On a Monday morning at a home for the elderly in Cologne, Germany, a nurse asked 73-year-old Mr. K. about his weekend. “Oh, my wife and I flew to Hungary, and we had a wonderful time!” he replied. The nurse paused—Mr. K.'s wife had passed away five years ago, and he had not left the home in months. Was he trying to impress her? More likely, Mr. K. was confabulating, a phenomenon in which people describe and even act on false notions they believe to be true.

For confabulators, even physical evidence proving them wrong is not enough to unseat their inaccurate beliefs. Neuropsychologist Morris Moscovitch of the University of Toronto coined the term “honest lying” to describe this condition. Confabulations can consist of wildly untrue statements—claims of being abducted by aliens—but also can consist of memories from long ago, as was the case with Mr. K. They are often autobiographical. Patients easily toggle between rational thought and their false beliefs, unable to differentiate between the two.

Confabulation is a common phenomenon that can stem from numerous dysfunctions in brain mechanisms. Mr. K., for example, suffered from Alzheimer's disease. Another common cause is Korsakoff's syndrome, a form of amnesia sometimes seen in chronic alcoholics. Other triggers include aneurysms or brain trauma that damages regions associated with memory or sensory perception. Yet even healthy people engage in a mild form of confabulation. In an effort to maintain a coherent narrative, we sometimes explain away unusual phenomena without ever becoming conscious of our own fibbing.

In recent years several compelling theories have emerged to explain aspects of confabulation, although a complete picture remains elusive. One overarching theme is that no single brain network or region is responsible for this form of dishonesty. Rather numerous brain dysfunctions can manifest in this one way. In my 20 years of work as a speech therapist at a neurological rehabilitation clinic, I find that many of my patients confabulate. Although currently our understanding is too incomplete to offer hope for treatment, research on confabulation has produced insights into one of the most fundamental questions about the brain: how it is that we construct our personal sense of what is real—and what is not.

Making Memories

Confabulators such as Mr. K. construct false memories out of the fragments of genuine recollections. So to understand how past moments can breed fictions, we need to understand the basics of memory.

Memories of the past serve one core function: to inform and guide future behavior. When we retrieve a memory, the recollection becomes temporarily unstable as its constituent pieces are reassembled into a conscious thought. While it is taking shape, however, the memory is open to tinkering. Indeed, research has shown that memories can be heavily influenced by the present and thus easily distorted.

This seeming flaw is also what permits us to recombine bits of the past to imagine new scenarios. “We have this imperfect system for representing the past in the service of a much stronger system for imagining the future,” says neuropsychologist Asaf Gilboa, who researches confabulation at the University of Haifa in Israel. In a 2010 study, Gilboa and his team summed up confabulation as a confluence of problems related to memory retrieval.

To efficiently call up memories, the mind relies on schemas, which are complex sets of associations among familiar situations, places and information. According to Gilboa, many confabulating patients have difficulty distinguishing between schemas. These mental scaffolds allow us to recall clusters of past events and information rather than becoming lost in irrelevant details. When trying to remember a recent visit to the doctor's office, for example, we can retrieve a schema of the office and then sift through our experiences there to find the correct memory.

In a recent test of this idea, Gilboa and his colleagues examined confabulators with damage to the ventromedial prefrontal cortex (VMPFC), an area just above the eyes associated with decision making and the regulation of emotions. (Clinicians had already noticed that injury to this area can give rise to honest lying.) The researchers asked the confabulators as well as healthy participants to imagine their bedtime routine. They then showed them a series of words and asked whether the words were relevant to bedtime. The participants then repeated the task, this time after imagining a visit to a doctor's office.

The researchers wanted to know whether the confabulators and healthy subjects differed in how well they could distinguish between the two scenarios. So they asked the subjects to rate how relevant a word used in the bedtime condition was to the doctor visit. Confabulators answered correctly only 60 percent of the time, whereas healthy individuals were right 95 percent of the time. The finding, which is not yet published, suggests that the VMPFC plays a role in helping us decide which memories are relevant to a given situation.

Of course, no one brain area works in isolation, and the VMPFC is no different. It communicates closely with parts of the limbic system, a collection of structures deep in the middle of the brain that play a leading role in our emotional life and our ability to form memories. Neurologist Armin Schnider, who has been studying confabulation at the University of Geneva, suggests this brain network serves as a reality-control mechanism that helps people distinguish between reality and fantasy as well as the past and the present.

In one 1999 study that sought to bolster the theory, Schnider and his colleague Radek Ptak compared confabulating patients, individuals with amnesia and healthy people as they viewed a succession of images. Most pictures appeared only once, but some of them turned up a second time. During several rounds of testing, subjects were asked to focus on a target image; for example, in the first round they might be instructed to look for photographs of airplanes, but in the second round they were to focus only on baby carriages. Participants were expected to press a button whenever they recognized the target image.

The amnesia patients and the healthy participants had no trouble with this exercise. Confabulators, however, were unable to distinguish between the rounds. The more the experiment was repeated, the more frequently confabulators identified earlier target images. Brain scans revealed that these participants showed lesions in, among other places, the VMPFC. This area was intact in the amnesic patients, whose brain damage was limited to other areas. Schnider suggests that the capacity for memory—embedded in the limbic system—had to co-evolve with a mechanism for checking whether a memory matches reality. Without such a mechanism, he adds, “Memory would be dangerous.” To further explore this idea, Schnider hopes to examine whether these structures are underdeveloped in an immature brain, which might help explain young children's ability to move easily between reality and fantasy.

Although Gilboa's and Schnider's explanations sound similar, the neural machinery of each is likely to be distinct. In a 2011 study, Schnider found that monitoring a memory for its precise content—Gilboa's theory—produced significantly different brain activity than judging whether a memory relates to the present reality. Ultimately several brain malfunctions might independently trigger confabulation.

Everyday Invention

Indeed, people need not have suffered brain damage to unintentionally invent stories. Healthy individuals also occasionally confabulate when called on to explain a choice. Psychologist Petter Johansson and his colleagues at the University of Lund in Sweden demonstrated this phenomenon in a study published in 2005. The researchers showed 120 male and female subjects photographs of young women and asked them to choose which they found most attractive. The participants were then asked to explain their selections. Unbeknownst to the study participants, the investigators had secretly switched the pictures at this point, so that the subjects were in fact offering justifications for a different picture. Only about a third of participants noticed the switch. The rest dreamed up completely plausible explanations. Johansson calls this phenomenon “choice blindness.”

The mechanisms underlying this phenomenon remain largely mysterious, but work by neuroscientist Michael S. Gazzaniga of the University of California, Santa Barbara, suggests that these kinds of confabulations arise from the language centers in the left hemisphere of the brain, especially Broca's area in the frontal lobe, Wernicke's area in the temporal lobe, and Geschwind's territory in the parietal lobe. Studies from numerous research groups show that these regions produce a constant stream of verbal explanations for our behavior based on the information collected and processed in other parts of the brain. Gazzaniga concludes that there is a “human tendency to generate explanations for events.”

Disruptions in the brain's language circuits can also give rise to pathological confabulation, independent of the brain's memory circuits. This observation dates back to research from 1965 by the pioneering neurologist Norman Geschwind. In the cases he considered, a brain lesion or other abnormality interrupted communication between the left hemisphere's language areas and the right hemisphere's association areas, which integrate stimuli into a coherent model of the environment. In an attempt to weave a consistent narrative, the left hemisphere will fabricate explanations.

American neurobiologist Roger W. Sperry documented this effect in his famous experiments in the 1960s on so-called split-brain patients, who had suffered from intense epileptic seizures. In an attempt to quell these patients' seizures, neurosurgeons had severed the main bridge connecting the brain's right and left hemispheres, called the corpus callosum.

On a special projector, Sperry showed a comical picture to the left eye of test subjects, which is governed by the right hemisphere; the left hemisphere perceived nothing. Several of the test subjects laughed, but they were unable to explain why they found the picture funny. They nonetheless attempted to explain away their laughter—for example, by claiming that they found the projector amusing.

In some cases, such as when patients with disruptions in their language-related circuits also endure damage in the prefrontal cortex (perhaps including the VMPFC, as mentioned earlier), confabulators may produce fairly grotesque explanations. As one of my patients once told me, “I can't move my right arm at the moment because the doctor put it in the refrigerator.”

Bringing all these brain findings together, we can surmise that there are two basic systems in our brain: a creative mechanism in the language areas of the left hemisphere, which produces explanations for our experiences and memories, and a control circuit in the VMPFC and limbic system, which tests the plausibility of these concoctions. This dynamic between creativity and control is an important component of our thought processes. Both mechanisms occur unconsciously and usually remain in equilibrium. In confabulating patients, however, the brain's supervisory circuits have gone haywire. They are no longer able to rein in their fantasies.

Even when these fantasies are completely unrealistic and illusory, patients nonetheless believe them. This phenomenon reminds us how deeply anchored is our need for coherence, causality and stability—even among those who have severe brain damage.