by E. Paul Zehr
Ghosts in the Iron Machine: Phantom Limbs and Phantom Pain Syndromes
Our somatosensory and motor brain maps have been reinforced and structured within us since we were in utero. Our nervous system has been carefully calibrated and tuned to us and our experiences throughout our lives, and it continues to work very well throughout our entire lifetimes. However, there are two dramatically different outcomes when the integrity of the maps can be challenged and plastic changes can occur. One is with decreased use, the other with increased use. And what represented the best and most powerful example of decreased use for the nervous system better than the complete amputation of a limb? A limb amputation is without doubt the most traumatic thing that can happen to the sensory and motor systems. On the sensory side, inputs that have been there since life began are suddenly gone. On the motor side, muscles that used to be easily activated by the nervous system no longer exist. This creates a huge mismatch between the sensory and motor systems as well as a mismatch between expectations and outcomes. It turns out that the nervous system isn’t very good at forgetting about body parts that it used to have and this results in some strange things.
Let’s begin first with the sensory maps we have. Think about the connection between neurons in the somatosensory cortex and receptors on your index finger. If there is a tragic and traumatic accident and a limb had to be amputated, there would no longer be an index finger with receptors to be connected to the nervous system and brain. However, the neurons in the somatosensory cortex still do exist and are expecting and awaiting information from the body. There can be an “expansion” of the maps such that areas that were nearby the now-disconnected brain regions take over and make use of the neurons in the old area. This means that there is remodeling of the cortical maps and is a useful response to accommodate the needs of the nervous system. Your nervous system doesn’t like to have a gap in the map of the relationship between the body and the brain.
One odd—but not uncommon—effect of missing limbs is phantom limb and the related phantom pain syndrome. Sometimes the representation in the brain of the amputated limb does not fade away and instead gets taken over by some other areas and persists despite the limb amputation—enter the phantom limb. People suffering from phantom limb syndromes can have the distinct sensation that the limb exists and is there. They can feel itching and tingling and perceive weight in the limb, the exact opposite feeling of a leg that has fallen asleep. Phantom limb syndrome would be merely a nuisance if it were all there was to losing a limb. Unfortunately, what often goes along with it is a phantom pain syndrome. This means exactly what it sounds like: the person feels painful sensations that seem to be coming from a limb that no longer exists! As you might guess, this is a very troubling thing to experience.
How can you treat pain in a limb that doesn’t physically exist? The work of Vilayanur Ramachandran and colleagues at the University of California at San Diego has taken an interesting approach to this—by tricking the brain into thinking the missing body part does exist. If a split mirror is set up (often a “mirror box” is used), someone sees the other side of their body on the mirror side. Figure 6.1 shows me using a mirror box at the lab of my friend Richard Carson at Queen’s University Belfast in Northern Ireland. Notice that it appears from the reflections that I have two arms, but one is the reflection of the other. My other arm is hidden behind the mirror. This has been used with amputees with phantom limb syndromes, including phantom limb pain, to treat the symptoms. If participants carefully study their movements and do different tasks with the intact limb while looking in the mirror, they will perceive that the amputated limb is actually moving and feeling sensation. The best look at this is in panel C in the figure.
I can tell you that it really did appear to me that I was staring at my right arm even though it was only the reflection of my left. It is a very powerful illusion. (Of course, in a real scenario, I would have been asked to remove my watch from my left wrist. It kind of takes away from the effect.) Using something like this, over time, the sensation of the phantom limb will often be reduced or disappear. This may seem pretty wild, but it appears to be grounded in the fact that of all our senses, our nervous system puts the most weight on vision. If vision informs us that something is happening, we almost always “believe” that over information from other systems. So, in this way the visual system can be used to trick the brain into realizing that there isn’t any phantom limb to feel phantom pain.
Figure 6.1. (opposite) The author using a “mirror box” apparatus shown from three different perspectives. Notice the appearance from the reflections creates an illusion that I basically have two arms but one is the reflection of the other. My other arm is hidden behind the mirror. This can be best appreciated from panel C. Courtesy Richard Carson.
Tony experienced something similar to this, described in “This Year’s Model” (Invincible Iron Man #290, 1993). In this story, Tony Stark underwent a procedure to help “fix” his nervous system degeneration. One of the outcomes of this procedure—which involved a much closer link to his armor—was enhanced sensation. While lying in bed in the recovery room, Tony remarks that “there are additional … side effects. I’m experiencing alterations in perception. Everything is too sharp, too clear. Sights, sounds, textures are overwhelming.” There is also the possibility of creating other weird outcomes associated with certain neurological syndromes such as “synaesthesia,” which is also known as sensory substitution. This means you get a different sensation than you expect coming from some input. An extreme example would be if you could taste words instead of hearing them. In Part 6 of the 2007 Hypervelocity story arc, Tony described experiencing something similar, “a sensory backwash of synaesthesia hits one splinterself—as I taste and touch and smell unlocked datafiles.” The closest story that hints at phantom pain (or probably more of a neuralgia—pain from inflamed nerves) was found in the Ultimate Iron Man 2006 graphic novel written by novelist Orson Scott Card. In this arc Tony is shown as having a skin condition that resulted in almost constant pain. Sadly, as part of the recurring theme of alcohol use and abuse in Tony Stark’s life, he found that alcohol consumption reduced the painful sensations. But at considerable cost—as we will delve deeply into in the next chapter.
Is There Space in Shellhead’s Brain to Store a Skin of Iron?
What would happen if we tried to dramatically increase the representation of the body in the brain without subtracting something? We would need to know this to determine the feasibility of fully integrating a nervous system with an Iron Man suit of armor. We don’t have any Iron Man acolytes to bring into the lab to study and get the definitive answer on this. Clearly we can predict that dramatic changes in how the nervous system works would result from connection to the neural interface with the Iron Man suit. The tantalizing question is whether the cost of this adaptation is possibly the ability to use the body. That is, will the connections for the suit be so strong that they replace the normal representation of the body? I think the answer has to be yes. If you take this far enough, it suggests that Tony’s protracted use of a suit of armor could lead to his inability to control his human body. Based on what we discussed above and how finely tuned our nervous systems are throughout our lives, Tony Stark’s bodily control would be very robust. However, given the extreme nature of the kind of implant needed to connect with the nervous system, a direct brain-machine interface would certainly be the kind of—very unnatural—scenario in which this could be overridden.
Some recent work by Karunesh Ganguly and Jose Carmena at the University of California in Berkeley has addressed a related point. They were interested in getting at what sort of long-term changes in the brain might occur with the use of a prosthetic limb in a brain-computer interface. Recordings were made from neurons in the motor cortex of a monkey while it performed a task that involved reaching toward a target. Each session involved recording from up to a hundred neurons (upper motoneurons) from the motor cortex of each monkey. The researcher
s then watched what kind of activity occurred when the monkeys did a very simple reach.
To get a basic idea of what the monkeys did, imagine lifting your elbow straight out from the side while keeping your hand at the same height as your elbow. (Yes, lots of these kinds of experiments involve arm movements like this. Partly this is because they are easier to control. Partly it is because the way the motor maps of the brain are oriented, it is easier to get at the cells for the upper limb.) Now place your hand out in front of your body and imagine straightening your arm out and flexing your arm to bring your hand back. You could try reaching to different places on a flat space right in front of your chest (a two-dimensional, side-to-side movement). These simple kinds of movement are the ones that the monkeys were trained to do by looking at a cursor that would light up to indicate to them what direction to move to next. They would then repeat these trials each day over a two- to three-hour training period for almost three weeks. During the reaches, the electrical activity of the neurons in the motor cortex was recorded.
From these recordings, it was possible to build a map of activity related to the direction that the arm moved. This makes sense, because the neurons were from areas of the brain controlling muscles of the arm. They should be active during reaching with the arm. In fact, the neurons are specially “tuned” (meaning they fire at higher rates) to certain movement directions, and this information can be detected from the brain and used to determine what the brain is trying to do with the arm. Normally, recordings like this would just give information about the output of the motor cortex. However, if you recorded that information with a specialized computer program, you could use it as a control signal for something else. Like a robot arm, for example. Or an Iron Man suit, eventually. At this stage, though, Ganguly and Carmena used this signal to control a computer display of cursors for where the monkey arm would be if it actually moved. This may start to get a little confusing.
The image at the top of figure 6.2 shows how the monkey moved its arm using the brain-controlled cursor. That means the cursor was moving based only on the commands from the brain—without arm movement itself. Think about this carefully—this means the monkey was controlling the computer cursor to move using a command that would normally come from doing the same movement with its arm. The monkeys were so well trained that they knew what to do to move their arms when a light appeared and generated the same brain activity to move the arm even though the arm itself didn’t move anywhere. This is full brain control that is exactly reminiscent of what would be needed for an Iron Man interface to work directly into the nervous system.
In the top panel of figure 6.2 are data from Ganguly and Carmena’s monkey study. Taking this idea toward the long-term effect of using the Iron Man suit of armor as a brain-machine interface is shown in panels B and C. The motor cortical map we saw in chapter 3 is redrawn here (panel B) so you can clearly see that integrating the exoskeletal suit of Iron Man armor into the brain means putting it on top of a normally “full” map! Panel C shows figuratively where the Iron Man suit might be represented in the brain.
Figure 6.2. Continued use of a neuroprosthetic (like the Iron Man suit of armor) will lead to plastic changes in the cortical maps of the body. Changes in brain activity shift when a monkey learns a reaching task (A). The human motor body map, raising the question where would the Iron Man suit of armor go? (B). How the Iron Man suit of armor would have to be somehow incorporated into the normal body map (C). Panel A courtesy Sedwick (2009); panel B modified from Penfield and Rasmussen (1950); panel C courtesy Patrick J. Lynch.
It is an open question about how good could the brain representation of the neuroprosthetic actually become. Could it become strong enough and stable enough to become a real memory or an “engram” of a new map? This was the main meat of the Ganguly and Carmena study, and the answer seems to be yes. With practice over almost three weeks of training, the monkeys seemed to form stable cortical maps for the use of the prosthetic. It is surprising (to me at least!) that the neural activity seemed to be easily recalled and used, very stable, and very robust. These are the same features that would normally be described when discussing long-term memories. In fact, they use the term “prosthetic motor memory” to describe this outcome. Ganguly and Carmena go on to predict that with “improvements in technology neuroprosthetic devices could be controlled through effortless recall of such a motor memory in a manner that mimics the natural process of skill acquisition and motor control.” These may well turn out to be prophetic words as we move technologically closer to a true Tony Stark / Iron Man neural interface. This could bring us face to face with the idea of embodiment of an entirely new body.
Can Tony Really Become One with Iron Man?
It is worth also touching on another strange outcome of wearing a suit of armor. That is again the idea of aftereffects. What I mean here by “aftereffects” is the continuation of a sensation or a perception after whatever you are doing to cause the change in perception has finished. This is directly related to the issue of the aftereffects of wearing a suit of armor that we discussed in chapter 5. So I mean effects that … continue after. Probably the best example of an aftereffect that you have likely experienced is a playground merry-go-round. When you spin around and around, you activate neural circuits related to balance information from the inner ear that carry on having effects well after you stop spinning. Recall trying to run forward after spinning on a merry-go-round. Although you sure do try to go forward in a straight line, you tend to deviate to one side despite your best efforts. Another example of similar vestibular aftereffects can be detected by being on an oceangoing boat all day with the sea rolling under you. If you do that and then sit down on a stable object like a chair at the end of the day, you often find that you can “feel” the roll of the boat even now that you are on land. This is where the idea of getting your “sea legs” comes from. Also, your nervous system is pretty specific about this. If you were sitting most of the time while at sea, the effects will be largest when you sit down later. What would it “feel like” to be Iron Man when the armor was off?
Let’s go back to that concept of “embodiment” in relation to prosthetic limbs. The idea of embodiment is that the artificial bit—the prosthetic—becomes so fully integrated into the person and her perception of her body that there truly is no line dividing the two. In a clever experiment with an unusual outcome, a team of Swedish scientists headed up by Henrik Ehrsson recently provided an excellent example of this. They were able to create a stunning illusion in a group of upper limb amputees, which had been shown before in people without amputation and is called the “rubber hand illusion.” They created a sensation of embodiment that a rubber hand was actually a real hand attached to the stump where the amputated limb used to be. The experiment was very simple (as are most clever and truly illuminating scientific studies) and basically involved hiding the stump from view while placing a rubber hand in view of the participants (figure 6.3). Next, the experimenters used small paint-brushes to rub simultaneously the index finger of the rubber hand and the stump of the amputee for two minutes. Later, just the rubber hand was rubbed—but sensation was felt in the hand!
Despite the fact that a similar illusion works well in able-bodied persons, the researchers were skeptical that it could work after amputation. They were therefore greatly surprised when amputees identified generally the same illusions. In fact, strong illusions were found in one-third of the participants. Interestingly, the illusions were more powerful the sooner after amputation that the tests were done. The most important point, though, is that this clearly shows both the tremendous plasticity of somatosensory maps and how they can be changed after damage. And, how vision can trump our other senses—like we talked about above for phantom limbs and phantom limb pain. By the way, they also examined anxiety in participants using skin responses and then plunged a syringe (panel C) into the rubber hand (which was not part of the body, remember). The participants had physiological responses of
anxiety that would normally be present if the hand were part of their body. The strict relevance is that prosthetic limb designs that include sensors on the digits could be used to activate skin areas on the intact stump. Over time, this work suggests that the sensation from the artificial sensor on the prosthetic would become integrated into the perceptions of the person such that they are one with the body (enter, embodiment).
Figure 6.3. The “rubber hand” embodiment illusion. The stump of the amputee is stroked at the same time as a rubber hand that is in sight (A). Eventually the amputee perceives that the stroking of the rubber hand (with no activity on the stump) is actually coming from the stump (B). The illusion becomes so strong that in some people plunging a syringe into the rubber hand evokes physiological responses from the amputee (C). Photograph: Christina Ragnö, courtesy Ehrsson et al. (2008).
This raises the interesting idea that early versions (that is, before full nervous system integration) of an Iron Man suit of armor should have sensors on the fingers, hands, toes and whatever other body surface that activate skin areas on Tony Stark’s body. In this way, Tony would really come to embody Iron Man in the way he declared in Iron Man 2, “the suit of Iron Man and I are one.”
