Chapter 10
If there is one skeptical explanation for near-death experiences that has captured the public imagination more than any other, it is the “dying brain” hypothesis. The idea is straightforward: when the brain is shutting down—starved of oxygen, flooded with unusual chemicals, and rapidly losing its grip on normal function—it misfires. Those misfires produce the vivid experiences people later report as NDEs. The tunnel, the light, the sense of peace, the encounter with deceased loved ones—all of it, on this view, is nothing more than the final spasms of a brain in its death throes. It is a compelling narrative. It sounds scientific. And it is the single most popular explanation offered by those who deny that NDEs point to the reality of consciousness beyond the body.
Michael Marsh makes this the backbone of his entire book. In his introduction, he lays down his central thesis in clear terms: extra-corporeal experiences “are likely to be generated by metabolically disturbed brains especially during the period when they are regaining functional competence.”1 That sentence is the engine driving everything else in Out-of-Body and Near-Death Experiences. For Marsh, NDEs are not mysterious. They are not windows into another reality. They are the predictable byproducts of brains that have been knocked offline by some clinical crisis—cardiac arrest, severe blood loss, respiratory failure—and are now rebooting. The strange experiences happen during the recovery phase, as the brain claws its way back to full conscious awareness.
Marsh is careful to emphasize that he is not talking about dead brains. He does not want to be caught in the trap of trying to explain how a brain with no activity can generate experiences. Instead, he shifts the focus. The experiences, he insists, are generated as the brain recovers. He compares this process to the phenomenology of dream states, arguing that ECEs share much in common with the subconscious mentation that occurs during hypnagogic states (falling asleep) and hypnopompic states (waking up).2 On his view, the NDE is something like a very intense, very vivid waking dream—produced by a brain that is transitioning from unconsciousness back to full awareness.
He puts it plainly in his conclusion: the phenomenology of ECEs is “more akin to subconscious oneiric experience than to robust, credible accounts of an authentic, veridical glimpsing of the hereafter.”3 The word “oneiric” simply means “dreamlike.” For Marsh, that is all an NDE really is: a dream that the brain spins up while it is getting back on its feet.
Susan Blackmore, the British psychologist and outspoken atheist, developed a slightly different version of this idea in her influential 1993 book Dying to Live. Blackmore proposed that each element of the NDE—the tunnel, the light, the life review, the feeling of peace—could be explained by specific neurological processes occurring in a dying brain. The tunnel, she argued, is produced by the random firing of neurons in the visual cortex as oxygen levels drop. Because more neurons serve the center of the visual field than the periphery, random firing would create a bright spot in the center surrounded by darkness—giving the impression of rushing through a tunnel toward a light.4 The feeling of peace might be caused by endorphins (the body’s natural painkillers) flooding the brain under stress. The life review might result from random seizure activity in the temporal lobe, which houses our memory systems.
Blackmore was honest enough to admit that much of her hypothesis was speculative. She wrote openly that when it comes to what actually happens in the brain at the point of death, “a first approximation to an answer is simply to say that we do not know.”5 But she believed the available neuroscience pointed firmly in the direction of a physical explanation. No souls. No spirits. No afterlife. Just a brain doing strange things as it shuts down.
Fischer and Mitchell-Yellin take a somewhat different angle on the same basic position. They do not develop a detailed neurological model the way Blackmore does. Instead, they argue philosophically that we should not be too hasty to abandon physicalism (the view that everything about the mind can be explained in terms of the brain and body). They suggest that our ability to measure brain activity is still quite primitive, and that what looks like an inactive brain today might, with future technologies, turn out to have been active all along. “It seems quite reasonable to allow for the possibility that future technologies will be able to detect brain functioning in cases where current technologies tell us there is none,” they write.6 In other words, even if we cannot currently explain how a brain in cardiac arrest could generate an NDE, that does not mean no physical explanation exists. Give science more time.
Taken together, these three positions—Marsh’s metabolic recovery model, Blackmore’s dying brain hypothesis, and Fischer and Mitchell-Yellin’s philosophical promissory physicalism—form the skeptical consensus. NDEs are brain-state phenomena. The brain is the whole story. And any appearance to the contrary is the result of our incomplete understanding of neuroscience, not evidence for the survival of consciousness beyond the body.
That is the argument. I want to give it a fair hearing, because it deserves one. Marsh is a medical doctor with a D.Phil. from Oxford. Blackmore is a trained psychologist with decades of research. Fischer and Mitchell-Yellin are sharp analytic philosophers. These are serious people making serious arguments. I respect the challenge they bring.
I should note, too, that the dying brain family of explanations is not monolithic. Marsh’s version is more sophisticated than Blackmore’s in some respects. He does not simply say the brain is shutting down and misfiring. He proposes a specific temporal sequence. In what he calls the “early phase” of the ECE, the recovering brain begins to revascularize—that is, blood starts flowing back into brain regions that had been starved. This process, Marsh argues, begins with the visual cortex, producing the tunnel and light phenomena as the visual system comes back online. The early-phase phenomenology, he says, is “predicated on, and conditioned by, premorbid cognitive constructs in subjects’ minds of what dying and the afterlife would be like.”54 In other words, the brain pulls from its pre-existing mental files—everything the person has ever thought, imagined, or believed about death and the afterlife—and plays it back as a kind of dream movie during recovery.
In the “late phase,” full conscious awareness progressively returns. The person begins to regain awareness of their body, their surroundings, and their earthly responsibilities. This explains, Marsh says, why NDErs often describe a reluctance to “come back”—they are experiencing the transition from a pleasant, unfiltered subconscious state to the harder reality of waking life.55
Marsh is also insistent that we should not be looking at dead brains to explain ECEs. He writes: “I am immediately freed from the straitjacket of brokering dead brains, but instead, offered the means of engaging with considerably more fruitful neurophysiological opportunity and possibility.”56 This is a clever rhetorical move. By placing the experience in the recovery phase rather than in the arrest itself, he sidesteps the problem of explaining consciousness in a brain with no measurable activity. Or so he thinks. As we will see, the move does not work as well as he hopes.
But the dying brain hypothesis has problems. Deep ones. And they go all the way to the foundation.
The dying brain hypothesis sounds elegant in the abstract. It has a certain just-so quality to it: the brain is shutting down, so of course it produces strange experiences. What else would you expect? But when you press on the details—when you hold the hypothesis up against the actual evidence—it starts to crack in places that matter.
The first crack is the selectivity problem. If NDEs are caused by the physiological processes of a dying brain, then everyone whose brain undergoes those processes should have an NDE. A dying brain is a dying brain. The hypoxia, the neurotransmitter floods, the cortical disinhibition—these are universal events. They happen to every person who goes into cardiac arrest. Yet the evidence tells us that only a minority of cardiac arrest survivors report NDEs. In van Lommel’s landmark Lancet study—the largest prospective study of NDEs at the time—344 consecutive cardiac arrest survivors were interviewed, and only about 12 percent reported a deep NDE.7 The British study led by Sam Parnia and Peter Fenwick found a similar figure: roughly 10 to 11 percent of survivors reported memories from the period of cardiac arrest.8 Kenneth Ring’s earlier estimate put the number somewhat higher—around 30 percent—but that still means the majority of people who come close to death do not report an NDE.9
Think about what that means for the dying brain hypothesis. All 344 patients in van Lommel’s study suffered cardiac arrest. All of them experienced the same physiological crisis: their hearts stopped, blood flow to the brain ceased, and within seconds their brains were oxygen-starved. If the dying brain is the explanation, why did only a fraction of them have NDEs? The chemistry was the same. The physiology was the same. The crisis was the same. Yet the outcomes were wildly different.
Key Argument: If NDEs were caused by the physiology of a dying brain, we would expect all dying brains to produce them. Yet only a small minority of cardiac arrest survivors report NDEs. The dying brain hypothesis cannot explain this selectivity.
Marsh never addresses this problem squarely. He builds his entire argument on the premise that the recovering brain generates these experiences, but he offers no explanation for why only some recovering brains do so. The selectivity problem is a serious blow to any purely physiological account.
The second crack is the coherence problem. A brain in severe metabolic distress does not produce clear, organized, structured thought. It produces the opposite. Decades of medical research on the effects of hypoxia (oxygen deprivation), hypercarbia (excess carbon dioxide), and other metabolic disturbances paint a remarkably consistent picture: when the brain is compromised, thinking becomes confused, fragmented, and disorganized. Patients experiencing hypoxia report mental laziness, irritability, difficulty concentrating, slowness in reasoning, and trouble remembering.10 As oxygen levels drop further, confusion deepens until consciousness is lost entirely. There is nothing lucid about it. Nothing coherent. Nothing memorable.
NDEs are the mirror image of this. They are consistently described as clear, vivid, structured, and profoundly meaningful. Experiencers do not report confusion or disorientation. They report heightened awareness. They describe thinking more clearly than they have ever thought in their lives. Many say the experience felt “more real than real”—not less real, but more. Parnia and Fenwick, both physicians, put the contrast bluntly: “Any acute alteration in cerebral physiology such as occurring in hypoxia, hypercarbia, metabolic, and drug induced disturbances and seizures leads to disorganized and compromised cerebral function,” whereas “NDEs in cardiac arrest are clearly not confusional and in fact indicate heightened awareness, attention and consciousness at a time when consciousness and memory formation would not be expected to occur.”11
This is devastating for the dying brain hypothesis. The hypothesis predicts confusion. The evidence shows clarity. The hypothesis predicts fragmented experience. The evidence shows structured, coherent narratives. The hypothesis predicts that a compromised brain should produce compromised experience. The evidence shows the opposite. Something is very wrong with the model.
The third crack is the one the critics most want to avoid: the veridical evidence. Even if you could somehow explain the tunnel, the light, and the feeling of peace as products of a dying brain, you still have to explain how a person with no measurable brain activity could accurately report real events happening in the physical world around them—events they should have had no way of knowing about. The dentures case from van Lommel’s Lancet study, the Pam Reynolds case, the dozens of verified cases catalogued in The Self Does Not Die—these are not vague feelings or dreamlike impressions. They are specific, detailed, verifiable observations that were later confirmed by medical staff and third parties. No version of the dying brain hypothesis has ever explained how a malfunctioning brain could produce accurate perceptions of external reality.
And there is one more weakness worth naming. Marsh frames his entire thesis around the recovery phase. The experiences happen, he says, as the brain regains functional competence. But recovery from any significant insult to the brain is not a graceful process. It is messy. It begins with confusion, disorientation, and amnesia. Even recovery from a simple faint involves a period of grogginess and muddled thinking. Recovery from cardiac arrest is far more severe. Chris Carter puts the problem sharply: it is very difficult to see how experiences during a confused mental state could result in the clear, detailed, and lasting memories that NDErs consistently report.12 Marsh’s recovery model has the same problem as the dying brain model: it predicts confusion where the evidence shows clarity.
We touched on the selectivity problem above, but it deserves a much fuller treatment, because it strikes at the very heart of the dying brain hypothesis. If NDEs are caused by the universal physiology of a dying or recovering brain, then they should be universal. They are not.
Van Lommel’s 2001 study, published in The Lancet, was a prospective study—meaning the researchers did not go looking for people who had had NDEs and ask them about it. Instead, they interviewed every single survivor of cardiac arrest in ten Dutch hospitals over a defined period. Out of 344 patients, 62 (18 percent) reported some NDE elements, and 41 (12 percent) reported a deep or core NDE. After adjusting for patients who had experienced more than one cardiac arrest, the researchers estimated the true frequency at roughly 5 percent.13
The British study by Parnia, Fenwick, and colleagues followed a similar design. Over one year, they interviewed all 63 survivors of cardiac arrest at a general hospital. Seven patients (11 percent) reported memories, and four had full-blown NDEs.14
These numbers are striking. In both studies, the vast majority of cardiac arrest survivors—more than 80 percent—reported nothing at all from the period of their arrest. No tunnel. No light. No deceased relatives. Nothing. Their brains went through the same physiological crisis, the same oxygen deprivation, the same metabolic disturbance. But only a small fraction had NDEs.
Van Lommel and his team investigated whether any physiological factor could account for the difference. They compared NDE patients with non-NDE patients on every measurable variable they could find: duration of cardiac arrest, duration of unconsciousness, the need for intubation, the administration of medications, previous NDE knowledge, religious belief, education, and many others. The result? No physiological or demographic factor could explain who had an NDE and who did not.15 The dying brain was the same in both groups. Something other than dying-brain physiology was selecting who would have the experience.
This finding alone should give any honest skeptic serious pause. If the dying brain were the explanation, you would expect to find some physiological marker that correlates with the occurrence of NDEs—lower oxygen levels, longer cardiac arrest times, different medications, something. Van Lommel found nothing. Parnia found nothing. The dying brain hypothesis predicts a correlation that does not exist.
I want to drive this point home, because it is one of the most powerful arguments against the dying brain hypothesis, and Marsh never deals with it adequately.
We know, from decades of medical research, exactly what happens when the brain is deprived of oxygen. The effects have been documented in thousands of laboratory experiments, in high-altitude studies, in military aviation research, and in clinical settings. Carter summarizes the research of R. A. McFarland, who studied the effects of hypoxia on mountain climbers at extreme elevations. The climbers reported “mental laziness, heightened irritability, difficulty in concentrating, slowness in reasoning, and difficulty in remembering.”16 That is what oxygen deprivation does to the brain. It degrades performance. It muddies thinking. It does not sharpen it.
In the old days, it was common practice for medical students to be demonstrated the effects of oxygen deprivation firsthand. Students would breathe through a carbon dioxide absorber into a spirometer filled with air, slowly depleting the oxygen without realizing it. Their performance at assigned tasks would deteriorate progressively until they lost consciousness. In thousands of such experiments on thousands of people, not a single person ever reported an NDE.17 Not one. The brain was starved for oxygen, just as the dying brain hypothesis requires. But the result was always the same: confusion, declining performance, and loss of consciousness. Never a tunnel. Never a light. Never a life review. Never anything remotely resembling an NDE.
Now compare that with what NDErs actually report. Here is the contrast that should keep every proponent of the dying brain hypothesis awake at night: at the very moment when the brain should be at its worst—when oxygen is plummeting, when neural activity is grinding to a halt, when every medical textbook says consciousness should be impossible—NDErs report the most vivid, clear, and meaningful experience of their lives. They describe thinking with perfect clarity. They describe perceiving with superhuman acuity. They describe an experience so overwhelmingly real that normal waking life seems pale and dreamlike by comparison.
Penny Sartori, a critical care nurse who conducted a five-year prospective study of NDEs in her intensive care unit, kept meticulous records of the drugs administered to her patients. Of those who reported NDEs, 26.7 percent had received no sedatives at all, and 66.7 percent had not been given painkillers.18 Meanwhile, a large proportion of her total sample—including those who did not have NDEs—had received painkilling and sedative drugs. Nearly all of the patients who had hallucinations (as distinguished from NDEs) had been given such drugs. Sartori concluded that drugs contribute to confused, hallucinatory experiences but actually seem to inhibit NDEs rather than cause them.19 Read that again. Drugs that muddy the brain produce muddied experiences. But NDEs, which are crystal clear, tend to occur in the absence of such drugs. The pharmacological evidence points away from the dying brain hypothesis, not toward it.
Insight: The pharmacological evidence runs directly counter to the dying brain hypothesis. Drugs that alter brain chemistry produce confused, hallucinatory experiences. NDEs, by contrast, are clear and structured—and are more likely to occur in the absence of brain-altering medications. If the dying brain were the cause, we would expect the opposite pattern.
To really understand why the dying brain hypothesis fails, you need to know what actually happens inside the brain when the heart stops. I want to walk you through this carefully, because the medical details matter.
The human brain weighs about three pounds. It makes up roughly 2 percent of the body’s mass, but it uses about 15 percent of the body’s oxygen supply. It is an extraordinarily hungry organ, entirely dependent on a constant flow of oxygen-rich blood from the heart. When the heart stops—when it goes into cardiac arrest—that blood flow ceases. Immediately.
Within seconds, the brain begins to lose function. Consciousness is typically lost within about ten seconds of cardiac arrest.20 The electroencephalogram (EEG)—a device that measures electrical activity in the outer layer of the brain, the cerebral cortex—shows a dramatic slowing of brain waves within an average of 6.5 seconds after circulation stops. Within ten to twenty seconds, the EEG goes flat.21 A flat EEG means no measurable electrical activity in the cerebral cortex. The cortex is the part of the brain that handles higher cognitive functions—thinking, reasoning, perception, memory formation. With a flat EEG, these functions are, by every measure we have, offline.
Breathing stops soon after, because the brain stem (the lower portion of the brain that controls automatic functions like breathing and heart rate) is also losing oxygen. Within minutes, brain stem reflexes are lost as well—the gag reflex, the pupil reflex. The pupils become fixed and dilated. At this point, the patient meets two of the three clinical criteria for death: no cardiac output and no spontaneous breathing. In the clinical setting, patients frequently meet all three criteria, including fixed dilated pupils, due to the loss of brain stem function.22
If circulation is not restored within three to five minutes, irreversible brain damage typically begins. If it is restored through CPR (cardiopulmonary resuscitation), the brain does not snap back to normal like a lightbulb being switched on. Recovery is gradual, often taking minutes or even hours. And during that recovery period, the brain is confused. Disoriented. Patients emerging from cardiac arrest typically experience a period of post-resuscitation confusion. Even patients recovering from a simple faint go through a confusional state. An acute cerebral insult—which is what cardiac arrest is—produces an even greater period of confusion.23
This is the landscape in which the dying brain hypothesis must operate. Marsh claims that NDEs are generated by brains recovering from metabolic insult. But recovery from metabolic insult is, by definition, a confusional process. It does not produce clarity. It does not produce heightened awareness. It does not produce the kind of structured, detailed, and profoundly meaningful experiences that NDErs consistently report. The medical evidence flatly contradicts the theory.
One of the most informative comparisons comes from military aviation research. When fighter pilots undergo rapid acceleration during combat maneuvers, the blood is displaced downward, away from the brain. If the acceleration is severe enough—typically around five times normal gravity—the brain is deprived of oxygen and the pilot loses consciousness. This is called G-LOC: gravity-induced loss of consciousness. The U.S. Air Force has studied this phenomenon extensively, using a ground-based human centrifuge to safely reproduce the conditions under controlled circumstances.
Over a thousand episodes of G-LOC have been documented. The effects are well known. Consciousness is lost within about six seconds of acceleration stress above tolerance. The period of unconsciousness lasts an average of twelve seconds. During roughly the last four seconds, convulsions are observed in about 70 percent of subjects. After regaining consciousness, there is a period of confusion and disorientation lasting about another twelve seconds.24
Some G-LOC subjects (about 35 to 40 percent) report short dreams during the unconscious period—brief episodes that Navy physician James Whinnery called “dreamlets.” These dreamlets typically involve family members, pleasant settings, and mundane activities. One subject dreamed of being at home with his mother and brother. Another dreamed of floating in a blue ocean. A third dreamed of being in a grocery store, unable to move his arm to reach a carton of ice cream.25
Supporters of the dying brain hypothesis have pointed to G-LOC as evidence that oxygen deprivation can produce NDE-like experiences. Blackmore and others noted that G-LOC subjects sometimes report tunnel vision, floating sensations, and euphoria—elements that superficially resemble parts of the NDE.26 Marsh himself draws on the centrifuge data to support his thesis that ECE phenomenology can be produced by brains recovering from hypoxic insult.27
But look at the actual details. The differences between G-LOC episodes and NDEs are far more significant than the similarities. Carter catalogs them systematically.28 First, tunnel vision (a narrowing of the visual field caused by falling blood pressure in the eyes) is not the same as perceiving a tunnel. NDErs describe entering a structured tunnel—often with walls, often leading somewhere. Tunnel vision is simply the loss of peripheral sight. Second, the “dreamlets” are described by the subjects themselves as indistinguishable from ordinary sleep dreams. They are fragmentary, often illogical, and frequently incorporate the physical convulsions the subject is undergoing—one man dreamed he was fishing and jerking his rod because his arms were convulsing. NDEs are the opposite: lucid, structured, and consistently described as utterly unlike dreams. Third, no G-LOC subject has ever reported a life review. Fourth, the dreamlets do not feature deceased relatives or a being of light. Fifth, the experience never ends with a decision to return—a hallmark of the NDE. Sixth, G-LOC episodes do not produce the lasting personality changes and value shifts that consistently follow NDEs.29
Perhaps the most telling testimony comes from Allan Pring, a British Royal Air Force pilot who experienced both anoxia at high altitude during his flying career and a near-death experience years later. Pring was in a unique position to compare the two. He reported that the NDE was nothing like the anoxia. The NDE was “crystal clear,” he said, and remained vivid for fifteen years. The anoxia was confusion and nothing more.30 When someone has experienced both a genuinely oxygen-deprived brain and an NDE, and tells you they are nothing alike, that is powerful testimony against the dying brain hypothesis.
Here is where the argument becomes most pointed. The dying brain hypothesis, in any of its forms, requires the brain to be doing the work. Whether you call it a dying brain (Blackmore), a recovering brain (Marsh), or a brain whose activity we simply cannot yet detect (Fischer and Mitchell-Yellin), the fundamental claim is the same: the brain is generating the experience. But what happens when the evidence shows that the brain is not generating anything at all?
As we have seen, the EEG typically goes flat within ten to twenty seconds of cardiac arrest. A flat EEG means no measurable cortical activity. The cerebral cortex is the structure most closely associated with higher cognitive functions—thought, reasoning, perception, and memory. With a flat EEG, the brain is, by every standard medical measurement, incapable of producing conscious experience.
Yet NDEs occur during this window. In fact, the evidence from van Lommel’s study and others suggests that many NDEs begin very early in the cardiac arrest, well before any resuscitation efforts are underway. Jeffrey Long, a radiation oncologist who has collected and analyzed thousands of NDE reports, made a striking observation: the substantial majority of people who have an NDE during cardiac arrest report seeing their physical bodies well before CPR begins. Once CPR is initiated, the content and flow of the NDE does not change—suggesting that blood flow to the brain is not affecting the experience at all.31
The authors of The Self Does Not Die have compiled more than thirty published cases of confirmed consciousness during cardiac arrest, many involving veridical extrasensory perception.32 These are not vague impressions. They are cases where patients accurately described events that occurred while their brains showed no measurable activity—events that were subsequently confirmed by medical staff and other witnesses. The sheer number of such cases, the authors argue, makes it unreasonable to continue doubting the reality of the phenomenon.
Key Argument: More than thirty published cases document confirmed consciousness and often veridical perception during cardiac arrest, when the EEG was flat and no measurable cortical activity was present. The dying brain hypothesis cannot account for complex, lucid consciousness in the absence of the brain activity it claims to require.
The materialist response to this evidence has typically been to suggest that perhaps some residual brain activity, too faint to be detected by current EEG technology, is responsible for the experience. Fischer and Mitchell-Yellin lean heavily on this line of reasoning: future technologies might reveal activity we currently cannot measure.33 It is a tempting move, but it has serious problems.
First, it is a promissory argument. It asks us to reject the current evidence in favor of evidence that does not yet exist. That is not science. Science works with the evidence at hand. You cannot dismiss a powerful body of data by gesturing toward a hypothetical future discovery that might, someday, possibly explain it away.
Second, even if some residual neural activity exists below the threshold of EEG detection, the relevant question is not whether any activity exists at all, but whether the specific kind and amount of activity required for complex conscious experience exists. Van Lommel has addressed this point carefully. As he notes, consciousness requires the coordinated functioning of large portions of the brain working together—the brainstem, the cerebral cortex, the hippocampus, the thalamus, and other structures. A flat EEG indicates that the cortex, at minimum, is not functioning in the coordinated way that neuroscience tells us is necessary for conscious experience.34 Even if some faint, isolated neural firing persists somewhere deep within the brain, that is a far cry from the kind of unified, large-scale cortical activity that contemporary neuroscience considers the necessary condition of consciousness.
As the authors of The Self Does Not Die argue: the question is not whether materialists have proven that remaining brain activity must be linked to an NDE. The question is whether there can be sufficient brain activity during cardiac arrest to make an NDE possible as a product of brain activity. On the evidence, the answer is clearly no. Abundant medical evidence supports the conclusion that cardiac arrest cannot be coupled with enough cortical activity to explain the complex, coherent contents of NDEs.35
Chris Carter, in Science and the Near-Death Experience, has provided what I consider the most thorough demolition of the Blackmore version of the dying brain hypothesis in print. Carter takes Blackmore’s model apart element by element, and the results are not kind to her theory.
Start with the tunnel. Blackmore proposed that random firing of neurons in the visual cortex, caused by oxygen deprivation, would produce a bright impression in the center of the visual field that spreads outward—creating the tunnel effect. Carter points out several devastating problems.36 First, reports of a tunnel are conspicuously absent from most non-Western NDE accounts. If the tunnel is a hardwired physiological response to anoxia, it should show up everywhere. It does not. Even in Western accounts, a tunnel or passage through darkness is reported only about 30 percent of the time. Second, Blackmore attempted to test her prediction with a small study of eight Indian NDErs, but her own data did not support her theory. Not one of the eight subjects reported the pattern her model predicted—a bright spot in the center of the visual field. Critics described her handling of the data as “torturing data until they give you the answer you need.”37
Then there is the matter of the deceased relatives. Blackmore attributed the various elements of the NDE to random neural disinhibition—essentially, the brain’s circuits firing without their normal controls. But as William Serdahely pointedly asked: if the brain is randomly firing neurons, why does it almost always call up images of deceased loved ones rather than living ones? If the brain is simply pulling memories at random for comfort, you would expect it to reach for the most recently accessed and most emotionally comforting memories—which would far more likely be of living family members. Yet NDErs almost universally encounter the dead, not the living. And in some cases—as we have seen in earlier chapters—they encounter deceased relatives they did not even know had died.38 Random firing does not explain this. A dying brain randomly accessing stored memories does not selectively choose only the dead.
Blackmore’s hypothesis also makes predictions that later research has disconfirmed. She predicted, for example, that the dying brain should consistently produce a tunnel experience. But in Jeffrey Long’s survey of over 613 NDErs, only about one-third reported going through a tunnel.39 She predicted that people who “dream in a bird’s-eye view” should be more likely to have out-of-body experiences. Sabom tested this prediction and found no difference between the dreaming patterns of NDErs with and without out-of-body components.40
Perhaps the most fundamental problem with Blackmore’s approach, however, is one that Miller highlights. Her hypothesis is built largely on “what ifs” rather than established facts. She proposes mechanisms that could, in theory, produce certain NDE elements. But proposing a mechanism is not the same as demonstrating it. To confirm her hypothesis, her speculations would need to be tested. She admitted as much. And over the decades since Dying to Live was published, many of her proposed explanations have been researched and found wanting. As a summary article by leading NDE researchers in 2009 concluded: “From our review of the characteristics of Western NDErs, we found little evidence to support previously proposed biological, psychological, or sociological explanations as the sole cause(s) of NDEs.”41
The dying brain hypothesis rests on an assumption that is so deeply embedded in modern neuroscience that many people do not even realize it is an assumption. That assumption is the production model of consciousness: the idea that the brain produces consciousness the way a factory produces goods. On this view, consciousness is an output of neural activity. No brain, no consciousness. Damage the brain, damage consciousness. Shut the brain down, and consciousness disappears entirely.
This model is not the only option. There is an alternative that has a long intellectual pedigree and that fits the NDE evidence far better: the filter model, sometimes called the transmission model. On this view, the brain does not produce consciousness. Instead, the brain filters, transmits, and channels a consciousness that exists independently of it—somewhat the way a radio receives and transmits a signal that originates elsewhere. Damage the radio, and the signal gets distorted. Destroy the radio, and the signal can no longer be received in that location. But the signal itself continues to exist.
This idea is not new. It was proposed in the late nineteenth century by the pioneering psychologist William James, and it was developed further by the philosopher Henri Bergson. More recently, it has been revived by scholars like Edward Kelly, Emily Williams Kelly, and their colleagues in the landmark volume Irreducible Mind.52 The filter model does not require any supernatural leaps. It simply proposes that the relationship between brain and consciousness is one of correlation and mediation, not production and identity. When the brain is functioning normally, consciousness is filtered through it and shaped by it. When the brain is damaged, the filtering is disrupted. And when the brain shuts down entirely—as in cardiac arrest—the filter is removed, and consciousness may actually function more freely, not less.
The filter model explains something that the production model cannot: why consciousness becomes clearer and more vivid during NDEs when it should, on the production model, become confused and degraded. If the brain is a filter that normally constrains and limits consciousness, then removing that filter would result in exactly what NDErs report—an experience of heightened awareness, expanded perception, and extraordinary clarity. The production model predicts that a failing brain should produce failing consciousness. The filter model predicts that a failing brain should release consciousness from its normal constraints. The NDE evidence fits the filter model. It does not fit the production model.
Now, I am not claiming that the filter model has been proven. It is a hypothesis, and like all hypotheses, it needs further testing. But here is the important point: the dying brain hypothesis is not the only game in town. The critics often write as though physicalism is the default position and anyone who challenges it bears the full burden of proof. But physicalism itself has never explained how the brain produces consciousness. No one has ever identified the mechanism by which electrochemical signals in neurons give rise to subjective experience—the felt quality of seeing the color red, or hearing a melody, or tasting chocolate. This is the hard problem of consciousness, and it remains entirely unsolved. If physicalism cannot even explain how a healthy brain produces ordinary consciousness, we should be cautious about using it to explain away the extraordinary consciousness reported during NDEs.
Marsh, to his credit, acknowledges the difficulty of defining consciousness. He writes that “a definition of consciousness or conscious-awareness is difficult because the concept is obstinately abstract, although much is known of the mechanisms underpinning the experiences of consciousness.”53 But he then proceeds to build his entire argument on the assumption that consciousness is a product of brain activity, without ever establishing that assumption. He simply takes it for granted. That is a significant gap in his reasoning. You cannot use the production model to explain away NDEs unless you have first demonstrated that the production model is correct. And no one has done that.
The filter model is not a retreat into mysticism. It is a scientifically respectable hypothesis that accounts for the full range of NDE evidence more effectively than the production model. It explains the selectivity problem (the filter may be more easily removed in some individuals than others). It explains the coherence problem (removing the filter produces clarity, not confusion). It explains consciousness during flat EEG (the filter is offline, but consciousness continues). And it is consistent with the veridical evidence (consciousness, freed from the filter, can perceive without the limitations of sensory organs). The dying brain hypothesis, built on the production model, cannot match this explanatory power.
Everything I have said so far targets the general features of the dying brain hypothesis—the selectivity problem, the coherence problem, the flat-EEG problem, the failure of Blackmore’s specific predictions. But none of these problems is as lethal to the hypothesis as the veridical evidence.
We have covered this evidence in detail in earlier chapters, so I will not repeat every case here. (Readers looking for the full treatment should turn to Chapters 4 and 5.) But I want to make the connection explicit, because the dying brain hypothesis must be evaluated against the strongest evidence, not just the average case.
Consider what the dying brain hypothesis requires. It requires that the brain, in a state of severe metabolic compromise, generates an experience. Fine. Suppose we grant, for the sake of argument, that a compromised brain can produce vivid subjective experiences. Suppose we set aside the coherence problem. Even then, the hypothesis requires that those experiences be internally generated—that is, produced by the brain from its own stored information, memories, and neural patterns. A dying brain can only work with what it already has.
But veridical NDEs involve information the brain did not have. When a patient accurately describes a conversation that took place in a room they were not in, a piece of medical equipment they had never seen before, or the actions of a nurse they had never met—while their heart was stopped, their eyes were taped shut, and their brain showed no measurable activity—the dying brain hypothesis has nothing to say. A brain cannot produce accurate perceptions of events it has no access to. Not even a healthy brain can do that, let alone a dying one.
Janice Holden’s 2007 analysis of every reported case of apparently veridical perception during an NDE since 1975 found 107 such cases. Using the most stringent criteria—where even a single inaccurate detail would classify the case as inaccurate—she found that only 8 percent involved any error. Thirty-seven percent were determined to be completely accurate by independent, objective verification.42 That is a remarkable record for experiences that the critics insist are hallucinations. If NDEs were truly hallucinations generated by a dying brain, you would expect a far higher error rate. You would certainly not expect 92 percent accuracy—the figure reported in van Lommel’s own data set.43
The dying brain hypothesis simply cannot account for veridical perception. This is its Achilles heel. Every other problem I have identified—the selectivity problem, the coherence problem, the pharmacological evidence, the flat-EEG data—is damaging. But the veridical evidence is fatal. A dying brain cannot see what it has no means of seeing.
This is probably the most common fallback for proponents of the dying brain hypothesis, and Fischer and Mitchell-Yellin lean on it heavily. The argument goes like this: neuroscience is a young science. Our tools for measuring brain activity are still crude. Maybe in the future, we will discover brain processes that can explain NDEs. Therefore, we should not abandon physicalism just because we cannot currently explain these experiences.
I understand the appeal of this argument. Humility about the limits of our current knowledge is a good instinct. But there are serious problems with using it to dismiss the NDE evidence.
First, as I noted above, this is a promissory argument. It asks us to reject the evidence we have on the grounds that future evidence might tell a different story. You could use this argument to dismiss any inconvenient data about anything. No matter how strong the evidence is, someone can always say, “Maybe someday we’ll find a different explanation.” That is not a scientific rebuttal. It is a way of avoiding the question.
Second, the argument cuts both ways. If we are humble about the limits of neuroscience, then we should also be humble about neuroscience’s ability to tell us that consciousness is entirely produced by the brain. Physicalism itself is an assumption that neuroscience has not proven. No one has ever explained how physical processes in the brain produce subjective conscious experience. This is what philosopher David Chalmers famously called “the hard problem of consciousness”—and it remains unsolved. If we do not yet fully understand the brain, then we are not in a position to dogmatically assert that the brain is all there is.
Third—and this is the key point—the argument from future discoveries cannot address the veridical evidence. No future brain scanner, however sophisticated, will explain how a patient with no measurable brain activity accurately described events happening in another room. The problem there is not one of measurement precision. It is a problem of basic physics. A brain locked inside a skull, with no sensory input, no functioning visual cortex, and no blood flow, cannot perceive events happening elsewhere. Better technology will not solve that problem. Only a different model of consciousness will.
Common Objection: “Neuroscience is still young. Future discoveries may explain NDEs in purely physical terms.” Response: This is a promissory argument, not a scientific one. It asks us to discount present evidence in favor of hypothetical future evidence. Moreover, even if future technologies could detect previously unmeasurable brain activity, they would still not explain how a brain with no sensory input could accurately perceive events happening in distant locations. The veridical evidence requires a different explanation, not just better instruments.
A skeptic might argue that the experiences do not actually occur during the period of cardiac arrest, but rather in the brief moments before consciousness is lost or during the recovery phase afterward. Marsh in particular has pushed this line. He argues that the experiences are generated during the “reawakening” phase, when the brain is regaining its functional competence.44
There are two responses. First, the transition to unconsciousness during cardiac arrest is extremely rapid. Blood flow ceases and consciousness is typically lost within ten seconds. As Carter observes, this window is so short that it is difficult to see how a complex, structured NDE could be generated during it.45 Furthermore, any sort of insult to the brain typically results in amnesia—a loss of memory around the event. The severity of the amnesia is actually used clinically to assess the severity of the brain injury. A brain going offline does not lay down clear memories in its final seconds of function. It loses them.
Second, the recovery hypothesis faces the problem we have already identified: recovery from any acute cerebral insult is via a state of confusion. Even recovery from simple fainting involves grogginess and disorientation. Recovery from cardiac arrest is vastly more severe and prolonged. Parnia and Fenwick point out that after cardiac arrest, the brain does not function normally for an extended period. EEG activity may not return to normal for several minutes or even hours, depending on the duration of the arrest.46 How could clear, lucid, detailed experiences be generated during a confusional recovery phase?
The veridical cases make this objection even harder to sustain. In many documented cases, the patient’s out-of-body observations correspond to events that took place at a specific, identifiable point during the cardiac arrest itself—not before or after. In van Lommel’s famous dentures case, the patient observed specific details (including where the nurse put his dentures) during the earliest stages of his resuscitation, before he had been fully stabilized. The nurse confirmed the details. The timing of the observation locked the experience to a moment when the brain, by every medical measure, was not functioning.47
This is a common skeptical refrain, and it sounds reasonable on the surface. We should not build scientific conclusions on mere stories. But the objection mischaracterizes the NDE evidence.
An anecdote is an unverified personal story. The veridical NDE cases we have been discussing are not mere anecdotes. They are reports that have been corroborated by third-party witnesses, confirmed by medical records, and in many cases investigated by researchers shortly after the event occurred. When a patient describes specific details of their resuscitation—details that are confirmed by the medical team who performed it—that is not an anecdote. It is corroborated testimony. It is evidence.
Moreover, the NDE evidence is not a handful of isolated cases. As of The Self Does Not Die, there are over 100 documented cases of verified paranormal phenomena during NDEs, drawn from thirty-nine different publications by thirty-seven different authors or author teams.48 There are four major prospective studies (van Lommel, Parnia, Sartori, Greyson) conducted in hospital settings using rigorous methodology. There are thousands of NDE reports collected by researchers like Jeffrey Long. And there are over 900 articles on NDEs published in scholarly, peer-reviewed journals.49 This is not a collection of campfire stories. It is a serious, growing body of empirical evidence that the skeptics have consistently underestimated.
Some readers may be aware of a study published in 2013 by Jimo Borjigin and colleagues, which detected a surge of neurophysiological activity in the brains of rats after cardiac arrest.50 This study received significant media attention, with headlines suggesting that the “dying brain” had been caught in the act of producing NDEs. Marsh would likely welcome such a finding as support for his thesis.
But the study requires careful evaluation. First, it was conducted on rats, not humans, and we should be cautious about extrapolating from rodent brains to human consciousness. Second, the burst of activity lasted only about thirty seconds and occurred in a very specific context (the rats had been anesthetized before cardiac arrest was induced). Third—and most importantly—detecting a surge of neural activity does not demonstrate that the activity produces conscious experience. We already knew that dying brains do things. The question is whether what they do can explain NDEs. The surge documented by Borjigin and colleagues involved gamma-wave oscillations, which some researchers associate with conscious processing. But no one showed that the rats were having NDEs. No one showed that the activity could produce veridical perception. No one addressed the selectivity problem, the coherence problem, or any of the other difficulties we have identified. Van Lommel and Greyson published a critique pointing out these and other limitations.51
The Borjigin study is interesting. It adds to our understanding of what happens in the brain at the point of death. But it does not rescue the dying brain hypothesis from its fundamental problems.
A skeptic sympathetic to Marsh might press this point. If NDEs are genuine glimpses of an afterlife or a real state of disembodied consciousness, shouldn’t they all follow the same pattern? Marsh himself emphasizes the idiosyncratic nature of NDE narratives. He writes that the phenomenology is “emphatically idiosyncratic, as each testimony clearly demonstrates.”57 The variability, for Marsh, is evidence that these are brain-generated experiences colored by individual cognitive and cultural content, not genuine encounters with an objective reality.
This is a reasonable observation, but it proves less than Marsh thinks. Consider an analogy. If you asked a hundred people to describe a sunset they had each seen from a different location, at different times, under different weather conditions, you would get a hundred different descriptions. Some would mention colors the others would not. Some would focus on the clouds, others on the horizon. Some would describe it in poetic terms, others in matter-of-fact language. The descriptions would be idiosyncratic. But no one would conclude from this that sunsets do not exist or that each person invented the sunset in their own mind.
The NDE evidence actually shows a remarkable pattern beneath the surface variability. The core features—the out-of-body experience, the tunnel or passage, the brilliant light, the encounter with deceased relatives, the sense of peace and love, the life review, the border or point of no return—recur across cultures, across religions, across age groups, and across historical periods. The variations occur in the details and the cultural “dressing” of the experience: who the being of light is identified as, what the border looks like, what language the deceased relatives speak. This is exactly what we would expect if the experiences are real encounters filtered through individual cognitive frameworks. The core experience is universal. The interpretation is personal. Variability does not disprove reality. If anything, it supports it.
Moreover, as we noted in Chapter 8, the cultural conditioning objection works in reverse. If NDEs were entirely generated by the brain from stored cultural expectations, we would expect them to match those expectations perfectly. But they often do not. Many NDErs report elements that surprise them or contradict their beliefs. Atheists encounter a being of light they were not expecting. Christians meet deceased relatives they did not know had died. Children too young to have absorbed cultural NDE scripts report the same core features as adults. The experiences regularly exceed what the experiencer’s brain “should” produce based on prior knowledge and expectations.58
I want to raise one more point that the dying brain hypothesis simply cannot touch. Shared death experiences are cases in which a healthy bystander—someone sitting at the bedside of a dying loved one—reports experiencing elements of an NDE at the moment of the patient’s death. They report the tunnel, the light, the sense of leaving the body, and sometimes even seeing deceased relatives of the patient, including people they themselves had never met.
The dying brain hypothesis has nothing to say about shared death experiences. The bystander’s brain is not dying. It is not hypoxic. It is not flooded with endorphins or carbon dioxide. It is a perfectly healthy brain, functioning normally, in a person who is not in any medical danger. Yet they report the same kinds of experiences that NDErs report. If NDEs are produced by dying brains, what is producing the experience in a brain that is not dying? We covered shared death experiences more fully in Chapter 9, but I mention them here because they represent a category of evidence that is simply incompatible with any version of the dying brain hypothesis. You cannot explain a non-dying brain’s experience by appeal to the physiology of a dying brain.59
The dying brain hypothesis is the most popular skeptical explanation for NDEs, and at first glance, it seems like a natural fit. Brains under extreme stress do strange things. We all know that. But when you look at the evidence carefully—when you hold the hypothesis accountable to the actual data—it fails at every critical point.
It cannot explain why only a minority of cardiac arrest survivors have NDEs. It cannot explain why NDEs are lucid and coherent when a dying or recovering brain should produce confusion. It cannot explain the pharmacological evidence showing that drugs which alter brain chemistry inhibit rather than produce NDEs. It cannot explain consciousness during flat EEG. It cannot explain the veridical cases.
Marsh deserves credit for attempting a sophisticated neurological account. He is not offering a quick dismissal. His model of the “recovering brain” is more nuanced than Blackmore’s original dying brain proposal. But nuance does not solve the fundamental problems. Whether you call it a dying brain or a recovering brain, you are still asking a compromised organ to do something it should not be able to do—produce the clearest, most vivid, and most accurate experiences of a person’s life at the very moment when it is least capable of producing anything at all.
The evidence does not support the dying brain hypothesis. It challenges it. And the challenge is not subtle. It is direct, specific, and backed by decades of carefully gathered data from prospective studies, verified cases, and medical records. The dying brain hypothesis is not wrong because we want it to be wrong. It is wrong because the evidence says so.
In the chapters that follow, we will examine the specific neurological mechanisms that have been proposed as explanations for individual NDE features—oxygen deprivation, carbon dioxide, the temporal lobe, the temporo-parietal junction, and others. Each of them faces the same fundamental problem: the physiology predicts one thing, and the evidence shows another. The dying brain hypothesis is the roof under which all of these explanations shelter. And if the roof is broken, the whole structure is compromised.
The brain is an extraordinary organ. No one disputes that. But the evidence from near-death experiences suggests that consciousness may be more than what the brain produces. It may be something the brain mediates, channels, and transmits—but does not create. And when the brain goes offline, consciousness does not necessarily go with it. That is what the NDE evidence is telling us. The critics have their hypothesis. We have the data. And the data wins.
↑ 1. Marsh, Out-of-Body and Near-Death Experiences: Brain-State Phenomena or Glimpses of Immortality? (Oxford: Oxford University Press, 2010), p. xvi.
↑ 2. Marsh, Out-of-Body and Near-Death Experiences, chaps. 5 and 7. Marsh draws extended parallels between ECE phenomenology and dream-state mentation, arguing that the subconscious imagery produced during ECEs is “highly reminiscent of the phenomenological bizarreness and illogicalities characteristic of dream-state modes” (p. xx).
↑ 3. Marsh, Out-of-Body and Near-Death Experiences, p. 261.
↑ 4. Susan Blackmore, Dying to Live: Near-Death Experiences (Buffalo, NY: Prometheus Books, 1993), chap. 4. Blackmore’s model of the tunnel experience is built on the theory that random firing of cortical neurons, concentrated in the area of highest neural density (the foveal representation in the visual cortex), would produce a bright center fading to darkness at the periphery.
↑ 5. Blackmore, Dying to Live, quoted in J. Steve Miller, Near-Death Experiences as Evidence for the Existence of God and Heaven: A Brief Introduction in Plain Language, chap. 3 and Appendix 5.
↑ 6. John Martin Fischer and Benjamin Mitchell-Yellin, Near-Death Experiences: Understanding Visions of the Afterlife (Oxford: Oxford University Press, 2016), chap. 4, pp. 31–32.
↑ 7. Pim van Lommel, Ruud van Wees, Vincent Meyers, and Ingrid Elfferich, “Near-Death Experience in Survivors of Cardiac Arrest: A Prospective Study in the Netherlands,” The Lancet 358 (2001): 2039–2045. The study found that 62 of 344 patients (18%) reported some NDE elements, with 41 (12%) reporting a deep NDE. After adjusting for multiple events, the estimated true frequency was approximately 5 percent.
↑ 8. Sam Parnia, D. G. Walker, R. Yeates, and Peter Fenwick, “A Qualitative and Quantitative Study of the Incidence, Features and Aetiology of Near Death Experiences in Cardiac Arrest Survivors,” Resuscitation 48 (2001): 149–156. Seven of 63 survivors (11%) reported memories; four had full NDEs.
↑ 9. Kenneth Ring, Life at Death: A Scientific Investigation of the Near-Death Experience (New York: Coward, McCann and Geoghegan, 1980). Ring’s estimate of approximately 30 percent closely matches Michael Sabom’s finding of 27 percent. See Chris Carter, Science and the Near-Death Experience: How Consciousness Survives Death (Rochester, VT: Inner Traditions, 2010), chap. 7.
↑ 10. R. A. McFarland’s research on the International High Altitude Expedition to Chile, as cited in Carter, Science and the Near-Death Experience, chap. 11.
↑ 11. Sam Parnia and Peter Fenwick, “Near-Death Experiences in Cardiac Arrest: Visions of a Dying Brain or Visions of a New Science of Consciousness?” Resuscitation 52 (2002): 5–11. Also cited in Janice Miner Holden, Bruce Greyson, and Debbie James, eds., The Handbook of Near-Death Experiences: Thirty Years of Investigation (Santa Barbara, CA: Praeger, 2009), 228.
↑ 12. Carter, Science and the Near-Death Experience, chap. 13. Carter notes that recovery from any sort of insult to the brain is via a state of confusion, and even recovery from simple fainting involves a confusional process.
↑ 13. Van Lommel et al., “Near-Death Experience in Survivors of Cardiac Arrest,” 2039–2045. See also Carter, Science and the Near-Death Experience, chap. 7, for discussion of the adjusted frequency estimates.
↑ 14. Parnia et al., “A Qualitative and Quantitative Study,” 149–156.
↑ 15. Van Lommel et al., “Near-Death Experience in Survivors of Cardiac Arrest,” 2039–2045. Van Lommel and his colleagues reported that no physiological, pharmacological, or psychological factor could be identified as the cause of NDEs in their sample.
↑ 16. Carter, Science and the Near-Death Experience, chap. 11, citing McFarland’s research from the 1930s.
↑ 17. Carter, Science and the Near-Death Experience, chap. 11. Carter notes that in “thousands of such experiments on thousands of people, no one ever reported an NDE.”
↑ 18. Penny Sartori, The Near-Death Experiences of Hospitalized Intensive Care Patients: A Five-Year Clinical Study (Lewiston, NY: Edwin Mellen Press, 2008), 235.
↑ 19. Sartori, Near-Death Experiences of Hospitalized Intensive Care Patients, 237, 281. Sartori concluded that “drugs inhibit the NDE, or the recall of it, as opposed to causing it.” She noted that “in some of the most vivid NDEs no drugs were administered at the time of the experience.”
↑ 20. A National Institutes of Health publication (2010) places loss of consciousness during cardiac arrest at approximately 10 seconds. See also Pim van Lommel, “Setting the Record Straight: Correcting Two Recent Cases of Materialist Misrepresentation of My Research and Conclusions,” Journal of Near-Death Studies 30, no. 2 (2011): 107–119, at 113.
↑ 21. Carter, Science and the Near-Death Experience, chap. 13, citing the medical literature on EEG changes during cardiac arrest. The initial slowing of EEG waves occurs within an average of 6.5 seconds after circulation stops. Within ten to twenty seconds, the EEG shows a flat line.
↑ 22. Parnia et al., “A Qualitative and Quantitative Study,” 149–156. Parnia and colleagues note that cardiac arrest patients by definition exhibit two of three criteria for clinical death, and usually develop all three in the clinical setting.
↑ 23. Carter, Science and the Near-Death Experience, chap. 13. See also Parnia and Fenwick, “Near-Death Experiences in Cardiac Arrest,” 5–11.
↑ 24. Carter, Science and the Near-Death Experience, chap. 11, summarizing the U.S. Air Force centrifuge research. See also James Whinnery and A. M. Whinnery, “Acceleration-Induced Loss of Consciousness: A Review of 500 Episodes,” Archives of Neurology 47 (1990): 764–776.
↑ 25. Carter, Science and the Near-Death Experience, chap. 11, quoting Whinnery’s descriptions of the “dreamlet” experiences reported by centrifuge subjects.
↑ 26. Blackmore, Dying to Live, chap. 4; Susan Blackmore and T. Troscianko, “The Physiology of the Tunnel,” Journal of Near-Death Studies 8 (1989): 15–28.
↑ 27. Marsh, Out-of-Body and Near-Death Experiences, pp. 76–78. Marsh draws on the centrifuge and syncope data to support his thesis about ECEs arising from brains recovering from hypoxic insult.
↑ 28. Carter, Science and the Near-Death Experience, chap. 11. Carter’s point-by-point comparison of G-LOC episodes with NDEs is one of the most thorough in the literature.
↑ 29. Carter, Science and the Near-Death Experience, chap. 11. Carter identifies six major differences: (1) tunnel vision versus perceiving a tunnel, (2) convulsions in G-LOC but not NDEs, (3) no life review in G-LOC, (4) no deceased relatives or being of light, (5) no decision to return, and (6) no lasting personality changes.
↑ 30. Carter, Science and the Near-Death Experience, chap. 11, quoting Allan Pring’s testimony.
↑ 31. Jeffrey Long, interview with Alex Tsakiris on Skeptiko, cited in Titus Rivas, Anny Dirven, and Rudolf Smit, The Self Does Not Die: Verified Paranormal Phenomena from Near-Death Experiences (Durham, NC: IANDS Publications, 2016), chap. 3.
↑ 32. Rivas, Dirven, and Smit, The Self Does Not Die, chap. 3. The authors document more than thirty published cases of confirmed consciousness during cardiac arrest.
↑ 33. Fischer and Mitchell-Yellin, Near-Death Experiences, chap. 4, pp. 31–32.
↑ 34. Van Lommel, “Setting the Record Straight,” 113–116. Van Lommel argues that “the issue is not whether there is any non-measurable brain activity of any kind whatsoever but whether there is measurable brain activity of the specific form, and in different neural networks, as regarded by contemporary neuroscience to be the necessary condition of conscious experience.”
↑ 35. Rivas, Dirven, and Smit, The Self Does Not Die, chap. 10. The authors conclude that “abundant medical evidence exists supporting the conclusion that an acute cardiac arrest cannot be coupled with sufficient EEG activity to explain the complex contents of NDEs.”
↑ 36. Carter, Science and the Near-Death Experience, chap. 11.
↑ 37. Allan Kellehear, Ian Stevenson, Satwant Pasricha, and Emily Cook, cited in Carter, Science and the Near-Death Experience, chap. 11. Their critique was directed at Blackmore’s Indian tunnel study.
↑ 38. William Serdahely, as cited in Carter, Science and the Near-Death Experience, chap. 13. Serdahely’s question about why a disinhibited brain would selectively produce images of deceased rather than living loved ones remains unanswered by dying brain theorists.
↑ 39. Jeffrey Long, Evidence of the Afterlife: The Science of Near-Death Experiences (New York: HarperOne, 2010), cited in Miller, Near-Death Experiences as Evidence for the Existence of God and Heaven, Appendix 5.
↑ 40. Michael Sabom, cited in Miller, Near-Death Experiences as Evidence, Appendix 5. Sabom found no difference between the dreaming modes of NDErs with and without autoscopic (out-of-body) experiences.
↑ 41. Holden, Greyson, and James, The Handbook of Near-Death Experiences, 225. Bruce Greyson, Edward Kelly, and Emily Williams Kelly authored this review of explanatory models.
↑ 42. Janice Miner Holden, “Veridical Perception in Near-Death Experiences,” in Holden, Greyson, and James, The Handbook of Near-Death Experiences. Holden found 107 cases from 39 publications by 37 author teams; only 8 percent contained any inaccuracy, while 37 percent were verified as completely accurate by independent sources.
↑ 43. Pim van Lommel, Consciousness Beyond Life: The Science of the Near-Death Experience (New York: HarperCollins, 2010), 20. Van Lommel reported that 92 percent of NDE accounts in his research were completely accurate.
↑ 44. Marsh, Out-of-Body and Near-Death Experiences, p. 91. Marsh argues that coherent cognitive functioning “does occur (and could only occur) during that terminal revitalizing process” as the brain reawakens.
↑ 45. Carter, Science and the Near-Death Experience, chap. 13.
↑ 46. Carter, Science and the Near-Death Experience, chap. 13, citing Parnia and Fenwick. In cases of prolonged cardiac arrest, EEG activity may not return to normal for several minutes or even hours.
↑ 47. Van Lommel et al., “Near-Death Experience in Survivors of Cardiac Arrest,” 2039–2045. See also Rivas, Dirven, and Smit, The Self Does Not Die, Case 3.7 (“The Man with the Dentures”), for detailed analysis and the nurse’s confirmation of the patient’s observations.
↑ 48. Holden, “Veridical Perception in Near-Death Experiences,” in The Handbook of Near-Death Experiences. See also Rivas, Dirven, and Smit, The Self Does Not Die, which expanded the catalog to well over 100 cases.
↑ 49. Miller, Near-Death Experiences as Evidence for the Existence of God and Heaven, chap. 2. Miller notes that over 900 articles on NDEs were published in scholarly literature prior to 2005, appearing in journals such as Psychiatry, The Lancet, Critical Care Quarterly, Journal of Near-Death Studies, American Journal of Psychiatry, British Journal of Psychology, Resuscitation, and Neurology.
↑ 50. 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 (2013): 14432–14437.
↑ 51. Pim van Lommel and Bruce Greyson, “Critique of Recent Report of Electrical Activity in the Dying Human Brain,” Journal of Near-Death Studies 41, no. 2 (2023). Van Lommel and Greyson point out that detecting neural activity in dying rats does not demonstrate that the activity produces conscious experience, and note several methodological limitations of the study.
↑ 52. Edward F. Kelly, Emily Williams Kelly, Adam Crabtree, Alan Gauld, Michael Grosso, and Bruce Greyson, Irreducible Mind: Toward a Psychology for the 21st Century (Lanham, MD: Rowman and Littlefield, 2007). This volume provides a comprehensive scholarly defense of the filter/transmission model, drawing on the earlier work of William James, Henri Bergson, F. W. H. Myers, and others.
↑ 53. Marsh, Out-of-Body and Near-Death Experiences, p. 98.
↑ 54. Marsh, Out-of-Body and Near-Death Experiences, pp. 95–96. Marsh describes “early-phase” phenomenology as predicated on premorbid cognitive constructs, and “late-phase” phenomenology as involving the progressive reassertion of full conscious awareness.
↑ 55. Marsh, Out-of-Body and Near-Death Experiences, p. 96.
↑ 56. Marsh, Out-of-Body and Near-Death Experiences, p. 96.
↑ 57. Marsh, Out-of-Body and Near-Death Experiences, p. 96.
↑ 58. See the discussion of cross-cultural NDE evidence in Chapter 8 of this volume. See also Jeffrey Long, Evidence of the Afterlife, chaps. 8–9; and Carter, Science and the Near-Death Experience, chap. 9.
↑ 59. See Chapter 9 of this volume for a fuller treatment of shared death experiences. The phenomenon is discussed in Raymond Moody, Glimpses of Eternity (New York: Guideposts, 2010); and in J. Steve Miller, Deathbed Experiences as Evidence for the Afterlife, vol. 1.