Towards a fundamental theory of consciousness

    The origin and nature of conscious experiences - sometimes called the Latin word qualia - has been a mystery to us from the earliest antiquity until recently. Many philosophers of consciousness, including modern ones, consider the existence of consciousness to be an unacceptable contradiction to the fact that, in their opinion, there is a world of matter and emptiness that they declare it an illusion. In other words, they either deny the existence of qualia in principle, or claim that they cannot be meaningfully studied using science.

    If this judgment were true, this article would be very short. And under the cut there would be nothing. But there is something there ...


    If consciousness cannot be comprehended using the tools of science, it would only be necessary to explain why you, I, and almost everyone else are so sure that we generally have feelings. However, a bad tooth caused me to flux. A sophisticated argument to convince me that my pain is illusory will not save me one bit from these torments. I do not sympathize with such a dead-end interpretation of the connection between soul and body, so I will probably continue.

    Consciousness is all that you feel (based on information from the sensory sensory organs), and then you experience (due to perception and understanding).

    A melody stuck in my head, the taste of a chocolate dessert, a boring toothache, a love for a child, abstract thinking and the understanding that one day all sensations will end.

    Scientists are gradually approaching the solution to the mystery that has long worried philosophers. And the culmination of this scientific research is expected - a structured working theory of consciousness. The most striking example of the application of this theory is a full-fledged AI (this does not exclude the possibility of the emergence of AI without a theory of consciousness, but on the basis of existing empirical approaches in the development of AI)

    Most scientists take consciousness for granted and seek to understand its connection with the objective world that science describes. A quarter of a century ago, Francis Crick and other cognitive neuroscientists decided to put aside philosophical discussions about consciousness (which had been worrying scientists since at least Aristotle's time) and instead set off in search of its physical traces.

    What exactly in the extremely excitable part of the brain substance gives rise to consciousness? Having learned this, scientists can hope to come closer to solving a more fundamental problem.
    In particular, neuroscientists are looking for neural correlates of consciousness (NQF) - the smallest neural mechanisms that are collectively sufficient for any particular conscious experience in sensations.

    What needs to happen in the brain for you to experience toothache, for example? Do some nerve cells have to vibrate with some kind of magical frequency? Do I need to activate some special "neurons of consciousness"? In what areas of the brain could such cells be located?


    Neural correlates of consciousness

    In determining the NQF, the “minimum" clause is important. After all, the brain as a whole can be considered NKS - it gives rise to sensations every day. Nevertheless, the location can be designated even more accurately. Take the spinal cord, a 46-cm flexible tube of nerve tissue inside the spine that contains about a billion nerve cells. If, as a result of an injury, the spinal cord is completely damaged up to the cervical zone, the victim’s legs, arms and body are paralyzed, he will not be able to control the intestines and bladder and will be deprived of bodily sensations. Nevertheless, such paralytics continue to know life in all its diversity: they see, hear, smell, experience emotions and remember as well as before the tragic incident radically changed their lives.

    Or take the cerebellum - the "little brain" in the back of the brain. This brain system, one of the oldest in the evolutionary sense, is involved in the control of motility, body position and gait, and is also responsible for the clever execution of complex sequences of movements.
    Playing the piano, typing on the keyboard, figure skating or climbing - all of these activities involve the cerebellum. It is equipped with the most famous neurons called Purkinje cells, which have antennae, fluttering like a coral sea fan, and conceal complex electrical dynamics. And the cerebellum contains the largest number of neurons , about 69 billion (mainly star-shaped cerebellar mast cells ) - several times morethan all other parts of the brain combined (remember - this is an important point).

    What happens to consciousness if a person partially loses a cerebellum as a result of a stroke or under a surgeon’s knife?

    Yes, almost nothing critical to consciousness!

    Patients with such damage complain of several problems, such as less fluent playing the piano or less nimble typing on the keyboard, but never the complete loss of any aspect of their consciousness.

    The most detailed study on the effect of cerebellar damage on cognitive function has been extensively studied in the context of post-stroke cerebellar affective syndrome . But even in these cases, in addition to spatial coordination problems (above), only non-critical violations of the executive aspects of management, characterized by perseveration , distraction and a slight decrease in learning ability, were established.


    The extensive cerebellar apparatus is not related to subjective experiences. Why? An important clue is its neural network - it is extremely uniform and parallel.

    The cerebellum is almost completely a chain of direct distribution: one row of neurons feeds the next, which, in turn, affects the third. There are no feedback loops that resonate back and forth within the framework of electrical activity. Moreover, the cerebellum is functionally divided into hundreds, if not more, of independent computing modules. Each of them works in parallel, with separate input and output that are separate and not overlapping each other, which control movements or various motor or cognitive systems. They hardly interact with each other, whereas in the case of consciousness, this is another indispensable characteristic.

    An important lesson that can be learned from the analysis of the spinal cord and cerebellum is that the genius of consciousness is not born so easily anywhere in the nervous tissue. Something else is needed. This additional factor lies in the gray matter, which makes up the notorious cerebral cortex - its outer surface. All available data indicate that neocortical tissue is involved in the generation of sensations .

    You can narrow the area of ​​the focus of consciousness even more. Take, for example, experiments in which the right and left eyes are exposed to various stimuli. Imagine that a photograph of Lada Priora is visible only to your left eye, and a picture of Tesla S is visible only to your right. It can be assumed that you will see some new car from the overlays of Lada and Tesla on each other. In fact, within a few seconds you will see Lada, after which it will disappear and Tesla will appear - and then it will disappear, and Lada will appear again. Two pictures will succeed each other in an endless dance - scientists call this binocular competition, or rivalry of the retinas. Ambiguous information from the outside enters the brain, and it cannot decide: is it Lada or Tesla?

    If at the same time you are lying inside a tomograph that records brain activity, scientists will note activity in a wide range of cortical zones, collectively referred to as the “posterior hot zone”. These are the parietal, occipital, and temporal regions of the back of the cerebral cortex, and they play the most important role in tracking what we see.

    It is interesting that the primary visual cortex, which receives and transmits information from the eyes, does not reflect what a person sees. A similar division of labor is also observed in the case of hearing and touch: the primary auditory and primary somatosensory cortex do not directly contribute to the content of the auditory and somatosensory experience. Conscious perception (including images of Lada and Tesla) give rise to subsequent stages of processing - in the back hot zone.

    It turns out that visual images, sounds and other vital sensations originate within the posterior cerebral cortex. As far as neuroscientists can judge, almost all conscious experiences originate there.


    Awareness Counter

    For operations, for example, patients are anesthetized so that they do not move, maintain stable blood pressure, do not experience pain, and subsequently have no traumatic memories. Unfortunately, this is not always possible to achieve: every year, hundreds of patients under the influence of anesthesia are more or less conscious.

    Another category of patients with serious brain damage due to trauma, infection or severe poisoning may exist for years without being able to talk or respond to treatment. To prove that they feel life is an extremely difficult task.

    Imagine an astronaut lost in the universe who is listening to the attempts of the flight control center to contact him. A failed radio does not broadcast his voice, which is why the world considers him missing. Something like this can describe the hopeless situation of patients whose damaged brain deprived them of contact with the world - a kind of extreme form of solitary confinement.

    In the early 2000s, Giulio Tononi of the University of Wisconsin in Madison and Marcello Massimini first applied a method called zap and zip to determine if a person was conscious or not.

    Scientists applied a coil with wires in the shell to their heads and sent an electric shock (zap) - a strong charge of magnetic energy that caused a short-term electric current. This excited and slowed down the partner cells of neurons in the connected areas of the chain, and the wave resonated along the cerebral cortex until the activity died out.

    A network of electroencephalogram sensors mounted on the head recorded electrical signals. As the signals gradually propagated, their tracks, each of which corresponded to a certain point below the surface of the skull, were transformed into a film.

    The entries did not demonstrate some typical algorithm - but they were not completely random.

    Interestingly, the more predictable the flashing and fading rhythms were, the higher the likelihood that the brain was unconscious. Scientists measured this assumption by compressing the video data using an algorithm that archives computer files in a ZIP format. Compression provided an assessment of the complexity of the response of the brain. Volunteers who were conscious showed a “perturbation difficulty score” of 0.31 to 0.70, with the index falling below 0.31 if they were in a deep sleep state or under general anesthesia.

    The team then tested zap and zip on 81 patients who were either minimally conscious or insane (in a coma). In the first group, which showed some signs of non-reflective behavior, the method correctly showed that 36 out of 38 are conscious. Of the 43 patients in the “vegetable” state with whom relatives at the head of the hospital bed have never been able to establish communication, 34 were classified as unconscious, and nine more were not. Their brain reacted similarly to those who were conscious, which meant that they were also conscious, but were not able to communicate with family members.

    Current research aims to standardize and improve the methodology for neurological patients, as well as extend it to patients in psychiatric and pediatric wards. Over time, scientists will identify a specific set of nerve mechanisms that give rise to experiences.


    By and large, we need a convincing scientific theory of consciousness that will answer the question under what conditions a particular physical system - whether it's a complex chain of neurons or silicon transistors - experiences sensations. And why is the quality of the experiences different? Why does the clear blue sky feel different than the rattle of a poorly tuned violin? Do these differences in sensations have any specific function? If so, which one? The theory allows us to predict which systems will be able to sense something. In the absence of a theory with verifiable predictions, any inference on the subject of machine consciousness is based solely on our inner instinct, which, as the history of science has shown, should be relied on with caution.

    One of the main theories of consciousness is the theory of the global neural workspace (GWT), put forward by psychologist Bernard Baars and neuroscientists Stanislas Dean and Jean-Pierre Shanzhe.

    To begin with, they argue that when a person is aware of something, many different areas of the brain gain access to this information. Whereas if a person acts unconsciously, the information is localized in the specific sensory-motor system (sensory-motor) involved. For example, when you type quickly, you do it automatically. If you are asked how you do this, you will not be able to answer, because you have limited access to this information, which is localized in the nerve chains that connect the eyes with fast finger movements.

    Global accessibility generates only one stream of consciousness, because if any process is accessible to all other processes, then it is available to all of them - everything is connected with everything. Thus, the mechanism of suppressing alternative paintings is implemented.
    Such a theory well explains all sorts of mental disorders, where malfunctions of individual functional centers associated with patterns of neural activity (or the whole area of ​​the brain) introduce distortions into the general flow of the “working space”, thereby distorting the picture in comparison with the “normal” state (of a healthy person) .


    Toward a Fundamental Theory

    The GWT theory claims that consciousness stems from a special type of information processing: it has been familiar to us since the inception of AI, when special programs had access to a small, public data warehouse. Any information recorded on the “bulletin board” became available for a number of auxiliary processes - RAM, language, planning module, recognition of faces, objects, etc. According to this theory, consciousness occurs when incoming sensory information recorded on the board is transmitted into many cognitive systems - and they process data for speech reproduction, storage in memory, or performing actions.

    Since space on such a bulletin board is limited, at any given moment we can have only a small amount of information. The network of neurons that transmit these messages is supposedly located in the frontal and parietal lobes.

    As soon as these meager (disparate) data are transmitted to the network and become publicly available, the information becomes aware. That is, the subject is aware of it. Modern machines have not yet reached a similar level of cognitive complexity, but this is only a matter of time.

    GWT theory claims that computers of the future will be conscious

    The General Information Theory of Consciousness (IIT), developed by Tononi and his associates, uses a completely different starting point - the experiences themselves. Each experience has its own special key characteristics. It is immanent, exists only for the subject as a "master"; it is structured (a yellow taxi slows down while a brown dog crosses the street); and it is concrete - different from any other conscious experience, as a separate frame in a movie. In addition, it is solid and specific. When you sit on a bench in a park on a warm clear day and watch the children play, the various elements of the experience - the wind blowing hair, the feeling of joy from the laughter of the little ones - cannot be separated from each other without the experience ceasing to be what it is.

    Tononi postulates that such properties - that is, a certain level of awareness - have any complex and conjugate mechanism in the structure of which a set of causal relationships is encrypted. It will feel like something coming from within.

    But if, like a cerebellum, a mechanism lacks complexity and conjugation, it will not be aware of anything. As this theory goes,

    consciousness is an inherent random ability associated with such complex mechanisms as the human brain.

    The theory also deduces from the complexity of the underlying interconnected structure a single non-negative number Φ (pronounced “fy”), which quantifies this awareness. If Ф is equal to zero, the system does not recognize itself at all. Conversely, the larger the number, the greater the inherent inherent random power of the system and the more it is conscious. The brain, which is characterized by colossal and highly specific connectivity, has a very high f, and this implies a high level of awareness. The theory explains various facts: for example, why the cerebellum is not involved in consciousness or why the zap and zip counter really works (the numbers produced by the counter are F in a rough approximation).
    The IIT theory predicts that an advanced simulation of the human brain based on a digital computer cannot be conscious - even if its speech cannot be distinguished from human speech. Just as a simulation of a large-scale gravitational attraction of a black hole does not deform the space-time continuum around a computer that uses a code, a programmed consciousness will never give rise to a conscious computer. Giulio Tononi and Marcello Massimini, Journal of Nature 557, S8-S12 (2018)

    According to IIT, consciousness cannot be calculated and calculated: it must be built into the structure of the system.

    The main task of modern neurobiologists is to use the increasingly advanced tools at their disposal in the study of the endless connections of the various neurons that form the brain to further outline the neural traces of consciousness. Given the confusing structure of the central nervous system, this will take decades. Finally, formulate a basic theory based on fragments of existing ones. A theory that will explain the main puzzle of our existence: as an organ that weighs 1.36 kg and resembles bean curd, it embodies a sense of life.

    One of the most interesting applications of this new theory, in my opinion, is the possibility of creating an AI that has consciousness and, most importantly, sensations. Moreover, the fundamental theory of consciousness will allow us to develop methods and ways to implement the faster evolution of human cognitive abilities. Man - the future.


    Main source

    Also popular now: