What Actually Causes Deja Vu?

You walk into a room you have never visited, and for two or three seconds you are certain you have stood in that exact spot before, seen that exact arrangement of furniture, heard that exact sentence. Then it passes, and you are left slightly rattled, unable to explain what just happened.

That experience has a name - deja vu, French for "already seen" - and unlike a lot of things people ask psychologists about, it is not a fringe phenomenon. In surveys reviewed by the psychologist Alan Brown, about 60% of people report having had it at least once in their life, and it is more common in younger adults and in people with higher socioeconomic status and more education, which turns out to be an important clue.

What deja vu is not

It is worth clearing the ground first. Deja vu is not evidence of a past life, a parallel timeline, or a glitch in some simulated reality. It is also not, for the overwhelming majority of people who experience it, a symptom of anything wrong with the brain. A small minority of cases are linked to temporal lobe epilepsy, where deja vu can occur as an aura right before a seizure - but those episodes tend to be longer, more intense, and repetitive in a way that ordinary deja vu is not. If deja vu is happening to you frequently, lasting an unusually long time, or coming with other unusual sensations, that combination is worth mentioning to a doctor. For the rest of us - the majority - it is a brief, harmless quirk of memory.

The leading explanation: familiarity without recollection

Human memory runs on at least two separate processes that usually work together. One is recollection: you remember a specific past event, with some sense of when and where it happened. The other is familiarity: a bare feeling that something is known, with no accompanying memory to explain why.

Normally these two travel together - you feel familiarity because you can also recollect the specific memory that is causing it. The psychologist Anne Cleary has argued that deja vu is what happens when familiarity fires on its own, disconnected from any retrievable memory. Something about the present scene - its layout, its shapes, the arrangement of objects in it - closely resembles something you encountered before, closely enough to trigger a feeling of "I know this," but not closely enough, or not accessible enough, for you to actually locate the matching memory. The result is recognition with nothing behind it: pure familiarity, no recollection, and a strange couple of seconds while your mind searches for a memory that either does not exist or cannot be found.

Where the near-match memory might come from

If deja vu is a near-miss between the present and something stored in memory, an obvious question is what that stored something actually is. It is not always a memory of something that really happened to you while awake: some people trace a given episode of deja vu to a dream rather than a waking memory, or say it could have come from either - meaning the source scene your brain is half-matching against might be something you only ever saw in your sleep. That also means people who pay closer attention to their dream recall and their capacity for mental imagery in general may simply have more raw material lying around that a new scene can partially match, which would help explain why the experience is so unevenly distributed across people rather than universal.

Testing a feeling you cannot summon on demand

The obvious problem with studying deja vu is that nobody can make it happen to order in a laboratory - it is fleeting, unpredictable, and typically over before a person even finishes noticing it. Cleary's team got around this with a clever workaround: they built virtual-reality scenes that shared the same spatial layout as an earlier scene the person had viewed but that used different objects, so the layout was familiar even though the person could not recognize having seen that specific scene. People navigating through the new scene reported deja-vu-like sensations more often when the hidden layout matched an earlier one, exactly as the familiarity-without-recollection idea would predict. It is a laboratory sketch of the real thing rather than the real thing itself, but it is one of the few pieces of direct experimental evidence for any explanation of deja vu at all.

What we still do not know

Honesty matters here more than usual, because deja vu research runs into a hard limit: almost everything we know comes from people describing a feeling after the fact, or reacting to an artificial approximation of it in a scanner or a headset. Nobody has caught a spontaneous, real-world deja vu experience in progress inside a brain scanner. The familiarity-without-recollection account is currently the best-supported idea, and it fits nicely with related sensations - some people report a mirror-image experience called jamais vu, where something completely familiar suddenly feels foreign, which would come from the same machinery misfiring in the opposite direction. But calling it settled science would overstate how much direct proof exists.

Why it fades with age

One of the more consistent findings is that deja vu becomes less frequent as people get older, which seems backward at first - older adults have more memories to half-match against, not fewer. One proposed explanation is that the effect depends on the brain regions that quickly and automatically compare a new scene to a big database of past ones, and that this fast automatic matching process, rather than deliberate remembering, gets less active with age. It also appears more often in people who travel more and consume more media, which fits the general idea: more novel-but-similar scenes in your life means more chances for something to almost, but not quite, match.

What this means for you

The next time it happens, you now have a more interesting question to ask than "have I lived this before": what about this specific layout - the furniture, the conversation's shape, the way the room is arranged - resembles something else you have actually seen? You usually will not find the match; that is the point of the phenomenon. But recognizing deja vu as your memory system doing a slightly overzealous pattern-match, rather than a message from anywhere mysterious, is itself the most useful thing research on it has to offer.