Chapter 8
Conscious and Unconscious Cognition
In this chapter, I discuss the relationship between conscious and unconscious cognition; how much conscious control we have over our behavior; and Piaget’s concept that, for an idea to be conscious, it must tend not to contradict the person’s dominant understanding. What we are conscious of at any moment is very limited. But what we access unconsciously is probably vast. A constructivist view helps clarify how and where this vast information is stored and how we gain access to it. Finally, I examine problems that haunt my thesis, outlined in chapter 4, that the sense of being conscious is constructed from the facilitation of cortical circuitry activity by the RAS.
Most Cognition Is Unconscious
Piaget (1973, 1976), Freud, and many other students of psychology, before and after them, concluded that most cognition is unconscious. For example, we are seldom conscious of the processes involved when we come up with the right idea, or when we use the right words in the right order. Perhaps more interestingly, we are seldom conscious of the processes involved when we use the wrong words--the so-called Freudian slips.
The unconscious is not a container full of intact ideas and emotions, as has sometimes been thought. Rather the unconscious consists of organized schemes that are processed outside of consciousness. These unconscious schemes and the ways in which they are organized may be revealed in altered states of consciousness, such as dreams or intoxicated states, and by repetitive behaviors, including transference phenomena--behaving toward a person (such as a psychotherapist) as if that person were someone from one’s past. Unconscious schemes and the ways in which they are organized may be revealed by themes that underlie what one says in unguarded discourse. Contiguous ideas that at first appear to be inconsistent will generally reveal consistency, once we know more about their missing--that is, their unconscious--organizations and content. Under ordinary circumstances, we have little or no conscious access to the complex organizations of schemes, or to the diversity of their content, even though they underlie what we think, say, and do.
Libet’s Work
Libet (1985a) showed how constrained our conscious control of voluntary action appears to be. Subjects were asked to remember the position of a dot on a fast-moving clock whenever they decided to flex a finger. Subjects chose when to flex the finger: The behavior, although limited, was under voluntary control. The movement of the finger, the position of the dot, and the subject’s electroencephlogram (EEG) were recorded with great precision.
In 1964, Walter found that when a person intends to make a movement, the movement is preceded by a negative electrical shift in his EEG recording (Cooper, 1987). This shift in the EEG is now called a readiness electrical potential (RP).
Libet found that each subject’s RP began about half a second before she was conscious of her intention to flex her finger, as measured by her report of the position of the dot. After the subject became conscious of her intention to flex her finger, an additional fifth to a fourth of a second elapsed before she did so. 69 Thus, there was half a second between the onset of unconscious processing--measured by the RP--and consciousness of intention. A subject then had only a fifth of a second of consciousness of intention during which she could choose to act. 70 On occasion, an RP was recorded, but the subject did not move her finger. When queried later, she reported that she had intended to act, but had changed her mind.
In itself, flexing a finger is a very limited act. It is powerful, however, if we consider our ordinary flow of thought and intentions, especially any selection from our unconscious store of possibilities. There may be a concatenation of many, perhaps even overlapping, fifths of seconds in which to veto, divert, or carry through our voluntary acts.
Libet (1985b) also showed that we fool ourselves into thinking that our consciousness of a sensation is instantaneous. Recording directly from of the cerebral cortex of patients undergoing brain surgery, he found that a single light touch of the skin instantaneously (in 10 to 20 thousandths of a second) evoked an electrical response in the somatosensory area (S1) of the cerebral cortex. However, the patient was not conscious of being touched. In order for the patient to be conscious of the touch stimulus, the stimulus had to be intense enough to result in a half second of activation of the cortex.
Libet found that direct stimulation either of S1 of the cerebral cortex or of the medial lemniscus by an electrical current must also last for a half second before the person feels a tingling sensation in the skin. The medial lemniscus is a neural tract that feeds directly--that is, with virtually no delay--into S1. Direct stimulation of the medial lemniscus by the electrical current resulted in an instantaneous evoked electrical response at S1, much as touch of the skin does. However, direct electrical activation of S1--the cerebral cortex itself--resulted in no evoked response.
Libet stimulated the skin and S1 at different times while the patient reported which stimulus was felt first. 71 He varied the order of the stimuli and the time intervals between them.
When Libet stimulated the skin as late as a fifth of a second after he stimulated S1, the patient reported that she felt the touch before she felt the tingling. When Libet stimulated the skin half a second after he stimulated S1, the patient reported that she felt the touch and the tingling at the same time. When Libet stimulated the skin more than a half second after he stimulated S1, the patient reported that she felt the tingling first.
A person’s consciousness of touch appears to be tied to the evoked response in S1, which is instantaneous, not to the half second of S1 activation that precedes consciousness of touch. Libet concluded that the consciousness of being touched is referred back in time to the moment when the evoked response occurred.
Under ordinary circumstances, unconscious processes are at work before we are aware of our intention to act voluntarily and before we are able to experience a sensory stimulus. In these two instances, the mechanics of our becoming conscious--what must be done unconsciously before we become conscious of an intention or of a sensation--are unknown. It is strange perhaps, but in some seemingly ordinary circumstances our unconscious processes are running things for a half second before we know what is going on inside and before we know what is going on outside.
The Interplay
Piaget tended not to think that the boundaries between conscious and unconscious cognitive processing are entirely distinct. In addition, he contended that repression and defensive distortions--mechanisms that keep parts of cognition unconscious--are not restricted to emotion-laden cognition, as was central to Freud’s theory (Piaget, 1976). Piaget found that repression and defensive distortions operated even when the child was dealing with cognition that had little emotional charge.
Piaget (1976) proposed that for unconscious cognition to become conscious, the cognition must ordinarily be consistent with the person’s dominant understandings. 72 He proposed that when children’s perceptions were not consistent with their dominant understanding of physical cognition, they repressed or distorted their perceptions.
Piaget based his proposal on studies of children’s explanation of how they behaved and of how the object behaved as they learned a skill. In one such study, 4-year-olds learned to use a sling to throw a ball at a target. The children could often hit the target long before they were aware of how they behaved or how the ball behaved. When Piaget asked them what they did, the children explained that they had released the ball directly in front of them. Their explanation was the same one that a child would have given if he had released the ball from his hand. The explanation was consistent with the child’s dominant understanding of how he threw a ball to hit a target. At the same age, still failing to notice the difference between using a sling and throwing, a child might explain that he failed to hit the target because he did not throw hard enough or soft enough. He distorted his behavior to conform to his dominant understanding. A child who was a few months older and who had a beginning awareness of where he released the ball explained that the ball took a curved course to hit the target. He distorted his perception of how the ball behaved. At this age, a child might even say that he released the ball behind him--a major distortion of his behavior: The ball would have to pass through him. At the point that a child first recognized that he released the ball at nine o’clock--at the tangent of the circle--he might return to an earlier incorrect explanation, suppressing--or, as Piaget wrote, repressing--the correct explanation.
Piaget proposal that cognition may be subject to distortion or repression, whenever cognition is inconsistent with a person’s dominant ideas, broadened Freud’s position. Freud’s position was that distortion and repression of unconscious content took place when such content would invoke painful emotions were it to become conscious.
From Piaget’s study of children’s learning to use a sling to throw a ball, it is clear that unconscious--that is, automatic or mechanical--function may precede conscious understanding. Learning does not always happen the other way around, as when we learn to drive a car. After sufficient conscious, step-by-step practice, our driving becomes automatic. The steps that are involved become an unconscious process. Once this state is achieved, thinking about what we are doing--becoming conscious of the steps involved--may interfere with our performance. Later, if we seldom drive, our driving may no longer be automatic, and we must become conscious of some of the steps.
In any case, there is interplay between conscious cognition and unconscious cognition and learning. On the one hand, we may recognize a friend from the back of his head, although we never were conscious of having known what the back of his head looked like. On the other hand, we were certainly conscious of what we were doing when we painstakingly learned to drive a car. Yet once having learned, we may drive for the rest of our lives with little awareness of how we do it.
Is Consciousness Distinct From the Waking State?
As I proposed in chapter 4, if consciousness is constructed from the waking state--the facilitation of the cortical circuitry activation by activity of the RAS--is consciousness ever completely distinct from the waking state? Except when we are dreaming, we are never conscious without being awake. However, during certain epileptic seizures the person loses consciousness without being asleep. 73
Petit mal seizures are characterized by loss of consciousness while awake. They are most common in children. During a petit mal seizure, the child stops talking, may blink a bit, and is unresponsive for half a minute or so. When the child comes to, he has no memory of the episode. Repetitive, generalized EEG spike-and-wave patterns coincide with petit mal seizures. Petit mal seizures originate in the thalamus.
Psychomotor seizures are characterized by loss or diminution of consciousness while awake. During a psychomotor seizure the person is nonresponsive and exhibits automatic, often repetitive, coordinated behavior. The seizure may last for minutes. When the person regains consciousness, he has no memory of the episode. During a seizure, one of my patients moved a chair back and forth. Another patient ran during his seizures. One day, when he was not being closely watched, he ran into a lake and drowned. During a psychomotor seizure, the person’s EEG shows a focus of repetitive, high-voltage spikes in the temporal lobe of the cerebral cortex. Sometimes, the spiking becomes generalized and culminates in a grand mal seizure.
A grand mal seizure — that is, a tonic-clonic seizure — is what the general public thinks of as an epileptic seizure. These seizures begin with unconsciousness, followed by a generalized contraction of the muscles--the tonic phase. The tonic phase lasts about half a minute and is followed by 1 or 2 minutes of generalized jerking of the muscles--the clonic phase. After a seizure, the person usually sleeps. During a grand mal seizure the person’s EEG shows repetitive, high-voltage spikes throughout both cerebral hemispheres. The trigger of the seizure--an irritable focus--may be in the thalamus, as in a petit mal seizure, or it may be elsewhere in the brain. Prior to the seizure, the person may experience an aura--an odd sensation. The nature of the aura depends on the location of the trigger.
Sometimes, spiking from an irritable focus in the motor cortex of the brain triggers repetitive movement of a limb. The person does not lose consciousness but has no conscious control over this movement. Sometimes the spiking spreads, and a grand mal seizure follows.
It is reasonable to think of repetitive, relatively regular, high-voltage electrical activation of the brain as the enemy of consciousness — that is, as the enemy of activation by the RAS. The RAS appears to provide a more gentle facilitation of activation of neuronal circuitry of the cerebral cortex during the awake state. During grand mal seizures, the generalized electrical spikes apparently swamp ordinary cerebral cortical function. In petit mal seizures, repetitive and generalized high-voltage electrical spikes and waves apparently interfere with the ordinary transmission of electrical impulses between cortical neuronal circuits and within these circuits. In psychomotor seizures, the repetitive, high-voltage electrical spikes trigger the behavior that characterizes these seizures. Although the psychomotor seizure spiking is somewhat localized, it must involve circuitry that is critical to consciousness or circuitry that is widespread enough to interfere with the ordinary transmission of electrical impulses between cortical neuronal circuits and within these circuits.
To think of repetitive, relatively regular, high-voltage electrical activation of the brain as the enemy of consciousness is consistent with a phenomenon observed during the use of electroconvulsive treatment (ECT). ECT is used primarily to treat certain types of depression. In ECT, alternating electrical current is passed through the cerebral cortex in order to trigger a grand mal seizure. When the amount of electricity--that is, the voltage and the duration--is insufficient, the patient may become unconscious, but the patient will not have a grand mal seizure. Presumably, the electrical activation is sufficient to interfere with the ordinary transmission of electrical impulses between cortical neuronal circuits and within these circuits.
Where Are Our Memories, Where Are Our Schemes, and How Do We Find Them?
If most of cognitive content is unconscious, it is relevant to ask where the content is stored and how do we gain access to it. "On being asked what I did at seven o’clock this morning, I am obliged to deduce the answer, and it is unlikely that it was noted (on a record always kept up to date) in my unconscious" (Piaget, 1962, p. 187). Piaget was saying that he had to construct his memory of what he did at seven o’clock.
Whenever a memory is retrieved, I propose that it may be reconstructed from different components each time. Reasoning or problem solving may be seen as finding useful information. We perceive what we know and have felt. So where and how memories are stored and retrieved is central to all forms of cognition.
Recording a Memory
Patients suffering from Korsakoff’s syndrome are unable to remember anything new except for a minute or two, although, after they are in the hospital for an extended period, they remember how to find their way to the bathroom or to the dining hall. 74 When asked to respond to questions that call for information that they do not remember, they may confabulate — that is, they offer an answer that is reasonable, but inaccurate. When I did research on such patients, they did not remember having met me, even after I had encountered them many times. When I asked them to remember a three-digit address--for example, 382 Main Street — they could repeat it immediately, but forgot it after 2 to 3 minutes and might not remember that they had been asked to remember anything. Their ability to carry on a conversation, their level of consciousness, and their intelligence were normal (Malerstein & Belden, 1968). They had no particular difficulty in recognizing an old friend or in remembering their past.
Korsakoff’s syndrome is caused by vitamin B 1 deficiency. Korsakoff’s syndrome usually occurs in alcoholics, because alcohol places a demand on the body for B 1, and because many alcoholics do not eat well. Ordinarily, after a month to a year of adequate nutrition and no alcohol, 90 % of these patients recover completely. The other 10 % do not improve. One who did not improve was a Boston taxi driver (Talland, 1965). When he resumed driving his taxi, he took his wife with him. He could find his way around the old part of Boston, but not around the new part.
There is general agreement that patients who have Korsakoff’s syndrome have no cerebral cortical damage to account for their profound memory impairment. It is also generally agreed that memories are recorded primarily in the cerebral cortex. It follows that Korsakoff’s syndrome provides an unusual opportunity for studying memory processing.
Investigators disagree on which damaged subcortical structure causes Korsakoff’s syndrome (Horel, 1978; MacLean, 1990). Malamud and Skinner (1956) found that in over 95% of their Korsakoff’s cases, three small brain structures were damaged: the mammillary bodies, the periaqueductal grey, and the periventricular grey. The mammillary bodies--small bilateral structures located at the base of the brain--are part of Papez circuit. This circuit consists of the mammillary body and a series of neural tracts that connect to several other subcortical structures and then back to the mammillary body. The structures of the circuit have widespread connections to other parts of the brain, including the cerebral cortex and parts of the brain related to emotions and to the autonomic system.
I suggested in chapter 1 that storage of information--that is, memories and what to do with memories--is widespread and content addressable. Barbizet (1970) theorized that, in Korsakoff’s syndrome, interruption of Papez circuit--which is a potential reverberating circuit that could repeatedly distribute a pattern of stimuli throughout the brain--impairs rehearsal of new information. To remember new information, it is useful to rehearse it and to connect it to other old information. Barbizet’s proposal is consonant with the theory of remembering presented here. 75
Finding a Memory
When I see someone I know and want to address by name, or when I am trying to remember something or trying to solve a problem, I am often not conscious of any of the links that are part of the reconstruction. The name, the memory, or the solution just pops into my mind. At other times when I cannot immediately recall someone’s name, I cast about through my experiences with the person. I may think of where I first met him. I may ask myself what we did together when I last saw him, and so on. The name may then come to mind. At times, I may have to get in touch with my feelings about the person--whether I like or dislike him. A similar trolling takes place when I try to recall other memories or attempt to solve a problem. Usually I do not know why a particular idea or feeling calls up the answer.
Piaget (1962) recounted an example of his being aware of some of the constructive process in seeking a memory, a memory which at the same time solved a problem. That is, the seeking of the memory was an example of reasoning. He was in a hurry, and used his handkerchief to wipe some oil from the steering wheel of his automobile. Rather than soil his pocket, he wedged the handkerchief deeply into the crevice between the seats. The windshield control of his car was defective. He could only leave the windshield completely open or shut. One day it was too warm to leave it shut and raining too hard to leave it completely open. He tired of holding the windshield open with his hand and was unable to find an object to hold the windshield open. Then, he noted the angle formed by the windshield and an upright part of the car body. Seeing the angle as a solution, he found himself automatically reaching into an angle--the crevice--for his handkerchief.
Here are some other examples of being able to catch sight of the parts that go into retrieving a memory and problem solving. 76 I live in a neighborhood of many cars and few garages. I have parked on the street thousands of times in almost every conceivable location within a four-block radius of my house. I have several patterns that I follow when I search for an open parking space, though sometimes these patterns disintegrate on an evening when the search is extended and has become random. At the time when I park, I usually make no special effort to remember where I parked. I wish to get into the house after a long day, and I simply assume that I will find the car in the morning. Occasionally in the morning, I am at a complete loss. I am left to retrace my usual patterns of looking for a parking space. I may remember where I parked the night before only when I actually find the car.
I would like to detail the intermediate states when I cannot remember where I parked the night before. In one instance, after I left the front door and as I walked, I started to remember. My first thought was that I had parked in a very unusual place. Then, I immediately remembered that, as I parked, I commented to myself, "This spot has almost never been vacant" (in my 20 years in the neighborhood). I could then visualize the spot and the act of parking.
One morning before I left the house, as I was thinking about this problem, I remembered that the previous night I had backed my car downhill to park in a tight space between two cars. The car behind me was parked illegally on the corner. I could visualize the truck in front and the car behind. At first, I thought that the corner was the corner across the street from my house. Then I remembered that the actual parking space was on a similar corner a block away. It was only then that I recalled that I had not driven my car at all. I had driven my daughter’s car. Mine, in need of repair, was parked elsewhere. So, when I was remembering parking, I was not remembering the car that was put in a particular space, and the space itself was not the space that I had first remembered. The search for a car that was parked was more a search for a means of getting back and forth than it was a search for a particular car.
Another morning as once again I was baffled by the mystery of the parked car, I remembered that I had to give a message to one of my daughters. This message was not from someone at college, where she had been living. It was from someone here at home. As I remembered who had called out to me, I also remembered the content of his message, and what had happened the previous night. I had met him as I walked home from the parked car. I was then able to proceed to find the car.
In retrieval of a memory, these steps in remembering were cognitive samples that appeared in consciousness--samples of what I assume was a widespread unconscious reconstruction of the memory of the car and where I parked it. Each time, the car and its location were constructed out of different parts--an unusual parking space, a spatial configuration, a message for my daughter, and, I assume, many parts that I was not conscious of. The image, or memory, of parking the car may never be constructed of the exactly the same parts. Hence it is probably never constructed of the exact same neuronal circuits, or schemes, and these are necessarily widespread, because their contents are diverse. Of course, the parked car that I construct or reconstruct is not parked on the street so much as it is parked in my head, built less of steel and plastic than of my experiences of it.
Memories are not kept in some repository somewhere. Many investigators now agree that memories are constructed, and that each time we remember something, the memory must be reconstructed. 77 I also propose that the memory may be reconstructed of different parts, and that these parts are widespread.
What would consolidation be like in such a widespread system? How might one measure it?
Hoffman and McNaughton (2002) have begun to test the theory that memories are consolidated in widespread regions of the cerebral cortex--regions that maintain a relationship to one another, part of that relationship being temporal order. They implanted large arrays of electrodes in individual neurons in the left posterior parietal and motor areas, and the right somatosensory and prefrontal areas of a monkey’s cerebral cortex. They recorded cell activity during three time intervals--an initial rest period, a period when the monkey learned a sequential reaching task, and a second rest period. About 20% of the neurons in the four different areas of the cerebral cortex had electrical responses that were related to the reaching task. In the posterior parietal, in the motor, and in the somatosensory areas, activity of neurons during the learning period correlated with activity of the same neurons during the second rest period. This suggests that there is ongoing, widespread, though related, cortical activity after the behavior ceased. During the second rest period, the investigators also found that activity of the neurons of the posterior parietal, of the motor, and of the somatosensory areas correlated within each area and between each area. This finding suggests that these three areas maintain some sort of relationship both within each area and between each of these areas. Additionally, they found some sequential relationships between cells within the motor area and between cells within the somatosensory area, as well as between cells in the motor area and cells in the posterior parietal area. These findings suggest there is some order of organization between different parts of areas and between different areas. Finally, they found no correlations of activations of cells of the prefrontal cortex with any of the other measures. The fact that these findings were confined to certain areas, and that activations of cells in the prefrontal cortex did not correlate with any of the other measures, suggests that the post-task activations are selective. Although the activations are widespread, they are not helter-skelter throughout the cerebral cortex.
There is psychological, and now neurophysiological, support for the idea that memory is recorded in widespread neuronal circuit activation. Thinking and reasoning consists of finding the right memory in the right order for a particular set of internal and external circumstances. When I am trying to answer a question and making no progress, yet think that I might know the answer, like most people I know, I set it aside. I put it on the back burner. In a few days or a few weeks, an answer may occur to me.
It is hard to explain such events without proposing that unconscious processes were working on the problem in the interim.
In the next section, I will discuss two conditions that might be problems for my theory that consciousness is constructed from the waking state. These conditions are the dream state and the p ersistent vegetative state.
Dreams and the Persistent Vegetative States
When we dream, we are asleep. Yet we are conscious. Someone might ask, "How do you know that we are conscious when we are dreaming?" I know it, because when I wake up or when I am awakened, especially during a particular stage of sleep, I will generally give a report of what I was doing or watching and so on in my dream. If I was not conscious while I was dreaming, there is no reason to believe me when I say that I am conscious now while I am writing.
If consciousness is constructed from the waking state, how can we be sleeping and yet be conscious? How could my proposal that we develop our sense of consciousness from the waking state--the facilitation of cerebral circuit excitability by activation of the RAS--be valid? When we are not awake, what gives the attribute of consciousness to our dreaming cognition?
When someone is in a persistent vegetative state, that person is in coma — that is, unconscious. Yet the person has what appears to be a sleep-wake cycle. If, as I propose, consciousness is constructed from the waking state, how is it possible for someone to be in a vegetative state — to be unconscious — and yet to be awake some of the time?
Consciousness While Asleep: Dreams
First, I will discuss dreams — consciousness, while asleep. Many things about dreaming are understood better now than they were when Freud wrote The Interpretation of Dreams. We have learned that even people, who say that they never dream, dream four or five times a night. If they are wakened during a sleep state when we expect to elicit a dream report, they will report a typical dream. In the morning, they will not remember that they dreamt. Freud had no reason to think that we dream as often and as regularly as we do. He thought that dreams were instantaneous. Now, we know that each dream lasts 15 or 20 minutes. Yet, when it came to understanding the structure and content of dreams, Freud did not do badly.
To study sleep, investigators rely on the EEG, behavioral observation, and sleep deprivation. In sleep deprivation studies, they keep the person or animal awake for longer than is usual for the subject, and then note if the subject’s EEG or behavior changes.
Through the use of EEG recordings, we have learned that sleep occurs in five stages: rapid eye movement (REM) sleep — the stage from which dreams are readily elicited--and four stages of nonrapid eye movement (NREM) sleep. When we fall asleep, we pass through the stages in a relatively orderly fashion: stages 1 through 4 of NREM followed by REM. Thereafter, the cycle of NREM (usually skipping stage 1) and REM is repeated four or five times a night. At each succeeding stage of NREM, we are more deeply asleep — that is, more difficult to waken. When we are in REM sleep, we are wakened a bit more easily. The percentage of time spent in REM sleep decreases with age. REM sleep is sometimes called paradoxical sleep, because the brain waves are low voltage and desynchronized — an EEG pattern that looks the same as the waking EEG pattern. During REM sleep, the person is paralyzed, except for the muscles of the diaphragm and those that control the ear ossicles and the eyes. NREM sleep is sometimes referred to as slow-wave sleep because it is characterized by generalized low-frequency, high-voltage EEG waves.
At one time, it was thought that REM deprivation resulted in severe mental disturbance. This is no longer thought to be true (Rechtschaffen & Seigel, 2000). However, total sleep deprivation for 4 or 5 days may result in a clinical picture that resembles paranoid schizophrenia (Bonnet, 2000). This state abates after the person has had adequate sleep.
In a sleep laboratory, dreams may be elicited 80 % to 90 % of the time when a person is wakened during REM sleep (Vogel, Foulkes, & Trosman, 1969; Rechtshaffen & Siegel, 2000). What we usually think of as a dream is the kind of cognitive processing that is most often elicited when we are wakened from REM sleep, particularly from the last REM period of the night, or what we may remember in the morning.
In recent years, less distinction has been drawn between the kinds of reports that are elicited from subjects when awakened from REM sleep and the kinds of reports that are elicited from subjects when awakened from NREM sleep. Nonetheless, the report that is elicited when a person is awakened from NREM sleep is more like ordinary thinking, and it is generally elicited only about 50 %, or, at most, 70 %, of the time. NREM reports are shorter, less vivid, less emotional, and more coherent than REM dreams or what we ordinarily recall in the morning when we dream at home — a home dream.
Although there is no absolute distinction between reports elicited during REM and those elicited during NREM, they are qualitatively different. When I use the term dream, I am referring to home dreams or reports elicited during REM, particularly REM period that occurs near the time of waking. Such dreams may include intense emotions and usually involve visual imagery; they do not often involve spoken language. Typically, they are incongruous to us when we are awake. When we are dreaming, we may be in shifting, strange, or ill-defined locations, roles, and relationships. Our behavior and the situations that we are in during dreaming often seem impossible or uncharacteristic once we are awake. Nonetheless, strange as the dream may be, when we are dreaming, we generally accept the content of our dreams as real. 78 Why such dreams usually occur during REM is not known. In fact, following damage to the right lingual lobe — an association area between the visual and auditory areas — an elderly woman no longer dreamed and yet had normal REM cycling (Grimm, 2004). Apparently in an adult, dreaming and REM may be disassociated, and damage to the lingual gyrus may block dreaming.
Sleep and Dreams as Partial Shutdowns
Although we know more than we once did about sleep and dreaming, no consensus exists regarding the purpose of either one. Sleep is a potentially dangerous state. We are almost completely paralyzed during REM. Our sensory systems are not as responsive during sleep as they are when we are awake, although our diminished sensitivity is selective. For example, a sleeping mother may respond to her infant’s cry and not respond to the cry of another infant. Partial shutdowns of our motor and sensory systems leave us vulnerable. Yet we need to sleep.
The most common and most intuitive explanation is that sleep is restorative. Zepelin (2000, p. 85) argued that the "restorative theory cannot readily explain the dramatic interspecies variations in daily mammalian sleep quotas." Indeed, there is a wide variation — from the horse that sleeps 3 hours a day, 1/2 an hour of which is REM, to the opossum that sleeps 18 hours a day, 5 hours of which is REM. However, the purpose of sleep could still be restorative if restoration depended on the variation in what had to be restored, and on the variation in the efficiency of the restoration for each species.
To Freud, the purpose of the dream was wish fulfillment — he believed that dreams were disguised wishes. He proposed that the dream process involved representation by opposites, condensation, displacement, and symbolism. 79 He proposed that dream content had manifest and latent meaning, and that dreams often included day’s residue — references to happenings of the previous day. Basically, Jung (1969) accepted Freud’s ideas about dreams. However, he proposed that the purpose of dreams was compensatory — that dreams provided a balance between the conscious and the unconscious. Jung posited that the Conscious and Unconscious being balanced was necessary for mental wellbeing.
Rechtshaffen and Siegel (2000) minimized the significance of day’s residue when they pointed out that subjects who were deprived of liquids for 24 hours often did not dream of thirst, and that only 30% of their dreams contained references to drinking. Wish fulfillment does not explain why we dream. However, the references to drinking in 30% of the dreams confirm s Freud’s proposal that dreams involve wish fulfillment, day’s residue, disguise or defense, and even representation by opposites — that is, being thirsty, but drinking in the dream. The references to drinking also fit Jung’s proposal that dreams may be compensatory.
Some dream researchers have postulated that we consolidate our memories during REM. Siegal (2001, p. 1063) concluded that "the existing literature does not indicate a major role for REM sleep in memory consolidation."
I think that we probably cognize — order, consolidate, and retrieve — all the time, whether we are awake or asleep. What varies is the quality of our cognizing that we are conscious of.
Animals other than mammals have been studied for clues as to why we sleep. The EEGs of fish when they appear to rest compared to when they are active are "difficult to determine and remain unclear" (Tobler, 2000. P. 78). When subjected to 4 days of light, carp appeared to rest more quickly than usual when the light was turned off. When subjected to 6 to 12 hours of light during a habitual rest period, perch exhibited increased duration of rest.
Tobler (2000) reported that reptiles, birds, and mammals show both behavioral and EEG sleep differences. When reptiles are kept awake by stroking, handling, or gently tugging on their leash, more stimulation is required to keep them awake as the time goes by, and when they are allowed to sleep, they become limp very quickly. The reptile’s EEG shows more spikes during the sleep state than during the awake state. Following sleep deprivation, EEG spiking increased during sleep. It has been suggested that the reptilian spikes and sharp waves have a functional similarity to slow waves in mammalian sleep. Reptiles appear to have awake and asleep states.
It is probable that both REM and NREM will be found in all mammals and birds, although the frequency and the amount of time spent in each state varies greatly. Most interesting are the EEGs of dolphins, seals, and manatees--mammals that have returned to the sea. They show high-voltage slow-wave sleep on one side of the brain while they show a pattern that resembles waking on the other side of the brain. In both seals and manatees, one-sided REM sleep has been recorded, and there is some suggestion that one-sided REM sleep occurs in dolphins (Zepelin, 2000). This one-sided sleep is a clue to why we sleep that I will return to.
Intuitively, many of us believe that the function of sleep is to rest the body and the brain. We feel refreshed on waking from a good night’s sleep and feel tired on waking from a poor night’s sleep. If we get no sleep for 24 to 48 hours, our thinking and behavior may be impaired. After more than 8 days of sleep deprivation, subjects may show nystagmus (a type of jerky movement of the eyes), their hands may shake, and their speech becomes slurred (Bonnet, 2000). The normal reflexive closing of the eyelid s when the cornea of the eye is touched may become sluggish, and other reflexes may become hyperactive. Clearly, our function is impaired when we are deprived of sleep.
Rats that were deprived of sleep, died in about 15 days (Bonnet, 2000). If the rats were allowed to rest their bodies, but were prevented from falling asleep, they did not survive longer: Rest of the body did not substitute for sleep. REM sleep-deprived rats lived only about twice as long. 80 I assume that humans would also not survive if they were subject to comparable conditions.
We appear to need to sleep in order to rest both the body and the brain--in order for each to recover neurophysiologically from being awake, just as muscles must recover physiologically from being active. However, if one of the purposes of sleep is to rest the brain, how can I account for the fact that the brain is very active during sleep? In fact, the metabolic rate of the brain is higher during REM sleep than it is during the waking state. Because the brain is very active during sleep, some investigators have discounted the idea that one function of sleep is to rest the brain.
One way to make sense of this situation is to assume that it is deleterious to brain function to shut down the entire brain in order to allow it to rest. If maintaining information in the brain depends on maintaining ongoing relationships between circuits, or schemes--as I have described, and as studies such as Merzenich’s have found--then a complete shutdown would indeed be damaging. With a complete shutdown, all past information and what to do with it would be lost. As I noted earlier in this chapter, information in the brain appears to be content addressable--not physical point localizable--as is information in books and in computers. One can close a book, open it, and go to the exact same spot as often as one likes. It is still there in black and white. One can shut down a computer and reopen it later as many times (until parts of it wear out) as one likes and institute the same command to find the letter a or the number 2. One can do this because every piece of information and what to do with it has a discrete physical location or set of locations--a mechanical switch, or set of switches, that is on or off.
It is plausible that the compromise solution that evolved for the brain of mammals and of birds was to rest different parts of itself at different times. As I just noted, dolphins, seals, and manatees--mammals that have returned to the sea--rest one side of the brain at a time. Land-based mammals may have found a different method of resting their brains, a part at a time. Or as Tobler (2000, p. 76) said, "sleep may be a local phenomenon leading to recuperation of those regions that were most active during waking." 81
If this is true, our stages of sleep, including REM or dream sleep, can be seen as partial shutdowns. I propose that cognitive processing is taking place at all times. Depending on what is not shut down, one or another sample of cognitive processing is exposed--that is, may be conscious. During dreaming, access to the motor system is blocked, and certain inhibiting cognitive-motivational and sensory systems are diminished. If dreams are merely a sample of cognitive processing during a partial shutdown of the brain, it isirrelevant to ask what the purpose of dreaming is.
I think that dreams give us a glimpse of how things tend to be catalogued and processed unconsciously. The cataloguing is more orderly when we are awake or in the reports we give when we are awakened from NREM sleep (or when we are awakened from the first REM period of the night). I think that a dream--with a piece here and a piece there and two seemingly disparate pieces that are usually tied together to form a story--is simply a sample of our unconscious cognitive-emotional processing. 82
I propose that we cognize using widespread areas of the cerebral cortex at all times, but we consciously cognize differently in our different states. 83 The rules for processing and cataloguing, and our access to the cataloguing, vary with the state we are in.
My proposal is that the dream state is a partial shutdown during which we get a glimpse of the unconscious cataloguing process and of some of the unconscious content. Freud believed that the purpose of dreams was wish fulfillment. I propose that dreams have no inherent psychological purpose, that the purpose of dreams is physiological--the resting of certain parts of the brain.
Nonetheless, I think that Freud was correct when he said that the dream is the royal road to the unconscious--that is, the dream, especially the morning dream, is one of the best roads that we have to understanding unconscious cognition and emotion and how they are processed. Dreams offer us a glimpse of parts of our self that we ordinarily do not see.
The Persistent Vegetative State
Karen Ann Quinlan was 21 when her heart stopped beating and she stopped breathing following her ingestion of prescription drugs and alcohol (Kinney et al. , 1994). After she was resuscitated, her pulse and breathing returned. However, she never regained consciousness. She remained in a coma and on life support. Over the following year, she increasingly required the assistance of a pulmonary ventilator. With much legal difficulty and with national attention, her family succeeded in getting the court to allow her ventilator to be disconnected. After the ventilator was disconnected, Karen Ann continued to breathe on her own for a number of years. But she remained comatose.
She opened her eyes to sound and withdrew her limbs in response to a pinprick. But opening one’s eyes to sound and massive withdrawal to pinprick are built-in reflex responses.
By all measures she was unconscious. However, of particular interest here was the observation that, measured by her EEG, she had a sleep-wake cycle. Slow wave activity--the pattern that is characteristic of NREM sleep--cycled with Beta activity (13-30 waves per second)--a frequency that is common when a person is awake (Kinney et al. , 1994).
At the time of her death, Karen Ann’s brain showed widespread damage, which was probably unrelated to her original coma. The brain lesion that was thought to be responsible for her coma was bilateral damage to the intralaminar nuclei (ILN) of the thalamus (Kinney et al. , 1994). The ILN are the primary synaptic relay stations in the thalamus for the reticular activating system (RAS). The ILN cells connect to widespread areas in the cortex, bridging the RAS to the cerebral cortex. Damage to Karen Ann’s ILN varied: It was mild to the parafascicular nuclei, moderate to the centromedial nuclei, and severe to all other intralaminar nuclei.
Bogen (1995a, 1995b) proposed that Karen Ann’s coma resulted from the bilateral damage to the intralaminar nuclei. His proposal is basically the same as my proposal that essentially equates consciousness with the waking state, since impulses from the RAS pass through synaptic relays in the ILN before reaching the cerebral cortex. Bogen’s and my proposals differ in one respect, however. Bogen proposed that the subjective aspect of consciousness was also a product of the ILN--that is, that the self was a prewired component of the ILN.
In keeping with Piaget’s findings, I propose that consciousness of the self is developed. As I explained in chapters 2, 4, and 5, a semblance of self is first manifest in Stage 4 of the Sensorimotor Period, and a more definitive sense of self is manifest in Stage 6, although the self is more fully defined in the Concrete Operational Period.
I think Bogen was mistaken in attributing consciousness of self to the ILN. If consciousness is the product of RAS activity--the facilitation of activation of the cerebral cortical circuitry during the waking state--and if the activity in the RAS must pass through the ILN, then bilateral damage to the ILN would eliminate all forms of consciousness, including differentiated consciousness of self.
We are left with the question, if consciousness is waking-state schemes, as I propose, how do I reconcile Karen Ann’s being awake--that is, having a waking EEG--at times with the fact that she was continuously unconscious? There are several possibilities.
One possibility is that during the period in which Karen Ann had no heartbeat and was not breathing, much of her established neuronal circuitry became near random--that, unlike what happens during sleep, she was shut down too completely for too long.
Another possibility is that mechanisms other than the usual ones were able to induce a sleep-wake cycle in Karen Ann, but that their operation was insufficient to restore her consciousness. It is not unusual throughout the body, including the brain, to have backup systems to compensate for a system that is impaired. It is possible that when passage of impulses from the RAS to the cerebral cortex was blocked by damage to the ILN, another route between the two was found. There are known systems that could pinch-hit for the RAS’s main route. The RAS has extrathalamic relays through the hypothalamus and basal forebrain to widespread areas of the cerebral cortex (Jones, 2000). It seems less likely that the brain stem nuclei that secrete neurotransmitter chemicals would account for Karen Ann’s sleep-wake cycle. The raphe nuclei, which secrete serotonin, are active during deep sleep. The locus ceruleus, which secretes norepinephrine, is active during the awake state. Both sets of nuclei were intact in Karen Ann. However, Steriade (2000) found that bilateral removal of the locus ceruleus does not affect the waking-state EEG.
Kinsbourne (1995) offered an explanation of Karen Ann’s state that differed from Bogen’s. Kinsbourne suggested that she might have suffered from bilateral neglect. Kinsbourne’s explanation does not conflict with my proposal. If Kinsbourne’s explanation is correct, waking could remain integral to consciousness. Karen Ann would then have had a highly limited kind of consciousness.
A person who exhibits unilateral neglect does not process input from, or output to, one side of his or her world. U nilateral neglect of the left side is most easily recognized. A person who exhibits left-sided neglect may fail to put on his left sleeve when dressing, may draw a clock face with all the numbers crowded onto the right side of the clock face, and may be unable to see things situated to his left. The usual cause of left-sided neglect is damage to the right parietal region of the cerebral cortex. 84 Measured by auditory comprehension, Lecours et al. (1987) showed that strokes involving the left hemisphere resulted in right-sided neglect.
Although unilateral neglect is usually due to damage to the parietal lobe of the cerebral cortex, Watson, Valenstein, and Heilman (1981) reported a patient whose unilateral neglect resulted from unilateral damage to the thalamus. Apparently, Kinsbourne reasoned that if unilateral damage to the thalamus could cause unilateral neglect, then bilateral damage to the thalamus could cause bilateral neglect. A person who had bilateral neglect might experience no input or output. Would that person’s consciousness be confined to memories--a profound loss of consciousness--but not complete unconsciousness? How long would those memories last if there were no input or output? It is more probable, however, that bilateral neglect would not result in no input or output, but result in the person’s operating as if all of (visual) life is straight ahead--Balint’s syndrome--which is caused by bilateral damage of the parietal areas.
In chapter 4, I addressed two fundamental questions--how is conscious cognition derived from matter and how does conscious cognition influence matter? In chapters 4 and 5, I outlined my theory that consciousness is constructed from the waking state. In this chapter, I discussed the relationship between conscious and unconscious cognition, and the widespread, content-addressable memory system, which is largely unconscious. Finally, I explained my approach to the problems that dreams and the persistent vegetative state present for my theory that consciousness is constructed from the waking state.
In the next chapter, I will describe ideas based on Ahern’s and my work as clinicians. These ideas involve a theory of character structure formation that relates to developmental psychology, especially three of Piaget’s stages of cognitive development. The theory is consonant with the quality of the Preoperational Period, when the organizations of neuronal circuitry are uncommitted, and to a time during which complete myelination of the auditory tracts to the cerebral cortex could induce one of those organizations of neuronal circuitry to become stabilized.
69 Lau et al. (2004) replicated Libet’s findings using functional magnetic resonance imaging, which can locate brain activation more precisely than an EEG can. They found that conscious intention to move a finger was preceded by activation of three small areas of the cerebral cortex--the pre-supplementary motor area, the right dorsal prefrontal area, and the left intraparietal sulcus.
70 In an emergency or in a highly practiced event, such as starting a 50-yard dash, people respond faster. For example, the time between hearing the gun go off and leaving the starting blocks is much shorter than the time between deciding to move a finger and moving that finger. Presumably the conscious decision-making process is bypassed. Herbert Bengelsdorf, after critiquing Chapter 4, drew my attention to the fact that, if one touches a hot stove, one removes one’;s hand before one’s cerebral cortex could possibly experience pain.
71 The patient could distinguish the touch of the skin from the tingling sensation that resulted from electrical stimulation of S1 or the medial lemniscus. However, to make the patient’s reporting easier, touch of the skin and electrical stimulation of S1 were done on the same side of the body. Because the sensory area of one hemisphere of the cerebral cortex, in this case S1, is activated by stimuli from the other side of the body, touch of the skin and stimulation of S1 on the same side, resulted in sensations that were felt in different arms. This allowed the patient to report whether she felt the stimulus on the right or on the left.
72 His proposal is consistent with the fact that, in the course of development, primitive organizations are organized out of conscious understanding. This occurs, for example, when Stage-4 ordered cognition--such as striking before grasping--replaces the mutual assimilation-accommodation of Stage-3 cognition--such as striking while watching and watching while striking.
73 See Westbrook (2000) for a current typology of epilepsy and a more extended account of what is known about epilepsy.
74 Remembering something for a few seconds is currently referred to as working memory. Memory for new procedures--for example, finding one’s way around the hospital--is not as impaired as memory for new facts--for example, learning a new address--or for new episodes--remembering having met a new physician. In recent years, much research has focused on these different kinds of memory. For a review of such research, see Squire and Schacter (2002).
75 Now, I incline toward Barbizet’s proposal. Earlier, like Solms and Turnbull (2001), I proposed that an inability to sort memories accounted for both the memory defect and the confabulation in Korsakoff’s syndrome (Malerstein & Belden, 1968; Malerstein, 1969). Because confabulation is found in a number of conditions, including conditions that involve damage to various areas of the brain (Schacter & Scarry, 2000), confabulation probably has no one-to-one relationship to a single area.
Nonetheless, Schacter, Verfaellie, and Koustaal (2002) found that Korsakoff ’s patients and midtemporal-damaged patients responded with an incorrect theme word more often than controls when asked which words were included in a list of words. For example, when presented with a list of words such as "candy," "sugar," "taste," "sour," "good," these patients might respond with "sweet." When such lists were repeated, controls responded more accurately, than they had before, while midtemporal-damaged patients made about the same number of errors, suggesting that each test was a new test to them. Interestingly, Korsakoff ’s patients increasingly included theme words. This suggests that Korsakoff ’s patients retain the overall subject or meaning, but not specific markers. Confabulation retains the overall meaning or theme, but lacks the specifics.
In Korsakoff ’s patients, the primary lesions described by Malamud and Skinner, were thought by Malerstein & Belden (1968) to be negative reinforcers--that is, if they were stimulated, a person would find it noxious. More recently, it has been found that parts of the periaqueductal gray are positive reinforcers (Solms & Turnbull, 2002). Damaged reinforcers would impair precise remembering and could account for confabulation. Korsakoff’s syndrome remains an intriguing puzzle.
76 This account appears in an earlier work (Malerstein, 1986) in slightly different form.
77 Sometimes our reconstructions are faulty. In recent years, this has had serious consequences for persons convicted of molesting children based on those children’s memories many years later. Piaget (1962) had a vivid memory of having been kidnapped at the age of 2. His memory was based on a story that his nanny told to his parents. Some years later, the nanny confessed that she had made up the story to provide an excuse for bringing the boy home late. In this instance, Piaget’s vivid memory was constructed out of having heard the story.
78 If we make an effort to recognize that we are dreaming--that it is only a dream--when we are dreaming, we can learn to do so--a condition known as lucid dreaming. One of my colleagues reported that, when he dreams, he always knows that he is dreaming.
79 S ymbolism necessarily includes condensation and displacement.
80 The cause of death was not clear.
81 Siegel (2003) suggested that REM sleep allows neuronal receptors to recuperate from the constant activation by monoamine neurotransmitters during waking, and that non-REM sleep allows repair of cell membranes from damage by free radicals generated by metabolism of cells.
82This idea is similar to what Antrobus (2000) refers to as his connectionist theory.
Why the dream ties these pieces together into a story is a question that Fabrice Clement raised when he read this book. Maybe it is as Piaget proposed. To be conscious, even in a dream, the idea must conform somewhat to dominant understandings. Similarly, Bleuler (1952), in his classic studies of schizophrenia, proposed that hallucinations and delusions were not primary symptoms. Rather they dealt with--made some sort of sense of--the primary disturbance of associations.
83 This is true except perhaps in NREM sleep at the times when we get no report, or during epileptic episodes when regular repeated high-voltage electrical spikes interfere with the ordinary waking activity of neuronal circuit activation, and with the communication among circuits.
84Oddly, if only one point of light is flashed in the neglected left visual field, the person will see it. If two points of light are flashed in the same field, he will see only the one on the right (Driver & Vuilleumier, 2001). If he turns his head or his body, this modifies what he can see in the neglected visual field. It is as if the intact left side of the brain has been programmed not to respond to information that would ordinarily come to the right side of the brain: Visual information, which would ordinarily go to the damaged right visual system, remains the property of the right side of the brain and is to be ignored.
Without such early programming, would we see two visions of each scene and hear two separate sounds from our ears, rather than be sensitive to sudden changes in the periphery of our vision, or to the bilateral air vibrations that strike our ears to recognize where the vibrations come from? This inhibition of functions, could be initially based on which impulses arrive first to a cortical region. This possibility is supported by experiments done on chicks. Before a chick is hatched, exposing an eye to light stimuli results in that eye being dominant (Vallortigara & Rogers, 2004). If that is the case, the inhibition should potentially be reversible, and could account for the improvements in stroke victims that result from restraint of the intact side, or from resection of the damaged area of the brain.
< Chapter 7 | Contents | Part II >
Do you have a comment or would like to open a discussion? The author would be pleased to hear from you. Contact AJ Malerstein here.
