Suppose you are sitting in your living room on an ordinary evening and a figure walks in. You notice nothing odd about the figure. It looks solid and casts no glow. It might be wearing a brown coat, or a nurse’s uniform. It walks across the room and sits in the armchair, and the cushion seems to take its weight. Only when it vanishes, or fails to answer you, do you realize that nobody was there.
Experiences like this are more common, and more ordinary, than the folklore of ghosts suggests. In the 1970s two Oxford-connected researchers, Celia Green and Charles McCreery, asked how such an experience could be so seamless. Their answer was a radical proposal. They suggested that the figure is not the only hallucination. On their account the whole room may be one, including the armchair, the lamp, and the wall behind them.
They called this a metachoric experience. The term connects what looks like a ghost story to lucid dreaming, false awakenings, and out-of-body experiences, and it has shaped how researchers think about how the waking brain builds a world.
The Institute and the Survey
The story starts with the Institute of Psychophysical Research (IPR) in Britain, founded by Celia Green. In 1968 Green published Lucid Dreams, a study of dreamers who become aware that they are dreaming. She also published a collection of first-hand accounts of out-of-body experiences (OBEs). Sleep researcher Keith Hearne acknowledged that his initial knowledge of lucid dreams came directly from Green’s work, and neurologist Oliver Sacks described her as the first psychologist to systematically examine a large number of out-of-the-body cases.
The Institute’s method was to launch public appeals through the press and radio, asking people to write in with first-hand accounts of particular experiences. The institute’s first metachoric subject was lucid dreams, and in 1966 it launched an appeal for a second type of experience closely related to them, out-of-the-body experiences. The third major type was the apparition, the perception of someone or something “not really there”
The apparitions survey ran from roughly 1968 to 1974, with Green working alongside McCreery, an Oxford psychologist. The material was published as Apparitions in 1975. The size of the sample is reported inconsistently. The source material for this article speaks of about 1,800 respondents, while another account describes the book as a taxonomy built on a collection of 1,500 first-hand accounts. The likely explanation is that the number of respondents and the number of cases analyzed in detail differ. Either way, this was a large set of testimonies, gathered by systematic appeal, from people who had no obvious reason to invent anything.
What Green and McCreery Were Not Doing
The IPR survey differed in purpose from older ghost research. Earlier investigators, notably the Victorian-era Society for Psychical Research, mostly used apparitions to argue for or against telepathy or life after death. Green and McCreery were less interested in whether apparitions shed light on telepathy or survival, and more concerned with analyzing many cases to build a taxonomy of the different types, treating them simply as a form of anomalous perception or hallucination.
That shift in framing made the work influential well beyond parapsychology. The authors did not ask “Was that a real ghost?” They asked what the experience was like, how it was structured, and what it revealed about perception.
What the Survey Found
Several findings from the book have become touchstones in the literature.
The stereotype was wrong. Popular imagery gives ghosts a translucent, sheeted look. The survey found something different. Apparitions tend to be reported as solid rather than transparent, and can be so realistic that they deceive the percipient about their hallucinatory nature; sometimes insight only arrives after the experience has ended. According to the summary material this article draws on, fewer than half of witnesses realized at once that they were seeing something non-living, and between half and two-thirds reported at least some color, such as hair or clothing.
Most experiences were brief and close. According to the survey, most apparitions appear visually within three metres of the person, and many last less than a minute. These were not the prolonged hauntings of fiction.
Most were narrow in sensory scope. The summary figures attribute to the survey that about 61% of witnesses experienced the apparition through a single sense, usually sight, and that touch featured in only about 15% of cases. Those numbers come from a secondary summary and I could not verify them against the original book, so treat them as approximate. The broader pattern is that apparitions are usually visual events. They are rarely the full multisensory presence that fiction implies.
Not every apparition was human, or frightening. The survey also recorded animals and even inanimate objects. Self-selected samples tend to report a predominance of human figures, but apparitions of animals and objects are also reported. Most human figures are not recognized by the subject, and not all are deceased persons, since apparitions of living people are also reported. That last point cuts against the idea that these are messages from the dead.
The survey also identified a category the authors called “reassuring apparitions.” Green and McCreery identified a class of what they called “reassuring apparitions”. In these cases the person emerged feeling comforted rather than alarmed. The summary material puts these at about 4% of subjects. That is small, but it suggests the emotional character of apparitions is broader than horror.
The Puzzle: Why Are They So Well Integrated?
Take all these findings together and a puzzle appears. If an apparition is a hallucination, why is it so good? It does not flicker, and it does not look pasted onto the scene. According to the accounts, it tends to occupy space plausibly. It sits on chairs, stands behind tables, and does not walk through walls, at least until it vanishes.
The default picture of hallucination, and of ghost-seeing, is an overlay. The brain receives a real visual scene and somehow inserts a phantom figure into it. Green describes the assumption they were challenging as the idea of a hallucination as an isolated area of the visual field, generated erroneously by the brain and superimposed on the rest of the field, which is generated from actual input in the normal way.
The overlay model raises a hard question. Inserting a convincing figure into a real, moving visual scene is a difficult computational task. The figure has to be lit consistently with the room, occluded correctly by real objects, and positioned in three-dimensional space. Doing this while the eyes continue to feed in the real scene looks like a lot to ask of a brain error.
The Metachoric Hypothesis
Green and McCreery’s alternative was to stop treating the figure as the hallucination. Perhaps the entire perceptual environment is hallucinated. In Green’s words, rather than normal perception plus a finite hallucinatory element, the perceptual environment is entirely replaced by a hallucinatory one, at least as long as the apparitional figure is being perceived.
The word “metachoric” comes from Greek roots suggesting a change of place. The synopsis of Apparitions states the hypothesis directly: the authors advance the hypothesis that some, or even all, apparitional experiences are “metachoric”, meaning the whole of the visual field is hallucinatory.
Note the careful wording in that synopsis: “some, or even all”. The authors treated the metachoric hypothesis as a proposal that the evidence supported, not a proven certainty, and they left open how many cases it covered.
An Analogy
Think of a video game. In the overlay model, the game shows a real-world camera feed and pastes a cartoon character on top of it. In the metachoric model, the game has rendered the entire scene from stored data, including the room, and the character is simply one more element in the render. The character then obeys the scene’s physics because it belongs to the same rendering. The metachoric claim is that the brain can switch from the camera feed to the render without the player noticing.
Why Would the Room Be Rebuilt?
Suppose the brain is generating the whole scene. Where does the room come from? The natural answer is memory. You have seen your living room thousands of times, and the brain holds a detailed internal model of it. A metachoric experience would involve the brain running on that model instead of on live input, and doing so faithfully enough that you don’t register any change. This would explain a feature the authors emphasized: subjects reported no noticeable gap or break in consciousness. They felt fully awake before, during, and after.
The Evidence for Total Replacement
The evidence for the hypothesis is indirect, since nobody can observe a hallucinated room directly. It consists of features of the accounts that the overlay model handles awkwardly.
Spatial integration. Apparitions are typically reported as consistent with the surrounding scene. This is easier to explain if figure and room come from one process. The source summary makes the strong claim that ghosts “never” clip through furniture and cast realistic shadows. That is overstated. Some accounts do describe figures passing through objects, and the claim is better read as a statement of the general tendency. But the tendency was striking enough to matter to the authors.
Changes in lighting. Some witnesses reported that when the figure appeared, the whole room seemed to change in brightness, or was suffused with an unusual glow. If the room were still being seen through the eyes, a change of that kind would need a physical cause. If the room is being rendered internally, the brain can change the apparent lighting as easily as it changes anything else.
Continuity of consciousness. Witnesses often describe no discontinuity. There was no sense of drifting off, no blur at the edges of awareness, and no sense of a transition. That fits a mechanism that swaps one detailed scene for another rather than one that visibly degrades perception.
The evidence is suggestive rather than conclusive. Each of these features could in principle be explained in other ways, and witnesses’ memories of an unusual experience are fallible. Reviews of the book were mixed. Green’s Apparitions received mixed reviews. Some praised the documentation and its sober tone, and the Daily Telegraph called it an excellent piece of documentation, soberly treated. The theoretical leap was another matter, and it remains a hypothesis rather than an established finding.
The Unified Family of Metachoric States
The most far-reaching consequence of the theory is that it groups together experiences that had been studied separately. If an apparition involves the brain generating a complete environment, then it belongs in the same family as other experiences where that happens.
The IPR had already studied several of these. Green and McCreery ultimately proposed that four kinds of experience share the metachoric mechanism:
| Experience | What the mind is generating | State of the person |
| Apparitions | A replica of the current room plus an anomalous figure | Apparently wide awake |
| False awakenings | A replica of the bedroom, leading the person to believe they have woken | Actually asleep |
| Lucid dreams | A fictional or remembered environment, with awareness that it is a dream | Asleep |
| Out-of-body experiences | A replica of the room seen from an external, elevated viewpoint | Varies |
Later work by McCreery and Green added false awakenings to the category. McCreery later collaborated with Green on a book on lucid dreaming, which discussed the concept of metachoric experience further and added false awakenings to the category.
False Awakenings
The false awakening is perhaps the cleanest illustration. A person dreams that they wake up in their own bedroom. They may get out of bed, start their morning routine, or notice that something is subtly wrong. The bedroom in the dream is usually accurate down to small details. Then they truly wake up, sometimes discovering they were asleep all along.
For the metachoric theory, this is the brain rendering a replica of the real bedroom from memory while the person sleeps. The apparition case is the same kind of event occurring in someone who feels awake. The overlap is the appeal of the theory. If false awakenings show the sleeping brain can build a convincing copy of your room, it is a smaller step to suppose the waking brain occasionally does something similar.
Lucid Dreams and OBEs
Lucid dreaming, in which the dreamer knows they are dreaming, was Green’s first major topic. It is metachoric in the same sense, since the environment is entirely generated. What distinguishes it is the awareness. The lucid dreamer has, in effect, noticed the render.
Out-of-body experiences fit the pattern with a twist. The person perceives their own body from outside, usually from above. The metachoric account treats this as a hallucinatory replica of the environment rendered from an unusual viewpoint. One dimension that helps categorize a metachoric experience is lucidity. Whether the person recognizes the experience as unreal is itself a variable that cuts across all four types.
Where Would It Come From? McCreery’s Neuroscience
A hypothesis about perception invites the question of mechanism. If the waking brain can drop into a dream-like state, what triggers it?
McCreery’s answer draws on the physiology of sleep. The conventional route to sleep is relaxation: arousal falls, sensory input is dampened, and the brain slides toward dreaming. But McCreery pointed out a second, less familiar route. He suggested that sleep may be approached not only by the conventional route of low arousal and deafferentation, but also by the less familiar route of extreme stress and hyper-arousal.
This addresses a paradox in the apparition data. Many experiences occur when the person is alert or upset, not drowsy, which is not when you would expect sleep-like phenomena. If a sudden spike of arousal can push the brain toward a dreaming mode, then a witness might be fully awake one moment and inside a waking dream the next. The brain, in this account, temporarily loosens its dependence on external senses. It runs on its internal model of the environment, producing a scene so complete the person has no reason to doubt it.
This is a hypothesis about mechanism, and it remains speculative. Sleep science offers partial support. The related phenomena of hypnagogic and hypnopompic hallucinations, which occur at sleep onset and on waking, are well documented. They are considered normal and generally harmless; hypnopompic hallucinations occur in over 12% of people and hypnagogic ones in up to 37%. These states show that dreamlike imagery can intrude into apparent wakefulness. They are usually short, tied to the transition between sleep and waking, and only loosely comparable to the metachoric cases, where people report being fully alert.
The neuroscience is also less settled than older accounts suggest. Researchers used to argue that hypnopompic hallucinations came from REM sleep intruding on wakefulness, but further research has not supported this, and they seem to occur during early non-REM stages. That complicates any simple story in which the brain flips into “dream mode.” A modern version of the metachoric mechanism would have to be stated in terms of the specific sleep stage and arousal state involved, and the field has not agreed on the details.
How Common Is All This?
One reason the work matters is that it treats these experiences as ordinary. Attempts to survey general population samples suggest hallucinatory experiences of this kind are relatively common, with incidence figures of 15 to 25 percent commonly reported.
If those figures are even roughly right, apparitional experiences are not the province of the credulous or the unwell. Green and McCreery’s conclusion, as summarized in the source material, was that these experiences happen to entirely ordinary, sane people. That stance has a practical consequence: it invites people to report their experiences without fear of being labeled disturbed, and it gives clinicians and researchers a way to distinguish benign hallucinations from those associated with illness.
What the Theory Explains, and What It Doesn’t
A fair assessment has to weigh both sides.
What it explains well. The metachoric hypothesis handles several features of reports more naturally than the overlay model. It accounts for how well integrated apparitions are, for reported changes in lighting, and for the lack of any noticed transition. It provides a framework linking waking apparitions to dreams, lucid dreams, false awakenings, and OBEs, and it suggests that they share underlying machinery, which is a parsimonious position. It also removes the need to postulate that ghosts are physical or spiritual entities. The apparent realism of the experience is explained by the brain’s own capacity for realistic simulation.
Where it is weaker.
- The data are retrospective and self-selected. People who answer an appeal about ghost experiences are not a random sample, and memory of a striking event is prone to reconstruction. Details like “the room went dim” may be shaped by later retelling.
- The strong claim may not cover all cases. The authors’ own wording, “some, or even all,” reflects uncertainty. Some apparitions may be closer to the overlay type, and the evidence cannot always tell them apart.
- Mechanism is speculative. The arousal-driven route to a dream-like state is an interesting proposal, but it is not settled sleep science.
- It does not settle unusual cases. Reports of apparitions that convey accurate information the witness could not have known are the cases that originally motivated parapsychology. The metachoric theory is a theory of how the experience is built, and it is silent, or at least agnostic, on whether any experience carries information from outside the witness’s mind. Green and McCreery approached apparitions as perceptual phenomena and did not try to adjudicate those claims.
Why the Idea Matters Beyond Ghosts
The metachoric concept carries several lessons beyond the study of apparitions.
It reframes perception. Modern neuroscience increasingly describes ordinary perception as a form of controlled hallucination: the brain generates predictions about the world and corrects them with sensory input. If so, the difference between normal perception and hallucination is one of degree, and a metachoric experience is a case where the correction from the senses is temporarily switched off while the generative machinery keeps running. The 1975 proposal anticipated that idea and gave it an unusual set of test cases.
It shows how convincing an internal model can be. The fact that a person can be persuaded they are awake in a familiar room, when they are not, says something about how much of what we perceive is drawn from internal expectation. Everyone experiences a mild version of this whenever a dream feels real.
It offers a comforting and practical reading of unnerving experiences. Many people who have seen an apparition are frightened, and some worry about their sanity. The metachoric account presents these episodes as a rare but natural product of a healthy brain, in the same family as dreams, and it explains why the experience can be so realistic. That does not make it less strange, but it may make it less alarming. The “reassuring apparitions” in the survey are a reminder that the emotional tone of these events varies.
Conclusion
Celia Green and Charles McCreery’s survey began as an attempt to catalog ghostly encounters and ended as a theory of how the waking mind can build a world. The central claim is deceptively simple. What looks like a phantom added to a real room may be a phantom within a completely replaced one, produced by a brain that has quietly switched from perceiving to simulating.
The theory is not proven, and its evidence rests on memory and testimony, but it is coherent and testable in principle. It explains why apparitions look solid rather than gauzy, why they fit into the room as they do, and why a person can feel wide awake throughout. It also connects apparitions to false awakenings, lucid dreams, and out-of-body experiences, treating them as variations on a single capacity of the human brain.
Half a century on, the term “metachoric” is still used, and the questions it raises are still open. How does the brain decide when to trust its senses and when to trust its model? What happens when that decision goes subtly wrong? The ghost in the armchair turns out to be an unexpectedly good place to start asking.
References
- Wikipedia, “Charles McCreery”: https://en.wikipedia.org/wiki/Charles_McCreery
- Wikipedia, “Celia Green”: https://en.wikipedia.org/wiki/Celia_Green
- Wikipedia, “Apparitional experience”: https://en.wikipedia.org/wiki/Apparitional_experience
- Wikipedia, “Anomalous experiences”: https://en.wikipedia.org/wiki/Anomalous_experiences
- Celia Green, “Metachoric experiences” (blog, 15 Aug 2020): https://celiagreen.blogspot.com/2020/08/metachoric-experiences.html
- Celia Green, “Lucid Dreams as Metachoric Experiences,” Dream Network Journal (1 May 1982): https://dreamnetworkjournal.com/bcpoz2mameta/lucid-dreams-as-metachoric-experiences
- Mirror of the Wikipedia McCreery entry: https://hrvwiki.net/Wikipedia_Charles_McCreery
- AbeBooks, publisher synopsis of Apparitions: https://www.abebooks.com/9780241891827/Apparitions-Proceedings-Institute-Psychophysical-Research-0241891825/plp
- Sleep Foundation, “Hypnopompic Hallucinations”: https://www.sleepfoundation.org/how-sleep-works/hypnopompic-hallucinations
- “Out-of-body experiences, dreams, and REM sleep”: http://lucidity.com/REMOBE.pdf
- ScienceDirect, “Hypnagogic Hallucination” overview: https://www.sciencedirect.com/topics/medicine-and-dentistry/hypnagogic-hallucination
- Susan Blackmore, “Paranormal Belief and Interpretations of Sleep Paralysis”: https://www.susanblackmore.uk/conferences/paranormal-belief-and-interpretations-of-sleep-paralysis/
- Green, C. and McCreery, C., Apparitions (London: Hamish Hamilton, 1975)
- Green, C. and McCreery, C., Lucid Dreaming: The Paradox of Consciousness During Sleep (1994)
- Green, C., Lucid Dreams (1968)
- McCreery, C., “Waking dreams and other metachoric experiences,” Psychiatric Journal of the University of Ottawa, 15 (1990)