Why Do We Get Deja Vu? What the Brain Is Doing in That Moment

Recognition involves at least two processes. One retrieves the details of a past experience: where you were, who was there, what happened. The other produces the feeling of familiarity, the sense that something has been met before, without any details attached. You use the second one when a face in the street looks known but you cannot place it. The two usually agree. Deja vu is what it feels like when the familiarity signal fires for a scene that the retrieval system correctly reports it has never stored.

The brain's frontal regions, which check memory signals for plausibility, then step in and flag the conflict, which is why deja vu comes with the knowledge that it is wrong. Young children, whose checking systems are still developing, rarely report deja vu, and neither do people whose frontal function is impaired; they may simply believe the false familiarity.

Anne Cleary at Colorado State University has produced deja vu on demand using virtual reality. Participants explore a series of computer-generated scenes, and some later scenes share the exact spatial layout of an earlier one, with the same arrangement of objects, walls and paths, while every surface, colour and object is different: a garden laid out like the junkyard seen twenty minutes before. Participants report deja vu far more often in those scenes than in genuinely new ones, and they cannot say why the place feels familiar.

The finding supports the idea that familiarity can be triggered by the shape of a scene, its configuration, independently of its content. A hotel corridor that matches the layout of a school corridor from childhood might set off the signal with no conscious memory of the school attached. Cleary's later work found that people in the grip of deja vu also feel they know what will happen next, and are no better than chance at predicting it, which suggests the feeling of premonition is another output of the same misfiring familiarity.

The strongest evidence about where the signal comes from is medical. People with temporal lobe epilepsy often experience intense, prolonged deja vu as the first sign of a seizure, and when neurosurgeons stimulate the medial temporal lobe with electrodes during surgery, patients report the feeling directly. The structures involved, the hippocampus and the rhinal cortex beside it, are exactly the ones that memory research assigns to recognition and familiarity.

Some researchers therefore describe ordinary deja vu as a tiny, harmless, self-correcting misfire in the same circuit: a brief burst of activity in the familiarity region that does not spread into a seizure and that the frontal lobes immediately catch. That would explain why it becomes rarer with age, since the brain's spontaneous electrical activity settles over the decades, and why it is more common when people are tired, stressed or travelling, all of which disturb the circuits involved.

Older accounts proposed that one eye's image reached the brain a fraction later than the other's and was registered as a memory of the first, or that the two hemispheres of the brain briefly fell out of step. Neither holds up: people blind in one eye experience deja vu, and the timing differences involved are far too small. The idea that it is a memory of a dream, or of a past life, has no evidence beyond the feeling itself, which is precisely the thing in question.

One partial survivor is the split-attention theory, in which a scene glimpsed while distracted is briefly processed, then noticed properly a moment later and recognised as familiar, because in a sense it has been seen before. It may account for some cases, but it cannot explain the laboratory results, where nothing was seen twice. Where the evidence stands:

• •Familiarity and detailed recall are separate processes
• •Deja vu is familiarity firing without recall, caught by frontal checking
• •Scenes that share a layout with a forgotten place can trigger it
• •Stimulating the medial temporal lobe produces it directly
• •It is most common in the young, the tired and the travelling

Deja vu has a mirror image called jamais vu, never seen, in which something you know perfectly well suddenly feels strange and new. You can provoke a mild version yourself: write or say an ordinary word over and over and after a while it starts to look or sound odd, as if you are not sure it is a real word. Researchers who have studied this suggest it is the familiarity signal being turned down after too much repetition, the reverse of deja vu, where familiarity fires when it should not.

A third relative is the tip-of-the-tongue state, sometimes called presque vu, almost seen. Here you are sure you know a name or word, you may know its first letter or how many syllables it has, and yet you cannot retrieve it. Again the feeling of knowing and the actual retrieval come apart. Taken together, these experiences are useful to psychologists because they show that the sense of knowing is its own signal, produced separately from the knowledge it usually accompanies. Most of the time the two agree so closely that you never notice they are different, and these small glitches are the moments when you can see the seam.

The term deja vu, French for already seen, was suggested by the French researcher Emile Boirac in the late nineteenth century and entered medical writing in the 1890s, when doctors were describing the experience in patients with epilepsy and mental illness. For most of the twentieth century it stayed hard to study, because nobody could summon it on demand and each episode was over in seconds. Much of what is known about how often it happens and to whom came from questionnaires and from diary studies, in which volunteers carry a notebook and record each episode, with its circumstances, soon after it occurs.

For nearly everyone, an occasional episode lasting a few seconds is normal and needs no attention. Doctors take more interest when deja vu becomes frequent, arrives in clusters, or comes with other symptoms such as a strange smell or taste, a rising feeling in the stomach, a surge of fear, or a gap in awareness afterwards, because that pattern can signal focal seizures starting in the temporal lobe. Since awareness can be dimmed during such seizures, an account from someone who saw the episode, together with an EEG recording of the brain's electrical activity, often helps settle the diagnosis.

Surveys suggest that roughly two thirds of people have experienced deja vu at some point. It is reported most often by teenagers and young adults and becomes less frequent with age, which runs against intuition, since older people have more memories that a new situation could resemble. Researchers are not sure why, though changes in the brain's memory systems with age are one suggestion.

Studies reviewed by the psychologist Alan Brown found that people who travel more, and people with more years of education, tend to report it more often, perhaps because they encounter more new places and situations in which a feeling of false familiarity can arise. It also seems more common when people are tired or stressed, consistent with the idea that it reflects a brief glitch in the systems that judge familiarity.

Deja vu happens when the brain's familiarity signal fires for a scene the memory system has never stored, and the frontal lobes correctly flag the mismatch. It can be triggered by a layout that matches a forgotten place, it arises in the same temporal lobe circuits that epilepsy disturbs, and in a healthy brain it is a brief glitch in a system that works well the rest of the time.