning in 2013: a respect for the science and a clear-eyed sense of what it does and does not show.
Before walking into the data, we need to be clear about the burden of proof.
The NDE evidence the previous chapters laid out is not a single curiosity. It is a multi-layered phenomenon. Patients report being conscious during periods of documented brain shutdown. They report perceiving accurate, specific, externally verifiable details of events going on around them and even in distant rooms. People who have been blind from birth report seeing for the first time during their NDEs — and seeing things that turn out to be there. The reports come back in a consistent shape across cultures, decades, ages, and medical contexts. The reports are characteristically more coherent than ordinary memories, not less. And the people who have them are typically transformed by them in long-lasting ways.
An adequate physicalist explanation has to handle every layer of that — not just one. A successful answer cannot stop at “the dying brain produces some kind of experience.” That much was never really in dispute. The question is what kind of experience, with what content, of what quality, with what consistency, and with what relation to the verifiable outside world. The strict physicalist prediction is sharp: when the brain stops, experience stops. That prediction has been falsified across many independent studies and a database of thousands of cases.7 A retreat from strict physicalism to a softer version is allowed; what is not allowed is to declare victory by pointing at residual brain activity and stopping there.
So the burden is not light. To count as a successful answer to the NDE evidence, the dying-brain hypothesis would need to do at least four things.
First, it would need to identify a mechanism in the dying brain that could produce vivid, lucid, organized conscious experience, not just any electrical activity. Gamma frequencies are correlated with attention and cognition in healthy awake brains, but mere oscillation in the gamma range is not consciousness. A seizure produces lots of gamma. So does some background noise. The mechanism has to do more than tick the right frequency box. It has to produce the right kind of experience.
Second, the mechanism would need to scale to the actual reported duration of NDEs. Many NDEs cover what feels to the experiencer like minutes or hours of structured narrative experience — OBE phase, tunnel and light, life review, encounter with deceased relatives or a being of light, return.8 A thirty-second gamma surge is not an obvious match.
Third, the mechanism would need to be present whenever NDEs are reported. NDEs occur not only in cardiac arrest but in deep anesthesia (where gamma activity is actively suppressed by the anesthetic), in barbiturate sedation (likewise), in some non-life-threatening contexts (where the brain is not dying at all), and in some drowning and trauma cases that do not involve the kind of asystolic cardiac arrest Borjigin’s rats experienced.9 If the mechanism is specific to dying-brain physiology, it should not be reported in non-dying contexts. It is.
Fourth, and most importantly for this book’s argument, the mechanism would need to account for veridical content. A patient in cardiac arrest who, on returning, accurately describes the actions of his resuscitation team, the appearance of equipment never previously seen, conversations between staff in adjoining rooms, and clothing colors he had no normal way to know — these are perceptions of specific external facts. To say “the dying brain produced an experience” is not, by itself, an explanation. It is a description of the wrapper. The content inside the wrapper is what calls for explanation. And the content includes information about the world outside the patient’s skull.
Whatever the dying brain does in its last seconds, that activity is local to the brain. The information in a veridical NDE is not local to the brain. The brain’s electrical activity, however vigorous, is not where the patient’s accurate observations of distant rooms come from. A successful physicalist explanation has to bridge that gap. The dying-brain hypothesis does not bridge it. It does not even propose a bridge.
Hold these four conditions in mind. They are the criteria by which we have to evaluate Borjigin’s work, the follow-up human study, and the broader dying-brain literature. The question is not whether the dying brain does anything — of course it does. The question is whether what it does is enough to produce what NDErs actually report. Let’s look.
The 2013 paper, “Surge of neurophysiological coherence and connectivity in the dying brain,” is a careful piece of basic neuroscience.10 Nine adult rats were anesthetized with chloral hydrate and instrumented with six surface EEG electrodes and an ECG lead. After baseline recording, cardiac arrest was induced by an injection of potassium chloride. EEG and ECG were recorded continuously through the dying process. The investigators then analyzed the data for power across frequency bands, coherence between cortical regions, cross-frequency coupling (the tight relationship between, for example, gamma amplitude and theta phase that characterizes some forms of waking cognition), and a number of related measures.
Their finding, in brief, was that the rat brains showed elevated activity across multiple of these measures during a brief window after cardiac arrest. Gamma-band power dropped, then surged. Coherence across cortical regions increased. Cross-frequency coupling tightened. By many measures the brains looked, in those seconds, more cognitively organized than they had during the anesthetized baseline. Then, after roughly thirty seconds, the activity collapsed and the EEG went flat.11
The interpretive move in the discussion was modest. The authors noted that the gamma-band findings “may underlie the heightened conscious processing reported in survivors of cardiac arrest.”12 The word that mattered was may. A bridge was being proposed, not asserted.
I want to take this seriously. The Borjigin team did not produce nothing. They identified a real, replicable physiological phenomenon — a brief electrical surge in dying mammalian brains — that had not been carefully described before. They documented it with rigorous methods. They published it in a top-tier journal. They invited follow-up work. This is what good basic science looks like.
The question that immediately presses, though, is whether the surge can carry the explanatory weight the headlines piled on it. To answer that, you have to read past the surge itself and ask what it can and cannot do.
Three points stand out from the rat data alone.
The surge is brief. Roughly thirty seconds. Many NDEs — particularly the longer cardiac-arrest cases that drive the evidential argument in Chapter 12 — are reported as covering subjectively much longer windows. Some Pam Reynolds–style cases involve experiences subjectively lasting many minutes during clinically documented periods of cardiac arrest extending well past the thirty-second mark. The surge, on its face, does not span what needs spanning.13
The surge is in rats. Rats do not report experiences. We have no way of knowing, from the rat data alone, whether the surge in fact corresponds to any conscious experience at all in the rat. Gamma activity is correlated with consciousness in healthy waking animals. It is also present in some non-conscious states, including REM sleep in animals showing no behavioral signs of awareness, and during certain seizures in humans whose phenomenology is reported as fragmentary or absent.14 The surge is suggestive. It is not definitive.
The surge is local. Whatever it is, it is happening inside the rat’s skull. It is producing an electrical pattern that can be picked up at the scalp. There is no mechanism, in the rat data, by which this localized pattern would correspond to information about events outside the rat — events in adjacent rooms, events on hospital roofs, conversations the rat could not hear, details of medical equipment the rat had not seen. Whatever the surge is, it is a localized brain phenomenon. The veridical content of human NDEs is, by definition, not localized to the patient’s brain. The two phenomena are not in the same explanatory category.
Ten years after the rat paper, in 2023, the Borjigin lab published a follow-up in PNAS.15 This time the subjects were human. Four comatose patients had been removed from ventilator support after consent from family, in a setting where EEG was continued through the dying process as part of an end-of-life monitoring protocol. The team analyzed the EEG data from before and after withdrawal.
The findings were partial. Two of the four patients showed transient gamma-band increases following the withdrawal of life support, with elevated coherence and cross-frequency coupling reminiscent of the patterns seen in the rats. Two of the four did not. The authors offered the careful conclusion that “our results provide evidence that the surges of gamma power and connectivity observed in animal models can also occur in the human brain during the process of dying.”16 They emphasized the small sample, the heterogeneity, and the impossibility of correlating the EEG findings with any subjective experience — the patients did not survive to report.
This is, again, careful science. It also has limits the headline coverage tended to obscure.
Half the patients in the study did not show the surge. If the surge is the universal mechanism behind NDEs, this is a problem — or at least an empirical puzzle. NDEs are reported by something on the order of ten to twenty percent of cardiac-arrest survivors in the major prospective studies; the surge appeared in fifty percent of dying patients in this small sample.17 The two numbers do not match either way. They suggest that the surge and the NDE are not in a tight one-to-one correspondence, even setting aside the question of content.
The patients in the 2023 study were comatose. None were resuscitated. None reported anything. The study cannot tell us whether the gamma surge, when it occurred, was accompanied by any conscious experience at all. It tells us only that, in some dying human brains, the rat-like electrical pattern is found. The bridge from electrical pattern to conscious experience — to any conscious experience, never mind the kind of vivid, narrative, veridical experience NDErs report — is not supplied by this study. It is an inference.
Borjigin herself was clear about this. In interviews and follow-up commentary, she noted that her work “is consistent with the possibility” of NDE-like experience but does not establish it, and that the connection to specific NDE content remains entirely open.18 This is the appropriate scientific posture. It is also a much weaker claim than “NDEs explained.”
The wider literature on the dying mammalian brain has continued to develop, with the Borjigin findings sitting alongside other studies. Two strands are worth naming.
The first is the larger-mammal extension. A research line at Yale, led by Nenad Sestan, demonstrated in 2019 that pig brains, isolated from the body four hours after death, could be partially restored to molecular and cellular function using a perfusion system the team called BrainEx.19 The pig brains showed restored cellular activity, reduced cell death, and limited synaptic activity — though no sign of organized cortical activity that would suggest consciousness. The Sestan work is not a dying-brain study in the Borjigin sense; it is a postmortem-restoration study. But it has been folded into the popular dying-brain narrative as further evidence that “the brain does more after death than we used to think.” That much is fair. What is not fair is to read it as supporting the explanation of NDEs. The Sestan team explicitly stated that their preparation was designed to preclude conscious activity.20
The second strand is the AWARE-II finding on gamma activity during human cardiopulmonary resuscitation, reported by Sam Parnia’s team in 2023.21 The AWARE-II study, which is engaged at length in Chapter 12, included continuous EEG monitoring during in-hospital cardiac arrests in collaborating institutions. In a subset of cases, the team observed transient appearances of gamma and other normal-spectrum frequencies during CPR — sometimes minutes into resuscitation, in patients whose brains had been without normal circulation for extended periods. Parnia and his colleagues framed this finding cautiously, suggesting that it might be evidence of a transient neural activity associated with what they have begun to call “lucid dying” — a recoverable consciousness-related phenomenon in the dying process.22
It is important to be honest about what these findings show and do not show. They show that the dying brain is more electrically active — sometimes — than the older simple picture of immediate flat-EEG-and-nothing assumed. That is genuine new knowledge and worth taking seriously. They do not show that the activity in question is the source of NDE content. The Parnia group, in fact, are themselves substance-dualism-friendly in their interpretation: they read the gamma findings as compatible with the kind of lucid consciousness NDErs report but do not interpret them as a reductionist explanation.23 The same finding, in other words, gets read by Borjigin’s lab as evidence that NDEs may be brain-generated phenomena, and by Parnia’s lab as evidence that the dying brain may sometimes still be hosting the kind of conscious experience that, on the dualist reading, is the soul-experience while the brain is going offline. The data underdetermine the metaphysics.
Gamma activity refers to high-frequency electrical oscillations in the brain, roughly 30–100 Hz. In healthy waking people, gamma is associated with focused attention, perceptual binding (the integration of features into unified objects), and conscious cognition. But gamma is also present in seizures, in some sleep states, and in other contexts where the person is not in a state of clear conscious experience. Gamma is a useful but not definitive marker of consciousness. Saying “gamma equals consciousness” is like saying “movement equals walking” — sometimes yes, sometimes no, and the underlying question is whether the movement is the right kind.
Set aside the data for a moment and look at the proposed mechanism. The dying-brain hypothesis, in its strongest form, runs like this. As the brain’s circulation fails, normal inhibitory neurotransmission collapses. Cortical neurons, freed momentarily from inhibition, fire vigorously and synchronize across regions. This burst of synchronized firing — visible at the scalp as a gamma surge — produces, for a brief window, a state of unusual neural connectedness. In that window, the brain may generate an experience that to the experiencer feels like vivid, expansive, integrated awareness. As the cells lose their resting potentials, the activity collapses, and that is the end.
Borjigin’s commentators have suggested that this could explain a number of features of NDE phenomenology. The vividness might come from gamma. The sense of expansive awareness might come from cross-regional coherence. The life-review element might come from increased memory access during the burst. The ego-dissolution and tunnel imagery might come from the breakdown of normal inhibitory boundaries between brain regions. The apparent encounter with light might come from anomalous activity in visual cortex. Each piece of NDE phenomenology, the story goes, has a candidate neural correlate in the dying-brain pattern.24
This is a serious story. It deserves a serious response. The response, on careful examination, is that the story does some real explanatory work for some features and no explanatory work at all for others — and the features it cannot explain are precisely the features that drive the empirical case for substance dualism.
Before walking into the case against the hypothesis, it is worth pausing on Borjigin’s own framing of her work.
In a 2013 commentary in the Michigan press following the rat paper, Borjigin said, “We reasoned that if near-death experience stems from brain activity, neural correlates of consciousness should be identifiable in humans or animals even after the cessation of cerebral blood flow.” The reasoning is conditional. If NDEs stem from brain activity, then we should look for the neural correlates. The 2013 finding, on this framing, was a candidate correlate — a place to look for the mechanism if the mechanism is to be found in the brain.25
In a 2023 follow-up interview, Borjigin was asked directly whether her work explained NDEs. Her response was, paraphrasing the careful scientist, that the work had identified a phenomenon that could be relevant, that further work was needed, and that “NDEs themselves remain a phenomenon that requires careful interdisciplinary investigation.” She did not claim, and has never claimed, to have explained NDEs. The headlines did. She did not.26
This matters. The hypothesis as advanced by the actual researcher is much weaker than the hypothesis as it appears in the popular and apologetic literature. The version that gets cited in Christian-physicalist arguments and atheist-skeptic arguments alike is a stronger claim than Borjigin herself defends. It is worth the effort to keep these distinct. There is the modest scientific finding, on which Borjigin can be cited approvingly. And there is the strong popular reading, on which she cannot.
So — the data have been laid out. The hypothesis has been described in its strongest form. The actual researcher’s caution has been noted. Now the work of evaluation. There are five places where the dying-brain hypothesis runs into the data and does not survive the contact.
The first and most decisive failure is the one named in the framing section above. The dying brain, on Borjigin’s account, produces a brief surge of activity inside the patient’s skull. Whatever experience is generated by that surge — if any — is generated by neurons firing in the patient’s own cortex.
That mechanism cannot generate accurate information about events outside the patient’s skull.
This is not a subtle point. A neuron firing in your visual cortex does not, by firing, gain access to the contents of the room next door. A surge of gamma synchrony across your prefrontal regions does not, by surging, retrieve the appearance of medical equipment you have never seen. The dying-brain hypothesis is a hypothesis about experience, and an experience is not the same thing as accurate information about distant facts. The hypothesis explains, at best, the wrapper. The veridical content inside the wrapper has no candidate mechanism in the hypothesis at all.27
Recall the kinds of cases at issue. The Pam Reynolds case, treated at length in Chapter 12, in which the patient accurately describes the bone saw used in her surgery, the conversations between staff, and a number of operating-room details — from a position above the body, during a hypothermic standstill in which all measurable brain activity, including auditory-evoked potentials with clicks delivered through molded ear inserts, had ceased. The Maria’s tennis-shoe case, in which the patient accurately described a shoe on a third-floor exterior window ledge no one in the resuscitation room had any way of seeing. The Al Sullivan case, in which the patient accurately described his cardiac surgeon’s anomalous “elbow-flapping” gesture during surgery. The 1,114 distant-observation cases in the dissertation database.28
For every one of these cases, the dying-brain hypothesis offers nothing. It does not propose a mechanism by which the patient’s gamma surge could yield accurate information about a tennis shoe on a window ledge. It does not propose a mechanism by which neurons firing in Pam Reynolds’s isolated, hypothermic, electrically silent cortex could yield accurate information about the brand of bone saw on the surgical tray. The hypothesis simply does not address this category of evidence.
I want to underline this, because it is a category-error trap that even careful writers fall into. The phrase “NDEs are produced by the dying brain” trades on an ambiguity. It can mean “some of the phenomenological feel of an NDE may be partly shaped by dying-brain physiology,” which is at least a candidate for being true and is also fully compatible with substance dualism. Or it can mean “NDE content is generated by the dying brain,” which is the strong claim the hypothesis would need but cannot deliver in the veridical cases. The first claim is empirically modest. The second is empirically false — not contested, not difficult, simply false in any case where the content is verifiable and the brain is documented as offline.
Sit with one of the cases for a moment. Pam Reynolds was thirty-five years old in 1991, a folk singer from Atlanta with a basilar artery aneurysm so deep in her brain stem that any normal surgical approach would have killed her. The procedure her surgeon proposed, called hypothermic cardiac arrest, was at that point one of the most extreme medical interventions in modern surgery. Her body temperature would be lowered to roughly sixty degrees Fahrenheit. Her heart would be stopped. The blood would be drained from her body, including from her brain, so that the aneurysm could be operated on without rupturing. During this period — clinically and by every measurable definition — she would be dead. EEG flat. Brainstem activity gone. Auditory-evoked potentials gone, monitored continuously through molded earphones that emitted clicks at a hundred decibels and would have produced detectable cortical responses if any auditory cortical processing had been ongoing. Her eyes were taped shut. Her ears were occluded.28b
What she came back with, after the surgery, was a detailed report of events from a position above her body during portions of that procedure. She described the bone saw the surgeon used — a particular pneumatic Midas Rex with a distinctive shape — in terms specific enough that the equipment manufacturer’s representatives recognized her description. She described the conversation between the cardiac team and the neurosurgical team about the size of her femoral arteries. She described portions of music being played in the operating room (“Hotel California”). She described the appearance of the female cardiac surgeon’s hands and the gesture used in cutting the femoral vein. Each of these details was independently verified by surgical staff who had been present during the procedure.28c
“The thing I remember about that period was the noise that the saw was making. It was like a dental drill. I had a sense of being above where they were working. I could see what was going on. I remember being more aware than I have ever been. I remember thinking, ‘This is the strangest thing — I’m not afraid.’ Then I remember seeing this thing in [the surgeon’s] hand. It was like an electric toothbrush, but it had this little dent in it. And there was a case for it that looked like the box my father’s socket wrench set came in.”
— Pam Reynolds, interviewed by Michael Sabom, in Light and Death (1998), 41. The pneumatic Midas Rex bone saw to which she was referring, and the case in which it was stored, matched her description in detail. The saw was first introduced into the surgical field after she had been anesthetized, was at hypothermic standstill, and had clinically flat EEG.
Now ask the dying-brain hypothesis to explain this. It cannot. It does not propose, and could not in principle propose, a mechanism by which a gamma surge in Reynolds’s isolated, hypothermic, electrically silent cortex would yield accurate visual and auditory information about a bone saw she had never seen, removed from a case she had never observed, used by a surgeon she had not been awake to watch. Whatever was happening neurally during her standstill — including any version of the Borjigin surge — it was happening in a brain that was not receiving sensory input. It was happening in a brain whose auditory processing pathway had been confirmed silent by the clicks-and-evoked-potentials protocol the surgeon had specifically set up to monitor exactly that question. There is no candidate mechanism. The hypothesis simply does not address the case.
Multiply this case by the 1,114 distant-observation cases in the dissertation database, and by the parallel case files developed across the medical-NDE literature, and the structure of the problem becomes clear. The dying-brain hypothesis is not failing on a few odd anecdotes. It is failing on the largest single body of evidence the NDE field has produced.28d
“But maybe the patient picked up the information unconsciously, before the cardiac arrest, and the dying brain just retrieved it during the surge.”
This is a real fallback line, and it works for some cases. It does not work for the cases that drive the argument. Pam Reynolds had her ear canals filled with molded clicking earphones; the bone saw was first introduced after she was already in hypothermic standstill. The Al Sullivan elbow-flapping gesture occurred during surgery, after he was anesthetized. Maria had never been on the third floor exterior of the hospital where the tennis shoe was. AWARE-I’s veridical case involved a man whose accurate report was of events that occurred during his arrest, when his eyes were closed and he was, by every available measure, unconscious. The unconscious-acquisition-then-retrieval story explains the easy cases. It does not explain the hard ones. And the hard ones are where the argument lives.
The second failure has to do with the structure of NDE phenomenology. The dying-brain hypothesis predicts that NDE content should reflect dying-brain physiology — whatever that physiology happens to produce. There is no obvious reason it should produce a consistent narrative pattern across cultures, ages, religions, and decades. A burst of disinhibited neural firing in the dying brain of a Tibetan farmer in 1972, an American ICU patient in 2010, an Iranian housewife in 2018, and a four-year-old drowning victim in 1995 should, on a strict dying-brain account, produce four different sets of phenomenology shaped by four different brains, four different cultural backgrounds, four different sets of expectations, and four different stages of development.29
What is actually observed is an astonishingly consistent core pattern. OBE phase. Tunnel and light. Encounter with deceased relatives or beings perceived as luminous. Life review. Border or limit. Return. The Greyson NDE Scale picks up the same core elements across studies in Western, Iranian, Indian, Chinese, and tribal populations.30 The Tabatt nine-point analysis from the dissertation database yields the same picture: a striking concentration of cases at the high end of the scale, with most cases reporting most of the core elements, regardless of cultural background.31
This pattern is exactly what one would expect if the experiencer is a soul whose nature is shared across human persons and whose experience at the boundary of death reflects something universal about that boundary. It is exactly not what one would expect if the experiencer is a brain, with the experience shaped by local neurochemistry and cultural conditioning. Brains differ. Cultures differ. Souls, on the historic Christian view, share something fundamental.
The dying-brain hypothesis can try to absorb this. It can claim that the basic dying-brain physiology is itself universal across humans, so the gross phenomenology should also be universal. That is a coherent story but a weaker one than the headline suggests. It also gives up the ad hoc cultural-conditioning explanation that the same physicalist literature reaches for elsewhere when arguing that NDEs are a culturally constructed phenomenon. You cannot have it both ways. Either the experience is shaped primarily by universal physiology — in which case cultural-conditioning explanations of specific content fail — or the experience is shaped primarily by culture — in which case the universal physiology cannot explain the consistency. The dying-brain hypothesis, pushed to do real work, cannot keep its story straight.
The third failure is the most subtle and, I think, the most damning when one sits with it. Dying brains, in every other context where we have good data, do not produce lucid, organized, narrative experience. They produce confusion. They produce delirium. They produce fragmentation. They produce, in the late stages, the absence of experience.
This is well-established. The phenomenology of decompensating cardiac function as the brain begins to lose perfusion is documented in many clinical contexts — aviation hypoxia training, mountaineering accidents, fainting episodes, the early stages of stroke. The picture, in all of these, is dysphoric and disordered. Tunnel vision in the optical sense (narrowing of the visual field, not the structured tunnel of NDE reports). Cognitive fog. Loss of executive function. Inability to perform simple tasks. Confusion that the experiencer often cannot recognize as confusion. As things progress, awareness fades into nothing.32
NDE phenomenology is the opposite of this. NDErs characteristically report increased clarity, organized sequential narrative, enhanced cognitive function. They report thinking faster and more clearly than ever before. They report perceiving more, not less. They report the experience as more real than ordinary waking life, not less. They report the experience as ordered, not fragmented — with a structure, with a sequence, with a meaning.
This is, on the dying-brain hypothesis, exactly the wrong prediction. If NDEs are the product of a decompensating brain, they should look like other products of decompensating brains. They should be confused, dysphoric, fragmented, disordered. They are not. They are, by every measure researchers have used, structurally distinct from the phenomenology of cerebral hypoperfusion.33
The dying-brain hypothesis would need to explain why this single context — cardiac arrest — produces phenomenology that is the opposite of what every other context of cerebral hypoperfusion produces. The proposed explanation is the gamma surge. But the surge is brief, uneven across patients, and does not, in itself, predict structured narrative. Gamma activity is not narrative. Gamma activity is a frequency band. The leap from “there is gamma in the dying brain” to “the dying brain produces structured, lucid, narrative experience that is the opposite of every other dying-brain phenomenology we know about” is a leap that the data do not warrant.
The fourth failure is sharp and is, in my view, decisive on its own. NDEs are reported by people who have been blind from birth. Of those reports, a substantial proportion include accurate visual content — descriptions of medical equipment, identification of colors, recognition of facial features — that the experiencer has no normal way of knowing. The Vicki Umipeg case and the Brad Barrows case are the touchstones, but they are not isolated; the dissertation database includes thirty-three congenitally blind NDE cases in which visual content figured.34
For the dying-brain hypothesis, this category of case is fatal. A person who has been blind from birth does not have the developed visual neural architecture that the hypothesis would need to be active during the dying surge. Vicki Umipeg’s visual cortex never received normal patterned input. Whatever activity her dying brain produced in those regions, it could not, by anything we know about cortical development, produce structured visual content. Brains develop functional architecture by being exposed to patterned input during critical periods. A brain that never received that input cannot generate structured perception of objects it has never experienced.35
And yet the reports come back, and the reports are accurate. Vicki, during her NDE following a car accident, accurately described the room she was in (a hospital room she had never been in before, designed for sighted patients), her own body from above (a body she had never seen), and the appearance of medical staff and equipment. She was able to identify colors of garments worn by people in the room — colors she had never seen, never been described in any way that would let her identify them in the moment. The accurate elements of her report were independently verified.36
Vicki herself described the experience as the first time in her life she had ever “seen” in any sense, and she struggled afterward to make sense of what had happened. “I was so totally amazed,” she told Kenneth Ring during one of his interviews, “I felt I could understand colors for the first time. I knew what red was, what green was, what blue was — not as words but as experiences.” The phrasing matters. She was not reporting that some abstract knowledge of color had filtered to her through the descriptions of sighted people. She was reporting a perceptual experience, in real time, of colors as colors. And she was reporting it during a period in which her brain was, by every available measure, dying.
The dying-brain hypothesis has nothing to say about this kind of case except, perhaps, that it is somehow mistaken or fabricated. Both options are difficult: the cases are documented, with witness verification and contemporary medical records, and the experiencers report the visual content as the most clearly remembered part of the experience. Once you concede the cases, the dying-brain hypothesis collapses on this category. There is no candidate mechanism, on the hypothesis, that would generate accurate visual content in a brain that lacks the architecture to generate visual content at all.
Some defenders of the hypothesis have suggested that perhaps in blind NDErs the experience is not, strictly, visual at all — that the experiencers retroactively translate non-visual perceptions into visual language because that is how sighted people talk about perception. This is a clever move, but it does not survive contact with the case material. Vicki’s descriptions of color are not the kind of thing a non-visual perceiver would generate by retroactive translation. She does not say “I knew the texture of the wallpaper.” She says “I saw that the surgical lamp was emitting a blueish-white light.” The specifics of color, the way colors related to each other in her field of view, the architectural details of a building she had never been in — these are not the artifacts of post-hoc re-translation. They are the artifacts of perception.36b
This single category of evidence, treated alone, would already be difficult for the dying-brain hypothesis. Combined with the other failures, it is decisive. The hypothesis can be modified to absorb the easy cases. It cannot be modified to absorb cases like Vicki’s without becoming so empirically diluted that it no longer constrains anything.
The fifth failure is more recent and increasingly difficult for the hypothesis to absorb. NDEs are reported in contexts where the brain is not, properly speaking, “dying” in the cardiac-arrest sense at all — specifically, during deep anesthesia and high-dose barbiturate sedation. These are contexts where the brain is pharmacologically silenced rather than ischemically failing, and where the specific neural patterns Borjigin found in the dying rat brain are actively suppressed by the anesthetic.37
Modern surgical anesthetics — propofol, sevoflurane, isoflurane, the standard combinations — reduce gamma-band activity. They do this by design; suppressing the kind of high-frequency cortical synchrony associated with conscious cognition is a significant part of how they produce surgical anesthesia. Anesthesiologists rely on this suppression as one of the markers that the patient is, by every measurable standard, unconscious. The mechanism on the molecular level is reasonably well understood: agents like propofol and sevoflurane potentiate inhibitory GABA-receptor function in the cortex and thalamus. They turn the inhibitory dial up. The Borjigin mechanism, recall, requires that inhibitory tone fall away — disinhibition is what allows the gamma surge to emerge in her dying-rat data. So the two states are, at the level of mechanism, near opposites. Surgical anesthesia is engineered to do precisely the thing that would prevent the dying-brain mechanism from firing.37b
The Pam Reynolds case is the classic example. She was under deep general anesthesia, on barbiturate burst-suppression, and in hypothermic standstill, all simultaneously. Her core temperature was lowered to about 60 degrees. The blood was drained from her head. Her EEG showed silence. Auditory clicks at 95 decibels in molded earpieces produced no brainstem evoked response. Her brain was not surging. It was silenced about as completely as a living human brain can be silenced. And she had a richly detailed, partly veridical NDE that included specific accurate descriptions of the operating room, the surgical instruments, and overheard conversation.38
The dying-brain hypothesis does not predict NDEs in this state. The mechanism the hypothesis invokes — gamma synchrony from disinhibition — is precisely what the anesthetic is preventing. To rescue the hypothesis here, the physicalist has to introduce some other mechanism for the anesthesia cases, on top of the dying-brain mechanism for the cardiac-arrest cases, on top of yet another mechanism for the non-arrest trauma cases, and so on. The proliferation of mechanisms is itself a sign of empirical strain. Each new case demands a new ad hoc rescue, and the rescues do not unify under any common physiological story. The dualist account, by contrast, is parsimonious: the experiencer is the same in every context (the soul); the brain is the body’s instrument that comes off-line under different physiological conditions; the experiences reported are the soul’s experiences during the period of off-line. One mechanism. One subject. Many varieties of physiological context.
The case against the dying-brain hypothesis as a complete explanation of NDEs is, I think, decisive. The case against it as a partial explanation of some features is much weaker, and I want to be honest about where the hypothesis still does work.
It is plausible — not established, but plausible — that some features of NDE phenomenology are co-shaped by dying-brain physiology. The vividness of the sensory phase. Some aspects of the life review’s rapid, integrated memory access. Some of the temporal-distortion effects (the experience of time stretching or compressing). Some of the imagery of light and brightness. These could plausibly be partly shaped by the kind of activity Borjigin documented — gamma synchrony, cross-frequency coupling, increased connectivity in the dying mammalian brain. Some of the “flavor” of the NDE, in other words, might be the soul having an experience while wearing a dying brain as an instrument that is itself behaving in unusual ways.
This is exactly what we would expect on substance dualism. The Christian theological tradition has never taught that the body does not matter. The brain is the soul’s instrument during embodied life, the medium through which the soul interacts with the world. As that medium fails, the soul’s experience is not unaffected. When the body is fevered, the soul’s perceptions are colored. When the body is exhausted, the soul’s thinking is sluggish. When the body is dying, the soul’s experience may well take on the texture of a body that is dying. This is consistent with what John Cooper has called “holistic dualism” — a view in which body and soul are deeply integrated during embodied life even though the soul can exist without the body in the postmortem state.39
So — the dying-brain hypothesis is not without explanatory traction. It has some. It explains some texture of the NDE. What it does not and cannot explain is the content — the specific accurate observations of the external world that the experiencer had no normal access to. And the content is what carries the empirical weight in the dualism-physicalism debate. The texture, on its own, is metaphysically neutral. The veridical content is not.
The dying-brain hypothesis can plausibly co-explain some features of NDE phenomenology — the texture, the flavor, the way an experience colored by a dying brain might feel from the inside. It cannot explain the content — specifically the verifiable accurate observations of external events that drive the empirical case for substance dualism. The hypothesis is partial. It is not a substitute for substance dualism. It is, on careful reading, fully compatible with substance dualism.
One more move needs to be addressed before the assessment is complete — the broader popular framing that has done so much damage to the careful conversation about NDE evidence.
The category error in the headline reception of Borjigin’s 2013 paper was the move from “dying brains have a brief surge of activity” to “NDEs are explained.” This is not a small mistake. It is a substantial error in reasoning, and it has been repeated so often in the popular and apologetic literature that it has acquired a kind of inertia. It deserves to be named clearly.
To explain a phenomenon is not the same as to identify some activity that occurs in the same time window. If a man reports seeing his daughter’s face in a dream during a thunderstorm, and we observe that his auditory cortex was processing thunder during the dream, we have not thereby explained the dream. The thunder is in the time window. It does not explain the dream content. To explain the dream content we would need an account of why this man saw this daughter doing this thing — an account that connects the proposed mechanism to the actual content of the experience. The thunder gives us nothing for that explanation.
So with the dying-brain hypothesis. To observe gamma activity in the dying brain is to observe activity in the time window. It is not, by itself, to explain the experience that fills that window. To explain a particular NDE you would need an account that connects the proposed mechanism to the actual content of that experience — the specific deceased relative the patient encountered, the specific accurate description of the resuscitation, the specific moral content of the life review, the specific recognition of a luminous figure. The dying-brain hypothesis offers nothing in this connecting role. It provides a wrapper. It does not provide content.
Readers who follow the secondary literature on NDEs will recognize this pattern. A new physicalist study comes out. The headlines say NDEs have been explained. The paper itself is much more modest. The Christian physicalist writers reach for the headline. The careful response is the same in every iteration: take the actual paper seriously, ask what it actually shows, and notice that it does not reach the content layer where the empirical case lives. The pattern recurs because the structure of the situation does not change. The dying brain is local. The NDE content is not. Closing that gap requires more than a frequency band.40
So, what does an honest assessment look like?
The Borjigin findings are real science. They identify a previously underdescribed phenomenon — a brief electrical surge in dying mammalian brains — and they have been replicated and extended in careful follow-up work. The work is part of a broader, growing empirical picture of the dying brain that includes Parnia’s AWARE-II EEG findings, Sestan’s BrainEx pig work, and a developing literature on what Parnia has begun calling “lucid dying.” This is good science and should not be dismissed.
The dying-brain hypothesis as an explanation of NDEs, as it appears in the popular and apologetic literature, is a significantly stronger claim than the actual data support. It runs into the data and does not survive the contact at five places: it cannot explain veridical content; it cannot explain cross-cultural consistency; it cannot explain the lucid quality of the phenomenology; it cannot explain accurate visual perception in the congenitally blind; and it cannot explain NDEs in deep anesthesia and barbiturate sedation. Each of these is a serious objection on its own. Together they are a comprehensive failure.
What the hypothesis can do is explain some aspects of the texture of NDE experience — the way the phenomenology might be colored by a dying brain that is itself doing unusual things. This partial explanatory work is fully compatible with substance dualism and was always expected on a holistic-dualist account. The brain is the soul’s instrument; as the instrument fails in particular ways, the soul’s experience may show traces of those particulars.
Borjigin herself, on the careful evidence, does not claim more than this. The strong claim — that the hypothesis explains NDEs — belongs to the headlines and to the popular reception. It does not belong to the science.
This leaves the cumulative case against physicalism, developed in the previous chapter, essentially intact. The strongest single physicalist counter-explanation in the contemporary literature does not reach the strongest evidential category. The blind-NDE cases stand. The Pam-Reynolds-class cardiac arrest cases stand. The 1,114 distant-observation cases stand. The 89.96% statistic from the dissertation analysis — cardiac arrest patients reporting consciousness during documented brain shutdown — stands. The dying-brain hypothesis has not falsified any of this and has not even tried to.41
The next chapter takes up another set of physicalist explanations — the anoxic-hallucination, DMT, ketamine, and pharmacological-mechanism cluster — and asks whether these do better. The pattern, as Chapter 16 will show, is similar. Each captures something. None reaches the content layer. The cumulative refutation is built up over the whole stretch of Chapter 17.
I want to close with the pastoral question, because this book is finally not about a debate among researchers. It is about dying.
If you are sitting at a bedside — your mother’s, your husband’s, your child’s — and you have been wondering what is happening to the person you love, the dying-brain hypothesis presents itself as a kind of cold comfort. Whatever they are experiencing, the hypothesis says, is just the brain’s last fireworks. It is not real in any deeper sense. It does not bear on the question of what comes next. The person is fading; the brain is shutting down; the lights they may seem to see are sparks from a closing circuit, not anything beyond the body.
This account is offered, sometimes, with a kind of sober scientific gravity, as if it were the brave conclusion of unflinching inquiry. I want to say plainly that the science does not support it. The headlines do. The actual research does not.
What the actual research supports — cumulatively, across the case files we have walked through — is something much closer to the historic Christian picture. The dying brain is doing things. It is doing some interesting things. And alongside those things, in the same window where physicalism predicts no consciousness should be possible at all, the experiencer is reporting being conscious. Aware. In some cases, actively perceiving the room around them. In some cases, perceiving things they could not otherwise have perceived. In many cases, encountering something or someone they describe as more real than anything they have known before.
For the Christian sitting at the bedside, this is not threatening news. The historic tradition has always taught that the soul is in the hands of God at the moment of death. In manus tuas, Domine, commendo spiritum meum — into your hands, Lord, I commend my spirit. The patristic tradition wrote about angels coming for the dying soul; the medieval Ars Moriendi tradition prepared the dying for an encounter at the threshold; the funeral liturgies of every Christian tradition speak of the soul going to be with Christ.42 The empirical evidence is converging with what the tradition has always said. Nothing in the dying-brain literature undermines this picture; nothing actually does the work the headlines claimed it did.
If a hospice nurse, a chaplain, or a well-meaning friend tells you that the experience your dying mother described to you the day before she died was “just the dying brain”: it is not unkind, and not unscientific, to say back gently that the science does not establish that. The brain does some things in those last seconds. But what your mother described to you — the relative she saw, the love she felt, the sense of going somewhere — is not in any straightforward way reducible to those things. The careful science is more open than the headlines. So is the historic Christian hope.
For the Christian physicalist in the conditional-immortality conversation, the implications are more pointed. The contemporary movement toward Christian physicalism has often been defended in part by appeal to current neuroscience — the claim that “modern science has shown” that the brain produces consciousness, that there is no need for an immaterial soul, that the biblical anthropology should be reread in monistic terms. The dying-brain hypothesis has been invoked in this argument as evidence that the most striking apparent counterexample to physicalism — the NDE — has a naturalistic explanation. I have been arguing throughout this part of the book that this invocation does not survive close examination. The hypothesis does not deliver what the argument needs. The empirical ground for Christian physicalism is much weaker than its defenders sometimes suggest. And the historic Christian dualism — far from being the embarrassed remnant of a pre-scientific worldview — remains the best account of the empirical evidence as well as of Scripture and tradition.
For the dying patient and the family, the upshot is simpler still. The brain is precious; its slow shutdown is a real loss; the historic tradition has never said otherwise. But the brain is the instrument, not the player. As the instrument falls silent, the player is not silenced. The player is, by every empirical indication available to us, doing exactly what the saints have always said the soul does at the moment of death — encountering, perceiving, going on. That is not a fireworks show in a closing circuit. That is the beginning of the postmortem journey the rest of this book will walk through.
The science of the dying brain has given us new and interesting things to think about. It has not given us the explanation the headlines promised. It has not, in any serious sense, made the historic Christian picture less plausible. It has, in some interesting ways, made it more plausible — by giving us a richer account of what the body is doing while the soul is doing what souls do. The next chapter takes up the second large physicalist response and asks whether the anoxic-hallucination and pharmacological accounts do better. The third and fourth take up the remaining alternatives. The pattern, as we will see, holds throughout.
↑ 1. Jimo Borjigin, UnCheol Lee, Tiecheng Liu, et al., “Surge of neurophysiological coherence and connectivity in the dying brain,” Proceedings of the National Academy of Sciences 110, no. 35 (August 27, 2013): 14432–14437.
↑ 2. Borjigin et al., “Surge,” 14433–14435. The team analyzed power spectra across delta, theta, alpha, beta, low-gamma (25–55 Hz), and high-gamma (65–115 Hz) bands, and reported particularly striking changes in the gamma range and in coherence measures.
↑ 3. See particularly the paper’s discussion section, which frames the findings as “a possible neurophysiological foundation” for the “heightened conscious processing reported in survivors of cardiac arrest” — a candidate-correlate framing, not an explanatory one. Borjigin et al., “Surge,” 14436.
↑ 4. A representative sample of headline coverage immediately following the paper’s release: Bryan Nelson, “Near-Death Experiences May Be Caused by Surge of Brain Activity,” NBC News, August 12, 2013; Sarah Knapton, “Near-Death Experiences Could Be Caused by Surge in Brain Activity Just before Death,” The Telegraph, August 12, 2013; Brandon Keim, “In Dying Brains, Signs of Heightened Consciousness,” Wired, August 12, 2013. The pattern of coverage was broadly the same: scientists have explained near-death experiences. For a contemporaneous careful response, see Bruce Greyson, “Comments on ‘Surge of Neurophysiological Coherence and Connectivity in the Dying Brain,’” Journal of Near-Death Studies 32, no. 1 (2013): 47–52.
↑ 5. See, e.g., the citation pattern in Glenn Peoples’s podcast and online writings on Christian physicalism, where the Borjigin work is frequently invoked as the scientific basis for treating NDE evidence as not undermining a physicalist anthropology. Similarly in popular skeptical literature: see Sebastian Dieguez, “Near-Death Experiences,” in The Routledge Handbook of the Philosophy and Science of Mental Imagery, ed. Joel Krueger and Jen Smith (forthcoming), where the Borjigin findings are presented as central to the physicalist response.
↑ 6. Gang Xu, Temenuzhka Mashour, Florian Jurchescu, et al. (Borjigin lab), “Surge of neurophysiological coupling and connectivity of gamma oscillations in the dying human brain,” Proceedings of the National Academy of Sciences 120, no. 19 (May 9, 2023): e2216268120.
↑ 7. The 89.96% finding from the dissertation database is developed in detail in Chapter 12, with the methodological notes and timing-reconstruction qualifications. The cumulative pattern across the 5,278-case dataset is the subject of Chapter 14.
↑ 8. Bruce Greyson, “The Near-Death Experience Scale: Construction, Reliability, and Validity,” Journal of Nervous and Mental Disease 171, no. 6 (1983): 369–375. The Greyson Scale’s sixteen items capture a phenomenology that, in many cardiac-arrest cases, is reported as covering subjectively much longer windows than the documented period of arrest. See also Pim van Lommel, Consciousness Beyond Life: The Science of the Near-Death Experience (San Francisco: HarperOne, 2010), chapter 2, on subjective time-dilation in NDE reports.
↑ 9. See the discussion of medical-context distribution in Chapter 4 of the dissertation. NDEs are reported across a range of contexts including non-cardiac-arrest situations — severe trauma without cardiac arrest, drowning rescues with rapid resuscitation, deep anesthesia without arrest, profound hypoglycemic episodes, and others. The dying-brain hypothesis is most plausible where cardiac arrest is in fact present; the breadth of contexts is itself a difficulty for the hypothesis as a complete explanation.
↑ 10. Borjigin et al., “Surge,” 14432–14437. The methodological details that follow are drawn from the paper’s methods section.
↑ 11. The pattern is most clearly summarized in the paper’s Figure 2 (gamma-band power) and Figure 3 (cross-frequency coupling). The thirty-second window for the surge is approximate; the exact duration varied across animals, with the bulk of the elevated activity concentrated in roughly the first thirty seconds following cardiac arrest induction.
↑ 12. Borjigin et al., “Surge,” 14436.
↑ 13. See Michael Sabom, Light and Death: One Doctor’s Fascinating Account of Near-Death Experiences (Grand Rapids: Zondervan, 1998), chapters 3–7, for the Pam Reynolds case timeline. The hypothermic standstill phase — during which Reynolds had cessation of all measurable cortical activity, including auditory-evoked potentials — lasted considerably longer than the thirty-second Borjigin window, and Reynolds’s subjective report covers events spread across that broader interval.
↑ 14. See, on the gamma–consciousness relationship, Christof Koch et al., “Neural Correlates of Consciousness: Progress and Problems,” Nature Reviews Neuroscience 17, no. 5 (2016): 307–321; on the dissociations between gamma and conscious experience, see Lucia Melloni et al., “Synchronization of Neural Activity across Cortical Areas Correlates with Conscious Perception,” Journal of Neuroscience 27, no. 11 (2007): 2858–2865, with the careful caveats about the limits of frequency-band markers.
↑ 15. Xu et al., “Surge of neurophysiological coupling and connectivity of gamma oscillations in the dying human brain,” e2216268120.
↑ 16. Xu et al., “Surge,” abstract and discussion. The paper is unusually careful in its interpretive claims, repeatedly emphasizing that the gamma findings are compatible with but do not establish conscious experience in the patients studied.
↑ 17. The 10–20% range comes from a number of prospective studies, including Pim van Lommel, et al., “Near-death experience in survivors of cardiac arrest: a prospective study in the Netherlands,” The Lancet 358, no. 9298 (2001): 2039–2045 (~18%); Bruce Greyson, “Incidence and correlates of near-death experiences in a cardiac care unit,” General Hospital Psychiatry 25, no. 4 (2003): 269–276 (~10%); Sam Parnia et al., “AWARE—Awareness during Resuscitation—a prospective study,” Resuscitation 85, no. 12 (2014): 1799–1805 (~9% reporting some form of recallable awareness).
↑ 18. See Borjigin’s comments in the University of Michigan press materials accompanying the 2023 paper, and her interview with The New York Times, May 1, 2023, where she emphasizes that the work has not “explained” NDEs and that the relationship between the gamma findings and any conscious experience remains an open question.
↑ 19. Zvonimir Vrselja, Stefano G. Daniele, John Silbereis, et al., “Restoration of brain circulation and cellular functions hours post-mortem,” Nature 568 (2019): 336–343.
↑ 20. Vrselja et al., “Restoration,” 340. The protocol included neural-activity blockers specifically to prevent the development of organized cortical activity that might suggest consciousness in the perfused tissue.
↑ 21. Sam Parnia, Tara Keshavarz, Pavan Patel, et al., “AWAreness during REsuscitation - II: A multi-center study of consciousness and awareness in cardiac arrest,” Resuscitation 191 (October 2023): 109903.
↑ 22. Parnia et al., “AWARE-II,” results section and discussion. The team reports that gamma-band activity was observed in roughly forty percent of patients in whom EEG could be obtained during CPR, with the activity sometimes appearing many minutes into resuscitation. The framing of these findings within Parnia’s “lucid dying” programme emphasizes their compatibility with reported NDE phenomenology rather than reading them as a reductive explanation.
↑ 23. Sam Parnia, Lucid Dying: The New Science Revolutionizing How We Understand Life and Death (New York: Hachette, 2024), chapters 6–8. Parnia is explicit that the AWARE-II findings are best interpreted within a framework that takes the reality of conscious experience during clinical death seriously, rather than as evidence that NDEs reduce to dying-brain physiology.
↑ 24. See, e.g., Olaf Blanke and Sebastian Dieguez, “Leaving Body and Life Behind: Out-of-Body and Near-Death Experience,” in The Neurology of Consciousness, ed. Steven Laureys et al., 2nd ed. (London: Academic Press, 2016), chapter 20, for a developed account of how dying-brain physiology might be linked to specific NDE features. Blanke is more cautious than the headline literature about the bridge from physiology to specific content.
↑ 25. See the University of Michigan Health press release, “Heightened Brain Activity at Death,” August 12, 2013; the conditional logic in Borjigin’s framing is consistent across her published commentary on the 2013 paper.
↑ 26. See Borjigin’s commentary cited in note 18, as well as the careful interpretive framing in Xu et al. (2023), section 4 of the discussion. The contrast between Borjigin’s actual claims and the popular “Borjigin explained NDEs” reading is a recurring feature of the secondary literature.
↑ 27. The point is developed in Bruce Greyson, After: A Doctor Explores What Near-Death Experiences Reveal about Life and Beyond (New York: St. Martin’s Essentials, 2021), chapters 6–8, where Greyson argues at length that the local-brain hypothesis cannot bridge to non-local content. See also the dissertation Chapter 4, which develops the same argument with case-by-case examples from the database.
↑ 28. The Pam Reynolds case is treated in Chapter 12; the Maria’s tennis-shoe case in Chapter 10; the Al Sullivan case (with Bruce Greyson’s and Surbhi Khanna’s analyses) likewise in Chapter 10; the dissertation database’s 1,114 distant-observation cases are tabulated in Chapter 14.
↑ 28b. The medical and surgical details of the Pam Reynolds case are documented in Sabom, Light and Death, chapters 3–7. The hypothermic cardiac arrest protocol used at Barrow Neurological Institute by Dr. Robert Spetzler involved cooling the patient to 60°F (15.5°C), using an extracorporeal circuit to drain the blood from the body and brain, suspending all cerebral and circulatory function during the period the aneurysm was clipped, and then rewarming and reperfusing. Continuous EEG and brainstem auditory-evoked-potential monitoring confirmed the absence of measurable cortical and brainstem activity throughout the standstill phase.
↑ 28c. Sabom, Light and Death, 41–51, with the verification details from Dr. Spetzler and the surgical staff. For the broader independent assessment of the case, see Greyson, After, chapter 6; Janice M. Holden, “Pam Reynolds’ Near-Death Experience: A Reply to Augustine,” Journal of Near-Death Studies 25, no. 4 (2007): 211–226. The case has been the subject of extended skeptical critique (Keith Augustine, principally) and equally extended dualist defense (Holden, Greyson, others); the careful balanced assessment in Chapter 12 takes up that debate.
↑ 28d. The 1,114 distant-observation cases in the dissertation database are tabulated and analyzed in Chapter 10 and synthesized in Chapter 14. The breakdown by evidential strength on the dissertation’s three-dimensional scoring system showed approximately 30.7% of cases meeting Strong or Exceptional thresholds — that is, with high medical-context severity, high veridical-quality scores, and high corroboration scores.
↑ 29. See Allan Kellehear, Experiences Near Death: Beyond Medicine and Religion (Oxford: Oxford University Press, 1996), for the major cross-cultural treatment. Kellehear documents the core elements of NDE phenomenology across cultures with very different background assumptions about death and afterlife, finding a striking commonality at the structural level.
↑ 30. See, for the Iranian study, Surbhi Khanna, et al., “After-Effects of Near-Death Experiences: A Survey of Greyson NDE Scale Reports across Cultural and Religious Contexts,” Journal of Religion and Health 60, no. 3 (2021): 1837–1854. For the Indian and Chinese studies, see Satwant Pasricha, “A Systematic Survey of Near-Death Experiences in South India,” Journal of Scientific Exploration 7, no. 2 (1993): 161–171; Feng Zhi-ying and Liu Jian-xun, “Near-Death Experiences among Survivors of the 1976 Tangshan Earthquake,” Journal of Near-Death Studies 11, no. 1 (1992): 39–48. The cross-cultural consistency on Greyson’s sixteen items is one of the more striking findings in the literature.
↑ 31. The Tabatt nine-point analysis from the dissertation database is developed in Chapter 9, with the cross-cultural and cross-temporal distributions tabulated in Chapter 14.
↑ 32. The classic source on the phenomenology of cerebral hypoperfusion is M. J. Aminoff, M. M. Scheinman, J. C. Griffin, and J. M. Herre, “Electrocerebral accompaniments of syncope associated with malignant ventricular arrhythmias,” Annals of Internal Medicine 108, no. 6 (1988): 791–796. For the aviation hypoxia phenomenology, see the standard military-aerospace training literature; for fainting and other forms of cerebral hypoperfusion, see the broader neurology and emergency-medicine texts. The pattern across all of these is dysphoric and disordered phenomenology, in sharp contrast to the lucid, sequential, peaceful pattern characteristic of NDEs.
↑ 33. See particularly Janice M. Holden, “Veridical Perception in Near-Death Experiences,” in The Handbook of Near-Death Experiences: Thirty Years of Investigation, ed. Janice Miner Holden, Bruce Greyson, and Debbie James (Santa Barbara: Praeger, 2009), 185–211, on the sharp distinction between NDE phenomenology and the phenomenology of the various conditions to which NDEs are sometimes assimilated.
↑ 34. Kenneth Ring and Sharon Cooper, Mindsight: Near-Death and Out-of-Body Experiences in the Blind, 2nd ed. (Bloomington: iUniverse, 2008). The Vicki Umipeg interview is treated at length in chapters 1–3; Brad Barrows in chapters 4–5. The dissertation database’s 33 blind-NDE cases are developed in Chapter 11.
↑ 35. The classic developmental work on the visual-cortical critical period is David H. Hubel and Torsten N. Wiesel, “The Period of Susceptibility to the Physiological Effects of Unilateral Eye Closure in Kittens,” Journal of Physiology 206, no. 2 (1970): 419–436, and follow-up work that extended these findings to humans. The relevance for blind-NDE cases is that the cortical architecture required to generate structured visual perception does not develop in the absence of patterned input during the early critical period — making the dying-brain hypothesis face a developmental impossibility in this category.
↑ 36. See the verification details in Ring and Cooper, Mindsight, chapter 2; in Kenneth Ring and Madelaine Lawrence, “Further Evidence for Veridical Perception during Near-Death Experiences,” Journal of Near-Death Studies 11, no. 4 (1993): 223–229; and in Chapter 11 for the broader case-file documentation.
↑ 36b. The non-visual-translation defense is a sophisticated rescue attempt and deserves a serious answer rather than a quick dismissal. The cleanest empirical pressure on it is the specificity of color and spatial-relational reports in cases like Vicki Umipeg’s and Brad Barrows’s — reports of a kind that someone congenitally without visual cortical processing should not be in a position to generate by post-hoc translation. Ring and Cooper engage the proposal at length in Mindsight, chapters 4–5; my treatment here follows their analysis and the case-file detail in Holden, Greyson, and James, Handbook, chapter 5.
↑ 37. See, on the gamma-suppressing effects of standard surgical anesthetics, P. L. Purdon et al., “Electroencephalogram Signatures of Loss and Recovery of Consciousness from Propofol,” Proceedings of the National Academy of Sciences 110, no. 12 (2013): E1142–E1151; and George A. Mashour, “Top-Down Mechanisms of Anesthetic-Induced Unconsciousness,” Frontiers in Systems Neuroscience 8 (2014): 115. Mashour, notably, was a co-author on the 2013 Borjigin paper; he is also among the most careful researchers on the anesthesia-and-consciousness front, and his work in that adjacent area is part of why the anesthesia-NDE finding presents a sharp difficulty for a strong dying-brain hypothesis.
↑ 37b. On GABA-receptor potentiation as the principal mechanism of propofol and sevoflurane action, see Beverley A. Orser, “Lifting the Fog around Anesthesia,” Scientific American 296, no. 6 (2007): 54–61; and the more technical review by Misha Perouansky and Hugh C. Hemmings Jr., “Neurotoxicity of General Anesthetics: Cause for Concern?” Anesthesiology 111, no. 6 (2009): 1365–1371. The point for this chapter is structural: the same inhibitory tone that surgical anesthesia increases is the inhibitory tone that the Borjigin disinhibition mechanism requires to decrease. Engineering one state effectively prevents the other.
↑ 38. Sabom, Light and Death, chapters 3–7. Sabom’s detailed reconstruction of Reynolds’s anesthetic state, the burst-suppression EEG pattern produced by the Pentothal infusion, and the auditory-evoked-potential evidence is the gold standard for what we know about the case’s neurophysiological context.
↑ 39. John W. Cooper, Body, Soul, and Life Everlasting: Biblical Anthropology and the Monism-Dualism Debate, 2nd ed. (Grand Rapids: Eerdmans, 2000), chapter 7, on holistic dualism; the position is developed in conversation with both biblical and philosophical sources. See also Charles Taliaferro, Consciousness and the Mind of God (Cambridge: Cambridge University Press, 1994), for a related philosophical treatment.
↑ 40. The category-error point is developed at length in J. P. Moreland, The Soul: How We Know It’s Real and Why It Matters (Chicago: Moody Publishers, 2014), chapter 3, in the context of a broader argument that physicalist responses to NDE evidence routinely confuse correlation with constitution.
↑ 41. The cumulative refutation of physicalist responses is the subject of Chapter 17, which integrates the discussion of the dying-brain hypothesis with the parallel discussions of pharmacological explanations (Chapter 16) and other alternative accounts.
↑ 42. The Latin commendation, drawn from Psalm 31:5 (Vulgate 30:6) and prayed by Christ from the cross (Luke 23:46), is the central commendatory prayer of the Christian dying tradition across centuries and traditions. For the historic Ars Moriendi tradition, see Allen Verhey, The Christian Art of Dying: Learning from Jesus (Grand Rapids: Eerdmans, 2011), chapters 4–5, and for a contemporary recovery, Lydia Dugdale, The Lost Art of Dying: Reviving Forgotten Wisdom (New York: HarperOne, 2020).