In 1939, a group of 10 people between the ages of 10 and 43, all with epilepsy, traveled to the University of Rochester Medical Center, where they would become the first people to undergo a radical new surgery.

The patients were there because they all struggled with violent and uncontrollable seizures. The procedure they were about to have was untested on humans, but they were desperate—none of the standard drug therapies for seizures had worked.

Between February and May of 1939, their surgeon William Van Wagenen, Rochester’s chief of neurosurgery, opened up each patient’s skull and cut through the corpus callosum, the part of the brain that connects the left hemisphere to the right and is responsible for the transfer of information between them. It was a dramatic move: By slicing through the bundle of neurons connecting the two hemispheres, Van Wagenen was cutting the left half of the brain away from the right, halting all communication between the two.

In a paper he and a colleague published in the Journal of the American Medical Association in 1940, Van Wagenen explained his reasoning: He had developed the idea for the surgery after observing two epilepsy patients with brain tumors located in the corpus callosum. The patients had experienced frequent convulsive seizures in the early stages of their cancer, when the tumors were still relatively small masses in the brain—but as the tumors grew, they destroyed the corpus callosum, and the seizures eased up.

“In other words, as the corpus callosum was destroyed, generalized convulsive seizures became less frequent,” Van Wagenen wrote in the 1940 paper, noting that “as a rule, consciousness is not lost when the spread of the epileptic wave is not great or when it is limited to one cerebral cortex.” Based on the cases of the cancer patients—and some other clinical observations—Van Wagenen believed that destroying the corpus callosum of his patients would block the spread of the electrical impulses that lead to seizures, so that a seizure that began in the left hemisphere, for example, stayed in the left hemisphere.

The surgery worked for most of the patients: In his paper, Van Wagenen reported that seven of the 10 experienced seizures that were less frequent or less severe.

Between 1941 and 1945, Van Wagenen’s colleague, the University of Rochester psychiatrist A. J. Akelaitis, tested the patients to see if they had experienced any cognitive or behavioral changes as a result of the invasive procedure. After giving the patients a series of assessments—an I.Q. test, a memory test, motor-skills assessments, and interviews—he reported that most of the patients had the same levels of cognitive functioning after the surgery as before, and displayed no behavioral or personality changes. Though the brain hemispheres of split-brain patients had been disconnected, he wrote in a 1944 paper in the Journal of Neurosurgery, they were otherwise normal.

Or so it seemed.

* * *

When Michael Gazzaniga first learned about the Rochester patients as an undergraduate research intern in 1960, he was curious—and skeptical.

Gazzaniga’s timing was fortuitous: Roger Sperry, who headed the neuroscience lab where Gazzaniga worked at the California Institute of Technology, had begun split-brain research on cats and monkeys just a few years earlier. Sperry found that severing the corpus callosum of those animals had affected their behavior and cognitive functioning.

In one experiment with split-brain cats, for example, Sperry would cover one of the animal’s eyes and then teach it to differentiate between a triangle and a square. Once the cats learned to do that, Sperry switched the covering from one eye to the other and tested the them to see if they recalled their new knowledge. They didn’t. “The split-brain cat,” as one neurosurgeon wrote in an overview of Sperry’s work, “has to learn all over again.” As Sperry noted, this suggested that the two hemispheres were not communicating with each other, and that each was learning the task on its own.

If the Rochester patients’ left and right brains were also no longer communicating, Sperry and his colleagues believed, then they must be experiencing some sort of change, too.

The question was still bothering Gazzaniga by the time he returned to Sperry’s lab as a graduate student in 1961: What kind of change was it? Would human brains react the same way as those of the animals in Sperry’s lab?

“In monkeys,” Gazzaniga told me, “sectioning the corpus callosum led to the right hand not knowing what the left hand was doing. I wanted to know if we would see a similar result in humans.”

The researchers didn’t have to wait long to begin looking for the answer. In the summer of 1961, as Gazzaniga was preparing to return to Sperry’s lab as a graduate student, a young neurosurgeon at Caltech named Joseph Bogen approached Sperry about the opportunity to study a split-brain patient—and Sperry, who had been working exclusively with animals, seized the chance to work on his first human case.

The patient Bogen had in mind was a man in his late forties named William Jenkins, a World War II veteran who had been hit in the head with the butt of a German officer’s rifle after parachuting behind enemy lines. Jenkins’ doctors believed that this was the likely origin of the uncontrollable seizures he later developed; when he returned to the U.S. after the war and sought treatment, he discovered that no drugs worked to contain the seizures.

In 1961, as a last-ditch effort, Bogen suggested that he have split-brain surgery. Sperry assigned Gazzaniga to conduct some standard pre-operative neurological tests, and Bogen and a colleague performed the procedure in February of 1962. After a few months of post-surgery monitoring, Bogen found that the severity and frequency of Jenkins’ seizures had abated, but he still did not know if the surgery had produced other unintended consequences. So about a month after the surgery, Bogen sent Jenkins to Sperry and Gazzaniga for cognitive testing. In doing so, he kicked off a line of work that would turn the two men into pioneers of split-brain research, eventually earning Sperry a share of the Nobel Prize in 1981—and causing scientists to reconsider long-held ideas about the brain and the self.

The cognitive tests performed on the 10 original Rochester patients hadn’t tested each brain hemisphere separately; believing that this was one reason why the patients hadn’t shown any changes after surgery, Sperry and Gazzaniga decided to run tests for both the left and right sides of Jenkins’s brain.

In one of the first split-brain studies that the pair designed, published in August 1962 in the Proceedings of the National Academy of Sciences, Gazzaniga invited Jenkins into the lab and had him stare straight ahead at a dot. As he was staring ahead, Gazzaniga flashed a picture of a square on a screen to the right of where his eyes were staring, meaning the image would be processed by Jenkins’ left brain. (Because of the way the brain is wired, if a patient looks straight ahead, something quickly flashed to the left of his gaze will be processed by the right side of the brain, and vice versa. The brain’s hemispheres control activity mainly on the opposite side of the body—the left hemisphere controls the action of the right hand, for example, while the right hemisphere moves the left hand.)

When Gazzaniga asked Jenkins what he saw, Jenkins was able to describe the square. Then Gazzaniga tried the same thing on the other side, flashing the same image to the left of Jenkins’ gaze. When he asked Jenkins again what he saw, though, Jenkins said he saw nothing.

Intrigued, Gazzaniga pulled another image, this time of a circle, to flash on Jenkins’s right and left sides separately, as he had done with the square.

Instead of asking Jenkins to describe the object, though, he asked him to point to it. When the image was on Jenkins’ right side (left brain), he lifted his right hand (controlled by the left brain) to point to it. When the circle flashed on his left side (right brain), he lifted his left hand (controlled by the right brain) to point to it.

The fact that Jenkins was able to point to the circle with both hands told Gazzaniga that each of Jenkins’ hemispheres had processed the sight of the circle. It also meant that in the previous trial, both of Jenkins’s hemispheres had processed the square—even though Jenkins said, when his right brain processed the sight, that he saw nothing. At that point, scientists had known for about a century that language arises from the left hemisphere; given that, the researchers later reasoned, Jenkins could only talk about the square when its picture was flashed to his right eye (left brain). On the other side, even though Jenkins had seen the square, he could not speak about it.

Between 1962 and 1967, Sperry and Gazzaniga worked together to perform dozens of additional experiments with Jenkins and other split-brain patients. In one set of studies conducted in 1962 and 1963, Gazzaniga presented Jenkins with four multicolored blocks. Then, he showed Jenkins a picture of the blocks arranged in a certain order, and asked him to make the same arrangement with the blocks in front of him.

Because the right brain handles visual-motor capacity, Gazzaniga was unsurprised to see that Jenkins’ right hemisphere excelled at this task: Using his left hand, Jenkins was immediately able to arrange the blocks correctly. But when he tried to do the very same task with his right hand, he couldn’t. He failed, badly.

“It couldn’t even get the overall organization of how the blocks should be positioned, in a 2x2 square,” Gazzaniga later wrote of Jenkins’ left hemisphere in his memoir, Tales from Both Sides of the Brain. “It just as often would arrange them in a 3+1 shape.”

But more surprising was this: As the right hand kept trying to get the blocks to match up to the picture, the more capable left hand would creep over to the right hand to intervene, as if it realized how incompetent the right hand was. This occurred so frequently that Gazzaniga eventually asked Jenkins to sit on his left hand so it wouldn’t butt in.

When Gazzaniga let Jenkins use both hands to solve the problem in another trial, he again saw the two brain hemispheres at odds with one another. “One hand tried to undo the accomplishments of the other,” he wrote. “The left hand would make a move to get things correct and the right hand would undo the gain. It looked like two separate mental systems were struggling for their view of the world.”

* * *

The more information the split-brain researchers discovered, the more they wondered: If the two sides of the brain functioned so independently of each other, how do people—ordinary people and split-brain patients alike—experience a single, cohesive reality?

In a 1977 study with a 15-year-old split-brain patient from Vermont identified as P. S., Gazzaniga (then a professor at Dartmouth) and his graduate assistant Joseph LeDoux performed a visual test similar to the one Jenkins had undergone years earlier. The researchers asked P. S. to stare straight ahead at a dot, and then flashed a picture of a chicken foot to the brain’s left hemisphere and a picture of a snowy scene to the brain’s right hemisphere. Directly in front of the patient—so that he could process the sight with both hemispheres—was a series of eight other pictures. When the researchers asked him to point to the ones that went with the images he saw, P. S. pointed to the picture of a chicken head and a picture of a snow shovel.

So far, the results were as expected: Each hemisphere had led P. S. to choose an image that went along with the one that he had seen from that side moments earlier. The surprise came when the researchers asked him why he chose these two totally unrelated images.

Because the left hemisphere, which controls language, had not processed the snowy scene, they believed P. S. wouldn’t be able to verbally articulate why he chose the snow shovel. “The left brain doesn’t know why,” Gazzaniga told me. “That information is in the right hemisphere.” Neither hemisphere knew what the other had seen, and because the two sides of his brain were unable to communicate, P.S. should have been confused when Gazzaniga asked him why he had picked the two images he did.

But as Gazzaniga recalled in his memoir, P. S. didn’t skip a beat: “Oh, that’s simple,” the patient told them. “The chicken claw goes with the chicken, and you need a shovel to clean out the chicken shed.”

Here’s what happened, as the researchers later deduced: Rather leading him to simply say, “I don’t know” to Gazzaniga’s question, P.S.’s left brain concocted an answer as to why he had picked those two images. In a brief instant, the left brain took two unconnected pieces of information it had received from the environment—the two images—and told a story that drew a connection between them.

Gazzaniga went on to replicate the findings of this study many times with various co-authors: When faced with incomplete information, the left brain can fill in the blanks. Based on these findings Gazzaniga developed the theory that the left hemisphere is responsible for our sense of psychological unity—the fact that we are aware of and reflect upon what is happening at any given moment.

“It’s the part of the brain,” Gazzaniga told me, “that takes disparate points of information in and weaves them into a storyline and meaning. That it’s central gravity.”

* * *

In addition to answering questions of brain specialization, split-brain research also examined some of the ways in which the left and right hemispheres are autonomous agents. Jenkins’ left and right hands started fighting over how to arrange the blocks, for example, because the two hemispheres are—as Gazzaniga told me—“two separate minds, all in one head.”

As he further explained in Tales from Both Sides of the Brain: “The notion that there is an ‘I’ or command center in the brain was an illusion.”

Among psychologists, the idea wasn’t exactly new; figures like Sigmund Freud and William James had previously theorized about a “divided self,” with Freud arguing that the mind is divided into the ego, the superego, and the id. But split-brain research was arguably one of the first scientific demonstrations that the divided self has a real, physical basis—a demonstration that, in turn, raised new questions about the relationship between the mind and the brain.

“The demonstration that you could in effect split consciousness by splitting anatomy—by just making a tiny change in anatomy … It was one of the most remarkable results in neuroscience, with huge implications,” said Patricia Churchland, a philosopher at the University of California, San Diego, whose work focuses on the relationship between philosophy and neuroscience. “If you thought that consciousness and mental states were independent of the brain, then this should have been a real wake-up call.”

Helping to illuminate the relationship between the mind and the brain, according to the cognitive psychologist Steven Pinker, is one of split-brain research’s most important contributions to modern psychology and neuroscience. “The fact that each hemisphere supports its own coherent, conscious stream of thought highlights that consciousness is a product of brain activity,” he told me. “The notion that there is a single entity called consciousness, without components or parts, is false.”

* * *

Today’s therapies for seizures are more advanced than those of the mid-20th century, and split-brain surgery is now exceedingly rare—Michael Miller, a neuroscientist at the University of California at Santa Barbara who did graduate work with Gazzaniga, told me the last one he heard of was performed around 10 years ago. Many of the split-brain patients that Gazzaniga, Sperry, and their colleagues studied have passed away.

Though the research on split-brain patients has slowed dramatically, Miller believes that the field still has something left to offer. He’s currently working on a study currently working with a patient to answer the question: Does each hemisphere of the brain reflect on and evaluate itself in a unique way?

“We know that the two hemispheres have different strategies for thinking,” Miller told me, “and we’re curious about how that might change their reflection of themselves. Does the left hemisphere think of itself as a sad person while the right one think of itself as a happy person? We are having each hemisphere evaluate itself to find out.”

Miller’s study uses a test called the “trait-judgment task”: A trait like happy or sad flashes on a screen, and research subjects  indicate whether the trait describes them. Miller has slightly modified this task for his split-brain patients—in his experiments, he flashes the trait on a screen straight in front of the subject’s gaze, so that both the left and right hemispheres process the information. Then, he quickly flashes the words “me” and “not me” to one side of the subject’s gaze—so that they’re processed only by one hemisphere—and the subject is instructed to point at the trait on the screen when Miller flashes the appropriate descriptor. (For example, if the screen reads “happy,” an unhappy left hemisphere would lead a subject to point when Miller flashes “not me” to the right side of the subject’s gaze, and to stay still when he flashes “me.”) If the subject reacts differently on each side—in this example, if the subject points to the screen when “me” is flashed to the right hemisphere—then Miller believes there must be a disconnect between the self-concept contained in each side of the brain.

Miller’s research is ongoing. But, he said, if the study finds that each hemisphere evaluates itself differently from the other, it could add a new layer of understanding to how divided the mind really is.

“Split-brain patients give you a unique glimpse into a state of consciousness you wouldn’t see otherwise,” Miller told me.

“There is something quite unique in interacting with a split-brain patient,” he added. “All the interactions you are engaging in are with left hemisphere, and you can suddenly manipulate things to interact with right hemisphere and it’s a completely different experience. A completely different consciousness.”