The Disordered Mind

by Eric R. Kandel

  THE GLOBAL WORKSPACE

  After designing and conducting a series of experiments that used brain imaging to study visual perception, Baars introduced the theory of the global workspace in 1988.2 According to this theory, consciousness involves the widespread dissemination, or broadcasting, of previously unconscious (preconscious) information throughout the cortex. Baars suggested that the global workspace comprises a system of neural circuits that extends from the brain stem to the thalamus and from there to the cerebral cortex.

  Before Baars, the question of consciousness was taboo among most rigorous experimental psychologists because it was not considered a problem that could be examined scientifically. We now realize, however, that psychology has an enormous variety of techniques at its disposal for examining consciousness in the laboratory. Basically, an experiment can take any stimulus—an image of a face or a word—change the conditions a bit, and make our perception of it come into and go out of consciousness at will. For example, if I show you a photograph of a person’s face followed very quickly by a different image, which masks the face, you will not consciously perceive the face. But if I show you the same photograph for several seconds, you will readily perceive it consciously.

  This was a new, cognitive psychological understanding of consciousness. It synthesized the psychology of conscious perception and the brain science of neural signals being broadcast from the thalamus to the entire cerebral cortex. The two approaches are inseparable. Without a good psychology of the conscious state we can’t make progress in the biology of broadcasting information, and without the biology we will never understand the underlying mechanism of consciousness.

  The French cognitive neuroscientist Stanislas Dehaene extended Baars’s psychological model to a biological model.3 Dehaene found that what we experience as a conscious state is the result of a distributed set of neural circuits that select a piece of information, amplify it, and broadcast it forward to the cortex. Baars’s theory and Dehaene’s findings show that we have two different ways of thinking about things: one is unconscious, involving perception; the other is conscious, involving the broadcasting of the perceived information.

  Dehaene devised a way to image consciousness in the brain by contrasting unconscious and conscious processing.4 He flashes the words “one, two, three, four” on a screen. Even when he flashes them very quickly, you can see them. But when he flashes a shape just before and just after the last word, “four,” the word seems to disappear. It is still there on the screen, it’s still there on your retina, your brain is processing it—but you are not conscious of it.

  Going a little bit further, he then places the words just at the threshold of consciousness, so that half of the time you will say you saw them, and half of the time you will say you didn’t see them. Your perception is purely subjective. The objective reality of the words is exactly the same, whether you think you saw them or not.

  What happens in the brain when we see a word subliminally, below the threshold of consciousness? First, the visual cortex becomes very active. This is unconscious neural activity: the word we’ve seen reaches the early visual processing station of the cerebral cortex. Then after 200 or 300 milliseconds it slowly dies out without reaching the higher centers of the cortex (fig. 11.5). Thirty years ago, if asked whether an unconscious perception reaches the cerebral cortex, neuroscientists would have said no, because they believed that any information reaching the cerebral cortex would automatically enter consciousness. In fact, however, when a perception becomes conscious, something quite different occurs.

  Conscious perception also begins with activity in the visual cortex, but instead of dying out, the activity is amplified. After about 300 milliseconds, it becomes very large: it’s like a tsunami instead of a dying wave. It propagates higher into the brain, up to the prefrontal cortex. From there it goes back to where it started, creating a reverberating neural circuit of activity. This is the broadcasting of information that occurs when we are conscious of that information. It moves information into the global workspace, where it is accessible to other regions of the brain (fig. 11.6).

  Put simply, when you are conscious of a particular word, that word becomes available in the global workspace, a process that takes place separately from your visual recognition of the word. Although the word is flashed in front of your eyes for only a very brief moment, you can keep that word in mind with your working memory. You can then broadcast it to all of the areas that need it.

  Figure 11.5. Subliminal perception: activity in the visual cortex dies out before reaching higher regions of the brain.

  Figure 11.6. Conscious perception: activity in the visual cortex is broadcast to the prefrontal cortex, where it is available to other regions of the brain.

  The basic finding from brain imaging is the same. Conscious activity is restricted in what it can focus on: it selects only a single item at a time and broadcasts it widely across the brain. Unconscious processing of information, in contrast, can take place in many different areas of the cortex simultaneously, but that information is not broadcast to other areas. As you read these words, for example, you are aware of your surroundings—ambient sounds, temperature, and so on. That sensory information about your surroundings is processed unconsciously in the brain, but because the information is not broadcast widely, you are not consciously aware of it as you are reading.

  The experiments previously described demonstrate that information can enter our brain yet not give rise to conscious perception. Intriguingly, however, such information can affect our behavior, as we shall see. That is because the brain’s unconscious processing isn’t limited to sensory information. While the mere recognition of a word is occurring unconsciously, the meaning of that word is being accessed at much higher levels in the brain without our being aware of it. Other aspects of the word can also be computed unconsciously, such as its sound, or its emotional content, or whether we spoke it in error and want to catch the error. Similarly, when we see a number, we effortlessly tap into the mathematical systems of our brain. Scientists are still struggling to understand how unconscious processing works and how deep it can go.

  CORRELATION OR CAUSATION?

  How do we distinguish between something that is preconscious and thus correlated with conscious activity (the neural correlate of consciousness) and something that actually causes conscious activity? How does the brain encode the actual content of consciousness? To make progress on these questions we will need more fine-grained techniques.

  Daniel Salzman, now at Columbia University, and William Newsome of Stanford University have used electrical stimulation to manipulate the information-processing pathways in the brains of animals.5 The animals are trained to indicate whether dots on a screen are moving to the left or to the right. By stimulating just a tiny bit of the brain area that is concerned with visual movement, Salzman and Newsome can induce a slight change in the animals’ perception of which way the dots are moving. This change in perception causes the animals to change their minds about which way the dots are moving. Thus, the dots might actually be moving right, but when Newsome and Salzman stimulate brain cells that care about leftward movement, the animals change their minds and indicate that the dots are moving left.

  In parallel work in 1989, Nikos Logothetis and Jeffrey Schall examined binocular rivalry.6 Binocular rivalry describes the situation in which one image is presented to one eye and a different image is presented to the other eye. Instead of the two images being superimposed, our perception flips from one image to the other: we are only aware of “seeing” one image at a time. The same phenomenon occurs in animals. In their experiments, Logothetis and Schall trained monkeys to “report” these flips. They found that some neurons respond only to the physical image, while others respond to the animal’s perception of it. As we have seen, perception involves cognitive functions, such as memory, not simply responses to sensory stimuli. Logothetis and Schall’s study has spawned additional work, the gist of which is that the n
umber of neurons attuned to percepts, or mental representations of an object, becomes greater as information moves from the primary visual cortex to higher regions of the brain.

  Logothetis concludes from his and related work: “The picture of the brain that starts to emerge from these studies is of a system whose processes create states of consciousness in response not only to sensory inputs but also to internal signals representing expectations based on past experiences.”7 He goes on to state that “our success in identifying neurons that reflect consciousness is a good start” toward uncovering the neural circuits underlying consciousness.

  Although we are only beginning to study the biology of consciousness, these experiments have given us some useful paradigms for exploring different states of consciousness.

  AN OVERALL PERSPECTIVE ON THE BIOLOGY OF CONSCIOUSNESS

  It is tempting to conclude that the propagation of electrical signals forward into the prefrontal cortex—the broadcasting of unconscious information to the global workspace—represents consciousness, but consciousness is not likely to be that simple. Some of this broadcast activity does represent consciousness, but some of it may just represent associations.

  Suppose, for example, that someone who does not know who John Lennon was looked at a photograph of him. That person’s brain would go through the usual process of sending information from the visual cortex to the prefrontal cortex; as a result, she would see a pleasant-looking guy with round glasses and long hair. However, if that person did know who John Lennon was, she might associate Lennon’s image with the song “Eleanor Rigby” and with Paul McCartney, George Harrison, and Ringo Starr, the other Beatles. That additional brain activity is separate from perception of Lennon’s face: it recognizes the image of Lennon and associates it with memories. We make those associations unconsciously, but they nevertheless result from activity in the frontal areas of our brain in response to information sent from the visual system.

  One final, very important point concerns the fact that consciousness can operate largely independently of incoming stimuli. We generally envision the brain as receiving sensory input and producing outputs in response. That is often accurate, but consider this: even in complete darkness, with no visual stimuli, we maintain very complex states of activity that originate high up in the cortex and thus are top-down, or cognitive, in nature. Moreover, when we dream, we may be aware of highly colorful and emotionally arousing events, even though some signals from the external world may be blocked from reaching the cortex. Sometimes we think and plan while ignoring external events around us. Even when we daydream, imagining future events, our brain temporarily blocks sensory stimuli and instead plays with our internally generated ideas. These ideas and daydreams are generated independently, without input from external stimuli. To be sure, our brain can be brought down to earth by a loud noise or the smell of smoke, but while we are concentrating on our inner thoughts—as we often do—our brain keeps new sensory stimuli out.

  DECISION MAKING

  The ability to make good decisions is a critical skill that depends on both unconscious and conscious mental processing. In chapter 8 we discussed the important role of emotion in decision making. Here, we go beyond that to explore several ideas from cognitive psychology and biology that have advanced our understanding of how conscious and unconscious processes interact in decision making.

  Timothy Wilson, a cognitive psychologist, introduced the idea of the adaptive unconscious, a set of high-level cognitive processes similar to Freud’s preconscious unconscious.8 The adaptive unconscious interprets information quickly, without our being aware of it, which makes it vital to our survival. While we consciously focus on what’s happening around us, the adaptive unconscious lets part of our mind keep track of what’s going on elsewhere, to make sure we don’t miss something important. The adaptive unconscious serves a number of functions, one of which is decision making.9

  Many of us, when faced with an important choice, take out the proverbial piece of paper and make a list of plusses and minuses to help us decide what to do. But experiments have shown that this may not be the best way to make a decision. If you are overly conscious, you may talk yourself into thinking you prefer something that you really don’t like. Instead, you are best off when you allow yourself to gather as much information as possible about the decision and then let it percolate unconsciously. A preference will bubble up. Sleeping helps equilibrate emotions, so when it comes to an important decision, you should literally sleep on it. So there it is: our conscious decisions rest on information that is selected from our unconscious.

  Although the adaptive unconscious is a very smart, sophisticated set of processes, it isn’t perfect. It categorizes very quickly and can be a little rigid. One school of thought holds that this may account, in part, for prejudice. We react to a stimulus quickly, on the basis of past experience that may not apply to the new situation at hand. In such new situations, consciousness may step in and correct a snap judgment, saying, “Wait a minute. This quick, negative reaction may be wrong. I need to rethink it.” The adaptive unconscious works in tandem with consciousness to guide us in ways that make us the smartest species on earth. It would be interesting to see how far back we could trace these two mental processes that evolved to deal with different kinds of information.

  The biological role of the adaptive unconscious in decision making was revealed in a simple experiment by Benjamin Libet at the University of California, San Francisco. Hans Helmut Kornhuber, a German neurologist, had shown that when we initiate a voluntary movement, such as moving a hand, we produce a readiness potential, an electrical signal that can be detected on the surface of the skull. The readiness potential appears a split second before the actual movement.

  Libet carried this experiment a step further. He asked people to consciously “will” a movement and to note exactly when that willing occurred. He was sure it would occur before the readiness potential, the signal that activity had begun. What he found, to his surprise, was that it occurred after the readiness potential. In fact, by averaging a number of trials, Libet could look into a person’s brain and tell that he or she was about to move even before the person was aware of it.10

  This astonishing result might suggest that we are at the mercy of our unconscious instincts and desires. In fact, however, the activity in our brain precedes the decision to move, not the movement itself. As Libet explains, the process of initiating a voluntary action occurs rapidly in an unconscious part of the brain; however, just before the action is begun, consciousness, which comes into play more slowly, approves or vetoes the action. Thus, in the 150 milliseconds before you lift your finger, your consciousness determines whether or not you will actually move it. What Libet showed is that activity in the brain precedes awareness, just as it precedes any action we take. We therefore have to refine our thinking about the nature of brain activity as it pertains to consciousness.

  In the 1970s Daniel Kahneman and Amos Tversky began to entertain the idea that intuitive thinking functions as an intermediate step between perception and reasoning. They explored how people make decisions and, in time, realized that unconscious errors of reasoning greatly distort our judgment and influence our behavior.11 Their work became part of the framework for the new field of behavioral economics.

  Kahneman and Tversky identified certain mental shortcuts that, while allowing for speedy action, can result in less-than-optimal judgments. For example, decision making is influenced by the way choices are described, or framed. In framing a choice, we weigh losses far more heavily than equivalent gains. If a patient needs surgery, for instance, he is far more likely to undergo the procedure if the surgeon says that 90 percent of patients survive perfectly well, as opposed to saying that 10 percent of patients die. The numbers are the same, but because we are averse to risk, we much prefer to hear that we have a high probability of living than that we have a low probability of dying.

  Kahneman went on to describe two general systems of thoug
ht.12 System 1 is largely unconscious, fast, automatic, and intuitive—like the adaptive unconscious, or what Walter Mischel, a leading cognitive psychologist, calls “hot” thinking. In general, System 1 uses association and metaphor to produce a quick rough draft of an answer to a problem or situation. Kahneman argues that some of our most highly skilled activities require large doses of intuition: playing chess at a master level or appreciating social situations. But intuition is prone to biases and errors.

  System 2, in contrast, is consciousness-based, slow, deliberate, and analytical, like Mischel’s “cool” thinking. System 2 evaluates a situation using explicit beliefs and a reasoned evaluation of alternatives. Kahneman argues that we identify with System 2, the conscious, reasoning self that makes choices and decides what to think about and what to do, whereas actually our lives are guided by System 1.

  A clear example of the systems biology of decision making has emerged from the study of unconscious emotion, conscious feeling, and their bodily expression. Until the end of the nineteenth century, emotion was thought to result from a particular sequence of events: a person recognizes a frightening situation; that recognition produces a conscious experience of fear in the cerebral cortex; and the fear induces unconscious changes in the body’s autonomic nervous system, leading to increased heart rate, constricted blood vessels, increased blood pressure, and moist palms.

  In 1884, as we have seen, William James set this sequence of events on its ear. James realized not only that the brain communicates with the body but, equally important, that the body communicates with the brain. He proposed that our conscious experience of emotion takes place after the body’s physiological response. Thus, when we encounter a bear sitting in the middle of our path, we do not consciously evaluate the bear’s ferocity and then feel afraid; we instinctively run away from it and only later experience conscious fear.