By Oleksandr Kostikov
Could it be that all AI (artificial intelligence) developers are wrong?
Imagine you have a radio. You use it with ease, and sometimes, you make simple repairs. But what if you want to create a new radio receiver? To do this, you would have to fully understand the design details. You would need to understand not only how the volume control and power supply work, but also where the music comes from.
Likewise, creators of artificial intelligence need to first try to understand how natural intelligence works. Otherwise, instead of a radio, they may end up with a boring and monotonous player.
The main mistake
Everyone thinks that the human brain is a kind of biological computing machine that works according to certain algorithms, and if so, it means that intelligence can be compiled using two simple tools: high computing power and a set of complex algorithms.
In fact, everything is not at all that simple …
Neurophysiologists have thoroughly studied the “reflex,” a stereotypical reaction of a living organism to any external influence. The reflex is not just a scheme of functional activity. It is the basic principle of the organization of any nervous system of a biological type. By itself, it really resembles the principle of the algorithm.
But all this is true only if we are talking about a relatively simple nervous system. With human intelligence, everything is much more complicated.
What is the difference?
Reflex is the standard in neurophysiology. Its basis is the reflex arc, which can be schematically depicted as the response of motor neurons to a stimulus received from sensory neurons. As soon as a stimulus appears, the living organism acts. It is important to understand that the number of neurons participating in the reflex arc and the complexity of the motor reaction do not change the essence of the reflex in any way.
A reflex is always a clear and predetermined sequence: stimulus (neural network) reaction. With the help of a reflex, the nervous system acts according to a pre-selected pattern designed to respond to a specific situation. In other words, a reflex is a kind of stereotype or a tactic of actions within the paradigm of occurring events.
Reflexes in the human brain
The human brain, like the brain of any other living organism, often operates according to the traditional pattern of reflex responses, except in situations where consciousness or intellect is used to make decisions.
At this moment, everything changes!
If our brain uses an intelligent (intellectual) reaction scheme to respond to an external stimulus, the first thing that becomes necessary is to form an abstract image with which we identify the surrounding reality.
Simply put, we must first understand what is going on.
This is how we interpret what we feel. The created abstract scheme and the subsequent dynamic modeling of the development of the situation form a solution in the form of a sequence of actions that the brain transmits as commands to motor neurons.
Intellect always and in any situation creates its own reality and virtually stands above the process itself invented by itself. So, the mind can identify itself and realize that it really exists. The resulting effect of this model is the ability of our brain to create images and situations that are completely divorced from objective reality and even contradict sensory information entering the brain.
Simply put, the reflex is always inside the situation and is looking for the simplest and most effective way out, and the mind (intellect), on the contrary, is always outside and itself models its vision of the problem and the ways to solve it.
It turns out that intellect (stream of consciousness) and reflex are fundamentally different reaction schemes.
What does it mean?
This means that the human brain is a binary system consisting of two functional circuits for responding to arousal. Formally, the reflex system is a subordinate part, but in practice, reflexes are responsible for most of our nervous activity. The stream of consciousness can control only some actions. At the same time, the division of response schemes is so clear that most of the functions of the reflex system, such as controlling the work of the digestive system, cardiovascular, endocrine, and respiratory systems, remained completely inaccessible for conscious or intellectual control.
Mind and reflex are not just different functional schemes for organizing neural activity—they are, in fact, opposite models that cannot be mixed even when we force the brain to sequentially switch the reaction scheme from reflex to consciousness and back in the simplest situations.
You can check this with a simple experiment.
You need to sit near the monitor or just in front of the lamp and ask your assistant to turn on the light unexpectedly for you. In response to the flash, you need to press the button as quickly as possible, which will fix the time between the flash of light and your mechanical reaction in pressing the button. After short training repetitions, you will bring your reaction to automaticity and will be able to react at a speed of 200-220 milliseconds. This is a simple example of a reflex. Your brain uses the standard neural organization scheme for any living organism.
If after that, you try to apply your intellect and intentionally delay the keypress by a little, something amazing will happen. With all the efforts of your brain and no matter how hard you try, you cannot press a button in, say, 250 or 350 milliseconds. Even 400 and 450 (twice the interval) will remain completely inaccessible to you. The closest value you can intellectually give is at least 550 milliseconds.
The task, conditions of the external environment, and the brain remain the same. But a different functional system will cause a drop in the maximum reaction rate by almost three times. The huge, 0.3-second delay shows what it means to realize and turn on the intellect.
The significance of the intellect delay
In fact, this time difference, which almost no researcher pays attention to, is of great importance.
The speed of movement of a nerve impulse along large fibers of the human nervous system is approximately 100–120 meters per second. Therefore, during a delay of almost a third of a second, the signal will have time to travel 30–40 meters.
Such a delay is a death sentence in a competitive environment. Human consciousness or intellectual schema is hopelessly lagging the reflex. From the point of view of evolution, this is an unacceptable and extremely harmful change that can in no way appear in the conditions of a real struggle for survival.
The evolutionary process involves sequential change, when new organs or functions provide an immediate competitive advantage, which increases the chance of survival and procreation. Otherwise, new traits have no chance of gaining a foothold in the population.
But we do exist and think. How did it happen?
A reflex is a simple and extremely fast way of reacting, a kind of ideal tactician. On the contrary, the mind is a slow and, in fact, a mechanism for abstract problem solving, divorced from reality, a kind of strategist.
Schematically, we can represent the mind as a rider on a horse. The horse in this example will symbolize the reflex activity of the brain. Not being able to control the movement of the horse’s legs, the intellect controls only the direction of movement. While the horse (reflex system), in principle, is not able to understand the motivation and purpose of the rider (intellect).
Symbiosis instead of evolution
Such a binary system very clearly corresponds to the biological scheme of symbiosis when we are dealing with a binary organism. In this case, intelligence is not the result of gradual evolutionary changes, but an additional functional system that appeared suddenly and did not go through the standard process of evolutionary consolidation. The second or intellectual part of the human central nervous system behaves like a classical symbiont, which, in a limited volume, has received the ability to control the first (reflex) evolutionary formed part of the nervous system.
This has already happened in biological history.
In evolutionary theory, there are three proven cases of symbiotic interactions between two organisms, which eventually created biological constructs of a new type.
This happened when the cell nucleus, mitochondria, and chloroplasts appeared. Each time, evolution did not follow the path of gradual changes, but the method of combining ancient cells into one new biological system with completely new capabilities.
It turns out that intelligence is not at all a happy coincidence, but the result of a rather original—but at the same time standard—evolutionary technique.
This is a very important conclusion that will allow us to get an answer to the main question: how to create realistic artificial intelligence.
In a future article, I will analyze the energy and functional features of the intellectual system of our brain and try to understand the details of unification: symbiosis.
As a result, it will become clear to us how the technology will be created that can build strong AI of a completely new type.
About the author
Dr. Oleksandr Kostikov is a medical doctor by education. He previously worked in a research group dealing with molecular systems in human neocortex synapses at the Department of Physiology of the State Medical University and the Department of Physics of the Ivan Franko University in Lviv, Ukraine. He is now in Canada, where he is working on a new theoretical concept about the nature of intelligence but also allows that aims to to create a completely new unusual type of artificial intelligence.