by E. Paul Zehr
Using Nervous System Commands to Control Iron Man
Now let’s return to Tony Stark—someone with a fully intact body and nervous system—wearing a robotic suit to improve and amplify his normal abilities. If you think this through, you will realize that using the Iron Man suit could occur by tracking the nervous system commands and using them to control the suit. Doing this effectively takes the user’s muscles out of the equation. That is, it creates the same disconnect between nervous system and movement that exists after a spinal cord injury or stroke. We just finished talking about using signals in the nervous system to trigger muscle activity and devices like robot suits or artificial limbs. The next step is determining the feasibility of using that initial command signal—the one from the brain or spinal cord—to power motors and computers directly. This means thinking about what Doc Ock from Spiderman or Professor X / Charles Xavier from X-Men can teach Iron Man about connecting machinery to his nervous system. What would it mean for Tony Stark to engineer the Iron Man armor to be able to use this kind of control? Is it even possible, and, if it is, is it dangerous? To answer this we are going to do a little fast forward and then a rewind!
Figure 3.3. The “neuromimetic telepresence unit” that Tony uses to interface with his brain and to remotely control the Iron Man suit of armor (A), from the graphic novel War Machine (2008). Note the circled “neural access port” that is meant to penetrate Tony’s skull. Tony connects to the telepresence unit (and therefore controls the Iron Man suit) from his hospital bed (B) from “This Year’s Model” (Invincible Iron Man #290, 1993). Copyright Marvel Comics.
First, the fast-forward part. What we are focusing on here is the issue of somehow using a direct connection between the nervous system and a robotic device. This kind of connection was shown in Iron Man in its most extreme form back in March and April 1993 in “This Year’s Model” (Iron Man #290) and “Judgement Day” (Iron Man #291). These stories contain elements of the extended story arc captured in the 2008 Iron Man graphic novel War Machine in which Tony Stark had to fake his death. Jim Rhodes has stepped in to become a fill-in “silver” Iron Man (and later became War Machine). Tony then has to use a remote control Iron Man (the NTU-150, but I will call it “robot Iron Man”), which is controlled by a direct connection to his nervous system called a “neuromimetic telepresence unit” (hence the name NTU). This unit basically involves a direct link between activity in Tony Stark’s brain and activity in robot Iron Man. Included in the graphic novel is a detailed description of this telepresence unit.
The image shown in panel A of figure 3.3 comes from that manual. There is a lot of description in the seven-page pseudomanual printed in the novel! However, for our purposes, the piece I want to key on is the description of the actual headset the user must wear. It is of course called a “user interface headset” so the writers are taking a very literal view of how real scientists actually describe things! Anyway, as written in the manual, the headset “provides a direct electronic control channel” for the operator to use to control the robot Iron Man. This headset interfaces with the operator by “the neural port surgically implanted at the base of the operator’s skull just behind the right ear, transmitting commands and information between the Central Nervous System and the neuromimetic operating system.” The image in panel B shows the headset being interfaced (“jacked in”) to Tony’s brain and comes from “This Year’s Model” (Iron Man #290). In both images, I have circled the key neural link panel. Sounds absolutely like comic book fiction, right? Well, partly it is but it also is very much like an emerging phenomenon generally known as a “brain computer interface.” To explore this for Iron Man, let’s look at the real science behind this concept.
Signals from Ol’ Shellhead’s Head
Since this section of the chapter is about detecting some information from the brain that is then relayed to robot Iron Man, we need to also understand how your own nervous system works to produce and regulate movement. That is, where does the signal for movement come from and what does it look like? I am pretty sure that you would agree with me that there are lots of things going on in your brain at any given moment. You probably also recognize that your brain doesn’t exactly work like your computer or even your car engine. So, there isn’t a little port just sitting there ready to have something plugged into it so it can directly relay commands to a computer. Is it even possible to get specific and useful information from all the activity in the brain? Let’s investigate that “brain-computer interface” concept I mentioned a bit earlier. Get ready and hold on because we are about to dig deep into your gray matter.
Your brain contains about 100 billion neurons, and there are about 1 billion more living in your spinal cord. As I write this sentence there are about 7 billion people on earth. So, the number of neurons in your nervous system is about 15 times more than all the people on earth right now. If we think of activity of the neurons in the brain like individual people trying to talk to each other, we can ask ourselves this question: what—if anything—can we extract from a conversation among 101 billion people? Luckily all our neurons speak basically the same “language” and communicate in the realm of electrical signals. And, they don’t all talk at once and aren’t literally all connected to each other. Despite the fact that there are so many neurons with different levels of activity, amazingly we can get something consistent and resembling certain patterns.
Why is it that we can get anything to use as a signal to control things? When we make a purposeful movement, the commands start way up in our brains. Literally at the top, because the part of your brain that helps initiate movement really is at the physical top of your brain. (We will come back to this in more detail later in the chapter.) It is a bit oversimplified to say that areas of the brain are set up completely separate from other areas like isolated little kingdoms. However, different areas of the brain have very specialized functions, and it is usually shown as divided into frontal, parietal, occipital, and temporal lobes (also called “cortices”; figure 3.4). The labels in the figure within each lobe are meant to generally indicate the functions for those brain areas. When speaking about commands for movement, we are in the “motor system.”
A common story in physiology and neuroscience is that many of the discoveries about function of parts of the brain and nervous system have come from observing what happens when things don’t work well or when there are injuries. In other words, much of what we knew before imaging technology came from descriptions of how movement control was disordered after brain or spinal cord injury. The “Edwin Smith Surgical Papyrus” described motor control problems after head injuries in ancient Egypt—over 5,000 years ago. Even though people have known about the connection between brain injuries and motor control for millennia, for quite some time there were many controversies about how the nervous system itself worked. For example, it took a long time to establish that the cells in the nervous system were “excitable tissue.” That is, they convey information using electrical signaling (see chapter 2). This is very important for the issues involved with Inventing Iron Man, since many of the things we are discussing in this book have to do with interfacing electrical devices (like computers) with the basic signaling within the nervous system (which is electrical). However, in classical medicine, Galen (AD 129–199) suggested that nerves were hollow and worked in a kind of pump or pipelike system to convey commands in the body. The substance relaying commands to activate muscle would then flow into the muscles and make them go. This idea was also favored by famous French philosopher René Descartes (1596–1650)—he of “I think therefore I am” fame. However, cutting to the heart of the matter (there is a pun intended as you will read), Alexander Monroe (1697–1762) showed that cutting a nerve did not reveal a gushing or outflowing from the nerve. This would have to have occurred if the older ideas of Galen were correct, so Monroe’s experiment proved this wrong. Monroe thought maybe electricity might be involved instead.
Figure 3.4. The human brain showi
ng different areas of specialization in the cerebral cortex and the cerebellum. Modified from Mysid’s adaptation of the 1918 edition of Gray’s Anatomy.
This idea was met head on—with lots of controversy about “animal electricity”—by two very important Italian physiologists, Luigi Galvani (1737–98, from whose name we get galvanic current and the word “galvanize”) and Alessandro Volta (1745–1825, from whose name we get volts as a measure of electrical amplitude). Galvani showed that a frog leg could twitch even (shortly) after death if the nerves going to the leg muscles were electrically stimulated. The controversy arose because Galvani thought this electrical stimulation used electricity within the frog’s leg (e.g., animal electricity), whereas Volta thought that the frog’s leg was merely a conductor of electricity. So, the combined research of the two men was the first real description of the electrical nature of the nervous system. However (this bit is really important, so pay attention please), when the brains of different animals were stimulated with electricity, not much actually happened. This suggested that maybe the brain didn’t do anything specific and related to the control of movement.
In fact, Charlotte Taylor and Charles Gross have described how, up until the eighteenth century, the outer surface of the brain (known as the cortex) was actually considered to be a useless “rind.” By the way, this is actually what the root word “cortex” means in Latin. Some scientists correctly disagreed. Thomas Willis (1621–75), a professor at Oxford, and Francois Pourfour du Petit (1664–1741), a surgeon in the French army, both thought the cortex had an important role in movement control. In particular, from observing lesions in injured soldiers and from parallel experiments in dogs, du Petit noted that the outer surface of the brain was indeed very important for movement. These observations from hundreds of years ago helped show that the brain and nervous system were electrical in nature and that there were specialized parts of the brain, including those related to movement.
Clear evidence of specific functions in different parts of the brain had to wait until the excellent work of Paul Broca (1824–80). In 1861 he wrote about several patients who had difficulties in speaking. They all had damage to the left frontal lobes. This showed clearly that certain functions (in this case, speech) could be largely controlled and affected by very specific parts of the brain. You can roughly locate this part of your own brain by running your hand over the corners of your forehead as your skull moves back toward your ear. Anyway, it would take a bit more creative work after Broca’s research to convince people that parts of the brain participated in movement control.
It is often said that the human brain is the most complex organ. Measuring activity in such a complex organ is not as simple as you might imagine. Remember, there are 101 billion neurons to listen in on. And they have to communicate together in useful patterns in order to produce all the behaviors we are capable of. Technology has often been a limitation for this kind of measurement and only small numbers of neurons have been recorded. In 2007, MIT neuroscientists Timothy Buschman and Earl Miller conducted a study aimed at looking at attentional focus in monkeys. They recorded from up to five hundred neurons simultaneously in three different brain regions during different tasks of focusing on targets. This represented a huge advance in the ability to record a large numbers of neurons simultaneously.
Creating Commands from the Cortex
An important insight into the role of the cortex in movement control came from the work of John Hughlings Jackson (1835–1911). He was a British neurologist who studied patients with epilepsy. His clinical observations suggested to him that certain parts of the brain must be closely related to specific motor commands. He saw that during a seizure there was a consistent and organized spread of muscle contraction across the body. This made him think that certain parts of the brain should have specific actions in causing movement and that the whole system must be organized in a way reminiscent of the layout of the body. However, he had to wait until the work of Gustav Fritsch (1838–1927) and Eduard Hitzig (1838–1907) for confirmation. Fritsch, while working as a military surgeon had noticed that his efforts to treat a head wound would sometimes (accidentally) cause small contractions on the side of the body opposite to the injury. In 1870 Fritsch and Hitzig used electrical stimulation of the brain to generate detailed maps of the brain of the dog and showed clearly that movements could be created by stimulating certain brain areas. So, at this point it was known that electrical stimulation of certain parts of the brain (but not others) could evoke twitches in muscles of the body and that there was a kind of map of the body muscles represented somehow by the neurons in the brain. These studies also revealed that the control of activity in muscles is generally found on the opposite side of the brain. If you are using your right hand to turn the pages of this book, it is the cells in the motor cortex of the left side of your brain that are sending the commands. Also, if you choose to turn the page with your left hand, it’s the command cells in the motor cortex on the right side.
Canadian neurologist Wilder Penfield and his friend Edwin Boldrey followed up on this work of locating the centers for different functions in the human brain. They did a detailed stimulation exercise and found that they could generate a kind of “map” of the muscles of the body from stimulation of the brain. The basic concept is this: if you give electrical pulses of stimulation to the motor areas of the brain, you can trigger the output cells of the brain to relay commands to the cells in the spinal cord that activate muscle. By moving electrodes over the surface of the brain, movements in different muscles can be observed. Through painstaking effort, it is possible to create a kind of input-output map of the surface of the brain, which is weighted differently depending upon how much area (and therefore numbers of cells) on the brain are devoted to a particular part of the body. Think of how a huge city with 15 highway interchanges compares with a small village with no exits off a highway are represented on a road map.
Penfield and Boldrey’s work was the basis for the “homunculus” (little man) concept that describes the map of the muscles of the body on the surface of the brain (figure 3.5). The surface area of the body on the map is an indication of the number of brain cells controlling those muscles. These cells are found in the “motor execution” part of the brain shown in figure 3.4. We also have similar maps related to the sensory areas of the brain. In that case, the maps are created by recording activity of brain cells when different skin areas are activated. Understanding this is important in grasping whether it might be possible to tap into this system to control computers and robotic devices. To set the stage for that, I think it is probably useful to ensure that we understand how movement commands arise and are relayed.
Figure 3.5. “Map” of the neurons (upper motor neurons) in the brain used for activating muscle. The distorted shapes of the body part represent the relative number of neurons that control muscles in that part of the body. Modified from Penfield and Rasmussen (1950).
Peeling off the Shell
Now let’s press ahead and look at how the signal for the activation of Tony Stark’s muscles actually occurs. To begin, you have activity in the brain (there are a number of places where this occurs, actually, and we will come back to this shortly). We will focus right now on that part of the brain where we find the motor map we were just discussing. Activity from these cells is sent down to the spinal cord in the form of what are known as “action potentials” and then out to the muscles. Recall in chapter 2 we learned about the movement of sodium, calcium, and potassium ions in and out of cells and how this was linked to the electrical energy needed to move muscles. An action potential results when this energy rapidly rises and falls.
What if you decide you want to make a movement? It could be a motion as simple as picking this book up from the bookstore shelf. Or it could be as complex as trying to jump into the air and fly, in true Iron Man fashion. Let’s say it was the former, because we can deal with that directly. Also, I don’t support you actually jumping into the air and
attempting to fly around. As soon as you decided to pick up this book, a command moved through several relays and then finally arrived at the main movement output center of your brain—your motor cortex. This is the home of the neurons that send along the command to activate muscle down into the spinal cord. This part of the motor cortex is right beneath about the topmost part of your skull. You can get a rough idea of where it is by finding the top of your ear and running your fingers up over the topmost middle of the skull and then down the other side. The cells in the motor cortex found under your fingers relay the motor command and are known as the upper motor neurons. They are called this because these motor output neurons are at the physically highest (“upper”) part of the nervous system. After that, the relayed command to move arrives at the spinal cord level (these are the lower motor neurons) for the appropriate muscles.
So that is how you and I and Tony Stark deliberately contract our muscles. But that is just the direct output part. What we also want to understand is how we can detect that activity in the brain related to activating the muscles, but then kind of “short circuit” it so that we can use that brain command to activate a computer or a robot or a motor. Or maybe a computer-controlled, motorized robot (can you say Iron Man?). In so doing, we will answer the question of where the commands to start movement actually come from.
