The Ravenous Brain: How the New Science of Consciousness Explains Our Insatiable Search for Meaning

by Bor, Daniel

  Logothetis sampled some of these later regions in his monkeys performing his binocular rivalry experiment and found some surprising results. A step or two removed from the primary visual cortex is V4, which processes color, and MT, which processes moving images. Both of these areas were far more reflective of what the animal actually perceived, with nearly half the neurons here involved in the specifics of consciousness. But these were trumped by the inferotemporal cortex, which has nearly all its neurons flipping in activity according to whether the monkey is seeing one object or another.

  SCANNING CONSCIOUSNESS AS CANDLES BECOME FACES

  Therefore, the more specialized and advanced the visual processing region, the more refined and compound is the information it represents, and the greater is its role in consciousness. But the trouble with these single neuron electrode studies is that they are only sampling a tiny subset of the brain. So you can’t make any other claims about any other areas—for instance, the lateral prefrontal cortex mentioned in the blindsight imaging study above—as you have no data on them either way. Only brain-scanning, although far less focused than single neuron electrode studies, can look at the whole brain simultaneously and investigate which suite of regions are involved in consciousness.

  In humans, fMRI studies have again shown that the later, more refined visual regions, as opposed to the primary visual cortex, activate when we experience switching visual views. But now two other regions also light up at least as brightly: the lateral prefrontal cortex and the posterior parietal cortex.

  Binocular rivalry is but one of a range of visual experiments where our experience flips back and forth between two competing visual views (see Figure 6, bottom, for examples). For instance, there is the famous image of either a candlestick, or two faces in profile, where we never see both the candlestick and faces, only one or the other. Just as in the binocular rivalry experiments, if you record when people’s experiences switch, in this case to the candlestick or faces, you again see activity in a combination of later, advanced visual regions, and the prefrontal and parietal cortices—two areas that have no devoted interest in vision at all.

  Although little has been said before now of the posterior parietal cortex, it is almost always activated along with the lateral prefrontal cortex, forms a tight network with it, and almost certainly carries out a similar functional role. In fact, these regions coactivate so frequently that some people have assumed they are just one large processing network for advanced, flexible thought. Although there may be subtle functional differences between the posterior parietal cortex and the lateral prefrontal cortex, for the rest of this chapter I’ll assume that they are a unified “prefrontal parietal network.”

  So far, I’ve only looked at fMRI studies where the test subject’s experience switches between one view and another. Many studies have also explored how brain activity changes when we detect a stimulus compared with when it’s invisible. For instance, Stanislas Dehaene and colleagues showed subjects a rapid sequence of jumbled squares with nothing of note in them. In the middle of this sequence, though, there was a word. Sometimes the gap between the word and the next set of jumbled squares was too short for the volunteers to notice the word, and sometimes the gap was just long enough for the volunteers to be aware of the word. In the aware condition, compared with the unconscious one, the standard activation pattern occurred, with advanced sensory regions lighting up, along with the prefrontal parietal network.

  These results are not limited to vision. You see exactly the same combination of activity in advanced sensory regions and the prefrontal parietal network when subjects detect a touch or a sound, or even when they must use a combination of senses.

  It’s always useful in establishing a result to make your attack from as many different roads as possible and show that each one leads to the same destination. So another technique increasingly employed is not to ask about how activity in a particular brain region is linked with a given function, but to examine the physical size of that brain structure and see if that relates to some behavioral measure. Recent studies have done this with consciousness, again finding links with awareness and the prefrontal parietal network. For instance, Ryota Kanai and colleagues used another kind of image that flips between two experiences, this time with moving dots appearing to rotate in one direction or another. Kanai found that those individuals who reported more frequent switches in apparent rotation also had thicker parietal cortices—as if having more of this brain makes you notice more changes in your awareness.

  Another route of attack is to look at neurological patients: If the prefrontal parietal network is generally involved in all forms of awareness, then damage to these regions should cause some detectable drop in consciousness. That’s exactly what researchers are discovering. For instance, Antoine Del Cul and colleagues gave patients with damage to the prefrontal cortex a challenging task where they had to spot a very briefly presented number, which was immediately followed by a collection of letters designed to interfere with the detection of the number. Patients had a reduced experience of these hard-to-spot numbers, even though they weren’t much worse than controls at actually identifying the numbers if forced to guess.

  You can obtain similar results with the parietal cortex. For instance, Jon Simons and colleagues gave a memory test to patients with damage to their posterior parietal cortex in both hemispheres. First the patients heard seventy-two trivia sentences, such as “Al Capone’s business card said he was a used furniture dealer.” They had to guess whether the voice they’d just heard was male or female and whether the person behind the voice actually believed the sentence he or she had read out. This was basically to throw the participants off the scent, because, after all these sentences were read out, there was a surprise memory test, where the volunteers had to say whether they’d just heard a set of sentences (half of which were new). These patients were no worse than controls at guessing whether the sentences were new or old, but were significantly less confident in their judgments—as if their awareness of their own memories had faded, even if the memories themselves hadn’t.

  An intriguing alternative approach is to explore what happens as you slowly sap consciousness from a person—for instance, by varying the levels of general anesthesia. In one such fMRI study, Matt Davis and colleagues played volunteers various sentences in headphones. While the temporal region responsible for the simple and more processed sound components of speech was active regardless of the level of anesthesia, the prefrontal cortex switched off as soon as the volunteers entered sedation.

  Outside of the world of imaging, there are fascinating corroborative clues from evolution and the study of comparative anatomy. When researchers measured how much of the entire cortex was taken up by just the primary visual cortex in different primates, they found that humans had the lowest share compared to our primate cousins. Moreover, the acuity of our vision as a primate is poor. Our other senses (and the size of our primary sensory regions) are nothing to write home about either. Our sense of smell is particularly poor, for instance. But evolution has taught us the vital lesson that it’s not what you’ve got, but what you do with it. We take the relatively feeble raw data that comes through our senses and we analyze it brilliantly, deeply, extracting insights continuously. In stark contrast to our shrinking sensory areas, our prefrontal cortex has greatly expanded compared to that of chimpanzees and other primates, almost certainly so that we can extract so much more understanding in exchange for less input. And with the rich discovery of knowledge, so comes consciousness: Whether or not other animals are conscious, it is beyond doubt that human consciousness is the richest, to go along with our highly expanded analytical prefrontal cortex, and intriguingly, despite our diminished sensory regions.

  So the general picture here is that although our later, more refined sensory regions are involved in the specifics of our experiences, these regions need to be combined with our most general brain areas in the prefrontal parietal network if we’re actually g
oing to be conscious. But at present, while many consciousness researchers have been ingenious at linking different brain regions with awareness, most have stopped there, balking at giving some functional, mechanistic explanation for how these brain areas actually contribute to consciousness. It certainly needn’t be this way, and one simple approach is to look to the burgeoning research literature from those who, outside of consciousness research, have sought directly to understand what the prefrontal parietal network contributes to cognition. As I will show in the next sections, adding these clues to the mix is a powerful way to further elucidate the shape of consciousness.

  PATIENTS AND THE PREFRONTAL PARIETAL NETWORK’S OFFICIAL JOB

  The first case study to link the prefrontal cortex with complex human cognition was famously written up in 1935 by the prominent U.S. and Canadian neurosurgeon Wilder Penfield along with his colleague Joseph Evans. As well as being an important addition to the neuroscience literature, this article carried a moving emotional angle. Wilder Penfield’s only sister, Ruth, had suffered from headaches and seizures for many years that were caused by an underlying brain tumor. In order to try to save her life, Penfield saw no other option than to cut out most of her right frontal lobe, and so he performed the surgery himself. The operation initially was a qualified success and bought her an extra couple of years of relatively normal life, but then the tumor aggressively returned. There was little anyone could do. After her death, Penfield felt that she would have wanted to use her experiences to help mankind—and his eloquent description of the changes she went through after losing her right prefrontal cortex certainly did that, as it steered frontal lobe research into a very productive new direction.

  Although in many ways Penfield’s sister Ruth didn’t change following the surgery, she did experience some subtle problems. Her main complaints were that she felt “a little slow” and she could not “think well enough.” One specific example in the paper illustrated her problems vividly. She was due to cook a dinner for a guest—in fact, her brother Penfield himself. But although before the operation she would have had no trouble organizing and serving a complex set of dishes, now she found herself flummoxed. She was able to get started on one or two dishes, but after a long and frustrating attempt, she admitted defeat in putting the meal together as a whole. She just didn’t know how to organize herself to decide what to chop or heat next.

  This situation of being overwhelmed by too many options is highly reminiscent of how things fall apart when we become self-conscious of a previously overlearned skill, such as a tennis stroke. We quickly run out of space in our working memory with all these novel, non-chunked motor commands to coordinate, and our movements become clumsy. Here, likewise, Ruth’s main problem might well have been a shrunken working memory, so that even those normal situations and well-chunked sequences no longer fit. The result is sporadic chaos—a failure to perform anything taxing, novel, or complex.

  While this report is largely anecdotal, there is good evidence that patients with prefrontal cortex damage do indeed have a working memory deficit. For instance, Cambridge colleagues and I tested a group of patients, each of whom had a lesion in their prefrontal cortex, in one hemisphere on a standard test of working memory. Eight red boxes would appear on the computer monitor in front of the patients, then some of these boxes would blink blue in a sequence. Immediately after this, the volunteers had to touch the boxes in the order that they’d just blinked blue. Compared to a closely matched group with no brain damage, the patients were indeed impaired on this spatial working memory task, especially if they had damage to the lateral part of the prefrontal cortex—the main region that was implicated in consciousness in the previous section.

  STILL CONSCIOUS, BUT ONLY OF THINGS ON THE RIGHT

  I began this book relating how my father temporarily had suffered hemispatial neglect following his stroke. This condition, associated either with damage to the prefrontal or parietal cortex in one hemisphere, has classically been the most explored attentional syndrome, though to my mind it is also the most central consciousness deficit there is—far more relevant to consciousness, for instance, than blindsight. Say the damage is on the right. Then, for that patient, for some of the time, the left half of space simply doesn’t exist (the opposite side is affected because of the crossover wiring of the brain). It’s not a deficit of vision, because if you do a sight test, these patients can see everything fine on both sides, in both eyes. Sometimes, also, if there is something particularly striking on the left, they will perceive it without any problem. But more often than not, these patients will just become distracted and fail to attend to anything in the left half of space. The classic test of this condition is simply to give patients a long horizontal line on a plain piece of paper and to ask them to put a vertical line in the middle of the horizontal one. Normal subjects have no trouble being very accurate at this admittedly very straightforward task. But neglect patients will almost always place a mark obviously closer to the right edge, in fact around three-quarters of the way to the right, just as if the left half of the line didn’t exist.

  Patients may also only shave the right half of their face, eat from only the right half of their plates, and, if asked to draw a clock face, would write all the numbers squashed up on the right half of the circle, or just leave the numbers 7 to 11 out of their drawing entirely. Similarly, if asked to draw a house, they’d merely draw the right half, with lines unfinished, half the windows missing, only half a door present, and a precarious right half of roof looking as if it could collapse at any moment. In every case, they are completely oblivious to the striking flaws in their behavior.

  The condition extends to every aspect of the neglect patient’s world. These patients will ignore sounds or touches on the left. And the disorder can even reach into their imaginations: In a simple yet striking experiment, Edoardo Bisiach and Claudio Luzzatti asked hemispatial-neglect patients to imagine they were in a square that was very familiar to them, the Piazza del Duomo, in Milan. They were first told to imagine that they were facing the cathedral from the opposite side and to describe all the buildings they could see. The patients reported all the buildings in their mind’s eye, which were only those on the right of the square, from that perspective. They were then told to imagine they were on the opposite side of the square, facing outward from the cathedral, and describe all the buildings they could then see. They proceeded to fail to imagine or mention all the buildings they’d just listed a few minutes before, as these were now on their left in their mind’s eye, but did describe all the buildings they’d previously failed to notice! So even in imagination, these patients were completely ignoring the left half of space.

  The fact that you can either describe this condition as one of a profound attentional deficit or as an absence of awareness provides more ammunition for the view that attention is a critical component of awareness, functioning as its directional gatekeeper.

  To reinforce the point that hemispatial neglect is not purely a visual impairment, Margarita Sarri and colleagues carried out an experiment on neglect patients in which she would touch either their right or left index finger while they were in the fMRI scanner. Half the time the patient wasn’t aware of the touch on the left. The region of the patient’s cortex responsible for finger sensations was activated regardless of whether the patient was aware of the touch or not, but if the patient was aware of the touch, then the intact portions of the prefrontal parietal network lit up.

  CONSCIOUSNESS SHRINKING TO A SMALL POINT

  Habitually failing to be aware of half of your world is a profound enough impairment of consciousness, and yet there is an obvious question here—why aren’t these patients more impaired, if these regions are so critical for consciousness? Why, for instance, can they occasionally still be largely aware of the left half of space? The answer lies in how much of the collective prefrontal parietal cortex is damaged in these patients.

  In all the neurological cases I’ve described so far in this
chapter, the damage has only included a small proportion of the total volume of the prefrontal and parietal cortices in both hemispheres. Most of the time the damage was only in one hemisphere, so at most would have only affected half of the volume of this network of regions. It’s also likely that these brain areas, so flexible in function normally, would be more capable than any other region at adapting to the brain damage, with the intact parts of the network merely taking on more of the burden of processing. So what happens if the whole prefrontal and parietal network is damaged? Is such a person still conscious? Well, in humans I know of no case studies that have included such a profound loss of cortex—probably because such extensive damage would lead to death. However, there are rare reports of near complete damage to both prefrontal cortices or both posterior parietal cortices. Are these patients any more impaired than those I’ve described above? Unfortunately, they most certainly are.

  The posterior parietal cortex, although in function highly similar to the prefrontal cortex, is marginally more linked to spatial processes. Patients who have extensive damage to both sides of the top portion of their posterior parietal cortex have a rare condition known as Bálint’s syndrome, which completely knocks out their sense of space. Their experience of the spatial extent of the entire outside world seems to disappear, and patients report that “there is no there, there.” One feature of this syndrome is that patients are unable to perceive more than one object at a time. Even with this single object, they cannot locate it, nor can they tell if it’s moving toward or away from them. If they are looking at a crowded scene of multiple objects, once they recognize one object the others will become invisible. Sometimes, even perception of a single object is impaired. One patient, known as KE, could recognize colors and read words, but couldn’t do both—so if he read a word, he was unable to identify its color. Although this condition is skewed toward the spatial domain, it nevertheless reflects the importance of the posterior parietal cortex for working memory. KE, and patients like him, have their spatial working memory effectively reduced to one—and in some cases not even a single complete object can be held in these patients’ working memory—it will be stripped into its component parts, and only one of those parts will make its way into consciousness.