The neural basis of déjà vu

Reading a book for the second time can be an enlightening experience. At the same time, aspects of this experience can be confusing. During a second visit to a work previously read I suspect that to a degree everyone plays a game that involves trying to determine which parts of the story you remember well and which parts you completely forgot. But there are also parts that lie somewhere in the middle; it is these parts that boggle our minds by leaving us uncertain of whether or not the details are familiar to us. Perhaps a nuance in the storyline that strikes you as familiar is something you actually skimmed over and ignored the first time you read the book. You have some awareness of your ignorance and begin to question your feeling of familiarity.

This clash of evaluations lies at the heart of déjà vu, the experience of recognizing a situation as familiar while simultaneously being aware that the situation may not have actually occurred in the past. Chris Moulin and Akira O’Connor, researchers who have attempted to study déjà vu in their laboratory, have recently published a paper outlining the current state and challenges of scientific research on this inherently subjective phenomenon. They discuss two broad categories of recent research: déjà vu in clinical populations (e.g. associations with epilepsy and dementia), and déjà vu in nonclinical populations.

Importantly, Moulin and O’Connor point out that these categories may be distinct, and that caution should be exerted when making comparisons between the two. A lot of research on déjà vu in clinical populations is not actually study of déjà vu, but of a slightly different experience called déjà vecu (also known as ‘recollective confabulation’) in older adults with dementia. Déjà vecu instances involve inappropriate feelings of familiarity, like in déjà vu, but the feelings are not necessarily accompanied by awareness that it is inappropriate. The validity of extending evidence from studies on déjà vecu to the casual experience of déjà vu is questionable.

However, there has been a movement in experimental cognitive psychology toward studying déjà vu by generating the phenomenon in nonclinical populations. These studies use techniques such as hypnotic suggestion and familiarity questionnaires about images that are previously shown or not shown to subjects. There are few studies using these techniques, and the applicability to déjà vu experiences in everyday life is still being questioned.

So a solid scientific theory of déjà vu is still nonexistent. But there have been some recent neuroscientific investigations that have shed light on the neural basis of déjà vu. Deep brain stimulation and brain lesions studies both implicate areas in the mesial temporal cortex in the generation of déjà vu. Moulin and O’Connor argue that this doesn’t necessarily mean we can label this region as the ‘déjà vu cortex’ of the brain; rather, if we are to make progress in understanding déjà vu in the brain, we should examine how mesial temporal structures interact with whole neural networks during instances of déjà vu. For example, the authors hypothesize that “mesial temporal structures may aberrantly indicate a sensation of familiarity despite the rest of the hippocampo-cortical network indicating the overarching nonrecognition state that ultimately presides.”

In other words, when you’re re-reading a book or article that was edited with a minor detail after you read it the first time, your mesial temporal regions are telling you that the minor detail is familiar, but the rest of your brain is telling you that you never read that minor detail the first time. What’s happening in the rest of the brain, and why the brain decides to confuse you like this in the first place are questions for further research. That means that in the labs of Moulin, O’Connor and déjà vu researchers alike, it will be, in the famous words of Yogi Berra, “déjà vu all over again.”