Marvin Minsky "The Emotion Machine": Chapter 2 "We want to create a machine that would be proud of us"

§2-7 Communication of newborns and animals

"We want to create a car that would be proud of us."
- Danny Hillis, 1983

Baby Carol loves to explore, but she also likes being close to her mother, so when the distance between them grows, Carol tries to quickly close the distance. But as soon as she finds herself alone, Carol begins to cry and look for her mother. Exactly the same behavior also appears when Carol's mom is near, but Carol is experiencing anxiety or fear, for example, when a stranger approaches.

Naturally, this dependence stems from our infantile helplessness: not a single baby will survive if it leaves its parental care. Of course, this is happening, not because babies cannot move for long on their own, but because of the disadvantage that babies cannot follow their mothers for the first few months. Fortunately, people usually do not experience much harm from this disadvantage because we have developed a feedback system: Carol’s mother almost always realizes (to varying degrees at different times) what happens to her daughter, and therefore her attention quickly returns to the child. at the slightest suspicion if something goes wrong.

It is clear that the survival of each child depends on the relationship with the people who are responsible for their well-being. Therefore, in earlier times, it was often assumed that children attached to those who gave them physical care, and that is why many psychologists gave this chap to the name “Caregiver” instead of using a word like “Imprint”. But more systematic studies studying attachments suggested that this theory was wrong:

John Bowlby: “The fact that a child can become attached to other people of the same age or a little older clearly shows that the creation of feelings of affection can develop and be aimed at people who have not done anything to meet the physiological needs of the child.”

In this case, what factors determine the circle of persons to which our children are attached? First, Bowlby acknowledged that physical characteristics can play an important role, because they provide children with the opportunity to learn to be like well-defined individuals. But in the end he came to the conclusion that there are other more important factors:

• The speed at which a person responds, and
• Intensity of these interactions

This care usually falls on the parents of the child, but may also apply to children whose parents take special care in selecting a company for their offspring by carefully choosing a social circle from the offspring of their companions and friends, and especially from the personalities most interested in children. And when someone chooses a school for a child, this someone can find out everything about the staff and the curriculum, but also about the goals that the students pursue.
')
What happens when a child loses its Imprinter? It seems that the absence of the Imprinter creates a special kind of fear and a powerful impulse for finding your Imprint.

John Bowlby: “Whenever a small child moves away from his mother against his will, he feels misfortune. And if he is placed in a strange situation and strangers look after him, the feeling of a strong misfortune will increase. The way he behaves is a series of ordinary reactions. First, all the energy he spends on the search for his mother. After meaningless attempts, he despairs, but nevertheless remains concerned about the absence of the mother and longs for her return. After a while, he seems to lose interest in his mother and is emotionally removed from her. ”

Bowlby goes on to describe what happens after the mother returns:

“However, if the time spent separately is not too long, the child will not be removed from his mother. Sooner or later, after the reunion with the mother, the child's attachment is restored. From now on, for days, or weeks, or much longer, he insists on staying by her side. In addition, when the child feels that he can lose his mother again - he begins to feel very anxious. ”

We see a similar property of affection in various groups of related primates, such as chimpanzees, gorillas and orangutans, as well as in our more distant relatives - monkeys. It should also be noted the discovery of Harry Harlow that, having no other alternative, the monkey will be tied to an object that has no behavior at all, but has some characteristics related to “comfort”. It would seem that this confirms Bowlby’s view that attachment is not related to “physiological needs,” until we improve this definition and bring in the need of an infant, which Harlow calls comfort contact.

John Bowlby: “The very detailed observations made by Jane Goodall about chimpanzees in the Gombe River Reserve in Central Africa show not only that disturbing and suspicious behavior manifests itself in separated animals not only in captivity, but also in the wild, but this stress from separation continues throughout childhood chimpanzees. ... Until the young individuals reached the age of four and a half years, none of the travelers saw them without their mothers. After this age, such behavior is quite rare. ”
[John Bowlby, p. 59 Separation]

When the mother and the child share a great distance, they keep in touch with the help of the special sound 'hoo' ('hoo'), to which the parent or son responds - as Jane Goodall herself reports:

“When a baby ... begins to move away from his mother, he makes a sound anyway when he gets into difficulties and cannot quickly return to her. As long as the locomotion of the infant is not very well developed, the mother immediately reacts to this sound and immediately goes to help. The same sound is used by the mother when she gets to her babies and is going to protect him from potential danger, or, sometimes, when she tells her baby to hold on when the walk begins. Thus, the sound “huu” serves as a rather specific signal to restore the contact of mother and baby. ”

What happens in other animals? In the early 1930s, Konrad Lorenz noticed that newly hatched chickens, ducklings or goslings are tied to the first large moving object that they see, and they will subsequently follow this object. He called this phenomenon “imprinting,” because this phenomenon occurs with incredible speed and consistency. Below are some observations about this effect.

The chick quickly begins to follow the moving object.

• Imprinting occurs soon after hatching.
• The imprinting reaction ends a few hours after hatching.
• The effect of imprinting is permanent.

What moving objects are tied to a chicken? This moving object will usually be the parent, but if the parents have been removed, then the cardboard box, or the red ball, or even Konrad Lorenz himself may be the object of imprinting. Over the next two days, when the duckling follows the chosen parent, it somehow recognizes the object of its following and no longer follows the other geese. When the goose loses contact with his mother, he stops eating or exploring things, and instead will look for his mother and make peaking sounds, as a sign that they are lost. Then the parent responds with a special sound, and Lorenz remarked that this answer must be given quickly enough to establish imprinting. Later, this sound is not necessary, but initially it is necessary to protect the hatched chicks from binding to an unsuitable object, such as a moving branch of a tree.

These peaking sounds, as well as the 'huu' sounds that Jane Goodall had noticed, suggest that other interaction signals could evolve from similar signals. In any case, the chicks can feed themselves soon after hatching, so imprinting does not depend on eating behavior.

As for the end of the imprinting period, R.A. Hindi discovered that these chicks become fearful about moving unfamiliar things, which made him suspect that the imprinting ends when this fear replaces the further feeling of “following”. Similarly, many human babies demonstrate a long-term fear of strangers that begins to develop at the beginning of the second year of life.

The Bowlby study on young children showed that when they are deprived of imprints for more than a few days, they may show signs of loss of communication for a much longer time. He also gives similar results when other researchers have removed rhesus monkeys from their mothers:

“From all these studies, we can conclude that not only a one-time separation that lasts more than six days at the age of six months has tangible effects after two years in newborn macaques, but that these effects are proportional to the separation time. Thirteen-day separation is worse than six days; two separations lasting for six days are worse than one six day separations. ”
- Bowlby in Separation, p.72

It is noteworthy that even children (and monkeys) that are mistreated can remain attached to an aggressive imprint.

How much did attachment-based learning evolve from more ancient forms, such as imprinting? Of course, people are very different from birds, but babies of both share similar needs, and perhaps they got these behavioral patterns from early warm-blooded dinosaurs. For example, Jack Horner discovered that some dinosaurs built nests for eggs that have the structure of bird nests. Further progress in genomics can help us understand this link much deeper.

Returning to humanity, we must ask how babies distinguish potential imprints. Although some researchers reported that babies can learn to recognize the mother’s voice even before birth, usually newborns first learn, mainly through tactile, gustatory and olfactory contacts, and only then learn to distinguish voices and begin to respond to the appearance of the head or face. Initially, it can be assumed that this is done by finding things like eyes, nose, mouth, but there is evidence that the mechanism of perception is much more complicated.

Francesca Aserra: “Four-day newborns look longer at their mother’s face than at someone’s face, but not when the mother wears a scarf that hides the hair outline and the outer contour of the head.”

This researcher found that such babies react less to facial features and to more global characteristics. However, this does not happen until the 2nd or 3rd month of age, while children especially secrete faces. This suggests that our visual systems use different methods of analysis at various stages of development, and perhaps the very first such feelings serve to bind the child to the mother! In any case, Lorenz was amazed that his ducklings could not distinguish:

Konrad Lorenz: “The duck who remembers the image of a man will refuse to follow anyone except the man, but the duckling will not be able to distinguish between a small, slim young girl and a big old man with a beard. It is amazing that a bird, reared and impressed by man, should direct its behavior model not only in relation to one person, but to the whole species of Homo sapiens. ”

I do not think this behavior is very strange, because all the geese look almost the same to me. Perhaps more importantly, the sexual distinction of adults can be established at this early time of development, although it will manifest itself in behavior much later.

“A daw, for which a man has replaced a parent, will send his sexual instincts not to his former parent, but to any relatively stranger. Paul, in this case, is absolutely not important, but the object must definitely be a man. It would seem that the former parent is simply not considered a possible “spouse” or “spouse”. ”

Some studies have shown that after such contact, some of these birds will, of course, be able to perceive other members of their species. However, this is still a serious problem in increasing the number of endangered species, and the standard policy of minimizing human contact with chicks is adopted for this, so that their preferences for humans do not lead in the future to abandon mating with individuals of their species.

I wonder if such contacts can shape a person’s sexual preferences?
All this may help explain why our children developed infantile helplessness: children who left mothers too early did not have enough knowledge to survive, and therefore we had to develop the ability to increase the learning time from our Impressors, not hoping to make risky life experiments .

§2-8 Who are the Impressors?

Galka, seeing the pigeons next to a lot of food, painted herself white in order to join them. The pigeons, until the galka did not say anything, assumed that another pigeon came to them and missed the daw to the food source. But when once the jackdaw forgot that she needed to be silent, the pigeons expelled her, because of the unusual voice of the pigeons. When the daw returned to her tribe of daws - they kicked her out because this daw was different from the rest of the daws in her color. Therefore, wanting two outcomes, she received nothing.
- Fables of Aesop.

How many people can there be Impressors? Many young children have only one imprinter, while others may have two, three, or more. In the case when a child has several Impressors, are the links with them interchangeable or do they perform different functions and goals? If a child creates several sets of ideals, will this enrich his personality, or will this worsen his development because these inconsistencies do not allow to form a single image of his personality?

When do connections begin and end? Even small children soon begin to behave differently when they are in the presence of their mothers. However, as a rule, until the end of the first year, the child does not protest against separation from the mother and begins to learn worries about the signs given by the imprint about leaving, for example, when the imprinter begins to put on a coat. At this time, also begin to show fear of unusual things. Both of these fears will begin to diminish in the third year of the child, so much so that the child will be able to attend school. However, we do not see the same decline in the role of other similar feelings, both with the feeling of oneself and with applied feelings. These feelings persist for a longer time, and sometimes for life.

John Bowlby: “In adolescence, other adults may acquire the same value, which is equal to or even greater than the value of parents; moreover, the sexual attraction to people older than their age becomes more pronounced. As a result, the individual variation in behavior, which was before and so great, becomes even greater. In the first maximum deviation from the norm, there are adolescents who broke off the connection between themselves and their parents, on the other - those who remained too strongly attached to their parents and could not direct their attachments to other people. Between the extremes lies the vast majority of adolescents, whose attachments to their parents remain strong, but connections with others are also of great importance. For most people, communication with parents continues into adulthood, influencing behavior in various ways. Finally, in old age, when behavior that forms connections with people can no longer be directed at members of the older generation or even at people of their generation, and instead, this behavior is directed at members of the younger generation. ”

What happens to other animals? In those species whose members do not remain in the herds, the attachments remain until the offspring can live on their own. In many species, female behavior is different. For example, in some animals, the mother will actively crowd out the young individuals as soon as a new litter appears (possibly due to evolutionary selection against inbreeding), while in others the attachment will remain until puberty or even later. In the appendix (p. 182), Bowlby mentions a phenomenon that arises from such behavior:

“In females of various types of ungulates (sheep, deer, bulls, etc.), attachment to the mother can continue until old age. As a result, a flock of sheep or a flock of deer is built from a young mother who follows her grandmother herself, a grandmother follows her great-grandmother, and so on. Young males of these species, on the contrary, sever ties with the mother to achieve puberty. From this point on, they become attached to older males for the rest of their lives, except for a few weeks a year during the rut. ”

Of course, other species have developed different strategies that are better suited for different environments. For example, the size of the herd washes will depend on the nature and extent of predators, etc.

Why do we need imprints at all, and why should we be so interested in how our brain chooses them? Why not simply raise your goals in response to criticism or praise of any person? There is an excellent reason why we have developed selectivity in this matter: if any stranger can program our goals, you will be in danger, because strangers are less inclined than our relatives to take care of our well-being.

However, “well-being” can mean different things. For example, Bowlby argued that our connections mainly contribute to the physical safety of our children. Here is the argument of his position:

“This protection against predators is by far the most likely link building function supported by three main facts. First, an isolated animal is much more often attacked than those who remain in a group with other individuals of their own kind. Secondly, the creation of bonds is particularly easy in animals that, due to age, size or conditions, are particularly vulnerable to predators. Thirdly, such behavior is particularly pronounced in situations of anxiety when a predator is nearby. At the moment, no other theory fits all these facts. ”

Here, Bowlby's main job was to refute the then popular opinion that the main function of making connections is to find a reliable source of food. Instead, he argued that physical assistance (including nutrition) did not play a decisive role in attachment, and a sense of security, on the contrary, was the most necessary. I suspect that this theory has been largely proven for animals, but it is also suitable for people, considering how the formation of connections between people strongly contribute to the acquisition of high-level values ​​and goals.

---

For the translation, thanks to Stanislav Sukhanitsky, who responded to my call in the “previous chapter.” Who wants to help with the translation - write in a personal or mail magisterludi2016@yandex.ru

By the way, we launched the translation of another cool book - “The Dream Machine: The History of Computer Revolution” .

about the author

Marvin Lee Minsky (Eng. Marvin Lee Minsky; August 9, 1927 - January 24, 2016) - American scientist in the field of artificial intelligence, co-founder of the Laboratory of Artificial Intelligence at the Massachusetts Institute of Technology. [ Wikipedia ]

Interesting Facts:

• Minsky was a friend of critic Harold Bloom from Yale University (Yale University), who spoke of him as “sinister Marvin Minsky”.
• Isaac Asimov described Minsky as one of two people who are smarter than himself; the second, in his opinion, was Karl Sagan.
• Marvin is a robot with artificial intelligence from the cycle of Douglas Adams novels Hitchhiker's Guide to the Galaxy and Hitchhiker's Guide to the Galaxy (film).
• Minsky has a contract to freeze his brain after death in order to be “resurrected” in the future.
• In honor of Minsk named the dog of the protagonist in the movie Tron: Legacy. [ Wikipedia ]