On the nature of emotions

Understanding how the human brain works is impossible if you do not give a clear answer to the question: what are emotions. I propose a concept that is different from all classical ideas, but allows us to construct a comprehensive explanatory model. Within this model, it is possible to give an explanation of harmony, beauty and humor.

At one time I wrote the article "Emotions in humans and light bulbs in the robot" and "Emotional computer . " There I tried to describe the ideas of adaptive control, peculiar to the real brain. I will repeat the main idea.
It seems quite obvious that sensations, emotions and the state of “good - bad” should be associated with human behavior. What is this connection? Probably, anyone who is related to technology will immediately see control signals in sensations and emotions. It is quite obvious to associate a part of these signals with an impulse to certain actions. It is also obvious to link the other part with informing about the achievement of the result. Naturally, there is a place for negative emotions - this is information that something is wrong and the situation needs to be corrected, and for positive ones, this is information that the result has been successfully achieved. Immediately it turns out to be an obvious place of memory - it stores knowledge of past experience and allows you to choose the most appropriate path. If there is no past experience, then a planning mechanism can be introduced that will model the future and determine behavior. The stimulus for behavior turns out to be two - this is the removal of negative emotions and sensations in the present and the achievement of positive emotions and sensations in the planned future.


Figure 1. "Classical" scheme of the formation of the act
')
In the “classical” construction (Figure 1) everything is quite obvious. Main principles:
• Emotions, existing or forecasted, create motivation for action.
• Motivation dictates the desired result.
• Actions are planned in order to achieve the result prescribed by motivation.
• The result is compared with the plan; mismatch is signaled by negative emotions, and success is positive. Both lead to an adjustment of motivation.
• The results achieved, both successful and not, are stored in memory in order to use this experience in the future.

This scheme can vary in details and can be found in different interpretations. One thing remains unchanged - the “guiding and directing” role of emotions that create motivation. Indeed, in our life we ​​are constantly convinced that emotions and feelings often precede our actions. The remarkableness of the “classical” scheme is that it absolutely naturally falls on the everyday understanding of the reasons that impel us to act. This scheme is a balm for the soul of those who have always intuitively felt how it all happens and wanted to formalize it. This scheme is so obvious that its appearance and development was absolutely inevitable. In any situation there is a simple, clear to all wrong decision. In reality, this is not the case at all. Moreover, as is often the case with obvious at first glance statements, the mistake lies in the most important basic statement.

“After this, therefore, as a consequence of this” (lat. Post hoc ergo propter hoc) is a logical trick in which the causal link is identified with the chronological, temporal.

“After means because of” - it was this logical trap that sent the supporters of the “classical” model along the wrong path. The observation that emotions often precede actions has led us to assume that it is emotions that are their immediate cause. So just this statement and erroneous. Namely, it builds the entire "obvious model." Then how is it really? Let's see.

The assumption that “emotions are pushing towards actions” makes it necessary to build a “classical” model. In it, each element is not accidental, but dictated by the need to achieve compliance with what is observed in reality. However, we will decide on a bold step and discard the thesis “emotions are pushing”, we will proceed from the fact that emotions and sensations only assess what is happening, that is, a change in the state of “good - bad” does not directly affect behavior. So, it turns out that in this case there is a completely logical model.


Figure 2. The proposed scheme of formation of the act

This model works like this:
• Initially, all actions are a consequence of unconditioned reflexes.
• Everything that happens to us is evaluated by sensations. This assessment is reflex in nature and is determined by the state of the sensors.
• The general meaning of what is happening is evaluated by emotions.
• Feelings and emotions form the “good - bad” state.
• Every action that leads to a change in the state of “good - bad” is recorded by the memory. Memorized:
o “Picture” of what was happening.
o Action committed in these circumstances.
o What a change in the “good - bad” state it led to.
• As you gain experience, memory begins to “take control”. When a previously encountered situation is recognized, the memory forces one to perform an action that previously led to a positive change in the “good-bad” state, and blocks actions that are remembered as worsening this state.
• The force with which a single memory affects the performance or non-performance of an action depends on the degree of change in the “good - bad” state that has been memorized.
• The control actions from various memories relating to similar situations add up to each other.
• At every moment an action is automatically performed which, based on our experience, promises the greatest possible improvement in the state of “good - bad”.
• New experience, as soon as it is acquired, begins to participate in the formation of behavior.
The principal difference from the “classical” scheme is that only unconditioned reflexes and memory determine the current action. This act is “inevitable” under the circumstances and does not directly depend on our assessment of what is happening. Evaluation is important only for the acquisition of new experience. If, in the “classical” scheme, emotions induce actions, in our model, as in actual life, the current action does not depend on them. At first glance, this may not seem obvious. The reason is clear. If millions of our actions are committed against the background of emotions, the idea of ​​a causal relationship is unwittingly formed. I repeat once again: “after that it does not mean because of that”. If you watch TV for a long time, you may get the impression that weather forecasters control the weather.

The described idea is reinforcement learning, where emotions and evaluations of sensations act as reinforcement apparatus. The only thing that makes the emotions and evaluation of feelings - it affects our state of "good - bad."

Now we will try to formulate “obvious” postulates that publicly or unofficially underlie the “classical” understanding of emotions:
• Emotions of the person are various and rather complex. The complexity of emotions is the result of evolution. Each of the emotions was formed in the process of natural selection, carrying with it a certain expediency.
• A child is born with a genetically predefined set of basic emotions that are more or less the same for all people.
• Emotions of the child require formation. Passing certain stages of development, the child learns to apply the emotions inherent in it from nature.

In one way or another, these statements form the basis of all psychological theories. These statements seem to be a logical consequence of the evolutionary theory. And indeed, for each emotion, you can discern the kind of expediency it brings to behavior. And we are used to everything that is expedient - the result of natural selection. But we, modern people, are separated from monkeys by only a few million years, and from our rather wild ancestors who could not speak, even fewer, only some tens of thousands. Both are by the standards of nature - a moment. How could such a complex system of human emotions arise in such a negligible period? So, what we have to give up is the notion that emotions are the result of evolution and are inherited by us through the genome. What about then? To answer this question, you must answer another, perhaps the main question. So what are all the same emotions?
Our memory captures everything that happens to us, taking into account what kind of “good - bad” condition corresponds to this. When subsequently we meet familiar signs, they not only affect actions, but also cause a state of “good” if they are associated with “good” memories, and a state of “bad” if they are associated with “bad” ones. In essence, this is what we call anticipation and fear. At the same time, we remember not only “good” or “bad”, caused by sensations, but also “good” and “bad”, caused by anticipation and fear, but also by fear of fear and anticipation of anticipation and so on. So, all positive emotions are either anticipations of pleasant sensations, or anticipations of anticipations, and all negative ones are either fears of discomfort or fears of other fears. That is, all emotions are not the evaluations of nature, but only the result of our experience. That is, a child is born only with reflexes, sensations and the ability to evaluate sensations, everything else is the result of the formation of memory, in which a complex complex of anticipations and fears is created.

The basis of the building of emotions lies in the evaluation of sensations, which in the first stage generate primary emotions, which, in turn, give rise to subsequent, secondary ones. Primary emotions are fears and anticipations of the evaluation of sensations, secondary ones are fears and anticipations of primary emotions.

We must part with the idea that emotions are inherited. All the diversity of emotions that we observe is the inevitable consequence of the development of a child, adolescent, person in his natural and social environment. Each child anew, from scratch, creates for itself all existing emotions. There are no emotions resulting from natural selection. There are no particular emotional structures. There is only memory and its ability to influence the state of “good - bad”. Any emotion is fear or anticipation of something that has already happened to us in real or in our fantasies. Each memory carries information about any fear or anticipation.

That is, if something gave us pleasure, then all the signs that were present at the same time, will subsequently cause a “good” condition in us. Conversely, what made us “bad” will make us fear its signs.
The variety of all memories creates an aggregate of all fears and anticipations. In this aggregate, statistically stable reactions are manifested in situations united by a common meaning. These reactions, which are stable and similar for all people, are usually meant when they talk about specific emotions.

John Watson, together with Rosalia Rainer, made remarkable observations back in 1920, which went down in history as the “case of little Albert”. Experimenters traced the formation of fear emotions in an 11-month-old infant. Prior to the experiments, the child was completely indifferent to white mice. During the experiments, Albert was shown white mice and at the same time loudly banged the iron strip behind his back with a hammer, which caused the child to cry. Soon, scientists have achieved that Albert began to cry only after seeing white mice. Moreover, it turned out that the baby’s fear began to spread to white objects in general - white sheets of paper, a white rabbit, a white fur coat. Actually, this is a good illustration of how the experience of experiencing a state of “bad” caused by loud sounds is transferred to the accompanying signs. And already these signs themselves become the cause of the emergence of the emotion of fear.

If you dig deeper, it turns out that all human emotions can be explained through the formation of statistical generalizations of experience assessing what is happening. Moreover, it was possible to give an exhaustive explanation for love, harmony, beauty and humor, to believe harmony algebra, so to speak.

I do not expect that this article will be very convincing. For a slender and demonstrative presentation I needed to write a whole book. If someone is interested in understanding this in more detail, then the book is called “Colored emotions of a cold mind” (Alexey Redozubov). Download it here .