The most common objection to near-death experiences is that they are surely just a hallucination – a dying brain, oxygen shortage, a borderline dream. That objection can be tested surprisingly concretely, in a place one does not think of at first: in memory. Hallucinations, dreams and deliria leave a very particular trace – and near-death experiences, several independent studies show, leave an entirely different one.
The logic behind the argument
Not all memories are alike. A real perception is stored rich in sensory detail, context and emotion; a dream or a mere imagining, by contrast, fades quickly, stays patchy, goes blurry. This is exactly where the argument bites: if near-death experiences were mere hallucinations, dreams or deliria, they would have to be stored as such – faded, fragmentary and indistinct. This can be measured. And the result comes out differently from what the hallucination hypothesis would lead one to expect.
Thonnard et al. (2013): richer than real memories
The team around the neurologist Steven Laureys (Coma Science Group, University of Liège) gave experiencers the Memory Characteristics Questionnaire – an established instrument that captures the phenomenological quality of a memory: sensory detail, clarity, self-reference, emotional content. Four memory types were compared: of the near-death experience, of a real event, of a merely imagined event, and of an unconscious period (coma) without an NDE.
The result was unambiguous: NDE memories carried more of these characteristics than any other type of memory – more than imagined events, more than the coma, and even more than memories of real events. The authors concluded that NDEs "cannot be considered as imagined event memories." The study speaks of a kind of hyper-reality – memories that stand unrivalled in their density.
Palmieri et al. (2014): the signature of a real perception
An Italian group (University of Padua) went a step further and combined the memory analysis with EEG. While recalling the NDE memory the brain showed electrophysiological patterns – including theta activity – typical of recalling actually experienced episodes, and precisely not of imagined content. At the level of brain activity, an NDE memory thus behaves like the memory of something actually perceived.
Important for honesty's sake: the same authors stressed that this did "not necessarily" mean the experience corresponded to the external physical world. Synthesising both studies, they offered their own reading – that NDE memories are "hallucination-like memories of actually perceived hallucinations." We will return to that camp in a moment.
Moore & Greyson (2017): the confirmation
If the finding were a fluke of the first sample, it would fall apart on repetition. Lauren Moore and Bruce Greyson at the University of Virginia gave the same questionnaire to a larger group – with the same result: near-death experiences are remembered as "more real" than real events. The effect is therefore robust and reproducible, not an artefact.
Greyson (2007): unchanged over two decades
There is a second property by which inventions give themselves away: they change. Confabulations are embellished with each retelling, dreams blur, stories grow. Greyson tested this directly: 72 experiencers who had recorded their near-death experience on the NDE Scale in the 1980s did so again almost two decades later. The scores remained practically unchanged – the accounts were not embellished over twenty years. Such stability is the opposite of what one expects from a good story or a hallucination.
How to read the data – two camps
This is where it matters, and where the paths diverge. All four studies show the same thing: near-death experiences are not merely imagined or invented memories. What follows from that, researchers read differently:
- Physiological reading (Liège/Padua): the exceptional memory quality is a property of the encoding itself – emotionally and self-relevant events are stored especially deeply. The experience was "really perceived" but not "lived in reality": a really perceived hallucination.
- Survival-oriented reading (Greyson and others): this very pattern weighs against the hallucination hypothesis. Unlike imagined or invented material, NDEs are recalled with the phenomenological density of genuine perceptions and remain constant over two decades – both of which sit poorly with hallucination or confabulation.
Both camps share the same hard finding and argue only about its reach. Strikingly, even the physiological reading classifies the experience as "really perceived." That only relocates the real question: if, at a moment when the brain is maximally compromised – in cardiac arrest, say – such a clear, dense, stable perception arises and is stored: where is it perceived, and what does it refer to?
What the studies show – and what they do not
Let us stay honest: a vivid, stable memory at first proves only that an experience was subjectively real and emotionally significant – not automatically that it corresponded to an external event. The authors of the memory studies themselves expressly did not take that last step. So the memory argument does not refute "the brain"; it refutes one particular, often casually offered explanation: that near-death experiences are mere hallucinations, dreams or confusion. That convenient shortcut is precisely what the data do not support.
Whether the experience touches something real outside the brain is another question – decided not by memory alone but by veridical perception: by checkable observations during the phase in which the brain demonstrably should not have been working. Memory does not provide the proof – but it clears away one of the most popular objections and keeps the question open.
Sources:
• Thonnard M. et al. (2013), Characteristics of Near-Death Experiences Memories as Compared to Real and Imagined Events Memories, PLOS ONE 8(3):e57620 (doi).
• Palmieri A. et al. (2014), "Reality" of near-death-experience memories: evidence from a psychodynamic and electrophysiological integrated study, Frontiers in Human Neuroscience 8:429 (doi).
• Moore L. E. & Greyson B. (2017), Characteristics of memories for near-death experiences, Consciousness and Cognition 51:116–124 (doi).
• Greyson B. (2007), Consistency of near-death experience accounts over two decades: Are reports embellished over time?, Resuscitation 73(3):407–411 (doi).
Continue in our curated knowledge collection – see also the articles on veridical perception, on the out-of-body experience, on the psychological defence against near-death experiences and on the question of brain and consciousness.