That most of the patients eventually gained control of their ipsilateral arm was evident. It was true for the original series of patients in California and also for the East Coast patients tested later. It took several of us a great deal of research to figure out how. The problem has fascinated everybody. Myers and Sperry studied it in the cat and Trevarthen studied it in the monkey. Everyone took different pieces of the puzzle. I went after a very simple question: How could a monkey, with one hemisphere viewing the world, pick up a grape with the ipsilateral hand? The answer would be revealing.
Again, the overall puzzle was, Why, when it came to carrying out goal-directed behavior, did animals with their brains divided (and sometimes far more extensively disconnected than was ever done in humans) always seem like they were behaving in an integrated way? How, for example, was the left hemisphere of a deeply split rhesus monkey—a hemisphere disconnection that extended deep into the brain, down even into the pons*—able to control its ipsilateral left hand? Part of our problem was that we had made an assumption. Our assumption, to which we were all so committed, was that voluntary motor control originated from a central command center that had to directly connect to particular peripheral muscles. Because of that assumption, what we observed at first made no sense to us. We had been duped by our own false idea, and as we shed the assumptions, the workings of the brain began to look very different from what we had assumed. The notion that there is an “I” or a command central in the brain was an illusion.
I know this is hard to swallow, which is the very reason it took us so long to figure out. There is no one point where the director lives. No homunculus calling the shots. Dozens of studies finally revealed the reality: The animals were engaged in self-cueing.25 There was no command central. One hemisphere was reading the cues set up by the other to produce an integrated, effective behavioral outcome. Suddenly our whole view of how the brain might coordinate its parts was undergoing what amounted to a paradigm shift.
In order to uncover this strategy, we took high-speed films of split-brain monkeys, each with one eye closed, reaching for objects such as grapes. In these animals the optic chiasm was also divided, which means that information presented to one eye went only to the ipsilateral hemisphere. So, if we occluded the right eye (and I did this by various means including a specially designed contact lens), only the left hemisphere could see. Then we filmed how well the two hands retrieved grapes presented to the animal at the end of a wand. With visual information now restricted to the left hemisphere, the right hand, which it controlled, was quick and deliberate in retrieving the much-desired grapes. As the hand moved to grasp the grape, the posture of the hand properly formed in anticipation of retrieving the morsel of food (Figure 12).
FIGURE 12. Schematic reconstruction of the slow-motion movie used to examine split-brain function in monkeys. The film helped us determine how a split-brain monkey could control the arm and hand on the same side of the body as the brain hemisphere that enabled it to see the object it wanted to retrieve.
(Courtesy of the author)
When the animal, his vision still limited to the left hemisphere, tried to use the left hand, however, a different strategy was evident. Cueing was active at many levels. First, the monkey would orient its entire body toward the object. The left, seeing hemisphere had control over gross body posture and orientation. It could easily position the entire body in the correct orientation toward the desired point in space occupied by the grapes. As a consequence, through proprioceptive feedback mechanisms, which provide motion and position feedback from tendons and joints, the right hemisphere now knew, in a general sort of way, where the object was located. Then the left arm would reach out in the general direction of the object. The left hemisphere could initiate left arm movements by signaling, using a body movement, to the right hemisphere to “go.” As a result, the right hemisphere commanded the left hand to start off in the appropriate direction, which it now knew because of the proprioceptive feedback from the arm. In short, the right hemisphere knew roughly where the object was. Here was the fascinating part: The left hand remained limp rather than getting ready to grasp the object, because its controlling hemisphere had no idea what the object was. The right hemisphere couldn’t actually see it, and the left could not control the distal digits of the left hand. As a result, the hand always looked ill-posed for actually grasping the grape—until the grand finale: Eventually the hand bumped into the grape! At that moment, the somatosensory and motor systems of the right hemisphere were cued and clicked in. The left hand snapped to, formed the correct posture, and grasped the grape. It’s a lot like when we stick our hand into a dark drawer to pull something out; as soon as we feel it, we know how to grab it.
So, understanding why we did not see the right hemispheric specialization for block design in all the patients is not only comprehensible, but to be expected. W.J. had additional brain damage other than the callosal brain damage that prevented the ipsilateral sensory/motor system with its feedback mechanisms from working well. The other patients were more intact neurologically, and they quickly learned how to work this system. Years later, when I began to study the East Coast series of cases, we saw how immediately after surgery there was poor ipsilateral control of the arm and hand, which later improved after the hemispheres had learned how to cue each other.
It is impossible to keep at a single project 24/7, whereas it is fairly easy to be highly involved with life 24/7. The human split-brain work kept me busy, but not all the time. I had other research projects going, partly because it was possible that the human work would soon wrap up; then what?
While W.J. was the inaugural case for launching the surgically sectioned split-brain era, it later turned out that he was not the most interesting case. After a while we realized that he had quite limited functioning from his right hemisphere. The original studies on W.J., however, did answer basic questions about disconnection effects. We showed that visual information presented to one hemisphere didn’t transfer to the other. We explained the more complicated picture of the somatosensory system, and we worked out the capacity and limits of a brain’s ability to control the arms and hands. Finally, we were able to show how one hemisphere was superior to the other at carrying out three-dimensional reconstructions, the block design test. And as I say, we did that all rather quickly.26
While that work was getting done, we also became interested in exploring the mental capacity of the nonspeaking right hemisphere and finding out what else it could do. Did it have any language at all? Could it solve problems? Could it learn simple games? For more than two years we pushed and pushed on these issues.
It turned out that W.J.’s right hemisphere had only rudimentary cognitive abilities. While his right hemisphere could do what is called a simple match-to-sample test, its success on more complicated tasks was only as frequent as would occur by chance. So when we flashed a picture of a triangle to the right hemisphere, it could point to a picture of a triangle from a set of many possibilities. If we flashed a word of an object like “apple” or anything else, however, the right hemisphere was at sea. The simple capacity to match objects showed that it was functioning independently, but it wasn’t functioning at a high level that involved any kind of language. The linguistic potential of the right hemisphere only became evident as more cases came along.
So there we were. We knew that the cognitive capacities of W.J.’s right hemisphere were limited. Before the next cases, N.G. and L.B., came along, everyone got involved with other projects. We were at a point in the overall program where we all remained excited about our discoveries but were asking ourselves how long it would go on. At that time, we had no idea how rich split-brain research would become when more cases were added to the research pool. The new cases would prove to be game changers in an exciting way, but all of that was to come later. From Sperry’s perspective, while we had an exciting finding, he wasn’t about to relinquish his long-standing commitment to the field of developmental neurobiology.
As Bogen pointed out, it was no wonder he was a bit circumspect in the beginning.
Still, what do you do in your spare time? Well, in those days you easily did more research. Money wasn’t an issue. Time was there if you made it. And certainly the questions were there—being generated in large part by the successful human work. After studying W.J., one of the first questions I had was, Were memories for events in a monkey’s life laid down doubly, one in each hemisphere? Language-based memories in W.J. were clearly present in only the left brain. How would a monkey that had learned various visual discriminations while normal respond to querying each hemisphere for that knowledge after it had undergone split-brain surgery?
And how would I run those experiments? While Sperry’s laboratory was a rich community of personalities, it was also a very cooperative community. To get me going, Trevarthen and Sperry did the surgeries. Lois Bird, Sperry’s lab technician, helped me learn how to train monkeys. Reggie in the shop taught me how to build testing devices, even though he did most of the work. And my best friend at Caltech, Charles Hamilton, an advanced graduate student, taught me everything else. So they launched me on a stint of primate research of one kind or another for more than fifteen years, all while I continued the human work.
Working with animals, especially monkeys, is a demanding emotional experience. While monkeys can be aggressive and nasty, they usually aren’t. We cared deeply for the animals and following a surgery would stay with them until they recovered. We did all of this naturally and without instruction. But in many labs cruelty was built into the demands of the research. There has been a deep cultural shift since then in how animals are thought of, reflected now in lab practices. Today, animal research is run by professionals and is more carefully monitored than the monitoring that goes on in some human clinics.
The results of the experiment asking whether the same memory was stored in both hemispheres were clean and clear. Monkeys that were taught a visual problem appeared to keep only one copy of the memory. Cats, however, appeared to keep two copies, one in each hemisphere.27 Our testing with W.J. showed that there wasn’t any clear indication that the right hemisphere had copies of the memories that were associated with hearing or reading language, memories which the left hemisphere clearly possessed. Yet we knew that both hemispheres could recognize an object, even if naming it was beyond the right hemisphere’s abilities. At one level, it appeared that the monkeys and humans were similar, but on another level, humans and cats were similar. Of course, the problem got more complicated as more experiments rolled in.
The puzzle of where memories are stored is alive and well after all these years. In that original study with monkeys, we also showed that the hemisphere that possessed the memories varied. Sometimes it was the left hemisphere and sometimes it was the right. This suggested that there was not an underlying specialization in the brain responsible for the unilateral memory traces seen in humans. Some patients seemed to have bilateral language capacity, while many more did not. Over the years, and as the human split-brain cases added up, the amount of human variation was also notable. How this works is still a mystery.
THE NOBEL PRIZE
It is no secret that Roger Sperry and I had some difficulties later in my career. As the years rolled by, I chose to continue the split-brain approach in my research and published accordingly. Scientists frequently switch their approach and what they study and it aggravated him that I didn’t. We had exchanged some unsettling correspondence both in the early 1970s and mid-1980s. While sharing credit was not his strong suit, it also should be no secret that I never had anything but the highest regard for him. When he was awarded the Nobel Prize in 1981 for split-brain work, it was well deserved. Science magazine asked me to write an appreciation for him, which I did with enthusiasm (see Appendix I). I much prefer to have this statement be the public record of my thoughts about him than the sometimes rather ghoulish and misleading letters I received indicate.
CHAPTER 3
SEARCHING FOR THE BRAIN’S MORSE CODE
For every minute you are angry you lose sixty seconds of happiness.
—RALPH WALDO EMERSON
MY VIEWS CHANGED OVER TIME. The first two reports by Joe Bogen, Roger Sperry, and myself were mainly about W.J. Single cases are always interesting but hardly definitive. As more cases came under study, it was clear that while what we’d learned from W.J. set the stage for split-brain study in humans, it would not define what could be learned. The simple, straightforward picture at the start was energizing but far from complete. In fact, fifty years later we are still far from completely understanding the full neural and psychological consequences of severing the major neural cable in the brain connecting the two half brains. While callosal surgery for epilepsy has always been done infrequently, it has become even less frequent with the advent of other surgical strategies and better pharmacologic interventions.
Joe used to say all the time that science advances first by a major finding being unearthed, then years of other stuff being piled on top of it, smothering it with details that are distracting. From that perspective, the newer cases were complicating. For example, as I mentioned in the last chapter, they quickly revealed that one hemisphere could control both arms but not both hands. We then went on to find that because each hemisphere could control both arms, a touch to either side of the torso could be located and pointed to by either hand. This finding made it seem that sensory information from both sides of the torso projected equally to both hemispheres, which we would then find was not the case. We also observed that the right hemispheres could sometimes seem extremely smart and sometimes even uniquely smart, capable of some nonverbal skills. In short, a much more interactive and dynamic mental system seemed present, even though the interactions were being controlled by two utterly private and disconnected cognitive systems. We slowly began to realize that it was not going to be easy to understand whether we were examining the separate psychological processes of one isolated and disconnected half brain or being duped by the other half.
All of this was happening a couple of years into the Caltech testing program. As still more cases were added for study in Los Angeles (and eventually from different surgical centers throughout the country), the issue of the two hemispheres continuously interacting with one another in complex ways became ever more evident. In the mid-1970s the first new patients from the East Coast broadened our basic understanding even more. That development, however, was almost ten years off.
As the mid-1960s approached, I knew at some point I had to leave the nest. It was an unsettling thought, as Caltech remained my scientific heaven. For a long time, I simply ignored that reality and went about my business. My last couple of years in Pasadena were rich, and I was recently reminded how rich when I came across the series of films I had made. Seeing those films, along with a later video of Case D.R., one of our East Coast patients, triggered fond memories of identifying basic mechanisms in both series of patients.
One of those basic mechanisms had to do with emotions. Emotions color our cognitive states almost moment to moment. More primitive subcortical parts of the brain located below the callosum are heavily involved in the management of emotions, and many of those structures have interhemispheric connections. Could it be that emotions experienced by one hemisphere could be detected by or have influence on the opposite hemisphere?
Those questions began to form as we started to see the differences between W.J. and the second Caltech case, N.G. While studying N.G., we began to suspect that each hemisphere wanted to monitor the other. N.G. could control either arm from one hemisphere. We saw an instance in which one hemisphere, by initiating just the slightest head movement, could cue the other hemisphere with a solution to a task we had requested of the first. In a sense, the two hemispheres were cheating like two kids in a classroom. Once we realized what was going on, the findings all made sense. Imagine being tightly yoked to someone, even though each of you remains a completely independent person, just like highly comp
etent tango dancers. Any slight movement of the head by one gives cues to the partner of exactly what to do and when to do it. Of course, it should work that way. In our studies this meant that with practice the patients were going to become better and better at self-cueing.
BRAIN CUEING IS EVERYWHERE
All of this subtle communication between the surgically separated hemispheres was clearly evident in our cognitive testing. We dubbed it “cross-cueing.”1 Modular or separate systems cueing each other in order to generate purposeful and integrated behavioral outcomes seemed to be everywhere. We detected this early on in both the animal and human split-brain work at Caltech and saw it occur time and again when testing our patients over the next fifty years.
In one of the first observations, I was in the process of seeing if patients who spoke only out of the left hemisphere could name simple colored lights shown to both visual fields. In the early days, we were always concerned with whether or not basic visual information could transfer over from the right hemisphere and be described by the left hemisphere, perhaps via intact subcortical pathways.
During one such study, the patient N.G. displayed our newly discovered self-cueing strategy. The test was as follows: If any colored light came on in the right visual field, which projected to the left speaking brain, then there was no hesitancy; it was quickly named correctly. When a light came on in the left field, however, projected to the right brain, matters changed, though it wasn’t obvious right away. If we flashed the color green to the right hemisphere and N.G. spoke the word green, and the color had been “green,” the patient said nothing else, and we got ready for the next trial. At that point in the test, we did not know if the information about the light had somehow transferred over to the left hemisphere, if the left hemisphere was simply guessing, or if the right hemisphere was actually speaking.
Tales from Both Sides of the Brain : A Life in Neuroscience (9780062228819) Page 8