Pick a memory. It could be as recent as breakfast or as distant as your first day of kindergarten. What matters is that you can really visualize it. Hold the image in your mind.

Now consider: Do you see the scene through your own eyes, as you did at the time? Or do you see yourself in it, as if you’re watching a character in a movie? Do you see it, in other words, from a first-person or a third-person perspective? Usually, we associate this kind of distinction with storytelling and fiction-writing. But like a story, every visual memory has its own implicit vantage point. All seeing is seeing from somewhere. And sometimes, in memories, that somewhere is not where you actually were at the time.

This fact is strange, even unsettling. It cuts against our most basic understanding of memory as a simple record of experience. For a long time, psychologists and neuroscientists did not pay this fact much attention. That has changed in recent years, and as the amount of research on the role of perspective has multiplied, so too have its potential implications. Memory perspective, it turns out, is tied up in criminal justice, implicit bias, and post-traumatic stress disorder. At the deepest level, it helps us make sense of who we are.

The distinction between first- and third-person memories dates back at least as far as Sigmund Freud, who first commented on it near the end of the 19th century. Not for another 80 years, though, did the first empirical studies begin fleshing out the specifics of memory perspective. And it was only in the 2000s that the field really started picking up steam. What those early studies found was that third-person memories were far less unusual than once thought. The phenomenon is associated with a number of mental disorders, such as depression, anxiety, and schizophrenia, but it is not merely a symptom of pathology; even among healthy people, it is quite common.

Just how common is tricky to quantify. Peggy St. Jacques, a psychology professor at the University of Alberta who studies perspective in memory, told me that roughly 90 percent of people report having at least one third-person memory. For the average person, St. Jacques estimates, on the basis of her research, that about a quarter of memories from the past five years are third-person. (At least a couple of papers have found that women tend to have more third-person memories than men do, but a third study turned up no statistically significant difference; on the whole, research on possible demographic disparities is scant.) In certain rare cases, people may have only third-person memories. As you try to recall your own, be warned that things can get confusing fast. Perhaps you can call to mind early-childhood scenes that you picture from a third-person perspective. But it’s hard to know whether these are genuine memories translated from the first person to the third person, or third-person scenes constructed from stories or photographs. To some people, third-person memories are second nature; to others, they sound like science fiction.

Why any given memory gets recalled from one perspective rather than the other is the result of a whole bunch of intersecting factors. People are more likely to remember experiences in which they felt anxious or self-conscious—say, when they gave a presentation in front of a crowd—in the third person, St. Jacques told me. This makes sense: When you’re imagining how you look through an audience’s eyes in the moment, you’re more likely to see yourself through their eyes at the time of recall. Researchers have also repeatedly found that the older a memory is, the more likely you are to recall it from the third person. This, too, is fairly intuitive: If first-person recollection is the ability to adopt the position—and inhabit the experience—of your former self, then naturally you’ll have more trouble seeing the world the way you did as a 6 year old than the way you did last week. The tendency for older memories to be translated into the third person may also have to do with the fact that the more distant the memory is, the less detail you’ll likely have, and the less detail you have, the less likely you are to be able to reassume the vantage point from which you originally witnessed the scene, David Rubin, a Duke University psychology professor who has published dozens of papers on autobiographical memory, told me.

Less intuitive, perhaps, is the reverse: People are able to recall a scene in greater detail when they’re asked to take a first-person perspective than when they’re asked to take a third-person perspective. “Sometimes in a courtroom, an eyewitness to a holdup might be asked to recall what happened from the perspective of the clerk,” St. Jacques told me. But if her research is any indication, such tactics may blur rather than sharpen the witness’s memory. “Our research suggests that might actually be more likely to make the memory less vivid, make the eyewitness less likely to remember the specifics.”

Even without an examiner’s instructions, such an eyewitness might be predisposed to recall the robbery in the third person: Researchers have found that people often translate traumatic or emotionally charged memories out of the first person. This may be because first-person memories tend to elicit stronger emotional reactions at the time of recall, and by taking a third-person perspective, we can distance ourselves from the painful experience, Angelina Sutin, a psychologist at Florida State University, told me. It may also be a function of the information at our disposal. In charged situations, Rubin said, people tend to zero in on the object of their anger or fear. Take the bank-robbery scenario: The police “want the teller to describe the person who’s robbing them, and instead he describes in great detail the barrel of the gun pointed at his head.” He can’t remember much beyond that. And so, lacking the information necessary to situate himself in his original perspective, he floats.

This distancing effect has some fairly mind-bending potential applications, none more so, perhaps, than to the problem of near-death experiences. For many years, philosophers and psychologists have documented instances of people reporting that, in moments of trauma, they felt as though they were floating outside—usually above—their body. Rubin points out, however, that such reports are not in-the-moment descriptions but after-the-fact accounts. So he has a controversial idea: What in retrospect seems like an out-of-body experience may in fact be only the trauma-induced translation of a first-person memory into a third-person memory, one so compelling that it deceives you into thinking the experience itself occurred in the third person. The recaller, in this theory, is like a person peering through a convex window, mistaking a distortion of the glass for a distortion of the world.

Traumatic dissociations are dramatic but by no means isolated cases of what Rubin calls the “constructive nature of the world.” In a 2019 review article on memory perspective, St. Jacques noted that shifting your vantage and fabricating an entirely new scene rely on the same mental processes occurring in the same regions of the brain. So similar are recollecting the past and projecting into the future that some psychologists lump them into a single category: “mental time travel.” Both are acts of construction. The distinction between memory and imagination blurs.

At some level, people generally understand this, but rarely do we get so incontrovertible an example as with third-person memories. If you and a friend try to recall the decor at the restaurant where you got dinner last month, you might find that you disagree on certain points. You think the wallpaper was green, your friend thinks blue, one of you is wrong, and you’re both sure you’re right. With third-person memories, though, you know the memory is distorted, because you could not possibly have been looking at yourself at the time. If, without even realizing it, you can change something so central as the perspective from which you view a memory, how confident can you really be in any of the memory’s details?

In this way, third-person memories are sort of terrifying. But shifts in perspective are more than mere deficiencies of memory. In her lab at Ohio State University, the psychologist Lisa Libby is investigating the relationship between memory perspective and identity—that is, the way shifts in our memory play a role in how we make sense of who we are. In one experiment, Libby asked a group of female undergraduates whether they were interested in STEM. The students then participated in a science activity, some in a version designed to be engaging, others in a version designed to be boring. Afterward, when she surveyed the undergrads about how they’d found the exercise, she instructed some to recall it from a first-person perspective and others from a third-person perspective. The first-person group’s answers corresponded to how interesting the task really was; the third-person group’s corresponded to whether they’d said they liked STEM in the initial survey.

Libby’s takeaway: Each type of memory seems to have its own function. “One way to think about the two perspectives is that they help you represent … two different components of who you are as a person,” Libby told me. Remembering an event from a first-person perspective puts you in an experiential frame of mind. It helps you recall how you felt in the moment. Remembering an event from a third-person perspective puts you in a more narrative frame of mind. It helps you contextualize your experience by bringing it in line with your prior beliefs and fitting it into a coherent story. Memory is the—or at least a—raw material of identity; perspective is a tool we use to mold it.

Maybe the most interesting thing about all of this is what it suggests about the human proclivity for narrative. When we shift our memories from one perspective to another, we are, often without even realizing it, shaping and reshaping our experience into a story, rendering chaos into coherence. The narrative impulse, it seems, runs even deeper than we generally acknowledge. It is not merely a quirk of culture or a chance outgrowth of modern life. It’s a fact of psychology, hardwired into the human mind.