A person experiencing hemispatial neglect would only see the trees on one side of the tracks. China Stringer Network / Reuters

When I was a graduate student 25 years ago, I met a man I’ll call Mr. X, who had recently recovered from a stroke to the right side of his cerebral cortex. Like many people with a neurological disability, he was intensely curious about his condition, and on his own initiative he visited the neuroscience lab where I worked.

He was a friendly, engaged, intelligent man of about 60. He wasn’t allowed to drive, but he had no obvious trouble walking, speaking, or interacting with people. I couldn’t see any lingering symptoms in a casual conversation. He shook hands firmly. He could take care of himself.

With his permission, us eager graduate students put him through a standard clinical assessment. This type of assessment is designed to be fast and low-tech for easy use at a hospital bedside. We took a large sheet of paper, filled it with scattered horizontal line segments about an inch long, and put it on a table in front of him. Then we gave him a marker and asked him to put a vertical mark through every horizontal line on the page.

I remember he laughed and said, “This again!” before confidently crossing out every horizontal line on the right-hand side of the page, but none on the left. It looked almost like a joke, but it wasn’t.

We pressed him to try again—to scan the page and make sure he got every target. After staring intently, he said, “Oops,” and crossed out another two lines near the middle of the page. He still had missed every line on the left.

Then, as he watched, we rotated the page until the left was on the right. He gave a start, and laughed in astonishment. We had done a magic trick, as far as he was concerned. Suddenly he was aware of all the lines he had missed. They were on his good side now. He had no idea where they had come from, but he could finally cross them out.

We were pretty sure he didn’t have a problem with his eyes. If someone is blind in a part of the visual field, they know to move their eyes and head to scan the whole world. If someone presses you to point to every object in the room, you would remember to turn around and point to the objects behind your head even though they were originally out of your vision. But Mr. X was different. He suffered from a syndrome called hemispatial neglect. No matter how hard he was pressed to try, he had no concept of the left side of space.

I wish we had tried a few more experiments on Mr. X. The more you dig into the neglect syndrome, the more you see the seams in the mind. In many patients, if you give them the same kind of task such as cancelling lines on a page, but the lines are drawn on a wall a couple of yards away and the patient is given a laser pointer, the neglect goes away. The patient can process both sides of the wall perfectly, just not on a piece of paper in front of them.

Even more bizarre, if you give the same person a long stick instead of a laser pointer, the neglect comes back. They touch the distant targets on the wall with the stick, but they ignore the targets on the left side of the wall. Laser pointer, no problem. Stick, the left side of space disappears.

Results like this show that the brain breaks space into at least two parts: the space of things you can reach with your hands or a hand-held tool, and the space too far away to reach. In some patients, the reaching and tool-using space is damaged and the neglect is limited to it. In other patients, only the distant, unreachable space is affected. Some patients suffer both. Our Mr. X probably had a neglect limited to the near space, since he was able to navigate quite well walking around.

One of the most beautiful and revealing experiments on neglect showed that it is not just about the world around you, whether near or distant, but also about the world inside your imagination. The study involved patients whose far space was affected. If you ask one of those patients to close his eyes and imagine standing at the north end of a familiar city square, he can describe the square from memory. But he’ll list only the buildings on his right side, without noticing that he’s given an incomplete account. If you ask him to imagine standing on the South end of the same square, now he’ll suddenly remember the opposite set of buildings, on the right side of his new perspective, while forgetting the first set of buildings now on his left. This experiment doesn’t test lines on a page or on a wall, or any part of the world actually in front of the patient. The whole test lies in the patient’s memory and imagination. His brain simply can no longer handle the left side of space. Of course patients in these situations often notice that they’ve given contradictory answers. It distresses them, but they can’t understand what has gone wrong. The very concept of the left side of space has been erased.

Everything I’ve described so far about neglect involves failing to notice or react to objects on the neglected side. Those objects are erased from consciousness. But they are not erased all together. Neglect is much more twisted than a simple mental blank. It turns out the patient does process information from the neglected side, but doesn’t know it. That hidden processing was revealed in a particularly poignant way in one study. Pictures are projected on a screen in front of a neglect patient. One picture shows a house. The neglect patient can see it’s a house and answer questions about it, although he is likely to miss the windows or the bushes on the left side. If asked to draw the house, he’ll probably draw only half of it. He says it’s a very nice house, and he’d be happy to live in it.

The next picture shows the same house, but this time with flames shooting out of the windows on the left side. Again, the patient knows it’s a house. He can describe it, and as expected, he neglects to mention the flames. He’s unaware that this house is on fire. But if you ask him about his emotional reaction, he’ll say that he doesn’t like the house. There’s something very wrong with it and he wouldn’t want to live there.

The visual information is getting in. Brain imaging studies show that the visual cortex is active to the highest levels. He can process stuff on the left side. He just can’t attach the property of consciousness to it.

Hemispatial neglect is the purest known disruption of consciousness caused by brain damage. You can process information on the bad side and react to it in subtle and even very complex ways, but you’re not conscious of it. One of the lessons from neglect is that, despite the common intuition, consciousness is not a single, unified entity. It breaks apart along spatial lines—left versus right and near versus far.

Damage to a lot of brain areas can cause neglect, but the most severe and long lasting cases are caused by damage to a specific area, one of the most widely connected nodes in the brain, the junction between the temporal lobe and the parietal lobe—more or less just above your ear. This temporoparietal junction, or TPJ, is connected to social networks, cognitive control networks, attention networks, and memory networks. It’s a poorly understood but obviously crucial nexus of information in the brain.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.