People who report ghosts are, overwhelmingly, telling the truth about what they experienced. That is the starting point, and it is what makes the subject interesting rather than embarrassing. The question is not whether they saw something but what the brain does with a dark corridor, a cold draught and a story it has already been told.
Expectation
The most reliable predictor of a haunting experience is knowing in advance that the place is haunted. Studies at sites with reputations — old prisons, theatres, castles open to the public — have found that visitors told a location is “active” report more sensations there than visitors told nothing, in the same rooms, on the same nights. The brain is a prediction engine. Given a prompt, it fills in the pattern.
This is not gullibility. It is the same mechanism that lets you read a smudged word or hear your name across a noisy room. Expectation lowers the threshold for detection; in an ambiguous environment, more ambiguous things get detected.
Pareidolia and the search for faces
Human vision is tuned to find faces and figures, because for most of our history the cost of missing one was higher than the cost of a false alarm. Shadows, folds of cloth, patterns in old plaster and the play of a torch beam across a wall all present the visual system with just enough to work with. The face at the window in a hundred photographs of old houses is usually a reflection or a curtain; the figure at the end of the corridor is usually a coat.
Sound works the same way. The brain assembles words from noise if it has been told words are there, which is why recordings of empty rooms can seem to contain voices once someone has suggested what they say.
The body’s own signals
Cold spots, pressure on the chest, the sense of being watched and the feeling that something is standing just behind you all have well-documented ordinary causes: draughts through old buildings, carbon monoxide from faulty heating, infrasound below the range of hearing produced by wind, machinery or organ pipes, and the paralysis and hallucination of the moments between sleep and waking, which has been described identically in every culture that has left records. None of that makes the experience less real to the person having it. It explains why the same experiences cluster in the same kinds of places.
Memory
A haunting is rarely reported in the moment. It is told afterwards, and every telling reshapes it. Details migrate from other stories, ambiguous sensations become specific, and the narrative acquires the structure that stories need — a beginning, a build, a climax. Memory is reconstructive, not archival. By the third telling, an honest witness may be describing something they did not quite see.
What this leaves
It leaves the places. Strip out the psychology and you are left with a real building where something terrible happened, which is the thing that made people expect to see ghosts there in the first place. The fear is a response to history. That is the part worth investigating.
