Our Senses: An Immersive Experience - Rob DeSalle 2018
The Monkey’s Unculus: Tactile and Balance Sensory Capacity in Animals
“Nothing we use or hear or touch can be expressed in words that equal what we are given by the senses.”
—Hannah Arendt, philosopher
As the fetal brain develops in mammals, an incredible number of brain cells are produced every minute. In humans, that number is 250,000 cells, such that by the time a baby is born, there are more than 100 billion cells in its brain. How these brain cells are arranged and how they enervate the rest of our bodies is an essential part of the story of the senses. It is significant that most of the human brain is dedicated to bringing in and processing sensory information. How these nerve cells of the brain are wired and what they do depends on how the brain develops. Where certain sensory functions are arranged in the brain is a matter of mapping them. And how they are mapped is a fascinating story.
Brain researchers scrape for every research trick they can get. Part of the reason for this scrappiness is that observations of the brain’s structure are best facilitated when the brain is outside the skull, and this requirement all but eliminated studying an active brain in this way in the early days of brain research. One researcher who overcame this limitation took an ingenious approach to understanding the localized function of the brain in patients undergoing brain surgery. Two small regions in the brain, called the sensory and motor strips, were the focus of Dr. Wilder Penfield. Penfield died long before the movie Hannibal was released, so he never had the chance to see it, but he did know, as Dr. Lecter says in the penultimate scene of the film, “You see, the brain itself feels no pain, if that concerns you, Clarice.” And Lecter then demonstrates this cool brain fact by feeding the not-so-lovable Agent Krendler parts of his own brain. Penfield used this knowledge of the lack of sense of pain of the brain to accomplish a tour-de-force study in brain science.
Penfield was a brain surgeon, and before he would perform the actual surgery, when a patient’s skull was opened and the sensory and motor strips were visible and accessible to touch, he would mechanically massage the strips in a sequential manner. Because the patient was every bit as awake as Krendler was during that scene in Hannibal, Penfield was able to ask the patient what happened after the stimulation. So, for instance, if he stimulated a certain part of the sensory strip, the patient might say, “I feel a tingling in my fingers.” Penfield kept painstaking notes on more than 120 patients that he examined in this way (over his career he examined more than 1,000 patients using this approach), and in the end he was able to map both of these neural strips quite exquisitely and accurately. “Map” is a perfectly apt name for what he did, but the way he visualized the products of his brain tickling is somewhat nightmarish. By estimating the amount of neural tissue dedicated to sensing in one case and to controlling motor activity in another, he was able to commission an artist, Mrs. H. P. Cantlie, who in 1937 started to draw very strange-looking beings called homunculi (the plural of the word “homunculus”).
A homunculus for the sensory cortex would look a bit different from a homunculus for the motor strip, because our motor and sensory abilities do not directly mirror each other. We use different amounts of the cortex to control how our fingers move compared with what we use to sense the outer world with them. It turns out, though, that Cantlie’s drawings were often inaccurate. Psychologist Richard Griggs points out that Cantlie’s drawings sometimes indicate that they are based on data for the left hemisphere, but the drawings of the homunculus for these data are for the left side of the homunculus, which is a mistake because functionally the left side of the brain controls the right side of the body. Griggs also notes that even though Cantlie never drew a female homunculus, in the late 1980s a female left breast started to appear in the classic drawings. The appearance of this left breast only partially corrects the bias of Penfield’s homunculus toward male subjects. In fact, only about one-tenth of the patients he examined were specifically noted to be female. In 2012, four female neuroscientists discussed this sex bias and the lack of a hermunculus in the literature. Their point was that males and females have some extreme differences in their sensory makeup that should be appreciated. Even with this sex bias, we can say that humans have a bizarre-looking sensory homunculus. Because the data are derived mainly from males, I will use the male pronoun forms. His hands look huge, his lips and tongue are supersized, and his penis is, well, really gigantic. He has extremely large feet and a torso that looks emaciated. This strange-looking being indicates that a large proportion of the sensory strip in the human brain is dedicated to our digits, lips, and genitalia (a hermunculus’s genitalia would be about as big as her real foot).
Since the late 1940s so many mammal homunculi have been described that they could fill a small petting zoo—platypus, rabbit, shrew, bats, mice, cat, dog, and monkey, for example (fig. 3.1). Ratunculus, platypunculus, and simunculus (rat, platypus, and monkey, respectively) are just a few of the strange names given to the even stranger-looking drawings. Perhaps the most revolting of all of the supersensed-unculi are those of mammals that burrow in the ground or that use the sensory apparatus of their snouts to interpret the world. The star-nosed mole is a bizarre-looking mammal in person, let alone as a molunculus. This burrowing mammal’s snout has twenty-two projections (eleven on each side) that radiate outward and are numbered 1 through 11, with the eleventh being a small ray at the bottom of the apparatus. As this mole moves along in a burrow, the rays are continually reaching out and touching objects it encounters. It is estimated that the rays can make more than a dozen “touches” in a second. Two unique structural aspects of the arrays ensure that this mole is deeply in touch with its surroundings. First, each ray has hundreds of smaller sensory structures called Eimer’s organs. These structures are distributed all around the ray and are enervated through a structure called tactile fovea. (In the case of sight, the fovea of the eye act to sharpen focus in the center of our visual field—see Chapter 8.) Ray number 11 is actually part of the tactile fovea. Whenever rays 1 through 10 encounter something really interesting to the mole, ray 11 kicks into action and explores the novel item with several touches itself.
Figure 3.1. Moleunculus (star-nosed mole), homunculus, and mouseunculus. The patches of cells in the mouseunculus head region are called barrels.
This bizarre animal begs the questions how and why. Of all of the predators on the planet, the star-nosed mole is, inch for inch, pound for pound, the most vicious, voracious, and velocious eater of all. If you think a cheetah is fast to the kill, this little creature can touch, identify, and eat a prey item in 120 milliseconds. There are online video examples. It turns out that, if this mole was even a half-second slower at this predatory behavior, it could not consume enough food to keep up with its rapid metabolism. The acuity and rapidity with which this behavior has evolved are stunning.
And then there is the poor naked mole-rat, which is neither a mole nor a rat. Naked mole-rats are more closely related to porcupines, chinchillas, and guinea pigs than to any other mammals. Recently they were placed into their own family—Heterocephalidae, which means they aren’t even a mole-rat. And if that wasn’t enough, they aren’t really naked, because they have many wiry like tactile “hairs” protruding from their bodies that look like a twelve-year-old boy trying to grow a mustache. But there is a heterocephalidunculus. In 2002, Kenneth Catania and Fiona Remple presented to the world the naked mole-rat homunculus, or heterocephalidunculus. This image was drawn by the modern-day Mrs. Cantlie, Lana Finch, and is a nightmarish picture that demonstrates this species’ way of life. Naked mole-rats have tiny eyes that were once thought to be nonfunctional, but a 2010 study revealed that they can indeed use their tiny eyes to detect light. Their hearing is acute, and they can hear the slightest rumbling of an insect in the tunnels they so adeptly construct.
Naked mole-rats are also social animals with a unique hierarchy among mammals. Each colony has a queen and a couple of reproductive males. The rest of the colony are in the worker social caste, and within it individuals assume a place in the hierarchy. One way to tell the queen from the others, besides producing pups every two months or so, is that she always gets to pass other members of her colony on the top in a tunnel. Nearly one-third of the somatosensory apparatus of naked mole-rate is dedicated to its teeth, which it uses to do almost everything except fornicate (and there is evidence that it uses its huge buckies in social interactions, and so perhaps even its large teeth are involved in sex).
What is most interesting about the naked mole-rat is its brain structure. It appears that the cortex of this species has been completely remodeled such that nearly all of the area of the cortex usually dedicated to vision has been converted to the tactile capacity of the teeth. This remodeling is a beautiful example of the plasticity of the mammalian brain, but little is known about how it works. Fortunately, how remodeling of the brain occurs in the context of the senses can be studied closely by use of the mouseunculus.
The normal mouse homunculus (mouseunculus) has the typical larger head and grossly exaggerated snout of homunculus mammals that rely on tactile mechanisms near their head for interpreting the natural world (see fig. 3.1). Rodents typically use the whiskers on their snouts to sense their immediate outer world. These whiskers are long sensory apparatuses that are continually being whisked back and forth and up and down to touch things. These long whiskers are attached to the snout by means of a set of sensors at the base of the hairs. The information from the sensors is transmitted to the neocortex of the mouse, and the information gathered and interpreted in what are called barrels in the somatosensory cortex. Dennis O’Leary and his colleagues at the Salk Institute asked how the touch sensory real estate in the brain is remodeled after shrinking regions of the brain dedicated to this sense.
They posited two possible outcomes when a brain region responsible for assembling touch sense information is reduced in size. First, the organization of the barrels, and hence the means by which the outside information is processed in the brain, could simply be miniaturized. Second, the barrel size and orientation could be altered to a different organization. The easiest way to approach this question is to breed mice with regions of the brain that are smaller than others, and specifically to breed mice with reduced somatosensory cortexes. There is a mutant in mice called small eye, which does result in a smaller somatosensory cortex. The mutation is caused by a lesion in a protein called PAX6 that is considered a master switch gene in eye development (a master switch gene is one that needs to be present for any downstream development to occur). Another function of PAX6 is that it controls growth, via gene regulation, in the brain. Unfortunately, small eye is also lethal in the embryonic stage, and simply breeding small eye mice wouldn’t work because they would die before they were useful in the experiment. O’Leary and his colleagues figured out a way to localize the PAX6 gene expression with the lesion to the somatosensory cortex, a spectacular feat in and of itself. With these localized mutants, they were able to obtain mouseunculus maps of mice with smaller somatosensory cortexes. From the mouseunculus so constructed, it is obvious that the rewiring of the brain involves not miniaturization but rather drastic remodeling of the existing plan. Add to this basic but elegant experiment an even more fine-tuned analysis, as explained in box 3.1, and one is led to the conclusion that the developmental trajectory of vertebrates hones the overall neural wiring of the brain. Altering the developmental pathways will drastically influence the sensory capacity of vertebrates.
Such studies shed light on how embryo development affects the limits of sensory perception. Development of the brain is an important factor in how our brains receive and process information from the outer world. Developmental studies can tell us where in the brain certain information is processed. But developmental biology can also tell us about how the organs that sense the outside world stimuli are structured.
We can barely see our noses. Try it. With both eyes open, you can vaguely see some of your schnozzola. Close one eye, and it is a bit better. What we can see more clearly is what is at the end of our noses. Horses in general simply cannot see the tips of their noses, or more appropriately, they have a conspicuous blind spot at the end of their snouts. Although we can see, just barely, the tips of our noses, we also have blind spots in our vision. For example, stare at the dot and the X below. Now cover your left eye and get your face very close to the page so that you can see both the dot and the X. Focus on the dot on the left and slowly move your head away from the page. The X on the right should disappear somewhere as you move your head farther from the page. For me, the distance from the page where this happens is about one foot or so from the page. This is the classical definition of a blind spot.
BOX 3.1 | BARRELS, BARRELETTES, AND BARRELOIDS
In the mouseunculus, some barrels are missing and others are reduced in size because of the localized PAX6 mutants (see text). The mouseunculi of these PAX6 mutant mice are very different from that of a normal mouse. The PAX6-deficient mouse also has other parts of its brain altered. In addition, the somatosensory cortex (the part of the brain that processes sensory input from other parts of the body), the hindbrain (the back of the brain), and the thalamus (part of the limbic system) are all also affected. It turns out that both the hindbrain and the thalamus have arrangements of patches of neural tissue similar to the barrels in the somatosensory cortex. These patches are even named similarly: in the hindbrain, as they are called barrelettes, and in the thalamus, they are called barreloids. How the barrels develop influences how barrelettes and barreloids develop and, in turn, how the thalamus and hindbrain develop. This result means that there is a hierarchy of signals occurring at the developmental level that fine-tune the ultimate neuronal structure in the brain. Altering the genes in the hierarchy, or how those genes are expressed, can have a huge impact on the sensory capacities of mammals.
Other vertebrates have similar blind spots, but our tiny blind spot is nothing compared to the blind area behind our heads. This huge blind spot we humans have serves as a reminder that we aren’t anywhere near the best in our visual proclivities compared to other organisms on this planet. As a species we have a well-defined but somewhat narrow capacity at sensing the outside world. For instance, although we are considered a very visual species, relying on vision for much of the information we need to exist, our range of visual perception is quite narrow compared to other organisms that sense light waves. Consider, for example, field of vision, an aspect of the sense that is a strength. Field of vision is defined as the amount of area that one can see by holding the head stationary. It is influenced mostly by the position of the eyes on the head as well as by how much the eyeballs can range within their sockets. Field of vision can be described as covering two directions: left/right fields and up/down fields. A subaspect of field of vision is how much of the total field is binocular, or stereo. Seeing stereo means that you are seeing an area with both eyes, allowing you to perceive the depth of what you are seeing. Humans have a pretty paltry 180 degrees or so of field of vision, and about half of that is in stereo, or binocular, vision.
As the position of the eyes is turned outward on the head, less of the field of vision of one eye overlaps with the other eye. So, in gaining peripheral vision, an organism gives up some binocular vision. Consider, for example, the pigeon, which has an almost 360-degree field of vision, but only a fraction (say, 30 degrees) of that in stereo. Dogs have a pretty good overall field of vision, nearly twice that of a human, but like a pigeon, a dog’s binocular vision is paltry in that it is about half of ours. It should be obvious that almost every variation of field of vision may have been tinkered with during the evolution of animal vision.
But what drives this fascinating characteristic of animal vision? Natural selection obviously drives some of it. Macaque monkeys, for instance, have a very limited field of vision. They can see a little more than 180 degrees. Their binocular vision is a little more than half of this, making for an animal that is not terribly adept at seeing a broad field of vision but very skillful at viewing the world in stereo. Why would a primate need to place a lot of its stock in stereovision? The answer to this question is in where the ancestral primate lived. The ancestral primate was most certainly arboreal, living in trees and adeptly navigating movement in that environment. Try the following, but be sure to have a net or a friend watching you. Close one eye and try to climb a tree (or a ladder). You’ll find it quite difficult to judge where to place your hands as you are deciding on your next move. If you feel yourself falling, please open both eyes and adjust. This little exercise demonstrates that stereovision gives us depth perception and makes it easier for us to interpret the physical structures we encounter. Other animals with acute stereovision are almost always those that need to navigate the three-dimensional world by relying on acute depth perception for hunting prey and avoiding predators. Organisms such as pigeons that have evolved to navigate a scary and deadly world of predators don’t need much stereovision. They simply need to know that something is sneaking up on them or streaking down toward them. So, natural selection plays a huge role in this phenomenon. But it is not the only part of the story.
Let’s consider the octopus (a mollusk), which has a phenomenal 360-degree field of vision and no blind spot. Why is there no blind spot in this animal? The answer comes from understanding the evolution of the eyes in both the human and octopus lineages. Eyes have evolved independently about twenty-five times in the history of animal life on this planet. This means that there have been twenty-five independent instances of light-sensing organs evolving in the more than half a billion years or so that animals arose on Earth. The eyes of vertebrates and of octopuses have evolved different structural quirks.
The light-sensing part of the human eye consists of a collection of cells (in vertebrates called rods and cones) that make up the retina of the eye. The retina is connected to the brain through a nerve bundle that in vertebrates passes in front of the retina. Although the part of the retina where these nerves traverse is minuscule, the nerves nevertheless obscure some of the visual field and hence produce the blind spot that occurs in all vertebrates. But in octopuses the eye evolved so that the nerves leading from the retina to the brain are attached to the backside of the retina and do not obscure light hitting the retina. Octopuses therefore have no blind spot. It would be excessively difficult for mammals with blind spots to evolve to overcome their blind spots. And because of the way that mollusk eyes evolved, they were destined to avoid that blind spot. In essence, the structural dynamics of how the vertebrate eye and the octopus eye develop constrain whether a blind spot will exist. Natural selection probably had little to do with the lack of a blind spot in the octopus, although the octopus may currently exploit not having a blind spot.
Scenarios like octopus’s lack of an eye blind spot are reminders of three important aspects of evolution when considering adaptations in nature. The first concerns what Richard C. Lewontin and Stephen Jay Gould called spandrels, after the architectural masterwork Saint Mark’s Basilica in Venice. After the cathedral was built, artists covered the inside surfaces of the awe-inspiring domes with scenes from the Bible. The artwork fits perfectly on the spandrels. One spandrel features a man pouring water from a large flask into the tapered end of the space at the bottom. It is easy to look at these painted surfaces and think that the spandrels were created explicitly to display art. Not a bad hypothesis, but incorrect, because the spandrels are there to hold up the massive domes of the cathedral, and the artwork, though it “adapts” to the space between spandrels and seems perfect, is an afterthought. So, too, the octopus’s lack of a blind spot, though it probably serves an adaptive response today, is an afterthought, driven instead by the architectural wiring of the eye’s nerve connections.
A second aspect of evolutionary forces us to recall that we have to resist interpreting everything we see in the natural world as an adaptation, when indeed many are not. In fact, sometimes the solutions to one challenge are compromises to solutions for other problems. On this theme of compromise, consider the eyes of arthropods.
Arthropods are a huge group of animals that include insects, and they have eyes that are called compound eyes. Compound eyes are an ancient structure, as evidenced by arthropod trilobite fossils with exquisitely preserved faceted eyes that are hundreds of millions of years old. Antonie van Leeuwenhoek, noted more for his observation of plaque from his teeth and the motility of his sperm through his famous microscope, was the first to describe the amazing complexity of an insect’s compound eye. His small microscope was a handheld device that was backlit by a candle to force light through the sample in the observation chamber. When he mounted an insect eye cornea on the device, he was stunned by what he saw. By moving the candle in back around a little, he was able to arrange it so that it caught the tissue at an angle such that he observed “the inverted images of the burning flame: not one image, but some hundred images. As small as they were, I could see them all moving.” He was seeing the light of the candle through the hundreds of little facets, called ommatidia, that make up the compound eye of an insect. Remarkably, each one of the ommatidia of the compound eye is wired to the insect’s brain. In addition, with more ommatidia, the more lenses there are and the smaller they get. Diffraction of light eventually becomes a problem, causing blurry focus or poor acuity.
The number of ommatidia among insects varies. Insects with tiny eyes have fewer ommatidia—for example, worker ants that rely mostly on smell to communicate have as few as six. Insects that rely heavily on detecting motion to hunt prey, such as dragonflies, have more than twenty-five thousand ommatidia. Compound eyes are very good at sensing motion because they can detect temporal changes at about two hundred images per second (the human limit is thirty per second, when everything starts to blur). The compromise here should be obvious—minuscule amounts of motion can be detected by the compound eye with lots of facets, but acuity is reduced in the process. Apparently, depending on the need for detecting motion versus acuity, insects have evolved specific numbers of ommatidia as a compromise to compensate one for the other.
The third evolutionary quirk involves how organisms develop. In some cases, the way that organisms develop constrains how morphologies are eventually formed in development. Because of the constraints of development, certain morphologies, even though they might be considered optimal, simply will not be possible to evolve. The placement of our eyes on our heads is constrained by the way the eye develops in vertebrates. The developmental control over the positioning of the eyes is more than likely involved in how the width of fields of vision in vertebrate organisms evolved. This very same developmental control is responsible for the constrained way that our eyes are placed on our heads and how the eyes of other vertebrates find their positions on the head during embryonic development.
In the early days of genetics, it was assumed that one gene was correlated to one enzyme. In fact, George Beadle and Edward Tatum were awarded the Nobel Prize in 1958 for this intriguing hypothesis, which probably rings true for simple single-celled organisms like bacteria. However, for more complex organisms, the story is quite different. The modern unraveling of the real nature of how genes control complex phenotypes probably started in Allan Wilson’s lab at the University of California at Berkeley in the 1970s. Wilson and his colleagues recognized that, although humans and chimps are incredibly different morphologically and behaviorally, their proteins are incredibly similar. What this meant to Wilson and his colleagues was that simple changes in the structure of proteins were not responsible for the broad morphological and behavioral difference between organisms. Instead, they hypothesized that changes in gene regulation were more important in producing phenotypic change in evolution than simple point mutations. Consider, for example, eye placement on the face of organisms and hence control over the field of vision in vertebrates.
One of the more important discoveries in biology in the past few decades was the discovery of how gene regulation works to pattern the vertebrate body. And in a way, this phenomenon is also about sensing, since the cells in a developing skull need to recognize where they are, over the developing field of other cells. The sensing is done in pretty much the same way that single-celled organisms sense their outer world, and that is through molecules that can signal the cell and give it a sense of where it is, which in turn is involved in telling the cell what to do. Signaling like this is similar to quorum sensing, only much more complicated, but the general idea of quorum sensing is there. Signaling molecules work by binding to other molecules in the cell. For some signaling systems that require the precise development of structures in a vertebrate body, the amount of signaling molecules present near a cell will dictate what the cell does. This is because signaling molecules work by gradient diffusion. In general, genes in cells have different concentrations of specific signaling molecules that they need to be turned on or regulated to start producing proteins. If there is variation in the concentration that gets a gene pumping out product, then a gradient of the signaling molecule will induce different outcomes at one end of the gradient (say, the low-concentration end of the gradient) as compared to the other end of the gradient (the high-concentration end).
This scenario is basically how the position of the eyes on the head is determined in vertebrate heads. The signaling molecule in question was first discovered in Drosophila melanogaster (the fruit fly) and was subsequently found in the genomes of vertebrates. Genes that produce proteins and interact with this signaling molecule pathway got named after hedgehogs. The embryos produced by mutants in hedgehog lesions result in stubby, hairy little creatures that don’t live past early developmental stages. In an orgy of silly gene naming—and Drosophila biologists are perhaps the worst of all silly gene namers—one of the important signaling molecules was named Sonic Hedgehog (Shh) after the video game and cartoon character. Other hedgehog genes like Indian hedgehog and desert hedgehog and even tiggywinkle hedgehog (see Beatrix Potter) were also coined. But we will concern ourselves with Shh here. An elegant way of explaining this difficult series of events that I use here has been presented previously by Thomas Jessel and is shown in box 3.2 and fig. 3.2.
BOX 3.2 | HOW TO MAKE A CYCLOPS
The signaling molecule Sonic Hedgehog (Shh) creates a gradient in the developing vertebrate embryo that controls a series of genes that in turn determine cell type in the developing brain and skull. The figure shows the gradient as being at the very anterior part of the face. The light grayish strips in the diagrams show where Shh is expressed. In the far-left panel, this signaling protein is turned on full blast in the normal developing embryo. It signals the production of all of the proteins and all are made (white, light gray, dark gray, and black), and the normal eye field knows where to develop above the black protein. The gray protein nearest to Shh needs the most Shh in order to be expressed, and the different-shaded gray, white, and black proteins need intermediate amounts. In the second panel, some of the Shh has been stripped away. When this happens, the light gray protein closest to the Shh protein doesn’t get expressed, as shown in the third panel. As more and more of the Shh gets stripped away, the white, light gray, and black genes don’t have enough Shh to turn on, and so their activity is ablated, as shown in the fourth panel. The fifth panel shows the result when all of the Shh gradient is removed, and the sixth panel shows that the eye field has moved to the very bottom of the developing brain and that both eyes have overlapped, producing a Cyclops-looking creature.
Figure 3.2. Thomas Jessel’s hedgehog gradient explanation for where eyes develop on the head. The different filled dots represent four proteins that need to be made for the eyes to be placed normally on the head. The gray strips near the bottom of the developing brain represent the expression amount of Shh, which controls the production of the four proteins. Where the eyes are pictured represents their final position after development. See box 3.2 for details.
This Cyclops phenomenon actually occurs in nature. Cattle or sheep that eat the false hellebore (plants in the genus Veratrum) ingest large amounts of alkaloids that are sequestered in the plants. It turns out that these alkaloids block production of proteins in the hedgehog pathway, hence producing a situation like the one in the far-right panel of the figure. The Cyclops produced by the knockdown of Shh production in these animals is striking, but it also gives us a way to visualize how the various placement of eyes on the heads of vertebrates might have evolved. Many genes are involved in the layout of the nervous system in the head and eyes and thus influence where the eyes are placed. But tweaking the signals that these genes are interacting with slightly is a perfectly logical and productive way to think about how nature can tinker with field of vision. Human development has settled on a specific field of vision connected sharply to how the nervous system and eyes evolved, and we are stuck with the relatively paltry field of vision we have. So, we are not very good with respect to field of vision compared to other animals, but at least we know why.