The Science of Déjà Vu: What's Actually Happening in Your Brain

Déjà vu is the feeling that you have lived through the present moment before, combined with the knowledge that you can't have. That second part matters. Researchers define it as a clash between a strong sense of familiarity and the certainty that the familiarity is wrong. If you really believed you had been in the room before, it would just be a mistaken memory.

It is also very common. In 2003 the psychologist Alan Brown reviewed the surveys published up to then and concluded that about two thirds of people say they have had it at least once. People report it most in their late teens and twenties, and less as they get older. It is reported more by people who travel a lot, who have more years of education, and who are tired or stressed at the time.

Illustration: déjà vu, the sense of having been here before.

What epilepsy taught neurologists

The first solid clue came from patients with temporal lobe epilepsy. For some of them, an intense déjà vu is the warning sign, or aura, that a seizure is starting. In the 1950s the Canadian neurosurgeon Wilder Penfield found that he could sometimes trigger the feeling by electrically stimulating the temporal lobe of patients awake during surgery.

A more precise result came in 2004. Fabrice Bartolomei and colleagues in Marseille stimulated different parts of the medial temporal lobe through electrodes already implanted in epilepsy patients for medical reasons. Stimulating the rhinal cortex, a strip of tissue next to the hippocampus that is involved in judging whether something is familiar, produced déjà vu far more often than stimulating the hippocampus itself. That supports the idea that déjà vu is a familiarity signal firing without a real memory behind it.

Familiarity without a memory

Psychologists distinguish between recognising something as familiar and recalling where you know it from. You can know a face in a crowd is familiar long before you remember that the person works at your bank. Déjà vu, on this account, is the first process firing on its own.

Anne Cleary at Colorado State University has tested what might set it off. In virtual-reality experiments, she showed volunteers a series of scenes and later showed them new scenes whose layout matched an earlier one, with the furniture and objects all different. People were more likely to report déjà vu for the new scenes that shared a layout with something they had seen but could not recall. The familiarity of the arrangement leaked through without the memory of where it came from. In a 2018 study in Psychological Science, Cleary and Alexander Claxton also found that people in déjà vu states felt strongly that they knew what would come next, but were no better than chance at actually predicting it.

A memory check that works

A different idea comes from Akira O'Connor and colleagues at the University of St Andrews. They produced something close to déjà vu in the lab with an old memory trick. Volunteers heard lists of related words, such as bed, pillow, night and dream, but never the word that links them, sleep. Later they were asked whether they had heard any words starting with "s", and said no. Then they were asked whether they had heard "sleep". Many felt that they had, while knowing they could not have.

When the team scanned volunteers' brains during this task, they expected activity in the memory areas of the temporal lobe. Instead, as they reported in the journal Memory, the activity during déjà vu-like moments was in frontal regions involved in decision-making and in detecting conflicts. Their interpretation is that déjà vu might be what it feels like when the brain's checking system catches a false sense of familiarity and flags it. On this view the experience is a sign of a memory error being caught, not of one slipping through. O'Connor has suggested this could also explain why déjà vu becomes rarer with age: if the checking system weakens, fewer false memories are caught and flagged.

The two ideas are not really rivals. One describes what produces the false familiarity, the other what happens when the brain notices it.

Jamais vu and déjà vécu

Jamais vu is the reverse: something completely familiar suddenly seems strange. It is easy to produce. Write the same common word thirty times and it often starts to look wrong. Chris Moulin, O'Connor and colleagues did exactly that in experiments, and found that most volunteers reported the word starting to feel unfamiliar, often after about thirty repetitions. The paper, titled "The the the the induction of jamais vu in the laboratory", won the Ig Nobel Prize for literature in 2023.

Déjà vécu, "already lived", is rarer and more troubling. People with some forms of dementia can experience a persistent feeling that they have lived through events before, to the point of refusing to watch television because they believe they have seen every programme. Unlike ordinary déjà vu, the person no longer recognises that the feeling is false.

For most people, occasional déjà vu is harmless. Episodes that come often, last longer than a few seconds, or arrive with other symptoms such as a strange smell or taste, a rising feeling in the stomach, confusion or lost time can be focal seizures, and are worth describing to a doctor.

Sources

1. Alan S. Brown, "A review of the déjà vu experience", Psychological Bulletin 129 (2003): 394–413. https://doi.org/10.1037/0033-2909.129.3.394
2. F. Bartolomei, E. Barbeau, M. Gavaret, M. Guye et al., "Cortical stimulation study of the role of rhinal cortex in déjà vu and reminiscence of memories", Neurology 63 (2004): 858–864. https://doi.org/10.1212/01.wnl.0000137037.56916.3f
3. Anne M. Cleary and Alexander B. Claxton, "Déjà vu: an illusion of prediction", Psychological Science 29 (2018): 635–644. https://doi.org/10.1177/0956797617743018
4. Josephine A. Urquhart, Magali H. Sivakumaran, Jennifer A. Macfarlane and Akira R. O'Connor, "fMRI evidence supporting the role of memory conflict in the déjà vu experience", Memory 29 (2021): 921–932. https://doi.org/10.1080/09658211.2018.1524496
5. Chris J. A. Moulin, Nicole Bell, Merita Turunen, Arina Baharin and Akira R. O'Connor, "The the the the induction of jamais vu in the laboratory: word alienation and semantic satiation", Memory 29 (2021): 933–942. https://doi.org/10.1080/09658211.2020.1727519