THE BIRTH OF THE DREAMING SELF FROM NEURAL AUTOMATA
Fred Alan Wolf
So long as the dream lasts
consciousness is unable to engage in reflection.
It is carried along by its own decline
and it continues to lay hold of images indefinitely.
- Jean-Paul Sartre
I did not know whether I was Chuang Tzu dreaming I
was a butterfly;
or a butterfly dreaming I was Chuang Tzu.
- Chuang Tzu
We have just looked at how the brain is capable of generating holograms and how the information contained in them can be both stored in glial cells and recovered whenever a specific reference wave is excited in the cortex. We have also looked at the requirement of memory for self-awareness and how in a dream the sense of self becomes diffuse and expanded. Taking as an example of dream awareness the experiences of a scientist who had taken the mind-altering drug ketamine, we have seen that the sense of "I" is strongly connected to body awareness and volitional control of one's actions. With diminished body awareness and volitional control, the self appears to expand or seems to lose its bearings in time and space.
Well and good, but there is still a major problem in the model: Where is the beholder in all of this? No matter what mechanism is finally arrived at, whether it is a holographic model, as I discussed in the previous chapter, or a neural-network model, as I discussed in an earlier chapter, or whatever, the major problem is still, how does the observer arise in all of this neural meat?
As the quote from Sartre above suggests, and from the reports of the lack of volitional control leading to a loss "of the sense of ones self-reflection ability under the action of ketamine, we are led to the problem of self and the apparent loss or diffusion of it sensed in dreams."
In fact we are led to consider the same problem of "self" even in waking awareness. Dream researcher Gordon Globus suggests that there is no difference between unreflective awareness in waking life and in dreaming life. The difference in whether we are dreaming or awake seems to be if we are aware of ourselves while we are conscious.(1)
The major theme of this book, that we dream in order to develop the self, would imply that dreaming is a ground from which the self-concept arises. Thus we could ask, is there a self present when one dreams? The answer seems to be neither yes nor no, but a wide and changing spectrum of maybes, or, in other words, self is perhaps like an electron; it is not a thing with fixed boundaries and properties, but a quantum system capable of existing in different states.
Evidence for levels of self-awareness comes to us through dream research and, as you saw in the previous chapter, through altered-consciousness-state research. Lucid-dream accounts presented in chapter 12 indicate that there is self-awareness when one is dreaming lucidly. Dr. Jayne Gackenbach's research not only indicates this, but also, through her work with transcendental meditators and others, that it is even possible to be aware of self while in stages of deep sleep as well as dreams. She calls this the "witness state." Perhaps this witness state is what is being referred to by the experience of Chuang Tzu in the chapter epigraph -- one loses any sense of "real" self and is only capable of wondering about the existence of a self from a vantage point that seems beyond any space-time confinement.
The Growing Self: Stages of Lucid Awakening
Jayne Gackenbach, a tutor for Athabasca University and part-time instructor for both the Department of Educational Psychology and the Faculty of Extension at the University of Alberta in Edmonton, Canada, is one of the major scholars in the study of lucid dreams. She has written several books, numerous papers, and chapters in books in the rapidly growing field of lucid-dreaming awareness. This research indicates that what we call a self has a structure dependent on levels of awareness and that in dreams we have the opportunity to study this structure in a manner similar to how a physicist looks at atomic structure in a study of matter. The dream is the experimental landscape of the movement of the mind just as "out there" wake reality is the experimental landscape of the movement of the body. In much of what follows I have liberally taken from Gackenbach's work and added my own thoughts.
In her review article(2) Gackenbach provides the viewpoints of several noted lucid-dream researchers. Without exception they all suggest that lucid dreams provide us with a new understanding of the relationship of self to consciousness. What emerges from these studies suggests that consciousness has levels ranging from little or no sense of self all the way up to full cognition and awareness of self in the world, whether that world is the dream or the wake world. Gackenbach has also indicated that there are five higher levels beyond lucidity.
Thus just as one's physical growth from infancy is matched by a growth in one's self-awareness, a similar sequence of mental growth, self-structuring, and greater self-awareness may be occurring while we are sleeping. The dream becomes an opportunity for the evolution of consciousness. As such, the dream is nature's experimental honing of the edge of consciousness, allowing it to come to grips with reality in its fullest expression through the development of the self-concept.
Along these lines Alan Moffitt and his colleagues at Carleton University in Canada have devised a nine-level Self-Reflectedness Scale to measure self-awareness present in dreams, based on earlier work by E. Rossi.(3) Rossi believes that dreaming aids the growth of personality and the development of personal competence, which strongly parallels my position. Going over the Self-Reflectedness Scale, as presented by Gackenbach, I have added my own observations in terms of five levels based on Moffitt et al.'s scale and upon the stages of personality or psychological growth of an awake individual.
Five Levels in the Dream
At level 0 or the ground level(4) the dreamer is not aware of being present in the dream. I would equate this level to the waking awareness of an infant or perhaps even fetal awareness. At this level we have pure awareness with little sense of identification of self and other. I would label this the floating consciousness or universal consciousness.
At level I the dreamer becomes involved in the dream. This may be akin to early waking childhood experiences when the child begins to differentiate herself or himself from the rest of the world. I would characterize the level as playful, but not fully self-conscious. In this regard I have my own observations of gifted children in California grade schools where I consulted several years ago. Children at grade levels four to six show remarkable curiosity and at the same time the innocence that most of us find so charming.
At level 2 the dreamer is able to think over an idea. Perhaps this would equate to preteen years or prepubescent years. At this level we have the beginnings of self-awareness, often marked by awkwardness and shyness and the beginning of the powers of reason. I have witnessed this level in children from grades seven to nine.
At level 3 the dreamer is aware simultaneously of the previous levels of participation and observation during the dream. I would equate this period with puberty and growth to adulthood. When we reach this level in growth, we are able to reflect on ourselves and determine our presences and our effect upon others. We can also think over what we have done in the world.
At level 4 the dreamer consciously reflects on the fact that she/he is dreaming. This would be the lucid state and, in comparison to personality growth, would correspond to spiritual or mystical awakening. This level has never been experienced by a large majority of people. Most seem to be stuck somewhere in level 3.
Harry Hunt of Brock University in Canada also believes that lucid dreams are reflective of a growth in self-awareness.(5) He sees dreams in terms of a multiplicity of distinct, qualitatively different experiences, again scaled on degrees of self-awareness. The ability to wake up in a dream is remarkably similar to the development of self-awareness in meditative traditions. During meditation and lucid dreaming a detached but receptive awareness develops usually accompanied by a sense of great well-being, positive outlook, and joy. The fully awakened "I" is by far more able to enjoy life and feel a general expansiveness. The humdrum of everyday life takes on "color" and appears in a state of wonder and awe.
I have already described, in a previous chapter, the University of California researcher Charles Tart's concept of lucid waking, and in this respect we also find agreement between Tart's concept and that of the other lucid-dream researchers, that lucidity indicates greater self-awareness.
Gackenbach believes that lucidity, when it occurs naturally, is a sign of the emergence of greater self-awareness and a manifestation of a higher state of consciousness. On the other hand, Stanford University researcher Stephen LaBerge believes that lucidity is achievable through training and, as I understand his work, does not necessarily indicate any growth in consciousness. It is just another "schema" (organizing principle) produced as are all other schemas by being raised up from what he calls a recognition threshold. In essence, lucidity is a cognitive tool that most individuals do not develop simply because they have a conceptual barrier against using this tool. They don't believe in it or find it necessary.
For Gackenbach lucidity is not the final point in the evolution of awareness. It is the starting point for an even higher level, reported to her by practitioners of transcendental meditation, called "witnessing," in which one passes through five additional "stages." During lucid dreams one is aware that one is dreaming, and one is still very much contained within a dream boundary of skin and body. There is a dream ego, as it were.
During witnessing, anew state of self-awareness is present. The dreamer becomes aware of greater detachment from the drama of the lucid dream. Choice remains, however, to enter into the dreaming persona or to step back and simply witness. Emotional content then withers. It is difficult to describe the witness experience. One is separate from the dream, and one does not really care what happens in the sense of having emotional charge.
Charles Alexander from the Maharishi University in Iowa and his colleagues have indicated that witnessing can occur not only in lucid dreams but also during deep sleep or in any other state of consciousness, and is therefore a fourth state of consciousness.
Alexander in a later article describes the difference between an ordinary lucid dream (OLD) and a lucid witness dream (LMD).(6) During an OLD the cognitive capacities normally present during waking are activated, and one functions volitionally from within the dream world. However, one's awareness is tied to a dream ego, a person that you recognize as yourself in much the same way that you recognize yourself while awake. Just as you are in the world and have your cares and woes, a similar situation arises in the OLD; you are absorbed by. the dream.
During a LWD one is amazingly conscious, but the objects of that consciousness are "content free," devoid of feeling or word descriptions, emotional charge, or any limitation. In fact this state of lucid witnessing can be present during an LWD or in waking consciousness, deep sleep, or meditation.
Gackenbach's five stages are marked by a growth from lucid awareness to, surprisingly, a similar stage as found in level 0 described above. The most descriptive feeling of this stage is one of unboundedness, expansiveness, a sense of light awareness. The sense of self in a body again is gone, but one is aware of this, while in level 0 the same sense of non-differentiation of self and other is present, but one is not aware.
Physical Correlation of Self-awareness in Dreams
Experimental data on lucid dreams comes to us primarily through the study of REM. LaBerge characterizes the Lucid state as one of high-density REM where the central nervous System (CNS) appears highly activated. LaBerge believes that this activation indicates a higher level of cognitive power.(7)
Studies by Ogilvie, Hunt, and associates also show an increase in alpha-wave generation during lucid dreams.(8)
Gackenbach has also discovered that during lucidity and meditation a higher level of EEG coherence occurs. She and researcher Tom Snyder have shown that during sleep the vestibular system within the brain stem is in some specific way associated with lucid dreaming.(9) This system is known to be important in spatial orientation during wakefulness. They conclude that lucid dreaming in terms of vestibular/spatial perspective is evidence of a continuing evolution of the brain system to incorporate spatial bodily movement. Such movements have been remembered or stored in the brain and thus have become incorporated in the brain's growth for perhaps millions of years. Memories of bodily movement patterns exist as archetypes, dance, mythology, and poetry.
Further supporting this is the remarkable observation of J. Allan Hobson, discussed in an earlier chapter, that in dreams we are usually continually moving. During sleep the decrease in sensory input and motor output along with increased cortical activity during REM causes a greater reliance on internally generated spatial-temporal reference frames, which will be useful during wake for spatial exploration.
Gordon Globus uses the concept of the mechanical connectedness of neural networks to account for both ordinary dreams and lucid dreams. (10) Accordingly, knowledge is carried or stored in the multiple connections between neurons in the brain. As such the brain is "constrained" or limited in the number of ways in which it can respond to any given stimulus. In chapter 7, I explained how neural networks worked in going over Hopfield's neural-network model and Crick's model of dream consciousness.
The goal of the brain is harmony, and given an input, it will suffer its remembered constraints, built up over early childhood development (see chapter 6 discussing Winson's work) and through evolution, to respond in any way it can to produce that harmony. There are many relatively stable states; one of them is dreaming while we are sleeping. I already mentioned the peculiar abhorrent states of neural nets known as fantasy, hallucination, and obsession. These, too, are relatively stable states, mini-valleys in the mountainous regions of the mind. If you remember, these states were wiped out using a negative biasing input to the connections between the "neurons" in Hopfield et al.'s model. (11) In their model a certain "energy" function was used.
This energy function appears to have a range of highs and lows, "energy" mountain peaks and valleys, distributed over a map where specific locations on this memory cartographic landscape are specific stable memory states of the network. With random input, the system jiggles and settles just as a pinball dancing on a pocketed surface of a pinball machine eventually lands in a pocket. The deeper and wider the pocket, the more likely the pinball will land there. This is akin to the network reaching a specific memory from a given random input.
The negative bias added to the connections tended to raise the floor of the pockets and narrow the widths of the spurious memories, so that jiggling tended to produce wanted or desired memories and the unwanted spurious states tended to vanish.
Globus sees the dream state in a similar manner. For him the mind has many possible states, just as an atom has many possible energy states, some more probable and some less. This mind does not and in fact cannot follow a causal rule. It operates by attempting to tune itself and thus arrive at a memory state. Dream generation is no different in this regard than memory recall during wakeful awareness when we try to cognate on our sensory experience. The difference between lucid and ordinary dreams is for Globus a question of tuning. Deliberate control or intent requires a constraining, highly tuned operation much like tightening a violin string so that it is constrained to vibrate at a high frequency. Meditation and lucidity are achieved by detuning. Yet in spite of all of this, the awareness of self is not lost.
What Globus calls tuning I have described in chapter 15 as the operation of intent. Intent, remember, manifests from vigilant observations, which result in constrained paths of evolution. The power of intent arises in a natural way from the self. The greater the self-development the higher is the power of intent. Thus to have greater intent and therefore greater ability to manifest change in the physical world according to desire, one needs to have greater self-awareness. This in turn arises from heightened processes of self-reflection. The amusing paradox of all this is that with heightened self-awareness, there is greater compassion and wisdom so that personal desires become less important. One dances in the dream like Siva.
Self-reflection in the Dreaming Neural Net
Is there anything in quantum physical terms that would help us to understand the phenomena of the levels of self appearing in dreams? I suggest there is, but a workable model has not been achieved to date. It is this that I will attempt here. My goal is to see if I can make a quantum physical model of levels of self-awareness. Although the model is based on principles that follow the rules of quantum mechanics, I also caution the reader that what follows needs to be read as speculation.
Let me first put the basic idea of the model into words. The key insight into self-awareness arises from the ability of a simple memory device, an automaton in the brain, to obtain images of holographically stored glial-cell memories and, most importantly, to also obtain images of itself. Each self-image is composed of a quantum-physical super- position of primary glial-cell images and the automaton containing those images. By superposition I mean that the images are added together producing a picture much like what you would see when looking at a doubly or multiply exposed photograph. Quantum physics shows us that we need these "multiple exposures" or super-positions in order to understand the simplest atom.
Included in the superposition is, however, something new although not forbidden by quantum rules. While the automaton "photographs" and remembers an image, it also photographs itself when it has an image. This is called a self-reflection. These self-reflective images are ordered according to a hierarchy based on levels of self-inquiry. Thus higher levels of the automaton's self-awareness are achieved by integrating images of itself on lower levels of the hierarchy.
Each level of the hierarchy can be viewed as a "reflection" of information obtained by self-inquiry on a lower level. When such a reflection occurs, we say that the device has conducted an inquiry, and as a result, a jump from a lower to a higher level takes place as the new self-reflective image becomes part of the device's record of previously obtained self-reflective images. Thus each jump upward integrates images from all of the lower levels, resulting in a sequence of complex images beginning with the lowest, simplest images and ending with the highest, most complex images.
These levels are distinguished by certain forms of information obtainable from lower levels by the self-inquiry process. The lowest or "zero" level consists of non-self-reflective images existing in the glial cells. Many images are possible including super-positions of images. According to the uncertainty principle of quantum physics, it is not possible for the automaton to obtain a single image and a superposition of images of an object at the same time and hold them both within its memory. This would be similar to attempting to observe the position and the momentum of a subatomic particle at the same time. Thus the automaton can obtain either a single image or a superposition of images, but not both.
I theorize that at the first self-reflective level, glial cell images and the records of the automaton containing those images are superimposed quantum physically, resulting in defined, bounded, emotional memories. At the second level, these emotional memories are integrated into thought forms. At the third level, these thought integrated into archetypes. At the fourth level, these archetypes are integrated into super-archetypes. In principle the process is never-ending. I can only guess that at the highest level one would achieve something called "pure" consciousness or "God" consciousness.
In the next section of this chapter I will go over this in greater detail. This may be difficult sledding for nonscientists. I have tried to put most of the mathematical thinking about this in footnotes, leaving the text as free from mathematics as possible. Although the model is based on quantum mechanics and may be difficult to follow, you should get some of the flavor of it. I suggest that the reader read as much of the next section as is comfortable and skip over anything too mathematical. I will summarize the ideas and apply them directly to dreams at the end of the section.
The Physics of Self-awareness
To keep this as simple as I can, I will assume that there are only sixteen primary images. In actual fact there are probably thousands of such images in the glial-cell memory. These sixteen can be superimposed into eight pairs (120 different pairings) where each pair constitutes a secondary image. These eight secondary images can be superimposed into four tertiary images (28 different pairs) and so on.
These images are states of quantum physical observables and must follow quantum rules of superposition. Consequently any level of composition of a superposition of states will be complementary to all other levels.(12) Thus the secondary images are complementary to the primary images and tertiary images. The tertiary images are complementary to both the secondary and primary images, and so on.
Although a single automaton cannot hold simultaneously multiple images consisting of complementary observations of another system, it can do so for complementary observations of images built up self-reflectively.(13) In other words, if the images contain the observer as well as the observed, the automaton can remember them. Images of objective "out there" observations cannot be held simultaneously because the simultaneous knowledge of objective complementary observables is in violation of the uncertainty principle. Thus any attempt to do the same thing with "objective" images is not possible.
In a very real sense an object in a state of self-reflection can hold both a truth and its opposite at the same time without paradox, while any attempt to determine the same thing about another object results in uncertainty or doubt. In brief, if I may jump anthropomorphically from the automaton to you, the more you can accept about yourself` and your shortcomings as well as your strengths, the higher is your level of self-awareness.(14)
To build the model we first need to consider the automata of the neural network in quantum physical terms. They are not just on-off elements. Only by allowing them to be quantum physical in their operation are they capable of self-reflectivity, That is, something in their quantum nature enables them to differentiate states that are self-generated from states that are based on observation of outside "systems.
Let me back up a bit. We also need to understand what is meant by the ordinary usage of the term automaton. This is simply a device that is capable of "remembering" information that it observes or measures. To reiterate: our quantum automata are different from ordinary automata. Each quantum automaton turns out to be able to (1) distinguish the states of an outside system and (2) determine the states of its own memory system. In simultaneously knowing states of itself and of another system, it has both the knowledge that "it knows that it knows" and "it knows" as distinguished from just the knowledge that "it knows."
The basic idea for this quantum automaton is not new or original with me. Physicist David Z. Albert in 1983 and in 1986 published two papers explaining how such automata might be constructed and used as computer memory elements.(15) Let me first present briefly Albert's ideas and then indicate how they might relate to the dreaming brain, specifically in regard to the problem of the self and how its different levels could arise.
Based on the parallel-worlds interpretation of quantum physics,(16) Albert indicates that certain types of memory devices called quantum automata can exist that are capable of remembering states of physical objects that they encounter, say, appearances of the objects in space, and self-reflective states as well. In other words, these devices can store information that includes the objective qualities of observed objects and at the same time qualities of themselves when they are "reflecting" the objects that they have encountered.
But when these automata attempt to observe those very same qualities in other automata, this attempt to "observe" those qualities disrupts states of the other automata. Thus it is made impossible for a second automaton to obtain some types of information from a first automaton without altering that information. This disruption may be, at the quantum physical level, the cause of disruption of scenes and dream characters during dreams. I shall have more to say about this later.
In other words, these automata can hold and observe information about themselves, without disrupting themselves, that they cannot determine in others without disrupting them 'and thereby changing their memories and causing them to "jump" into a different and usually random state.
I suggest that this self-reflective aspect of quantum automata, which exists as a necessary consequence of the parallel-worlds interpretation of quantum theory, is a deep clue to our self-conscious nature and that the dream is a laboratory in which the differences between self-reflective and unreflective perception (non-self-reflective) can possibly be measured.
Ordinary computer automata contain a means by which some bit of information can be obtained, stored, and a means by which they are capable of giving that bit of information to another device. They follow a set of rules called Boolean logic. Each ordinary automaton is therefore capable of remembering a single bit of data, 0 or I, for example. (In Hopfield et al.'s model the automata are called neurons and have states +1 and - 1.) We can think of this data bit as signifying the answer to a question: "0" would mean "no" and "I" would mean "yes." Or "I" could stand for "true" and "0" for "false."
The automata that Albert describes do not follow Boolean logic, nor are they able to causatively pass all of their information along to the next fellow. Because each automaton is quantum-mechanical in its operation, it can not only "remember" if a system it interacts with is "true" or "false" but can also keep in memory a superposition of both answers.(17) We might call this a "maybe" state.
It Is this "maybe" state that makes quantum-mechanical automata so curious and capable of remembering not only states of systems that exist outside of themselves but also states of themselves. At this point I wish to explain how this applies to dreams and consciousness in some detail using Albert's automata. However, I again caution the reader that the use I put them to and the ideas expressed about them are my own speculative thoughts.
Suppose that the object that the automaton interacts with, and thereby obtains a measurement of, is an ionic wave pattern holding a holographic image in a glial cell as described in the previous chapter. The problem is that this wave pattern can be obtained in several complementary ways depending on what information the automaton wishes to extract from the cell. Now suppose that the wave contains a holographic record of a superposition of images.
To put this in psychological terms, suppose that the superposition consists of sixteen specific "woman" images. Eight pairs of these images compose emotional states associated with women. As the pairs of images are "not real" but super-positions of "real" people, the result of remembering a specific pair produces a memory of an emotion, not the emotion itself. Four pairs of emotional images produce images of thought forms, not the thought forms themselves as experienced. Two pairs of thought forms produce the images of archetypes, not the archetypes themselves. A single pair of archetype images produces a super-archetype image. With a greater number of primal images, even higher super-positions could be generated.
We might think of the whole superposition as the super-archetypal "goddess" image and each primary image consisting of women that the person has met, including the person's mother, girlfriends, sisters, spouses, etc.
Any single image will correspond to a specific quantum physical state, while the superposition of images would also correspond to the complementary physical state of an emotion, thought form, archetype, or super-archetype. The automaton could measure and thereby obtain various combinations of images in groups of two, three, four, or more of these images. It could obtain a single image or all of the images together, thus fetching the complete super-archetype.
Suppose we label each image by the symbol "W1." "W" stands for "woman" and the "1" is a simple counter index. So W1 is woman number 1, perhaps the subject's mother, while W2 might be the subject's sister, and so on.
Suppose also that the superposition of two images, say W1 and W2 is also an image that we label E1. To make this a little dramatic, and perhaps reflective of some people's memories, let this superposition reflect some aspect of the person's life while living with these two women, mother and sister. Of course there are many images in a person's memory. We are only looking at two specific images. Perhaps the sister image reflects a younger crying child and the mother image reflects a hysterical woman. Now there is no remembered emotional content to these separate images. They are just pictures.
But the younger crying sister image, W2, and the hysterical mother image, W1, when taken together create the emotional image "unhappy woman," which we have symbolized by the letter "E1" (where "E" refers to emotional state). There may also be an E2 image consisting of the super-positions of, say, the first girlfriend image, W3 and the second girlfriend image, W4. These latter images taken together composing E2 might reflect another emotional state that the person observed in woman, say, a joyful woman image.
Other images could exist. A superposition of the two emotional images E1 and E2 would makeup a thought-form image, F1. Since this Image consists of both the unhappy and the joyous emotions, it could represent a state of feminine or motherly understanding. A superposition of thought-form images, say F1 and F2, would represent an archetype "goddess" image, G1, and a superposition of archetypal images, G1 and G2, would stand for the super-archetypal "woman" image, S1.(18)
Let us assume that an automaton can remember and record a single "woman" image, i.e., mother, sister, girlfriend I, girlfriend 2, etc. It does this because there is, according to the rules of quantum physics, a possible observable image of any of these women in the memory of the subject, i.e., as a record stored in the glial cells(19) Now which woman is remembered cannot be determined unless we know just what is contained in the glial cell.
How this is accomplished is the province of the automaton. It (and I haven't said yet what "it" consists of, but I will) interacts with the glial cell and obtains via this "measurement" the single image W1, provided that the glial cell had just this image. Once this interaction has occurred, the automaton and the cell are said to be coupled together, and the information contained in the glial cell is now also a record of the automaton.
But suppose that the glial cell contains a superposition of two woman images, W1 and W2, corresponding to the state of emotion E1. What happens if the automaton interacts with that cell?(20) Then, if it is under the instruction to obtain a woman image and according to quantum rules it can only obtain a single image, W1 or W2, it will do so, obtaining one or the other at random.
At least it would do so according to the Copenhagen interpretation of quantum physics in which only one image is allowed to be obtained. Only one of the above "women" images will be determined, and the other will mysteriously disappear. Lost forever, it would vanish from any memory record. This disappearance of all other images except one is called the "wave-function collapse" in the jargon of quantum physics. No one yet knows how this mysterious occurrence takes place.
A collapse of a wave function doesn't follow the rules of quantum mechanics. This collapse has never been observed, and in fact, just the opposite has been observed in all cases where an attempt has been made to measure it. This opposing observation shows that after the measurement, both possible images are still present as if they were in separate "worlds."
Now we come to the parallel "worlds" idea of quantum physics. In this model, in the case where the glial cell contains an emotional memory record, E1, but the automaton attempts to determine which woman, W1 or W2, is present, the automaton itself enters two possible "worlds." In one "world" it obtains the image record W1, and in the other "world" it obtains and records the image W2.
The idea here may sound like science fiction. But the use of the term world does not necessarily refer to a completely different world, but may only be a separate state or level of both the automaton and the glial cell when taken together. In the parallel "worlds" interpretation the superposition of the automaton and the glial cell in the two worlds is also a quantum-physical state, which means that it is, itself, a possible memory record capable of being measured by yet another automaton or even by the automaton itself.(21)
The fetching-of-an-image interaction alters the glial cell. Before it contained the record E1, but now, in one world it contains a single image, W1, and in another world it contains the single image W,. This is exactly what quantum rules predict even though this prediction sounds bizarre. It is the essence of the quantum rule "to observe is to disturb."
Now we come to an interesting aspect of all this. Remember, the glial cell physically exists in the brain and so does the automaton. It is just that as a result of their interaction, they are coupled together, and if interrogated separately about their memory contents, each would come up with a consistent story. In one version of the story (world II), the glial cell would yield a crying sister image and simultaneously the automaton would yield the same image. In another version of the story (world 1), the glial cell would yield the image of a hysterical mother and the automaton would likewise yield the same image.
How would such an interrogation occur? There are two ways: (1) interrogation by a second automaton, and (2) self-interrogation. And in each way of interrogation there are two complementary questions that can be asked: (a) What image exists separately in the glial cell and in your memory? This question is about the W, states. Or: (b) What image exists compositely? This question is about the emotional composite state of both glial cell and automaton.
Consider 1a, a second automaton brought in to attempt to determine the woman image in the glial cell and then compare it with the record in the first automaton. It finds that the identity image of the first automaton and the glial-cell record match exactly. How can this be? Because according to the quantum rules, it, too, will enter each of the parallel worlds and in each world no inconsistency will be found. For there is no world where the second automaton records the image W1 while the glial cell contains the image W2 and vice versa.
Or the second automaton can ask about the emotional state of the first automaton and the glial cell taken together. In that case it will know that there is an unhappy woman image present, but it will not know which image is present.(22) If it attempts to determine the woman image, it loses its knowledge of the emotional state. The image and the emotional states are complementary observables for the second automaton.
So everything fits together consistently. If a billion automata come along and do the same thing as the second automaton does, then they, too, will either enter into each of the two parallel worlds and find that a single image of woman is present, or they may simply agree with the second automaton that the system of the first automaton and glial cell are in an unhappy woman emotional state.
But now we are going to ask the first automaton to do what we asked of the second in lb above. That is to record and remember something that, it turns out, could not be done by an outside automaton. This record is the "emotional" state superposition of itself and the glial cell in both worlds taken together.(23) Now this may also sound strange. But before we get to this self-reflective state, consider a simpler question.
How could an automaton, in the first place, record a superposition of two images in the glial cell? It is here that we go back to the original concept as I first described it. I told you that the glial cell could contain a record of a single woman image or a pair of images, or three or four, or for that matter a superposition of all of the woman images taken together, the so-called super-archetype image, S. I told you that various combinations of woman images produced different emotional states associated with women.(24)
If the glial cell contained just the two woman images of mother and sister superposed, the automaton could have been asked to obtain the record of the emotional state, E. It would have found the state E1, the unhappy woman record. It is perfectly capable of obtaining either of these complementary records, emotional, E, or identifiable, W, women states.
But now we have asked it to look at its own state after it has been asked to find an identifiable woman image, W. Let's look at this state just before the automaton observes it itself. This state is a superposition of itself and the glial cell when each part of that superposition is in a 'single woman state. Taken together the combination reflects an emotional state of memory, but not of the "unhappy woman" state, which just refers to a superposition of images in the glial cell taken alone, but of itself remembering two separated-identity woman states each coupled to a glial cell memory in a separate world.
Surely this superposition reflects an emotional state, but what state could it reflect? We can guess that this state would be an emotional state associated with unhappiness. But who is unhappy? So far, no one is. The memory state of unhappiness exists, but until the automaton records that state in its memory, the automaton knows nothing about it and nothing emotional is experienced.
Now let the automaton look at itself.(25) It is here that we have the first level of self-reflection. The unhappiness is now "felt" by the combination of glial cell and automaton. Furthermore the automaton "knows that it knows" and "knows" at the same time. Now an experience and a memory are known by the automaton. In each world an image exists and a self-reflection of emotion exists that is due to an image existing in another world.
It is this self-knowledge that it can exist simultaneously in two worlds with separated images of mother and sister(26) that gives rise to the sense of "self." If the automaton attempts to interact with itself again and measure its emotional state, it will again find that it is the same "unhappy" state. And if it were to interrogate itself regarding which woman is unhappy, it would find that in each world, a single woman, either the mother or the sister, was unhappy. Thus it would contain records of itself in each world and of itself in both worlds together.
In each world the automaton contains a record establishing that it has a woman image and a feeling of unhappiness that it will attempt to objectify and associate with the image. However, it cannot completely separate the feeling from itself and identify that the feeling is due entirely to the image because the unhappiness is associated with its memory containing an image in another world. Undoubtedly an aura of uncertainty surrounds this state since it is a memory of something real and "unreal" at the same time as far as the automaton is concerned in each world. The other world for it reflects upon its world and in doing so provides a recognition of itself via this reflection, label this as a "level one" or a primary "unit" of self-reflection.
Suppose a second automaton attempted to obtain the total information found in automaton 1. Could it do so without altering the records of the first automaton? The answer is no. The attempt of the second automaton to obtain records of the woman image would put it into both parallel worlds where the separate woman images were held. This would alter the combined state of the first automaton and glial cell. The attempt to determine the emotional state of the first automaton and glial cell could also alter the memory of the first automaton's record. Thus automaton 2 could change the first automaton by attempting to find out what automaton I knows about itself.
This is an interesting aspect concerning asking and knowing.(27) The state of the system is unchanging so long as no one makes any further inquiries. However, to get information from the automaton someone has to ask. In the asking comes the inevitable disturbance regardless of who asks. The second automaton alters the system in inquiring, and even if the first automaton asks of itself, "What's going on?" the system is also altered. Thus the mere asking of a question changes the memory.
The only difference is not in who is asking, but in who is knowing in the first place. The ability of an automaton to know information depends on the identity, as it were, of the automaton. Taking this in human terms, you may thus know things about yourself that cannot be determined by an outside observer, but don't ask. It is this self-knowledge state consisting of information taken from two worlds that is peculiar to quantum automata and does not exist in classical memory elements. I suggest that at this remarkably simple level, we are seeing the origin of a delicate and primary self-concept. It comes about via secret knowledge, awareness of something that actually exists in parallel worlds simultaneously known only to the participator in both worlds.
Now suppose we consider the other "woman" images. Just as the unhappy woman image E1 was a superposition of hysterical mother W1 and crying sister W2 there is also an emotional state "happy woman," E2, consisting of images W3 and W4, taken together.(28) Thus in a similar manner there are thought-form memory states of the combined automaton and glial cell that can be created in the same way that the unhappy emotional state was created.(29)
In the overly simplified example taken above, there are four self-observations for the automaton in a hierarchy: super-archetype, archetype, thought-form, emotion, and the last, which is not self-reflected - identity. Following this line of reasoning with greater numbers of primary images, we could find a vast hierarchical number of self-reflection states. In the highest state, in each parallel world, a greater sense of self exists. The more images there are, the higher is the level of self-awareness possible, provided that as the automaton rises through the levels it does so by stopping at each level and taking in its own state of awareness as it exists in not only the world it happens to be in as determined by the observation of the glial cell (the lowest level of observation), but by taking into itself its memory states of itself in the other parallel worlds. These hierarchical images constitute an ever-growing and evolving self-awareness since they involve an increase of more and more complex self-reflecting images.
During an ordinary dream, processing would be taking place at the lowest levels of image production wherein no sense of self is present. This would correspond to the formations of non-self-reflecting states, level 0, where the dreamer is not aware of being present in the dream and the dream is marked by a sense of pure-awareness with little sense of identification of self and other. The images would be seen, but not experienced, since no self has been defined. This would constitute unconscious data processing. I say unconscious, meaning un-self-reflecting. The images would be there but they would have no meaning.
At level I the dreamer becomes involved in the dream. This would correspond to jumps between level 0 and level I and the formation of images mostly in the lowest level. When level I was reached, emotions would be felt during the dream, but as soon as the dreamer descended levels, only images would be present. By jumping between level I and 0 various emotions could be aroused. Meaning is also arising when the self appears. Moments when this occurs are seen as enlightened or aware. For in the moments when level I is achieved, the dreamer not only sees the images, she or he also possesses feelings about them that can be associated within space and time boundaries. These are perhaps primary body images. The first bodily images would be emotional in this model. Perhaps this is a clue to illness arising from emotional causes that probably took place during early childhood.
At level 2 the dreamer is able to think over an idea. Here thought forms arise, and jumps between levels 0, I, and 2 occur, resulting in a range of states of thinking, feeling, and observing. Thought adds much greater meaning to the emotions and the observations. Just as images come from sensory inputs, thoughts and feelings are capable of out-putting in terms of expression of words and feelings. Thought forms integrate or superimpose emotions and thus tend to have no emotional content per se. However, as thoughts are expressed in words, emotions can and often do arise, sometimes unexpectedly. This would be due to a jumping from level 2 to level I as a result of perhaps unconscious inquiry. Remember that asking about one's state of awareness in this model alters the state. Thus when words move us to emotional action, we are descending to a lower level of self-awareness.
At level 3 the dreamer is aware of the previous levels of participation and observation during the dream simultaneously. Here the sense of self more fully emerges as the dreamer deals with archetypal images as experienced by the dreamer. Again there is full access to the lower levels.
At level 4, the dreamer consciously reflects on the fact that she/he is dreaming. This would be the super-archetype state corresponding to lucid awareness during the dream.
At each level the device, for which, as you have noticed, I have not indicated its identity or materials, is able to focus on images from lower levels or on images of the highest level it is capable. The balance is quite delicate as I see it, and the tendency would be to descend levels more readily than to ascend them. Descent results in less self-awareness and therefore more automatic, mechanical behavior. Ascent results in greater choices, becoming aware of existence in other "worlds," and more complex imagery with a higher number of paradoxical features simultaneously knowable. Remember, any attempt to observe these images coming from different levels by another automaton destroys the images. It is only through the passive awareness within oneself that simultaneous knowledge of all of the images is possible.
Now what is this automaton? How big is it, and from what types of matter is it made? I have no answer in terms of matter and energy, so I am embarrassed as a scientist to say that the automaton is not a material device. The automaton is "consciousness," and what we have looked at is how consciousness operates in creating the "I" experience inside these "tubes of meat" we live in. This may be the first attempt to model what many mystics have known for a long time, the structure of consciousness in terms of levels of self-awareness.
Of course there would be higher levels requiring the formation of more complex images and, as I would imagine, greater mind-ability. This would be reflected in neuro-physiological data corroborating the observations of lucid-dream researchers.
At the highest level, possibly reachable by training as LaBerge suggests, or through meditation techniques as transcendental meditation practitioners indicate, we reach a state of "pure" awareness with surprisingly no images present. This may come about through some form of focusing on the highest-level image. What this image could be, I can only guess. We could call it God awakening from the dream.
The Dreaming Universe