Near-Death Experiences — notes by Justin C. Fisher, SMU Philosophy

The expression “Near-Death Experiences” (or NDE’s, for short) encompasses a variety of experiences that people report having had in extreme life-threatening circumstances. Widely publicized NDE reports often fall into one or more of the following themes:

When subjects report having these experiences, there are a number of hypotheses that could potentially explain these reports.

H1. Veridical. Sincerely reported, accurately recalled veridical experiences — the subject perceived a circumstance that really was the way they experienced it to be. (“Veridical” basically means “true”, same root as “verify”.)

H2. Hallucination. Sincerely reported, accurately recalled, non-veridical experiences — the subject had an experience as though of being in one sort of circumstance (e.g. floating above their body) but they weren’t actually in that circumstance.

H3. False Memory. Sincerely reported false memories that misrepresent their experience.

H4. Fraud. Insincere reports of experiences the subject knows they didn’t actually have.

Of course, some cases might involve a mix of these. Someone might have a veridical memory of seeing a first-responder and accurate memories of an ensuing hallucination, but these could be entwined with false memories implanted after the fact, and then further embellished to impress listeners.

Notice that H2 — H4 are all fully compatible with a naturalistic scientific world-view. We already know that hallucination, false memories and fraud often occur and that these are especially likely to occur in cases where a brain is operating under extreme duress, and/or where other people might be especially interested if you can come up with a good story to tell. It might be a bit interesting to try to determine what mix of H2 — H4 is applicable in particular cases, but overall this debate isn’t all that exciting, and probably shouldn’t be worthy of all that much attention except to hard-core neuroscience buffs.

The reason NDE’s have enjoyed so much popular attention is because many people think H1 provides a better explanation than the others for some reported NDE’s. If people really could veridically experience things unavailable to their physical sense organs during NDE’s, this would be a real challenge to our scientific understanding of how our senses and our minds work; and if people really could get a veridical preview of the afterlife during NDE’s, this would be of great import to religion and to anyone who cares about what happens after we die. It is because many people take these experiences to be veridical that NDE’s get written up in best-selling books with titles like The Proof of Heaven or The Boy Who Came Back From Heaven.

Prior Probabilities and Ockham’s Razor.

H2—H4 all involve well known and well understood processes (hallucination, false memories, fraud), and there is very strong evidence that our understanding of the world will need to include all three of these. In using these to explain reported NDE’s, we are not adding any sorts of complexities beyond those that we already know we need to embrace.

In contrast, H1 requires complexities far beyond anything we need to explain current scientific evidence. Scientific studies have given us a quite good understanding of how veridical visual perception works: light reflects off the objects, enters our retinas, gets detected by light-sensitive rods and cones, and triggers cascades of ensuing neural activation as our brains process this retinal image to calculate what objects we are seeing. Vision scientists already have a quite good understanding of how this process works. If NDE’s allow a sort of veridical vision via some process that doesn’t involve light entering the eyes, then this will require a much more complex story of how vision works than any story we’ve ever needed to account for all our numerous scientific studies on vision.

Ockham’s Razor tells us to avoid embracing such needlessly complex hypotheses, unless we’re forced to do so by very strong evidence. This tells us that we should demand very strong evidence before accepting H1. Unless H1 does a lot better at predicting surprising observations than H2 — H4 do, Ockham’s Razor says we should assign much higher probability to H2 — H4.

Spiritual or Religious Content

Many widely publicized NDE’s involve content that fits neatly with popular spiritual or religious views: many subjects claim to have experienced their souls rising up out of their bodies, going to a place where they were surrounded by light, being overcome with a peaceful feeling, and having conversations with loved ones or religious personas.

According to a hypothesis like H1 that holds that these subjects are veridically experiencing a taste of the afterlife, these are just the sorts of NDE’s subjects should have. In contrast, hallucination hypotheses like H2 don’t seem to predict that NDE subjects would have such a narrow range of experiences — shouldn’t some hallucinate that they’re walking down the street naked or dancing with a pink elephant? You’ll recall that the Prediction Principle tells us to shift our confidence towards hypotheses that make better predictions, and away from those that make worse predictions. If H1 does a much better job at predicting the spiritual or religious content of NDE’s, then the Prediction Principle would tell us to shift our confidence towards H1.

One response is to consider what happens in the brain as it becomes starved for oxygen and/or suffers other sorts of trauma. It turns out that many of these religious-seeming experiences can be triggered in brains fairly easily, and in circumstances where nothing truly life-threatening has happened so it isn’t plausible that the subject actually should have started a trip to heaven. For example, we know that as visual processing becomes starved for resources, peripheral vision appears to go dark first, giving you a “light at the end of the tunnel” experience. A feeling of peace arises as brain areas that are involved in actively struggling become starved of resources and abate their struggles. Dreams and hallucinations very often involve conversations, so H2 predicts that conversations should also sometimes occur in near-death hallucinations. These considerations suggest that H2 might not be all that bad at predicting the spiritual and religious-seeming content of NDE’s after all.

A second response is to question whether NDE’s consistently do have these religious or spiritual contents. To study this, it’s best not to rely on widely publicized anecdotes, but instead to have researchers basically lurk around hospitals and interview everyone who underwent emergency resuscitation to ask them what sorts of NDE’s, if any, they had. Such an experiment has been done with thousands of subjects, and it turns out that the range of NDE’s is much more diverse than it is popularly conceived to be, including people experiencing talking to living people (and hence shouldn’t have been in heaven). This diverse range of experiences, including many that clearly weren’t veridical, weighs strongly in favor of the hallucination hypothesis H2 over the veridical hypothesis H1.

In fact, it seems like the real question here isn’t “Why do NDE’s have religious or spiritual content?” The answer to that is that they often don’t. Instead the real question is, “Why do widely publicized NDE’s have spiritual or religious content?” Of course, the answer to that will involve sociological facts about why people prefer to tell one another some NDE stories rather than others. When somebody nearly dies and has an absurd hallucination, that’s not really news, so nobody passes that story on or writes best-selling books about it. But when somebody nearly dies and has an NDE that fits popular religious views, that sort of story can find an eager market, especially among religious readers, and it often gets spread far and wide. Because of this asymmetry, we probably shouldn’t be surprised that widely discussed NDE’s aren’t really all that representative of NDE’s in general.

Unknowable details and cherry-picking.

Believers in NDE’s sometimes claim that NDE’s allow subjects to gain knowledge of details that their ordinary senses shouldn’t have given them access to. E.g., in one famous case a subject reported being able to observe shoes hanging outside in a position that shouldn’t have been visible to her unless her vantage point had somehow risen up out of her body.

The veridical hypothesis (H1) predicts that NDE subjects would report such details, whereas the hallucination hypothesis (H2) predicts that people are much more likely to have false hallucinations than ones that coincidentally happen to match reality. If we observe a lot of cases like this, the Prediction Principle would tell us to start taking H1 more seriously, and perhaps even eventually to abandon H2.

One key worry here is that we don’t know how common cases like this are. For every case you hear about where an NDE subject guesses a surprising detail right, there are countless other cases where NDE subjects hallucinated false details and everybody wrote it off as an understandable part of their ordeal. To know whether NDE subjects get things right so frequently that they must have had some extra-sensory help, it won’t be enough to know that there are a few cherry-picked cases where someone got something right — even the Hallucination Hypothesis (H2) predicts there will be some cases like this. Instead, we would need a systematic study that collects both mistaken and correct guesses about details, and then we can see whether these guesses turn out to be correct a surprising number of times. We’ll talk about such studies in a moment.

There are also lots of worries about false memories (H3) and even fraud (H4) here. In many cases, an NDE subject makes an initial description of her experience, and then someone in the know describes a real event that sort of matches this initially reported experience. We already know from other evidence on false memories (see our notes on Eye-Witness Testimony) that this can be a very effective way of implanting false memories that the subject will then (either sincerely or insincerely) report as having been part of her original experience, but that we know were instead implanted later. To avoid these worries, it is important to collect subjects’ reports before they could have been corrupted with later suggestions of further details to implant, ideally by using “double-blind” studies where the subject doesn’t encounter anyone “in the know” until after she reports her NDE.

Here’s a study methodology that avoids these worries. Experimenters put letters atop cabinets in hospital rooms where they should be easily “visible” to someone floating up in an “out of body” experience, but wouldn’t be visible to patients otherwise. Then we can see whether patients who report out of body experiences are better than chance at guessing which letter was atop the cabinet in their room. In fact, experiments like this have been done, though it’s difficult because resuscitations are fairly rare, and only a tiny fraction (less than 2%) of those yield out-of-body experiences, and most of those subjects claim not to remember having seen the signs the experimenter had laid out for them. At this point, we don’t have good evidence of subjects being able to “see” things they shouldn’t, and we’ve been looking hard enough that we probably should have observed it by now if it really was real. So current evidence better supports the Hallucination Hypothesis (H2) than the Veridical Hypothesis (H1).

Hallucination when it shouldn’t be possible?

In one high profile case, Eben Alexander (2012) claims that his NDE of visiting heaven occurred while he was in a coma caused by a bacterial infection that “completely shut down” his cerebral cortex, making it be “stunned to complete inactivity” and “totally offline” in a way that should have rendered it incapable of producing hallucinations. If this were right, this might weight against a version of H2 which held that his brain was hallucinating the NDE at this time.

Even if it were true that Alexander’s cerebral cortex did completely shut down during his coma, this would not prevent his brain from having the NDE as his cortex was coming back online (Sacks 2012), nor from constructing false memories of his supposed NDE after he awoke (a version of H3).

But there’s strong reason to doubt that his cerebral cortex even was shut down throughout the period in question. There isn’t detailed brain-scan evidence verifying that his cerebral cortex was completely inactive. (What EEG evidence there is is unable to discriminate between inactivity and chaotic activity.) Furthermore, when an Esquire reporter, Luke Dittrich, tracked down details of the case, it turns out that Alexander significantly misrepresented it. Alexander described his coma as having been caused by the bacterial infection, whereas it was actually a medically induced coma that was maintained by his doctors to help promote his recovery, and one of his attending physicians reports that Alexander was conscious but delirious during that week. (We’ll see in the section on fraud that Alexander doesn’t exactly have a good track record when it comes to honestly portraying such things.)

False Memories.

We know from memory studies that memory isn’t some sort of passive “safety deposit box” in which items get stored away and then later get retrieved in exactly the same condition as when deposited. Instead, memory is a very active process by which we attempt to reconstruct a plausible narrative regarding what must have happened based on faint traces and clues that have been retained (perhaps with significant modification) from before.

Cases of NDE’s are cases where a brain was operating in far-from-ideal conditions, and whatever memory traces it laid down were likely to be unusual and jumbled. This is the sort of case where it’s especially likely that later attempts to remember may end up constructing a new narrative that differs significantly from the original experience, and it’s very likely that this newly constructed narrative will then sincerely be reported as having been part of the original experience.

As mentioned above, this makes it especially easy for subjects to implant details they learn later into their “memory” of an earlier experience, increasing the risk that they’ll later report having seen details they shouldn’t have been able to know.

This may also help to account for why subjects often report having seen their lives “flash before their eyes”. If your traumatized brain lays down a jumble of traces as it fights to survive, then your later attempts to reconstruct these traces may well gather together whatever diverse past experience seem to be somewhat linked to these traces, and then construct a narrative tying them together in a plausible way. Chaotic traces created by a resource-starved brain struggling for survival could easily get reconstrued after the fact as having been a coherent replay of life events.

Outright Fraud.

People who experience NDE’s are people who nearly died, and hence are often vulnerable and need further support. One way to help ensure that you’ll get that attention and support is to have an engaging story to tell. This is especially true if the story you tell fits people’s religious beliefs and can entice them to part with resources in the ways religions often get people to do. Hence, there can be strong incentives for people to manufacture engaging NDE stories even when they’re false.

Two recent NDE best-selling books are high profile cases that look likely to be fraudulent.

In one tragic case, a boy in a very religious family was seriously injured in a car accident. Paralyzed and debilitated, Alex Malarkey had only one remaining way to get people’s attention: telling stories of an NDE involving visiting heaven and conversing with Jesus. With his father, he co-wrote The Boy Who Came Back From Heaven, which became a best-seller, especially among Christian readers. Five years later, the boy recanted, admitting, “I did not die. I did not go to Heaven. I said I went to heaven because I thought it would get me attention.” As a consequence, the publisher pulled the books from the shelves, but only after reportedly selling a million copies. How many of those million buyers do you suppose are people who never learned of the fraud, and who still think of this as evidence of veridical experience of heaven?

Another recent high-profile case is Eben Alexander, the guy who claimed to have had an NDE at a time when his brain should have been unable to hallucinate. Unlike many believers in NDE’s, Alexander is a trained neurosurgeon who even worked at Harvard for a while. Presumably these credentials helped to lend further credibility to his best-selling 2012 book, Proof of Heaven: A Neurosurgeon’s Journey into the Afterlife.

I mentioned earlier that an Esquire reporter found that the true story of Alexander’s coma wasn’t the way he presented it. He also found that this is a recurring pattern in Alexander’s life. Alexander’s career basically involved a descent from more prestigious to less prestigious postings, following repeated accusations of malpractice. In one case, Alexander reportedly operated on the wrong spinal vertebrae in a patient, and when he later discovered this fact, went back and rewrote his medical reports to try to cover it up. This suggests that Alexander probably isn’t the most credible of witnesses. Dittrich reports that, by the time of his NDE, Alexander had effectively been prohibited from performing surgery, and was facing a multi-million dollar malpractice suit. That might seem to be a very good time for a lucrative career change, and writing a best-seller about NDE’s seems to be just what the doctor ordered.

Alexander hasn’t yet confessed to fraudulently reporting his NDE, and just because someone falsified patient medical records for their own benefit doesn’t automatically mean that they’d also falsify stories about NDE’s to sell millions of books. However, Alexander doesn’t seem to have that great of a record even on NDE’s. Alexander reports that his first NDE occurred during a near-collision with a fellow parachutist named Chuck. When Dittrich tracked Chuck down, Chuck had no memory of the incident. When pressed on the matter Alexander admitted falsifying the identity of the other parachutist, and could offer no evidence that the incident had even happened at all. Again, this isn’t quite proof of fraud, but it certainly isn’t helping Alexander’s credibility.

Admittedly, even if these two high-profile cases of NDE’s turn out to be frauds, that doesn’t mean that all other reports of NDE’s are fraudulent as well. However, it’s clear that there are incentives for fraud here, and it’s clear that these incentives sometimes do lead people to commit fraud. Knowing that should probably make you all the more inclined to take juicy reports of NDE’s with a big grain of salt.

The moral of this story.

A good general rule of thumb is to ask yourself, “Would I still be hearing this story, even if it were a fraud, or even if it were just a random fluke?” Sometimes the answer to these questions is “no” — there is no incentive for fraud, and/or the data has been vetted by gatekeepers (like journal referees) who would have been good at screening out false or spurious claims. In cases like that, you can often take a story at face value. In contrast, when the answer to these questions is “yes”, then the story should have very little evidential value for you, because you’d be likely to hear the story regardless of whether it was true or misleading. Juicy stories about NDE’s very often are in the “yes” category — these are the sorts of stories that people would be likely to pass on even if they were frauds or flukes. When a story like that gets passed on to you, you should basically just ignore it, because you already knew you’d be likely to hear stories like that regardless of whether they’re true.

 

Additional Reading.

 

My notes on Inference to the Best Explanation and the Prediction Principle: http://www.physics.smu.edu/pseudo/Induct/Fisher-Arguments-Notes.pdf

Prof. Cotton’s notes on False Memories: http://www.physics.smu.edu/pseudo/Eyewitness/

Report from the AWARE study, which interviewed thousands of resuscitated hospital patients to see what NDE’s they might have had. http://www.resuscitationjournal.com/article/S0300-9572%2814%2900739-4/fulltext

Nuanced discussion of the AWARE study, including its failure to find objective evidence of subjects seeing pre-placed visual targets atop shelves. http://theness.com/neurologicablog/index.php/aware-results-finally-published-no-evidence-of-nde/

Neuroscientist Oliver Sacks’ critical discussion of Eben Alexander’s Proof of Heaven and other NDE’s. http://www.theatlantic.com/health/archive/2012/12/seeing-god-in-the-third-millennium/266134/

NPR report on Alex Malarkey’s confessing to fraud in writing the Boy Who Came Back From Heaven. http://www.npr.org/sections/thetwo-way/2015/01/15/377589757/boy-says-he-didn-t-go-to-heaven-publisher-says-it-will-pull-book

Luke Dittrich’s Esquire article on Eben Alexander’s questionable reliability: http://www.esquire.com/entertainment/interviews/a23248/the-prophet/

Skeptics’ Dictionary article on NDE’s. http://www.skepdic.com/nde.html