On the question of the possible self-education of an intellectual quasi-biological system

In the famous work of Emelyanov-Yaroslavsky L.B. “Intelligent quasi-biological system. Inductive automaton "(M., Nauka, 1990) proposed a model of the interaction of neurons forming new control connections with the formation of a network of interactions by adding a new property, by changing the functional state of neurons (excitation), depending on the need to obtain additional energy resources, followed by using these control interactions to obtain a general property of network activity - the resource control system. A hypothetical assumption about the necessary energy inflow and its subsequent interneuronal redistribution was required to substantiate the idea of ​​the emergence and constant renewal of impulse neural inductance to maintain the activity of the neural network.Obviously, the author of the model understood how the neural mechanisms of a living protobrain could work in the early evolutionary stages. It was also assumed that in the future there was a self-assembly of active resource management networks into enlarged neural ensembles, followed by an increase in their functionality and the formation of intelligence.







However, the hypothetical assumption about the necessity of neural control of energy inflow contradicts the theory of cellular bioenergetics, the first law of which reads: “A living cell avoids direct utilization of energy from external resources when doing useful work. First, it transforms this energy into a convertible form of ATP, ∆μNa + (the difference in the electrochemical potentials of sodium) or ∆μH + (the difference in the electrochemical potentials of a proton) and then uses it in various energy-intensive processes. " The biological logic of any living cell in providing itself with energy is as follows - the purpose of glucose oxidation is to obtain ATP (adenosine triphosphoric acid). The way in which glucose is oxidized for energy is called glycolysis. The end product of glycolysis is pyruvic acid (pyruvate).Depending on the presence or absence of oxygen, acetyl-coenzyme-A (electrotransport metabolite) or lactic acid (lactate) are formed. Both of these substances, together with pyruvate, are included in the homeostatic substrate, which the cell utilizes in its internal metabolic processes and in this it does not need external regulation.



The regulation mechanism is the normalization of the concentration of substances by homeostasis itself, if, for example, the concentration of lactate rises above the norm, then reactions are triggered that convert lactic acid into pyruvic acid and the utilization of these excess in the same cycle of tricarboxylic acids (Krebs cycle) with the formation of ATP. Thus, it is explained that the importance of glucose for the cell is such that any slowdown through external control is unnatural, including for the neuron itself. In addition, due to the well-known circumstance of metabolism, the brain tissue depends on external energy sources and receives glucose together with oxygen from the local bloodstream, and glucose itself enters the bloodstream as a result of hydrolysis of carbohydrates in the gastrointestinal tract, from where it is transported to all cells of the body ,and although brain activity correlates with fluctuations caused by pulsation of the vascular bed, it is not a regulator of glycolysis.



Thus, the trophic function of the brain tissue by itself solves the original question "why does a neuron need a brain?" opens a detailed model reasoning about the possibility of the emergence in nature of a quasi-biological inductive automaton, as a computational superstructure over the physiological mechanism of cellular nutrition. An explanation is immediately required of how a neuron copes with fluctuations that provide an influx of nutrients and oxygen, this is an ancient physiological mechanism that arose even before a biological neuron had electrical and chemical synapses. The blood flow provides trophism and respiration in all other cells of the body, not only in neurons, and this community has structural reinforcement, as in other cells as in neurons,mainly from electrotonically excitable tissues (which are skeletal muscles and some specialized cellular formations such as glia), these cells have connexons. They represent a common type of contact between animal cells, which are slit bilayer structural elements formed by the membranes of adjoining cells.



This contact structure exists in the form of a water channel between the cytoplasm of two neighboring cells, in the lumen of which there are special connexin proteins, bending in an accessible way due to their own physicochemical interactions, as a result of which the channel lumen either closes or opens and one of the main driving forces This valve-like action of mobile protein molecules is mechanical fluctuations in blood flow, to which cell membranes are susceptible. Through the connexons, ions and also water-soluble signaling molecules are exchanged, and in those cases when one of the neighboring cells by its own metabolism cannot provide the phosphate synthesis of AMP, ADP or ATP, this type of convertible energy from other cells penetrates through the connexons.



And at this evolutionary level of cellular interdependence, the predominant way of transmitting information from cell to cell is direct signaling semi-chemical interaction. At this level of organization of intracellular cybernetics, the neuron itself does not have the initiating ontogenetic prerequisites for manifesting itself as an intelligent computational element. Since molecular target receptors with their sensory properties perform the functions of molecular recognition and formation of a signal about an object with which an intracellular interaction has occurred, and the structure formation going inside the cell (neuron) is not transmitted further. Substances are compacted into interaction complexes for those biological reasons that are rigidly determined by cell biology in the mechanism of stabilizing chemical transformations.All interacting particles of living matter having their own characteristic size (atom, molecule, macromolecule, organelle, cell, organism) simultaneously participate in different processes with different kinetic times. Therefore, they form hierarchies of organization of biological matter, each with its own lifetime, their connectivity into a single whole is due, among other things, to the methods of transferring information between chemical individuals (specific compounds) of which they are composed.their connectivity into a single whole is due, among other things, to the methods of transferring information between chemical individuals (specific compounds) of which they are composed.their connectivity into a single whole is due, among other things, to the methods of transferring information between chemical individuals (specific compounds) of which they are composed.



Based on these concepts, the biochemistry of the internal environment of the body is a complex polyhierarchical system (DNA, RNA, peptides, lipids, sugars, bioelectrolytes and other organics) with its various chemical and physicochemical processes underlying the vital activity of all types of tissues and organs. Biological polyhierarchicality is a generative machine of evolution, when other sign effects on a biological substance become demanded to control the physical states of the organism (tissues and organs), then chemical structures become subordinate elements in the transmission of information. The semi-chemical system of molecules initiates special changes in the cell, the sequence of which is prescribed in biochemical messengers, the concentration of which is strictly controlled by hormones,neurotransmitters and other extracellular biochemistry, the synergy of all these factors determines the strict periodicity of the role of messengers in cellular metabolism. The messenger molecules themselves, possessing an affinity for proteins, are needed to regulate interactions between intracellular proteins, which are located at a certain distance from each other, due to the formation of a spatial protein structure (folding).



Correct work of proteins in a cell is possible only with properly formed protein globules (three-dimensional structures), then on the outer surface of the molecule in the right place there will be such a conformation of the substance (active center), which contributes to the correct attachment of the protein molecule to the cell membrane. Taking into account that the active center is primarily a configuration of chemical properties, such as hydrophilicity, hydrophobicity, and electric charge, this determines the energy states of a protein's readiness to perform its specific function. In addition, the terminal sections of the neuronal membrane are stimulated in a special way by a receptor of a protein nature (for example, acetylcholine),which is an electron donor (ligand) giving additional electronegativity and at the same time serving as a coordination bridge for the advancement of an electric charge. This largely determines the functionality for the impulse discharge of a neuron during its mission of transmitting information in the brain.



In this regard, it is required to reconsider one more theoretical assumption of Emelyanov-Yaroslavsky LB, namely, "A discharge in a neuron is needed by the neuron itself." Why does a neuron need this loss of energy and how can it be replenished? Only in the case when the response of the structure to which the discharge is directed in terms of energy efficiency will be greater or equal in magnitude? An indisputable fact is that the neuron also has such structural elements as electrical (efaps) and chemical synapses, which are directly related to the electrical impulse transmission of information. Taking into account the fact that synapses with a chemical mechanism for the transmission of excitation and ephaps with their electrical mechanism for the transmission of excitation are similarly built into the mechanism of propagation of a bioelectric impulse along nerve fibers,than significantly differ from the passive channel-contact mechanism of connexons. The nerve fiber of axons and dendrites, as it were, insulates membrane contacts with different local potentials, while everything happens in a single bioelectrolyte medium with ionic electrical conductivity, where aqueous and organic bioelectrolytes are self-ordered by thermodynamics.



Both inside cells and in the intercellular space, and in the liquid medium of the bloodstream, bioelectrolytes represent the environment in which signal molecular structures are correlated and linked into one system, which only in this way can perform their biological functions inside and outside organelles and cell compartments. After all, "a cell is a flow-through reactor, in the cells of which (van't Hoff's quasi-boxes) thermodynamic self-organizing structures accumulate" (Vasnetsova, Gladyshev

"Ecological biophysical chemistry" p.61) and these functional events are recorded in the conformations of oligopeptides whose synthesis occurs in chemical synapses. The conformation that develops at the moment of oligomerization of the protein polymer determines its molecular reaction characteristics, the configuration of a substance with a large proportion of surface atoms predetermining the features of physicochemical properties is structurally formed.



Since in a neuron there is a variety of substances required for the homeostatic norm of its existence - more than 10,000 chemical individuals, but all these substances (neuroprotectors) obey supramolecular forces (temperature, pressure, time to establish intermolecular equilibrium) that carry out homeostatic rearrangements, then all variations in chemical composition can be displayed in the selected conformation of the memory molecule M-zeta (PKMζ), this does not require a strict quantitative and specific definition of the substances involved, it is enough that the memory molecule is marked with its own morphology, which is an integral property of organic matter and, at the same time, an indicator of sign belonging to the reflected set of intraneural events ...In this case, a metastable state of the PKMζ polymer develops, which satisfies the required times for maintaining homeostasis.



The passage of current through bioelectrolytes is accompanied by the transfer of matter; therefore, it seems that the only possible evolutionary transition occurred when the static substance of the nerve fiber appeared with universal impulse conductivity due to the built-in ligands and the transfer of matter ceased to be necessary for the transmission of signal information. With the emergence of chemical and electrical synapses, a kind of conditional "informational decoupling" of semi-chemical information in the DNA-RNA-protein ligament took place, where the sign information in a bound form is created and stored in the oligopeptides of the chemical synapse (Protein kinase M zeta, PKMζ), and the variable sign "volatile »Information of impulse currents slips over efaps for rapid rearrangements of bioelectrolytes and background chemical environments inside neurons.Only after this evolutionary step, nature began to build inductive automata with conditionally isolated circuits, which are known to us and are morphologically classified according to the types of topical arrangement of ephaps and chemical synapses relative to the soma of neurons and their other integral parts: axoaxonal, axodendritic, axosomatic, dendro-dendritic, dendrosomatic , somatosomatic.



Thanks to this variety of options for contouring connections, including feedback (if such is understood as the passage of potentials in the opposite antidromic direction), such options for the transmission of impulse coefficients have arisen and evolutionarily fixed, when the value of the ratio of potentials arising in the pre- and postsynaptic membranes during the process of excitation characterizes the functional properties of the neural automaton. However, this is enough only for the predicted by Emelyanov-Yaroslavsky L.B. resource management systems, with the proviso that these are information resources, not energy resources, as suggested by the author of the quasi-biological doctrine. But, an inductive automaton is not a thinking machine with its own psychophysics creating its own individual neuro-models correcting being, it is the simplest intelligence of interneural interactions.This is a lower class of real information systems that track semi-chemical signals for their subsequent conversion into signals of a different physical dimension, which exert an energetic stimulating effect on the structures they control.



Since the conformation of the memory molecule, together with the topology of chemical synapses, ephaps, and even connexons, form a single geometric set in the system of information interactions, the contours of their connectivity are a structural diagram of an automated control system for the physical capabilities of an organism (its homeostatic state of health). Neurons do not need reminescence about chemical reception (there is no need to duplicate the memory prescribed in DNA enhancers); neurons need information about permissible deviations in homeostasis. They synthesize this information in chemical synapses, obtaining such a folding of the PKMζ protein polymer, which is possible under those homeostasis conditions in which all the necessary multiple physicochemical parameters of oxygen partial pressure, glucose concentration, ... etc., coincide.and all this is reflected in the assembly of the folding of the memory molecules of the protein kinase M-zeta.



And already this synthetic information can be received by other neurons in the order of information exchange in the control circuits of deviations of homeostasis. This is because the accumulation of memory molecules increases the likelihood that, during streaming ionization within a chemical synapse, the PKMζ protein polymer will turn into a multiply charged ion of a characteristic semiotically identifiable species. The subsequent semi-chemical transformation depends on how the ionic charges move from the surface of the multiply charged molecule to other bioelectrolytic structures carrying impulse currents, as a result it will be a different signaling sign information of interneuronal communication (regulation, control, execution). The selection of which occurs in a probabilistic manner, where the probability structure is predetermined by neurotransmitters,it is logical to assume that in each case correlates of deviations are selected to obtain a working sample of inductive analogies close to the initial semi-chemical premises.



A neuron is faced with the task of obtaining information about deviations in its own autohomeostasis, therefore, it must either receive responses from other neurons with which it is connected by a dendro-axonal tree, or somehow redirect its own synthetic information to its own address. It follows from this that for solving this problem, the object of control of a neuron is the dendro-axonal tree of connections with other neurons, and the semi-chemical laws of control of the required energy-informational events apply to this complex system of connections. This means that inductive conversion of neuroimpulse into a signal under the influence of regulatory substances occurs at critical points of the dendroaxonal pathways. An example of such a substance is acetylcholinesterase,which completely and instantly hydrolyzes acetylcholine with the formation of acetic acid and choline, the transmission of the nerve impulse stops. This interruption is of decisive importance for the impulse neuron, during the interruption, inhibition develops, extinguishing the excitability of the neuron, functional activity decreases along with energy consumption, and autohomeostasis is stabilized. Considering that such an interruption may not be single, but multiple with the participation of other substances (adenosine, glutamate, dopamine, norepinephrine, serotonin, etc.) on the entire spectrum of dendro-axonal connections, including due to numerous repetitions, information acquires a different procedural resource form , - bionic. And to extract additional utility from this type of resource, an information device is needed that is different from an induction machine,one that builds functional neuromodels.



Literature:

Intellectual quasi-biological system Inductive automaton L.B. Emelyanov-Yaroslavsky MOSCOW "SCIENCE" 1990 - www.aha.ru/~pvad/f0.htm



LAWS OF BIOENERGY V.P. Skulachev, Moscow State University named after M.V. Lomonosov - nature.web.ru/db/msg.html?mid=1159125&s



ECOLOGICAL BIOPHYSICAL CHEMISTRY A.L. Vasnetsova, G.P. Gladyshev MOSCOW "SCIENCE" 1989 - search.rsl.ru/ru/record/01001483035



"Information and signals in molecular systems" Kruk N.N. - cyberleninka.ru/article/n/informatsiya-i-signaly-v-molekulyarnyh-sistemah/viewer



Perspectives of thinking modeling Alexander Lvovich Shamis ABBYY company, basic department of PHYSICAL TECHNOLOGY - samlib.ru/s/shamis_aleksandr_lxwowich/iskusstwennyjintellekt-mifilirealxnostx.shtml



Neural Thinking Models Ph.D., Senior Researcher, V.G. Strakhov - www.gotai.net/documents/doc-msc-018.aspx



To an introduction to psychology by L.B. Emelyanov-Yaroslavsky, V.G. Strakhov. - ailab.ru/media/kunena/attachments/82/____.rtf