Engraving of human brain © Anteromite/Shutterstock
On September 27, 1999, Dr. Clark Elliott, a professor of artificial intelligence and cognitive science, was in what seemed like a minor car collision: a skidding Jeep, a brief blackout, the "seeing stars" specific to a blow to the head. But soon after, he found himself in a whole new world, one in which the journey from office to parking lot could take more than an hour. Basic tasks like walking and remembering his children’s names -- or whether he had children at all -- could suddenly turn out of reach, and emotional foundations like a connection to God or loved ones were no longer to be felt. The Ghost in My Brain is the eight years Elliott lived with the supposedly incurable damage from his concussion, relying on his scientific training to cope with and record his experience -- until coming across the neuro-optometric methods that would ultimately save him.
Signature: Let’s talk about the storytelling aspect of your book.
Clark Elliott: I had two purposes in writing the book, one as an exposition of this new brain science because I thought that could do some good and is quite exciting in its own right, but also in writing the narrative because there are many other people living through this very same complex and sometimes quite troubling experience. From a writing perspective it was a challenge to incorporate both goals. From a structural point of view of the science, ordinarily you would present the essential foundations of brain plasticity itself, then some of the larger concepts -- each supported by detailed examples -- and lastly ground these in the details of what one might experience in each of these many areas as the result of a brain injury.
But that of course is not the order in which I experienced this in my life, and I wasn’t introduced to the concepts of neuroplasticity until I had lived with my experiences -- and captured them in my notes -- for eight years. I learned much of the science later. It was tricky to present the details in the order in which they happened along with the order the science demands. Consider, for example, this fabulous moment when the ghost returns: This is in many ways the high point of the story, and would make a great ending. But almost all of the science of my recovery came afterward. I pondered how to handle this for quite a while. It was actually my daughter Nell, who came up with the idea. She said, Dad, why don’t you present it as a detective story? And this worked beautifully.
I also always considered the structure of the book from a musical perspective -- the rhythm and arc of it -- so the narrative ends by going back to that concussion scene from the very beginning in a restatement of what we might consider recurring Rondo theme. There were several reasons for presenting that piece at the end. One important one was to bring us back to the fact that the book is not just a scientific treatise but essentially a shared story. I didn’t want us to lose track of the fact that there are so many people stuck in this same waking nightmare. And I wanted to say that we never know what the future can bring, so keep trying.
SIG: How did your work in AI help you understand what was happening with your brain post-injury?
CE: In the decade and a half before the crash, I had thought a great deal about cognitive aspects of the brain and how we turn sensory input into meaning in our heads, and how that meaning is represented. My original work was in computing emotions, which is a very small community. I was extremely fortunate to begin working with Professor Andrew Ortony, at Northwestern University. My thesis work was this giant program modeling people’s emotions. People said, Nobody’s done that work, these are huge unsolvable problems, why would you go into this? But I was really interested in this subject, so I got used to working in a vacuum, coming up with original ideas of how people described the emotions in their life. I analyzed 600 different stories that people told, then broke them apart according to the emotional fabric of the story. This was input into a large AI program that manipulated symbols that I thought were close to the way people felt about things in their own life. So even though I was so extremely humbled by the nature of my injury, I thought: At least my experience is interesting, and I was quite familiar with observing all these processes. You never know what might be useful, so I took careful notes of all these bizarre episodes that were happening to me.
My AI background gave me some insight into at least the structure of what was happening to me. For example, people have a lot of emotions that require that they have a relationship to someone else. For example if you are in a friendship relationship, then if something bad happens to your friend, you’ll feel sorry for them, but if you’re in an adversarial relationship with them, you might gloat instead. So to feel these emotions of sympathy and gloating, we need to be able to represent, I’m over here and you're over there. Because of the concussion, at times I’d lose the sense of where my body ended and the rest of the world began, and it was very difficult to support that sense of I’m here and that person is over there. Without that relationship you can’t feel sympathy, or gloating, or love in the normal way. I would just go flat that way at times of brain stress, because when I lost the ability to support the concept of relationship, I simultaneously lost the ability to experience those emotions in the usual way as well. But this was not a psychological problem, it was more like spatial geometry problem.
SIG: Difficulty processing speech comes up several times in your book -- with misunderstood words, understanding jokes, talking on the phone, and the overwhelming effect the toddler-talkiness of your youngest daughter has on you. You also talk about the sense of loss of humanness that concussives feel and for you that included losing your connection with God, which you’d always seen in terms of a dialogue. Are these two phenomena related?
CE: I would characterize those as three different problems. The first one is just the processing of sounds and speech, and this was a really big problem for me. Speech carries meaning in it: People says things to paint pictures in our heads so we know what they are thinking. This is actually a very complex process in our brain. Most of what people communicate with is not actually present in the words, but because we’re listening and we share a worldview, we know what they’re talking about. If you say, “I’m going out because the refrigerator is empty,” I know that you’re out of food, you haven’t eaten lunch, etc, even though you haven’t said any of that, because we share a worldview. But I couldn’t form these images fast enough to keep up with the speech. Then I would get a little bit behind and have to save what the speaker was saying and queue that up while also trying to process what they were saying now, and as that happens my brain is getting more fatigued, and then I’m getting farther behind and could attend even less to what the speaker is saying, in this overwhelming downward spiral. The reason it might be such a particular problem for a concussive is that you also can’t filter out the information: you can’t turn off the input so you’re forced to try to process it. It’s the equivalent of having five people talking to you at once, in this painful overwhelming sense that is like just going mad. This was a common and very difficult experience for me leading to social difficulties when I hung up the phone on someone, or walked out of the room in the middle of a conversation with a person who didn’t understand what was happening.
For the second part, the spiritual relationship I had always felt, I don’t think the salient feature was so much that it was a dialogue. It was the fact that I had lost the geometric/spatial ability to support the concept of a relationship between me and others: that is, I’d always felt spiritually connected to something that’s very essentially and persistently not me. That connection would come in the form of a dialogue, yes, but like with the flatness of emotions -- except in this case not temporary -- it was more the relationship part of my experience of God in my life that had gone missing. It didn’t change my faith, but it was certainly a less joyful and a more sterile existence. A tangible loss I felt every minute of the day. Interestingly it gave me a great appreciation for those highly spiritual people who say they’ve never had that visceral connection with God.
Another way to think of this is that the speech processing was real-time sequential processing of an ephemeral input stream, one word after another, but the problem with God was more the loss of a simultaneous connection in all directions at once.
The third, the sense of alienation, which was true for me, and I believe is quite common with other concussives as well, can be quite profound. Losing the sense of relationships and the ability to process speech can be pretty isolating just in themselves. Then add to these hundreds of other small ways in which we are no longer ourselves, and the end result is that you’ve just become this impoverished alien being walking around in what used to be a human body.
For example, one of the small things is the misuse of words. I had a whole file I called "Word Misuse," in which I kept hundreds of examples of which words I used incorrectly, and how it happened. When this happens over and over again, you just are this odd person with nothing quite working correctly. There are so many hundreds of thousands of these tiny breakdowns; there’s this machinery and you can’t inhabit it anymore, you can’t fit in this damaged machine. The machinery that supports me wasn’t there.
When the ghost returned, it was a little scary because up until then, I hadn’t had any hallucinations or inability to see the world around me. We later considered it might have been an odd form of Charles Bonnet syndrome, or a position-movement syndrome that a Swiss study recently reported. Interestingly I’ve had a number of people say quite emphatically, You’re talking about this shamanic traditional interpretation of the soul leaving the body because the trauma is too much for it to stay. I’m just reporting this, I don’t know much about it.
SIG: So it turns out our visual and spatial abilities play a special role in making thought and movement work. As does decision-making. Is there a brain relationship between visualization and decision-making?
CE: There is. I’m not an expert on the underlying neuroscience, but I definitely had my very clear experience that is: Our bodies don’t respond to what we tell them to do, or even our stated desires. They respond to visual/spatial representations that mark lower-level goals.
Let’s say: Put your hand on the desk in front of you and tell it to lift up off the desk; it’s not going to move. This is tricky, but if you say, I want you to move up off the desk, it doesn’t care about your desire either. But if you just relax, and allow the necessary images of movement and the images of the goals form naturally and gracefully, and without thought, your hand will move. Whatever it is that drives the motor system, it needs that visual spatial goal as an essential step, and I couldn’t create those goals because of visual system problems. Even your desires, your body doesn’t care about that. There's a step under the hood, before your body moves, where there’s a spatial sense of the room, and where you’re headed; the wanting gets translated into that feeling, and that feeling is very precisely spatial in nature. Like when you can’t remember nouns so much when you’re older, where before it was instantaneous, you’re perched, waiting for that name to rise out of the ether, but nothing’s coming.
Making decisions was just like that for me: I couldn’t form the spatial-ness of the problem, so I couldn’t make the decision. I didn’t know which choice to reach for, because I couldn’t see the options in a clear way in my mind’s eye. Telling your body in words doesn’t work because your body doesn’t care about words. Wanting something, you’re body doesn’t respond to that either. Both have to be translated into spatial form before it will move. Making decisions is similar: the decision is often represented as some spatial spectrum, when we can feel the decision to be made, our body will respond by choosing.
SIG: In the eight years between the car accident and finding Donalee and Zelinsky, you devise your own systems for navigating this strange world of brain injury. At one point you talk about a two-part system that’s of some practical help: a list of the things you are creating, dreaming, making up inside plus a list of what you’re seeing, recalling, and perceiving outside. How does this system help with, for example remembering the names of your children?
CE: Yes, this was just one trick where I had to manually re-create these mental structures that we usually just take for granted because they come to us so automatically: in this case, a simple list representing objects in the real world. But with concussion, much of that natural mental grace disappears and we have to reinvent even these basic tools of thought just to get through one day at a time. In this case you are talking about an instance where I was laboring away for more than three minutes just trying to remember who my children were.
We all have a natural, graceful way of linking our internal world, which is internally spatial in its own way, with the external spatial world to which you can actually point, say with your finger. When I was trying to make this list of my five children, I had to build everything up, think about things like how many elements do I need, what an element is, what a sequence is, and so forth, and then since I was trying to solve the problem of do I have children, I had to link those internal elements in my head (which were purely mental symbols) one by one to these young bodies walking around on Earth which of course were real. Unless you have this kind of injury, you would never think of having to manually link your internal world to the outer world in this way.
Certainty is also an important concept because with the loss of visual pattern-matching ability a concussive may also lose the critical sense of certainty that drives so many of our low-level process. For example, they may end up with something like OCD, or an inability to make decisions. After a long adventure I finally get home one night, but I don’t recognize my own house. I know that I am home -- after all my key fits and how many houses have two pianos in the living room anyway? But I have no sense of certainty that I am home. After all, if this were really my house, I would feel at home, and this room would mean something to me. It’s a difficult topic, the idea that knowing is not certainty, and that certainly is what our low-level brain processes depend on.
Let’s say I tell you, "You hit someone over the head and the body is in your backyard right now." Even if you thought OK, maybe I did it and blocked it out, after a while you’d figure out that isn’t true, you don’t do things like that, etc., and certainty would come. But with my kind of injury, despite the mental logic still being there, certainty might never arise, and you have to keep checking because there is no resolution, no way to move on.
SIG: One of the most frustrating parts of trying to understand what was happening to you was doctors missing the "important questions of cognitive breakdown." What are some of those questions?
CE: I know for a fact that there is a lot of neurological testing that makes mistakes in several ways, relative to concussion. In particular, a lot of tests for other problems or assessments are used because there has been years of data collected and there is some sense of how to interpret the results relative to norms. But many of these tests simply miss the point of concussion damage which can be quite complex, and which may be a different problem altogether. The neuro-optometric testing was much more appropriate in my case, and I am absolutely convinced should take precedence over much of the testing given to most concussives. That is, neuro-optometric testing should come very early in the process.
I want to make clear that I think doctors are doing a lot of good work, but this is a really complex problem, and I believe that they are getting this part wrong. They use established tests for dementia, personality, and etc, because there’s a good sense of what the results mean, and they have a norm to compare the results against. But this is a little bit like the joke of the guy who has all these people looking for his lost contact lens. Finally someone asks and he says, "No, I lost my lens over there by the fence, but I thought we should look here because the light is so much better." Yes, we know how to interpret the test results but they are not appropriate for concussion diagnosis and treatment. The neuro-optometric testing was much more appropriate in my case, and I am absolutely convinced should take precedence over much of the testing given to most concussives. That is, neuro-optometric testing should come very early in the process.
Here is an example of two things that can go wrong with traditional neuro-cognitive testing. A concussive is first given a Towers of Hanoi logic problem to solve. He struggles with it but still scores in the 99th percentile. He tells the tester, "I’m an Ivy League computer scientist. I eat this kind of problem for breakfast. Now in just five minutes my hands aren’t working right, and I can’t walk. I am experiencing strong nausea. If you gave me this simple problem again, I wouldn’t be able to complete it." The tester says, "That’s not part of the data. I’m not interested in what you have to say." Then he writes "99th percentile, no problem" in his report. Next the concussive is given a simple memory test which he fails completely because he is just completely non-functional after the Towers of Hanoi problem. He tells the tester, "I have no difficulty with memory at all. I can do this all day long with no effort, but the previous test wiped me out." The tester writes "Memory: complete failure" and ignores what the concussive says.
There is no accounting for how rapidly a concussive may lose function, and no accounting for interference between tests. There is no reporting of sometimes profound physical manifestations that can occur, but which in themselves are highly diagnostic.
SIG: Along with bringing greater attention to brain plasticity, your book could be used as a tool for improving dialogue between concussives and doctors. Are there ways you can recommend for using it as such?
CE: This is tricky. There is a good way to answer. There is very, very strong science behind what the best neuro-optometry can do for us. On my website there's a link to 700 scholarly articles. But getting people to read that research and acknowledge it and incorporate so many clinical findings is tricky.
There are two things I would love to see happen. I would love to see emergency rooms physicians making use of this research. I think that dialogue would be very helpful, if emergency room physicians would attend neuro-optometry conferences, and read the published research results. And I would love to have military doctors attend these seminars as well because they’re dealing with so many blast victims. There are so many head injuries in combat, and concussion from a blast's shock waves: Our soldiers may step on an IED and lose a leg but also suffer one of these highly troubling closed head brain injuries, or someone in front of them does and they’re not even physically damaged in a way you can see but they have a brain injury. And there are many cases of vehicles being in an explosion causing closed head brain injuries. There are a lot of these in our military.
I was fortunate to be invited to speak at the Neuro-Optometric Rehabilitation Association international conference a month ago. My experience is that these are very dedicated people. These numbers are guesses, but let’s say there are 1,000 active members and 400 show up any given year, and 200 are interested and qualified to deal with these major brain injuries, and you spread them across the United States...this gives you, what, two or three people with such expertise in the state of Tennessee? Yet there will be another 5,000 head injuries in the U.S. by this time tomorrow, and additionally there are 5-6 million people known to be dealing with long-term complications from head injuries. So this is a small community and it is understandable that their work is not so well known, yet the need for their expertise is staggering. These are brain scientists who just happen to access the brain through the retina.
My problem was neuro-optometric; I’d love to help move the medical community toward thinking of retinal brain processing as being critical to brain health, and also as a direct pathway into reconfiguring brains for the purpose of recovery. There are relatively few people working in these areas compared to the number of victims.
I have a message for some of our very brightest young people: In a recent survey I saw that something like 85 percent of doctors are quite unhappy with their career choice, and with how they are allowed to practice. I’m thinking for a bright young person, this is a study of the brain, and a way to really help people. The neuroscience of optometry is a huge growth industry. I would dearly love to see younger people go into this area. We need them.
SIG: Finally you find cognitive restructuring specialist Donalee Markus and neuro-optometrist Deborah Zelinsky, whose work based on "How the visual/spatial functions in the brain are integrated with the higher-order brain processing that makes us human," brings you back to your life. Can you give us an introduction here of the science of brain plasticity, including the "brain glasses" that finally helped you?
CE: Brain plasticity as a basic concept has actually been around for decades, it’s just that new research is taking us in new directions now. I want to give a strong disclaimer: This is not my area. I’m an AI professor, in many ways I’m just a guy who took notes. Having said that, I can talk about my experience. A very important point is that I lived with this for eight years, and I was at the point that I couldn’t manage anymore. I was within a few weeks of having to quit my job, losing my house, the custodianship of my children, disappearing into oblivion and becoming a ward of the state.
Nothing I did helped. You think if you just push yourself, you can habituate and your brain will get better, but it doesn’t. So after the initial period there was no improvement. Yet, within three weeks of first seeing doctors Markus and Zelinsky the ghost of my old self had returned and I was seventy percent better. If I hadn’t improved any more, I still would have gotten by. This is a marvelous testament to the brain’s plastic nature and being able to reconfigure itself if pointed in the right direction. The brain will find a way to get its job done if you can give it a jump start. For most concussives, this is what happens naturally, the brain figures out workarounds, and their symptoms recede. For others, like me, they won’t be so lucky. But even then, and even after a decade, given the right treatments, the brain is plastic and it can get an awful lot done in a very short time. The idea is that we borrow healthy brain real estate which takes over new jobs.
I believe I’m correct that Zelinsky found part of my retinal processing that was still pretty good. She angled light to slightly different parts of my retina, which were connected to brain pathways that were healthy, they just weren’t used to doing that job. I now had new neurological pathways that had to adjust during these early days of treatment. My hair’s standing up just thinking about it: It’s like watching your body waking up into spring after a long and bitter winter. It was like I had to go through being a baby, with my body dancing down hallways because I couldn’t walk. I remember standing outside my front door and watching my hand randomly wandering around in front of me as I rediscovered the concept of center which allowed me to finally put the key in the lock. It was sometimes intensely frustrating, but it’s like a pleasurable weirdness getting to know your body this way. It’s this delightful sense of finding your self again, and your place in the world around you. My understanding is that I was now seeing the world through healthy tissue, it’s just tissue that wasn’t used to performing these tasks and I was seeing it figure out its job and reconnect me to the world.
There is a scientific detail: Dr. Zelinsky very explicitly blurred my center vision. She was very specific in that she emphasized my peripheral vision to force me to get the context information to begin addressing my executive function difficulties, such as with planning. With my central vision blurred -- think of a dime-sized circle in front of you -- I was forced to get the spatial-ness, the meaning, of the whole visual scene. This allowed me, for example, to shop in the usual way again without being overwhelmed in stores. As a computer scientist, I can tell you that the visual rotations we perform throughout the day to make sense of the visual scenes around us, and especially to bring the two different scenes from our binocular vision into a single whole, and then to give to whole thing meaning is an extremely computationally expensive process. Ordinarily because we have such phenomenal processing power in our brains this is no problem. But when our visual systems are out of whack the computations become overwhelming and it is just exhausting for the brain to try and keep up.
When Zelinsky and Markus reconfigured my brain’s retinal processing I once again had the horsepower to gracefully extract meaning from the world around me. And I could hear again too, because I could process the sounds in the 360 degree space around me into my now healthy internal spatial and symbolic maps. I could give meaning to the sounds I was hearing, including speech. So, you’ll understand when I say: Thank you Doctor Donalee Markus and Doctor Deborah Zelinsky for giving me back my life, because this is exactly what they did.