It is possible that, similar to animals, humans also undergo a process of "immolation in memory" - unconscious learning through trial and error
Zvi Atzmon
Direct link to this page: https://www.hayadan.org.il/hizachrut20905.html
How do you know someone remembers? According to his responses. For example, she blushed when she heard the question "Do you remember what happened at the party?" Also an answer to the question "How much is seven times eight?" or "Who painted the Mona Lisa?" is nothing but behavior, whether it is expressed as a voice or in writing. But from the point of view of the questioner himself, remembering has another facet - an inner feeling, the recognition "I remember".
And here, in an article published about two weeks ago in the scientific journal "Nature", a group of researchers led by Larry Squire, one of the most important memory researchers, points to the existence of a learning process without a sense of memory - they call it "immolation in memory". It is about learning by trial and error, the kind that is usually attributed to the learning of animals. It is an engraving in the memory of a kind of habit, a custom. Squire is convinced that such processes play a fundamental role in humans. "These processes shape, without us knowing, our preferences, our tendencies and a whole spectrum of our interactions with the world", he claims.
The study was conducted on two amnesiac patients, identified by their initials, AP and GP. In both cases, the forgetfulness was caused by a viral infection of the brain, as a result of which the middle parts of the parietal lobes were severely damaged. Bilateral destruction of these parts (the hippocampus and parts of the brain related to it) causes a fatal and specific damage to memory, which is manifested in the inability to retain new "declarative" memories for the long term. Declarative memory is memory that can be explicitly expressed in words and numbers.
Such an amnesiac patient can meet for a long period of time with a particular therapist, and every day introduce himself anew. He can also read the same sheet of newspaper every day, without getting bored, until he is completely worn out. Short-term memory is not damaged - such a patient can have a meaningful conversation, but if his attention is diverted for a short time, not only will he forget what it was about, but he will forget that he even had a conversation. His memories of the past are also normal, apart from the memories of the immediate past (on the order of months) to the brain damage.
In a patient with such amnesia, long-term memories can be released, provided they are "non-declarative" memories: he can acquire new skills and new habits as a series of actions, but not as cognitive information that can be expressed in words or numbers.
AP suffered from forgetfulness following an infection with the herpes virus. Memory and brain researchers led by Squire mapped the damaged parts of his brain using magnetic resonance imaging (MRI), and conducted detailed memory tests. After it became clear that the parts of his brain that are essential for the creation of new declarative memories were indeed damaged, the researchers asked themselves: can we teach him new semantic knowledge? They showed him a series of short, three-word sentences (for example: "a shark killed an octopus") and tested his memory. The test was done in two ways: as an open question (complete: "Killing shark?") or as a closed choice question (Shark killing a fish or an octopus?).
Healthy subjects achieved good results after a short training series, while AP's achievements were very low initially. But when the training series was extended, his achievements improved significantly, although they remained quite low. So the researchers conducted another test: an open question in which one of the words was replaced by a synonym (for example: complete the sentence "killer shark?"). Among healthy people, replacing a word with a synonym, and even replacing both words with synonyms, hardly affected the results, but with AP, replacing a word with a synonym degraded his results to zero. It became clear, therefore, that even what AP learned he learned as a process, as a habit, "learned like a parrot", and could not make real use of the facts he learned, except for rigid repetition.
In the study that Squire and his colleagues just published, they sought to find out whether humans can learn things by trial and error - a way of learning that is very common among animals, and it was not clear whether it exists at all in humans. This question was asked to be examined using two amnesia patients - AP, whom they already knew well (but not them!) and GF, whose brain also had middle parts of the parietal lobes destroyed due to herpes infection.
The researchers presented pairs of items to each of the subjects. On the underside of one of the pair's items is written "correct". The subjects were asked to guess who is the "correct" one, then pick it up and check. There were a total of eight pairs of items. Healthy subjects reach very high accuracy after few repetitions. In contrast, the AP and LP required many repetitions to achieve good results - but they finally achieved good results. This is despite the fact that at the beginning of each test - even after multiple trainings - they did not remember at all what it was about, and had seen these items before.
Monkeys learn through trial and error to perform such a task successfully, but only after a very large number of repetitions, a number that is roughly similar to that required for AP and GP (of course, the monkey does not read which item is "correct" - but for each item "correct" He gets a reward, a favorite food).
Following a long series of trainings and tests, when AP and GP's achievements were already good, the researchers presented each of them with all 16 items together. They were asked to create two groups - a group of "correct" items and a group of incorrect ones. Healthy people, after learning what is "right", fulfill this task very easily. And what about AP and GP? Their "score" on the classification task was zero - their classification of items was random. It has now been proven that even when very high results are obtained in a memory test (far beyond those obtained in the three-word test), the ability to apply what you have memorized can be zero.
Squire and his colleagues hypothesize that the memory of which item is correct is instilled in their subjects through the basal ganglia - brain structures that are involved in the acquisition of habits, for example sending a hand to the electrical switch upon entering the room, even during a power failure. This is how memory became something that can be performed well, but without awareness, in a rigid framework and without the ability to implement.
The proof that a process of unconscious learning through trial and error is possible in humans may indicate that our choices ("What is the correct item?") and even texts that we treat as correct ("The shark killed the octopus", not the fish), can explicitly be a kind of habits , a fruit of learning that we are not aware of. We do not know how to criticize such things, and we treat them as if they were absolute truths.
The find may also shine a new light on one of the oldest and most difficult philosophical problems - the question of consciousness, recognition. Why are we, as living beings, endowed with consciousness, and not just acting in the world? Does consciousness allow a faster and more efficient way to learn and remember things about the world, and is this an advantage?
The finding may also open a new way of thinking about another ancient problem - the dream problem. Why dream? During the dream the hippocampus is paralyzed; This is the reason why it is difficult to remember dreams. Perhaps the dreams are used, then, as a kind of practice of behaviors, which become, through the kernels of the base, habits that we later apply in reality, without knowing why, as chosen by AP and JP?
* The author is the scientific editor of the journal "Galileo"
The brain savant
https://www.hayadan.org.il/BuildaGate4/general2/data_card.php?Cat=~~~245326158~~~58&SiteName=hayadan
