Chapter 17

The Timing Problem—When Do NDEs Actually Occur?

Imagine you are watching a movie, and right at the climax—the moment that makes the whole story come together—someone taps you on the shoulder and says, “That scene didn’t happen when you think it did. It was filmed months later.” Would that change what you saw? Would it undo the dialogue, the action, the meaning? Of course not. The scene is what it is, regardless of when it was shot.

That’s essentially the move skeptics make when they argue that near-death experiences don’t happen during clinical death. Their claim is that NDEs must occur either as the brain is shutting down or as it is booting back up—never during the period when measurable brain activity has ceased. If they can relocate the NDE to a time when the brain was still (or again) functioning, then they can explain it as a product of brain chemistry. No need for consciousness existing apart from the body. No threat to physicalism. The entire evidential force of NDEs, in their view, depends on a misunderstanding about when the experience really took place.

It’s a clever argument. It is also, on its face, a reasonable one. And it deserves to be taken seriously. Michael Marsh makes it explicitly in his critique of the Pam Reynolds case. Fischer and Mitchell-Yellin devote an entire chapter to it. And it shows up in virtually every skeptical treatment of NDEs published in the last forty years.

But here is what I want to show you in this chapter: the timing objection, even if you grant it every benefit of the doubt, still cannot explain the evidence. It fails on its own terms. And the closer you look at the science, the narrower the window for this objection becomes—until it effectively disappears.

A. The Critic’s Argument: NDEs Happen Before or After—Not During—Clinical Death

The timing objection is one of the most frequently repeated arguments in the skeptical literature on NDEs. At its core, the argument is simple: we do not actually know that NDEs occur during the period when the brain is non-functional. They might seem to occur during that period, but perception of time is unreliable, especially in altered states of consciousness. The NDE could have been generated before full brain shutdown or after the brain began to recover—and the patient simply perceived it as occurring during the crisis.

Marsh deploys this argument with particular force in his analysis of the famous Pam Reynolds case. Reynolds, you may recall, underwent a radical neurosurgical procedure called hypothermic cardiac arrest—a “standstill” operation in which her body temperature was lowered to sixty degrees Fahrenheit, her heart was stopped, the blood was drained from her head, and her brain showed no measurable electrical activity. By every clinical standard, she was dead. Yet she reported a vivid, detailed NDE, including accurate perceptions of her surgical procedure.¹

Marsh argues that Pam’s NDE most likely occurred not during the standstill phase, but during the rewarming period—specifically, between approximately 1:00 and 2:00 p.m., when her body temperature was still significantly below normal but her brain was beginning to regain function. He suggests that her recalled experiences, formalized in a few hundred words, would have lasted only a minute or two and could have been generated by a brain that was warming back up. He concludes that her brain would have been “too cold” before the rewarming period to have produced the experiences or laid down the necessary memory traces for later recall. In Marsh’s view, Sabom’s claim that these events occurred earlier—before or around the time they began rewarming her body from sixty degrees—is “completely untenable.”²

Marsh further contends that ECEs (his term for NDEs and OBEs) are “neurophysiologically grounded phenomena arising from brains metabolically recovering from various antecedent clinical crises.” In his model, the abnormal subconscious mentation experienced during these events is generated as the brain recovers its full neurophysiological integrity.³ This is the central thesis of his entire book: NDEs are brain-state phenomena produced by metabolically disturbed brains, especially during the period when they are regaining functional competence.⁴

Fischer and Mitchell-Yellin take a somewhat different but complementary approach in their treatment of the timing problem in chapter 3 of Near-Death Experiences: Understanding Visions of the Afterlife. They don’t focus on the medical details as much as the philosophical logic. Their key argument runs like this: just as a dream can seem to last for hours when it actually occurred in the few minutes before waking, and just as the events depicted in a dream can seem to happen at one time when they actually happened at another, the same could be true of NDEs. The apparent timing of a near-death experience may be different from the actual timing of its causes.⁵

Fischer and Mitchell-Yellin propose three specific reasons to doubt that NDEs occur when they seem to occur. First, the patients may have pieced together the conscious experiences at some time after the events represented in their experiences actually took place—perhaps as their brains were “coming back on line.” Second, even during impaired brain functioning, the brain may still have been recording information that the person could later recall in the form of remembered conscious experiences. And third, they argue that we should not place unwarranted confidence in our ability to measure brain activity—perhaps future science will discover that some brain function persists even when current instruments cannot detect it.⁶

So there you have it. The skeptical position, stated as fairly as I can put it, amounts to three claims: (1) NDEs probably happen before brain shutdown or during brain recovery, not during the crisis itself; (2) the felt timing of an NDE is unreliable; and (3) perhaps our instruments aren’t sensitive enough to detect the residual brain activity that generates these experiences.

B. Identifying the Weaknesses

The timing objection sounds reasonable at first glance. But it has several serious problems, and these problems only multiply the closer we look.

The “Recovery” Hypothesis Predicts the Wrong Kind of Experience

This is the single biggest weakness in the timing argument, and it is devastating. If NDEs are generated by brains recovering from metabolic crisis—whether from cardiac arrest, hypothermic shutdown, or any other severe insult—then we have very clear medical expectations about what those experiences should look like. A brain emerging from severe metabolic disruption does not produce clear, lucid, organized cognition. It produces confusion. Disorientation. Fragmented perception. Garbled memory. That is what the clinical literature shows us, over and over again.⁷

Think about what happens when someone wakes up from general anesthesia. Think about someone regaining consciousness after a fainting spell. Think about someone emerging from a coma. In every one of these scenarios, the pattern is the same: the person is groggy, confused, disoriented, and unable to think clearly. Memories from the recovery period are typically absent or badly garbled. This is not a controversial claim—it is standard neurology. As Parnia and Fenwick have observed, recovery from any sort of insult to the brain occurs via a confusional state, and it seems extremely difficult to imagine how experiences during a confused mental state could result in the kind of clear, organized memories reported in NDEs.⁸

And yet NDEs are the exact opposite of this. They are routinely described as hyper-lucid—“more real than real,” as many experiencers put it. Parnia and Fenwick noted that NDEs during 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.”⁹ An analysis of 520 NDE cases found that eighty percent of experiencers described their thinking during the NDE as “clearer than usual” or “as clear as usual.” Even more striking, people reported enhanced mental functioning significantly more often when they were actually physiologically close to death than when they were not.¹⁰

Stop and think about that. The closer someone was to actual death, the clearer their thinking became. That is precisely the opposite of what any brain-based model predicts. A recovering brain should produce muddled thinking. A brain in the throes of metabolic crisis should produce chaos. Instead, we get crystal clarity.

We can be even more specific. Consider what happens when patients have partial awakening during general surgery. This occurs in roughly 0.1 to 0.3 percent of general surgeries, and the experiences are consistently described as extremely unpleasant, frightening, and painful—and notably, they are not visual. They involve vague, fragmented sensory impressions without organized thought or narrative coherence. That is what a partially functioning brain produces under anesthesia.⁵² And yet Pam Reynolds, under far more profound brain suppression than ordinary anesthesia, reported vision that was “brighter and more focused and clearer than normal vision” and hearing that was “a clearer hearing than with my ears.”⁵³ Those descriptions are the exact opposite of what a partially functioning, metabolically compromised brain produces.

Or consider what we know from the G-force studies conducted by the United States military. When fighter pilots experience rapid acceleration, the resulting loss of blood flow to the brain produces what are called G-LOC (G-force-induced Loss of Consciousness) episodes. These episodes sometimes include brief dreamlike fragments—floating sensations, tunnel-like visuals, even flashes of light. Marsh and other skeptics have occasionally compared these to NDEs. But the comparison actually undermines the timing objection, because G-LOC experiences are uniformly brief, fragmentary, confused, and non-veridical. Nobody emerges from a G-LOC episode with an organized narrative of verified real-world events. Nobody describes their thinking during a G-LOC episode as “clearer than normal.” The contrast between G-LOC fragments and full NDEs is precisely the contrast between what a disrupted brain actually produces and what NDE researchers are documenting.⁵⁴

Penny Sartori, a nurse and NDE researcher who conducted her own prospective hospital study, has addressed the timing question directly. In her extended discussion of the question “When did the NDE/OBE occur?”, she concluded that the research confirms Fenwick’s point: it appears that consciousness can exist independently of a functioning brain. Sartori also dismissed the suggestion that blood flow during CPR could give the brain enough blood to sustain partial consciousness, noting the medical evidence against this possibility.⁵⁵

Marsh’s Rewarming Hypothesis Doesn’t Fit the Reynolds Evidence

Marsh’s specific argument about Pam Reynolds has its own set of problems. He claims the NDE occurred during the rewarming phase, when her brain was gradually coming back to functional capacity. But here is what he has to contend with: during the rewarming phase, Pam’s brain was still severely hypothermic. As Stephen Woodhead has pointed out, when Pam’s heart went into ventricular fibrillation during rewarming, her body temperature was only 27 degrees Celsius—a temperature at which consciousness is not considered possible during recovery from a lower temperature. She was still in cardiac arrest, still full of barbiturates, and still showed no auditory brainstem response.¹¹

Marsh’s timeline also cannot account for the beginning of Pam’s experience. As Sabom and Carter have both observed, the critical question about Pam Reynolds is not when her NDE began but when it ended. The experience clearly started during the early phase of surgery—Pam accurately described the Midas Rex bone saw, the case it came in, and the conversation about her femoral vessels—and she described her NDE as an uninterrupted, continuous experience from beginning to end.¹² If the experience was continuous and it began when Pam’s brain was under anesthesia with 100-decibel clicks blasting in both ears, then Marsh needs to explain how a recovering brain generated the earlier portion of the experience, too. He cannot have it both ways.

Even Pam’s own neurosurgeon, Dr. Robert Spetzler, was clear on this point. He stated that the observations Pam made were not available to her through normal sensory channels: the surgical instruments were covered and not visible until the patient was fully asleep, and given the clicking modules in each ear and the depth of anesthesia, there was no way for her to hear through normal auditory pathways.¹³

Fischer and Mitchell-Yellin’s Dream Analogy is Misleading

Fischer and Mitchell-Yellin’s argument that NDEs might be like dreams—perceived as occurring at one time but actually happening at another—is philosophically interesting but empirically weak. There is a crucial difference between dreams and NDEs that the dream analogy obscures. Dreams do not produce veridical perception of real-world events. You do not dream about your surgery and then have the surgical team confirm that your dream matched what actually happened. You do not dream about the location of a shoe on a windowsill and then have someone find the shoe exactly where you described it. Dreams are internally generated. NDEs with veridical content are tied to external, verifiable events.¹⁴

The dream analogy also breaks down because the content of the NDE matches events that occurred at specific, identifiable times during the crisis. If someone accurately reports that they saw two defibrillator shocks applied to their chest, and the medical record confirms exactly two shocks were administered during cardiac arrest, then the experience is locked to a specific window of time. You cannot simply relocate it to a different moment and wave away the match.

There is a further problem with the dream analogy that Fischer and Mitchell-Yellin do not address. Dreams and hallucinations produce experiences that the dreamer or hallucinator typically recognizes, upon waking, as having been unreal. People regularly say, “I had the strangest dream” or “I thought I was seeing things.” The felt quality of a dream, once you wake up, is recognized as different from waking reality. NDEs are the opposite. Experiencers overwhelmingly insist, not just during the experience but for years and decades afterward, that the NDE felt more real than ordinary waking life—not less. The research of Charlotte Martial and colleagues at the University of Liège showed that NDE memories have more characteristics of real memories than of imagined events. They are more vivid, more detailed, more emotionally intense, and more stable over time than either ordinary memories or memories of dreams and fantasies.⁶⁶

Jeffrey Long’s research through the Near-Death Experience Research Foundation (NDERF) adds another important piece. Long found that among NDErs who had also experienced dreams, hallucinations, and drug-induced altered states, the experiencers consistently and emphatically distinguished their NDEs from all other experiences. They did not describe NDEs as dream-like or hallucinatory. They described them as realer than real.⁶⁷ If NDEs were merely dreams that happened at a different time than they seemed to, we would expect them to share the qualitative features of dreams. They do not. They share the qualitative features of real experiences—and they contain verifiable content that matches real events.

The “Future Science” Argument Is an Appeal to Ignorance

Fischer and Mitchell-Yellin’s third argument—that perhaps future science will discover brain activity we cannot currently detect—is the weakest of the three. It is a textbook appeal to ignorance. One could use this move to dismiss any inconvenient evidence about anything. “We can’t currently explain it, but maybe someday we will, so we don’t need to take it seriously now.” That is not science. That is a promissory note written on a blank check. As Chris Carter has observed, any argument that rests entirely on the hope that future discoveries will vindicate current assumptions is not an argument at all—it is a confession that the current evidence doesn’t support the skeptical position.¹⁵

Moreover, the claim that our instruments might be missing some residual brain activity sounds more plausible than it actually is. Modern EEG monitoring is highly sensitive, and in a neurosurgical setting like the Pam Reynolds case, brain activity is monitored continuously throughout the procedure. As Karl Greene, the neurosurgeon involved in Reynolds’s care, emphasized: EEG activity is continuously monitored throughout neurosurgical procedures, and to ignore ongoing electrophysiological activity during such monitoring would constitute medical malpractice.¹⁶ The suggestion that some hidden brain activity was silently generating Pam’s experience without being detected by any of the monitoring equipment is not a serious medical hypothesis. It is special pleading.

C. The Pro-NDE Response: Why the Timing Objection Fails

The EEG Evidence: Brain Activity Ceases Within Seconds

To understand why the timing objection fails, you need to know what actually happens to the brain during cardiac arrest. Cardiac arrest (when the heart stops pumping blood) is the most common medical context for NDEs in the research literature. And the physiology is unambiguous: when the heart stops, blood flow to the brain ceases almost immediately. Within ten to twenty seconds, the EEG (the electroencephalogram, which measures electrical activity in the brain’s outer layer, the cerebral cortex) goes flat.¹⁷

The cerebral cortex is the part of the brain that contemporary neuroscience associates with higher cognitive functions—thinking, perceiving, remembering, reasoning. When the EEG flatlines, it means the cortex is no longer doing the things that, according to physicalism, are necessary for any kind of conscious experience.

But perhaps, the skeptic suggests, deeper brain structures could sustain consciousness even after the cortex goes silent. Pim van Lommel, the Dutch cardiologist who conducted the landmark Lancet study on NDEs, has addressed this directly. He points out that the issue is not whether there is any immeasurable brain activity of any kind whatsoever, but whether there is brain activity of the specific form regarded by contemporary neuroscience as essential for the experience of consciousness. And there is no sign of that specific kind of activity in the EEGs of cardiac arrest patients.¹⁸

Furthermore, animal studies using electrodes implanted deep inside the brain have shown that when cortical activity ceases as measured by EEG, activity in deeper brain structures disappears shortly afterward as well. The deep structures do not keep humming along on their own.¹⁹ And this makes sense: during cardiac arrest, blood flow to the entire brain stops, not just to the cortex. Without oxygen and glucose, no part of the brain can remain active for long.

So the window of opportunity for “residual brain activity” is extremely small. From the onset of cardiac arrest, you have perhaps ten to twenty seconds before measurable cortical activity disappears, followed shortly by the cessation of deeper brain activity. After that, the brain is functionally offline. And in cases of prolonged cardiac arrest—which can last five to ten minutes for out-of-hospital arrests—EEG activity may not return to normal for several minutes or even hours after successful resuscitation.²⁰

Let me put this in everyday terms. Imagine you are driving a car and the engine dies. For a few seconds, you might coast on momentum. But very quickly, the car stops. Now imagine someone tells you that during those few seconds of coasting, the car not only kept moving but actually accelerated to highway speed, navigated a complex series of turns, and arrived at a precise destination—all without an engine. That is essentially what the skeptic is asking you to believe when they suggest that residual brain activity can account for the complex, organized, veridical NDEs reported during cardiac arrest.

The Subcortical Hypothesis: A Desperate Retreat

When confronted with the flat-EEG evidence, some materialists have retreated to a secondary position: perhaps, they suggest, even though the cortex is offline, deeper subcortical brain structures might take over and generate consciousness. This hypothesis sounds more plausible than it is.

The problem is that, according to mainstream neuroscience, mental functions are strictly coupled with certain areas of the brain. The cortex is where the heavy cognitive lifting happens—perception, memory, reasoning, narrative construction. Subcortical structures handle more basic functions: regulating heart rate, controlling breathing, processing raw sensory signals. The idea that subcortical structures could suddenly assume the cortex’s job during cardiac arrest—producing the kind of complex, organized, narrative experience reported in NDEs—has no support in the neurological literature.⁵⁶

Even in cases where patients undergo long-term cognitive rehabilitation after brain damage, learning to access different brain areas to compensate for damaged ones requires extensive, intensive neuropsychological training over months or years. It does not happen spontaneously. It certainly does not happen in the ten to twenty seconds between cardiac arrest and cortical shutdown. The assumption that subcortical structures could spring into action and assume complex cognitive functions during cardiac arrest is, as the authors of The Self Does Not Die put it, expecting them to do so “as if by magic.”⁵⁷

Furthermore, during cardiac arrest, blood flow ceases to the entire brain, not just the cortex. Subcortical structures need oxygen and glucose just as much as cortical ones. Without blood flow, they go offline too. The subcortical hypothesis does not solve the problem; it merely relocates it to a different part of the brain that faces the same physiological constraints.

Van Lommel’s Landmark Study: What the Numbers Tell Us

The 2001 Lancet study led by Pim van Lommel remains one of the most important studies in NDE research, and it speaks directly to the timing question. Van Lommel and his team prospectively studied 344 consecutive cardiac arrest patients who were successfully resuscitated in ten Dutch hospitals. Of these, 62 patients (18%) reported an NDE. The study’s prospective design meant that patients were interviewed shortly after resuscitation, minimizing the opportunity for post-hoc reconstruction or information contamination.⁵⁸

Here is what makes this study so relevant to the timing question. All 344 patients had been clinically dead—no heartbeat, no breathing, and, given the well-established physiology of cardiac arrest, no measurable cortical brain activity within seconds of the arrest. All 344 patients had essentially the same brain state during their cardiac arrest. And yet only 18% reported NDEs. If NDEs were produced by the dying or recovering brain, we would expect a much higher percentage. The fact that most cardiac arrest survivors do not have NDEs, despite having the same brain conditions, suggests that the NDE is not simply a function of brain physiology.⁵⁹

Van Lommel’s study also included follow-up interviews at two and eight years post-cardiac arrest. The NDE memories remained remarkably stable over time—they did not degrade or change significantly, as one would expect if they were products of a metabolically confused brain. Ordinary memories from the same period showed the expected patterns of degradation. This further undermines the idea that NDEs are confabulations generated during brain recovery.⁶⁰

The Transition to and from Unconsciousness Rules Out the “Edges”

The skeptic’s best hope is to argue that NDEs happen on the “edges”—either as the brain is losing consciousness or as it is regaining it. But both of these options have serious problems.

First, the transition to unconsciousness during cardiac arrest is extraordinarily rapid. It can occur within a single second and often appears immediate to the person experiencing it. There is almost no time for a complex, organized, narrative experience to unfold.²¹

Second, any sort of insult to the brain typically results in a loss of memory around the event. Neurologists call this anterograde and retrograde amnesia—the inability to form new memories or recall events from the period just before and after the loss of consciousness. The length of this amnesia is actually used clinically as a measure of how severe the brain injury was. If a cardiac arrest survivor remembers events from just before or just after losing consciousness, that is itself medically unusual—and when the remembered events turn out to be accurate descriptions of things happening in the operating room, that takes us into extraordinary territory.²²

Third, as we have already seen, the recovery from any brain insult occurs via a confusional state. Even waking up from a simple faint involves a period of grogginess and disorientation. Waking up from cardiac arrest is far more severe. The idea that a brain emerging from the kind of profound metabolic shutdown involved in cardiac arrest could produce the hyper-lucid, organized, detailed experiences reported in NDEs—and do so in a way that matches real-world events—stretches credulity well past its breaking point.

Parnia’s AWARE Studies: Designed to Test Timing

One of the most significant developments in NDE research in recent decades is Sam Parnia’s AWARE (AWAreness during REsuscitation) study, a multi-hospital prospective investigation specifically designed to address the timing question. The study placed visual targets—pictures on shelves near the ceilings of resuscitation rooms—that could only be seen from an elevated vantage point. If a patient reported an out-of-body experience during cardiac arrest and could identify the hidden target, this would provide powerful evidence that the experience occurred during the arrest itself.²³

The first phase of the AWARE study, published in the journal Resuscitation in 2014, produced one remarkable case. A 57-year-old social worker at Southampton General Hospital suffered a cardiac arrest while being prepared for a catheter insertion. After resuscitation, he reported an NDE in which he saw people standing around him and witnessed them trying to defibrillate him twice. He described hearing an automated voice saying “Shock the patient.” He also described which people were present, where they were standing, and what they looked like. Review of his medical file confirmed that an automated external defibrillator (AED) had been used, that it does indeed produce voice commands, and that it was used exactly twice.²⁴

Italian anesthesiologist Enrico Facco, Professor of Anesthesia and Resuscitation at the University of Padua, analyzed this case and concluded that the AED’s acoustic algorithm demonstrated that consciousness persisted for at least three minutes during cardiac arrest, given that the acoustic command occurs once every three minutes.²⁵ Three minutes of consciousness during cardiac arrest is not a transitional experience at the “edges.” It is consciousness operating in the dead center of a period when, according to physicalism, no consciousness should be possible.

Now, the AWARE study also faced limitations. No patient happened to have their cardiac arrest in a room where a hidden visual target was properly positioned, so the experimental test of OBE veridicality through targets has not yet succeeded. But the non-experimental veridical evidence—patients accurately reporting details of their resuscitation that were later confirmed by medical records—remains powerful. As Bruce Greyson has noted, the kind of spontaneous veridical cases documented in studies like The Self Does Not Die are actually more evidentially important than what prospective target-based studies have so far produced.²⁶

The Borjigin Rat Study: Less Than Meets the Eye

In 2013, neurologist Jimo Borjigin of the University of Michigan published a study that made headlines around the world. Her team induced cardiac arrest in nine rats and observed a surge of brain activity lasting about thirty seconds after the hearts stopped. Media reports breathlessly suggested that this could explain NDEs—perhaps the dying brain produces one last burst of activity that generates the NDE experience.²⁷

The study was scientifically interesting. But it does not rescue the timing objection, for several reasons.

First, and most obviously, the study was on rats, not humans. We do not know whether rats experience anything at all during this surge, let alone anything resembling a human NDE. Extrapolating from rat EEG data to human conscious experience is a very large leap.²⁸

Second, the surge lasted only about thirty seconds and then transitioned to a permanently flat EEG. Thirty seconds is nowhere near enough time to account for the complex, extended, narrative NDEs reported by many patients—especially those involving travel through a tunnel, encounters with deceased relatives, life reviews, and visits to otherworldly environments. And the cases documented in The Self Does Not Die involve NDEs during cardiac arrest that clearly lasted far longer than thirty seconds, with veridical observations that span minutes, not fractions of a minute.²⁹

Third, van Lommel has pointed out a crucial distinction that Borjigin’s study overlooks. The rats in the study underwent induced cardiac arrest under controlled laboratory conditions. This is very different from the acute, spontaneous cardiac arrests that characterize the clinical settings in which most NDEs are reported. All the research in humans shows that from the onset of acute cardiac arrest, measurable brain activity lasts no more than an average of fifteen seconds, involves no spike in activity, and is followed by a consistently flat EEG. No spontaneous resurgence of brain activity has ever been observed in such patients.³⁰

Fourth, even granting that some brief surge could occur in humans (which remains undemonstrated), it could only begin to explain NDEs that started immediately after cardiac arrest, lasted no more than thirty seconds, and involved only perceptions available to the normal senses. The cases in The Self Does Not Die, in van Lommel’s research, and in Parnia’s studies clearly defy all three of these conditions.³¹

The Chawla Study and Other “Brain Surge” Claims

A related study by Lakhmir Chawla and colleagues (2009) examined the dying process in seven patients who were taken off life support. After the patients had no blood pressure and either no heartbeat or severely compromised heart function, approximately eighty percent showed a one-time resurgence of brain activity lasting at most three minutes.³²

Some materialists seized on this as evidence that NDEs could be explained by a last surge of cortical activity. But there are critical problems. Chawla himself acknowledged that he did not know whether the activity came from the cortex or the brainstem, and he did not even rule out the possibility of an external signal causing artifact readings. He recognized that the conclusions that could be drawn from his investigation were very limited. Loretta Norton and colleagues later suggested that the increased EEG activity before cardiac cessation could be an artifact caused by muscle contractions—meaning it might not reflect brain activity at all.³³

More importantly, these patients were in deep comas being gradually taken off life support. This is a fundamentally different situation from the acute, sudden cardiac arrests in which most NDE research takes place. As van Lommel has stressed, the gradual dying process involved in withdrawal of life support is not comparable to the sudden cardiac arrest that characterizes the standard NDE research context.³⁴

The Veridical Evidence Defeats the Timing Objection

Here is where the timing objection ultimately collapses. Even if you grant every skeptical assumption about timing—even if you suppose that the NDE occurred during some brief window of residual or recovering brain activity—you still cannot explain the veridical elements.

Consider the dentures man from van Lommel’s Lancet study. A patient in cardiac arrest was brought to the hospital, and a nurse removed his dentures during resuscitation and placed them in a drawer. The patient later identified the specific nurse, described where the dentures were placed, and accurately reported details of his resuscitation—all from a vantage point above his body. He had no pulse, no breathing, and dilated pupils throughout the resuscitation process.³⁵

Fischer and Mitchell-Yellin suggest that perhaps the patient pieced together this information later, as his brain was coming back online. But pieced it together from what? His eyes were closed. He was unconscious and being resuscitated. The specific nurse who removed the dentures later confirmed the man’s account. How does a recovering brain, in a state of post-cardiac-arrest confusion, accurately reconstruct the specific details of what happened during the arrest itself, from an elevated vantage point, without any normal sensory input?³⁶

Or consider the cases documented in The Self Does Not Die where patients described events happening in other rooms or other parts of the hospital—events confirmed by witnesses who were not present in the resuscitation room. No amount of residual brain activity in the patient’s skull can explain how they perceived events occurring in a completely different location.³⁷

Or think about Jeffrey Long’s observation from his research: the substantial majority of people who have NDEs associated with cardiac arrest report seeing their physical body well prior to the time that CPR was initiated. Once CPR begins, there is no alteration in the flow of the NDE, suggesting that blood flow to the brain from CPR is not affecting the content of the experience.³⁸ If the experience begins before CPR and continues uninterrupted through it, the timing of the experience is locked to the period of cardiac arrest—not the edges.

This is the fundamental problem the timing objection cannot solve. Relocating the NDE to a different time window does not make the veridical evidence go away. A brain that is shutting down cannot generate accurate perceptions of specific surgical procedures. A brain that is recovering cannot reconstruct events it had no access to through normal sensory channels. And no brain—whether shutting down, recovering, or fully functional—can perceive events happening in another room from an elevated viewpoint outside its own skull. The timing objection addresses when the NDE occurs. It does not address what the NDE contains. And the content is what makes the evidence so powerful.

NDEs During Monitored Standstill: The Pam Reynolds Evidence Reconsidered

The Pam Reynolds case deserves further attention here because it remains one of the most thoroughly documented NDEs in the medical literature, and Marsh’s timing argument against it is central to his book. We addressed Marsh’s specific critiques of the Reynolds case in detail in Chapter 5, so I won’t repeat all of that here. But the timing dimension warrants further discussion.

Dr. Karl Greene, a neurosurgeon with direct knowledge of the Reynolds procedure, wrote in a 2015 letter that from a practical standpoint, Reynolds’s entire conscious experience could be considered anomalous. He explained that such conscious experience does not typically occur under the influence of doses of barbiturates that markedly suppress brain activity, profound hypothermia that eliminates all spontaneous EEG activity, and circulatory arrest that produces complete loss of all electrophysiological activity.³⁹

Greene was also emphatic that the anesthesiologist would have detected, reported, and responded to any brain activity if it had occurred at any point during Reynolds’s procedure. EEG monitoring was continuous throughout. The auditory clicks used for brainstem monitoring were also continuous. To suggest that some hidden brain activity slipped through undetected is, in Greene’s view, not a credible medical hypothesis.⁴⁰

The authors of Irreducible Mind made the same point more broadly: even if we assume for the sake of argument that Reynolds’s entire experience occurred during the earlier stages of the procedure, brain activity at that time was still inadequate to support organized mentation, according to current neurophysiological doctrine.⁴¹ In other words, Marsh’s rewarming hypothesis doesn’t escape the problem. No matter where you place the NDE on the surgical timeline, the brain was not in a condition that should have supported conscious experience.

Reynolds herself reported that her experience was continuous and unbroken from beginning to end. She did not describe separate episodes at different times. She described one long, flowing experience that included both the surgical observations and the transcendent encounter—an experience she perceived as completely real throughout. As Sabom observed, this description of continuity is consistent with virtually all other reports of combined autoscopic and transcendental NDEs that he had studied over thirty years of research.⁴²

Cases That Lock the Timing to the Crisis

Some of the strongest cases in the NDE literature effectively nail down the timing beyond reasonable dispute. Consider cases from The Self Does Not Die in which young children who had cardiac arrests spontaneously began talking about an out-of-body experience during the clinical death event—without anyone having told them that they had almost died. In these cases, the children described their experiences immediately upon regaining consciousness, and the details matched what had happened during the arrest.⁴³

One particularly compelling example is the case of nine-month-old Mark Botts, documented by Melvin Morse. Mark suffered a cardiac arrest and was resuscitated. As he grew older, he began spontaneously describing what happened during his medical emergency—including specific details about where people were standing and what they were doing. His mother later confirmed the accuracy of these descriptions, asking skeptics how a child could describe things he had no way of knowing. Kenneth Ring and Evelyn Elsaesser Valarino, who independently investigated the case, confirmed that the impressions Mark received during his OBE of specific events in the physical world turned out to correspond with the facts.⁶¹ A nine-month-old infant does not reconstruct surgical memories from post-operative conversation. He was far too young to have absorbed information from discussions with medical staff. The simplest explanation is that he perceived what happened during his cardiac arrest, from outside his body, at the time it was happening.

Consider also cases where the specific content of the NDE is locked to a specific moment during the crisis. The AWARE study patient heard two AED voice commands that, given the device’s algorithm, demonstrated consciousness lasting at least three minutes during cardiac arrest.⁴⁴ Pam Reynolds saw her body jump from the defibrillator shocks—shocks that occurred during ventricular fibrillation at a body temperature of 27 degrees Celsius, a state in which consciousness is medically not expected. She knew her heart had to be stimulated twice to restart. As neurosurgeon Karl Greene stated in a National Geographic interview, she should not have known that. She was physiologically dead—no brain wave activity, no heartbeat, nothing.⁴⁵ Al Sullivan accurately described his surgeon’s unusual habit of pointing with his elbows during surgery—something that happened while Sullivan was under deep anesthesia with his chest open.

The Zemmar Case and the AWARE II Study: Recent Developments

In early 2022, a case report by Dr. Ajmal Zemmar and colleagues generated considerable media attention. The report described an 87-year-old patient who unexpectedly suffered a cardiac arrest while undergoing EEG monitoring. The researchers observed unusual gamma-frequency activity and speculated that it could support a last “recall of life” in the near-death state. Media headlines breathlessly announced that science had finally explained NDEs as a last burst of brain activity.⁶²

The reality is far more modest than the headlines suggested. Both Bruce Greyson and Pim van Lommel analyzed the case and concluded that it offers no materialistic explanation of NDEs. The measured cortical brain activity in this specific case did not match the level of normal cortical activity needed for the kind of complex consciousness reported in NDEs. Crucially, the patient in question died before having an opportunity to report whether he had experienced anything at all—we have no way of knowing whether this brain activity correlated with any subjective experience, let alone an NDE.⁶³

Greyson further stressed that the press release was heavily misleading because no surge in brain activity had been measured after the cardiac arrest started. The little activity that was measured happened before the patient went into cardiac arrest, not afterwards. So the case provides no reason to believe that brain activity persists or surges after the onset of cardiac arrest in a way that could generate NDEs.⁶⁴

The AWARE II study, published in its final form in Resuscitation in July 2023, added further relevant data. Among the patients who reported NDEs or other recalled experiences of death, none had usable EEG data. As NDE researcher Orson Wedgwood observed, none of the subjects who reported conscious recollections, including the six who had NDEs, had any EEG data at all, let alone data that showed markers of consciousness. It is therefore entirely false to claim an association between brain activity and NDEs based on these results.⁶⁵

What do these recent developments tell us about the timing problem? They confirm what we have been arguing throughout this chapter: there is no evidence that brain activity during cardiac arrest is sufficient to generate the complex, coherent, veridical experiences reported in NDEs. The Zemmar case shows some residual activity before cardiac arrest, not after. The AWARE II study shows that the patients who actually reported NDEs had no measurable brain activity at all. The skeptic’s hope that advancing science would vindicate the timing objection is not being fulfilled. If anything, the evidence is moving in the opposite direction.

In each of these cases, the content of the NDE is tied to events that occurred at a specific, identifiable time—a time during which the brain was demonstrably not functioning at a level capable of supporting the kind of complex, organized, veridical perception that the patient reported. The timing objection asks us to relocate these experiences to a different moment. But the content of the experiences locks them to exactly the moment when they should not have been possible.

D. Counter-Objections and Responses

“Maybe memories form before consciousness is fully lost and are later reconstructed”

A skeptic might press the point: perhaps the brain, in the moments before full shutdown, absorbs sensory information that is then reconstructed into a coherent narrative during recovery. This would explain the veridical elements without requiring consciousness during the arrest itself.

There are several problems with this suggestion. First, it requires an extraordinary coincidence: the brain would have to absorb extremely specific visual and auditory information during the few seconds before shutdown, store that information despite the complete metabolic disruption that follows, and then reconstruct it into a perfectly ordered, narratively coherent experience during recovery—a recovery process that normally produces confusion and memory loss, not crystal clarity. This is not a parsimonious explanation. It is an epicycle designed to save the physicalist hypothesis at all costs.⁴⁶

Second, many veridical NDE observations include events that occurred during the crisis, not before it. The dentures were removed during resuscitation. The AED commands occurred during cardiac arrest. Pam Reynolds’s body jumped from defibrillator shocks during ventricular fibrillation. These are not pre-shutdown observations. They are observations of events that happened when no brain was available to observe them through normal means.

Third, the elevated viewpoint reported in out-of-body experiences cannot be explained by pre-shutdown sensory intake. Before losing consciousness, the patient was lying on a bed or operating table. They did not have an aerial view of the room. Yet they consistently report seeing events from above—from the ceiling, from a corner of the room, from a vantage point that corresponds to no physical location their body occupied at any time.⁴⁷

“NDErs are simply unreliable reporters of when their experience occurred”

Fischer and Mitchell-Yellin lean heavily on the idea that people are bad at knowing when their experiences happened. Fair enough in general—we sometimes misjudge the timing of dreams or hallucinations. But NDE reports are not bare subjective impressions about timing. They contain content that is tied to specific events at specific moments. When a patient says, “I saw them shock me twice,” and the medical record confirms exactly two shocks during cardiac arrest, the patient’s report is not an unreliable impression about timing. It is an accurate observation of a real event that occurred at a real time. The timing is established by the content, not by the patient’s subjective sense of when the experience happened.⁴⁸

“Perhaps NDEs are reconstructed from information acquired post-operatively”

Marsh suggests that some of Pam Reynolds’s knowledge could have been acquired after surgery—from conversations with medical staff, from reading about the procedure, or from other post-operative sources. This is theoretically possible in some cases. But it faces a practical problem: in case after case, NDErs report specific details that they would have had no way to learn post-operatively. The dentures man identified the specific nurse and the specific drawer—information that would not be part of any post-operative briefing. Al Sullivan described his surgeon’s unusual elbow-pointing technique—a detail no one would think to mention to a patient after surgery. Pam Reynolds identified a female cardiologist involved in the discussion about her femoral vessels—information that would have been unusual to acquire after the fact.⁴⁹

There is another dimension to this objection that deserves attention. The post-operative information hypothesis requires a specific mechanism: the patient must acquire the relevant information through normal channels (conversation, reading, inference) and then unconsciously weave it into a false memory that feels like a first-person experience of directly witnessing the event. This is not a trivial psychological process. It requires (1) exposure to the specific information, (2) unconscious construction of a false narrative, (3) the false narrative having the phenomenal character of direct personal experience rather than secondhand knowledge, and (4) the resulting memory being so vivid and stable that it resists the normal degradation of false memories over time. Each of these steps is individually plausible for a small percentage of cases. But the conjunction of all four—happening reliably across hundreds of independently documented NDE cases in multiple countries over several decades—strains the hypothesis past its reasonable limits.

Moreover, in prospective studies like van Lommel’s and Parnia’s, patients are interviewed shortly after resuscitation, before they have had time to acquire information from other sources. And in the cases documented in The Self Does Not Die, many patients reported their experiences immediately upon regaining consciousness, before any debriefing or conversation had taken place. Some patients startled their medical teams by describing exactly what had happened during their resuscitation the moment they opened their eyes.⁵⁰

“Science is young. Maybe we just need more time to find the answer.”

This is Fischer and Mitchell-Yellin’s final fallback: the physical sciences are in their infancy, and our current state of knowledge is not representative of physicalism’s prospects. They ask us to be patient. Maybe one day a physical explanation will emerge.⁵¹

I respect the intellectual honesty of this move. But I also want to point out what it concedes. It concedes that, right now, the physical explanation is not adequate. It concedes that current neuroscience cannot account for the evidence. It concedes that the skeptical case rests not on present evidence but on future hope. That is not nothing. But it is not a refutation of the NDE evidence, either. It is an acknowledgment that the evidence currently points in a direction that physicalism cannot account for, paired with a hope that someday it will be able to.

Hope is fine. I am a Christian; I know a thing or two about hope. But hope is not evidence. And if we are following the evidence where it leads—which is what this entire book is about—then the evidence currently points away from the timing objection and toward the conclusion that NDEs involve genuine consciousness operating apart from the measurable activity of the brain.

Conclusion: The Clock is Not on the Skeptic’s Side

The timing objection is one of the more sophisticated skeptical arguments against NDEs. It’s not a bad question to ask: when, exactly, does the NDE occur? And the research community takes this question seriously. Parnia’s AWARE studies are designed precisely to answer it.

But as we have seen, the timing objection fails on multiple levels. The “recovery” hypothesis predicts the wrong kind of experience—confusion instead of clarity, fragmentation instead of coherence. The EEG evidence shows that the window for residual brain activity is vanishingly narrow. The Borjigin rat study and the Chawla study do not apply to the clinical contexts in which NDEs occur. The dream analogy breaks down because NDEs contain veridical content that dreams do not. The subcortical hypothesis fails because subcortical structures face the same blood-flow constraints as the cortex and have never been shown to produce complex cognition. The Zemmar case and the AWARE II study have not delivered the evidence materialists hoped for. And no matter where you try to place the NDE on the timeline, the veridical evidence remains unexplained. A brain that is shutting down, recovering, or offline cannot perceive surgical procedures, identify specific nurses, describe events in other rooms, or observe its own body from the ceiling.

I want to be clear about what I am not claiming. I am not claiming that every NDE occurs during flat-EEG cardiac arrest. Some NDEs occur during surgery under anesthesia. Some occur during severe illness. Some occur in circumstances where the exact brain state is not well documented. I am not arguing that we have perfect timing data for every case in the literature. What I am arguing is that (1) the evidence we do have consistently points to NDEs occurring during periods of severely compromised brain function; (2) the specific experiences reported during these periods are incompatible with what compromised brains are known to produce; and (3) the veridical elements of these experiences cannot be explained by any timing-based relocation of the NDE to a period of greater brain function.

The timing problem is real. But it is a problem for the skeptic, not for the NDE researcher. The evidence increasingly suggests that consciousness continues to function during periods when the brain cannot account for it. That is exactly what the veridical NDE evidence has been telling us all along. And it is exactly what we would expect if consciousness is not entirely a product of the brain—if there is, in fact, an immaterial dimension to human experience that survives the shutdown of the body’s most complex organ.

There is a theological resonance here that I do not want to overstate, but I also do not want to overlook. If the NDE evidence is pointing us toward the reality of consciousness functioning apart from the brain, then it is pointing in the same direction that Scripture has been pointing all along. The apostle Paul spoke of being “away from the body and at home with the Lord” (2 Cor. 5:8). The rich man and Lazarus were conscious after death (Luke 16:19–31). The souls under the altar in Revelation cried out with awareness and emotion (Rev. 6:9–11). The biblical picture of the conscious intermediate state—the soul existing between death and resurrection, aware, thinking, perceiving—is precisely the picture that the NDE evidence corroborates. The timing problem, far from undermining the NDE evidence, actually highlights the very phenomenon that makes NDEs so significant: consciousness apparently functioning when, by every materialist criterion, it should not be functioning at all.⁶⁸

Marsh, Fischer, and the other critics are right to press the timing question. Every good researcher should press it. But the data they are hoping for—data showing that NDEs are simply products of residual or recovering brain activity—have not materialized. Instead, every new study confirms that the brains of cardiac arrest patients are profoundly non-functional during the period when NDEs appear to occur. Every new case with veridical elements ties the experience more tightly to the crisis itself. And every new analysis of the quality of NDE memories shows that they are not the garbled products of a brain rebooting—they are among the most vivid, most stable, and most reality-like memories that human beings ever form.

The clock, it turns out, is not on the skeptic’s side. It is on ours.

In the next chapter, we will turn from the question of when NDEs occur to the question of whether we can trust what NDErs remember—the memory, confabulation, and narrative reliability objection. As we will see, the skeptical case gets no stronger on that front, either.