Comprehensive coverage

Brain research has revealed: this is how we interpret new situations based on existing information

Minor: The brain provides us with the wonderful ability to understand situations we have never encountered before. A new study deciphered the mechanism responsible for this.

The learning mind. Illustration: shutterstock
The learning mind. Illustration: shutterstock

The brain provides us with the wonderful ability to understand situations that we have never encountered before, for example recognizing a familiar person in a new place or understanding the meaning of the sentence: "I need to sit with you" as the sentence "I need to sit and talk with you" and this even though the main word in the sentence "to talk" does not appear at all . The mechanism responsible for performing this task has been an age-old mystery in the field of neuroscience.

Now researchers from the University of Colorado Boulder seem to have taken a step toward solving the mystery. The researchers demonstrated how human brains are able to process these new situations while relying on a method similar to the "pointer" system used in computers.

A "pointer" indicates to the computer where to locate information stored elsewhere in the system in order to process some variable. For example, a computer program is able to create a template for an e-mail letter that includes a recipient's name, if a suitable 'pointer' is placed immediately after the word 'to respect' in the body of the email. The voter will 'pull' from the database the name details of each of the selected recipients and insert them in the correct place.

A "pointer" process on the computer: the content of the third field from the top (variable a) will be taken from the value appearing in the third field from the top (variable b). [Image from Wikipedia]
A "pointer" process on the computer: the content of the third field from the top (variable a) will be taken from the value appearing in the third field from the top (variable b). [Image from Wikipedia]
In the new study published in the scientific journal Proceedings of the National Academy of Sciences, the scientists showed that the connections between the prefrontal cortex and the basal ganglia can play a role similar to computer pointers. During the study the researchers used sentences and words in an unfamiliar and illogical way and showed how the brain can understand the meaning, or role, of these words in a sentence.

According to Professor Randall O'Reilly, from the Department of Psychology and Neuroscience at the University of Colorado Boulder and co-author of the article, the findings of the study proved that human brains are able to understand sentences as a structure with variables - subject, verb, and sometimes also object - and that the brain is able to assign a wide variety of words to these variables and still Understand the structure of the sentence: "When in the past we tried to get computer models to perform this task we failed completely. The fact that the true meaning of the sentence can be understood even though it is missing the main word in it demonstrates the brain's ability to correctly process completely new data for it.

In the next step, the researchers added information to the computerized model similar to how the connection mechanism in the brain works. This allows the model to be trained to understand simple sentences using a defined set of words. After the practice period, the researchers fed into the model new sentences that use familiar words but in new forms and discovered that the model is able to understand the structure of the sentence.

However, while the findings show that it is possible that a pointer-like system does work in the human brain, the function it uses is not exactly the same as the system used in the computer model: "It is similar to the comparison made between an airplane wing and a bird's wing. Both are used for flight, but they work differently," explained O'Reilly. According to him, in the mechanism in the brain, it is necessary to teach and practice the pointer-like system to understand new sentences using familiar words, while the computer can be programmed to understand new sentences immediately without prior practice. "As our brain learns, it works better and better at processing these new types of data," he explained.

The news about the study

9 תגובות

  1. Joseph
    You have the right to believe in God, Bigfoot and the Tooth Fairy. All of these make sense to me. There are no confirmations for any of them, belief in any of them does not predict anything and they do not explain anything.

    But let's get back to consciousness. Do you think cockroaches have consciousness?

  2. All Western philosophers, including atheists, believe that there is consciousness: I think means I exist.
    I do not involve awareness of the existence of God in this panel.
    Whoever comes to claim that it is all about other thoughts, is not the mainstream yet.
    All in all, I say that we have not cracked the consciousness, and in my opinion, and I cannot prove exactly as the others here, it is a different process than the process of the senses. That is, they are not enough. The opposite view claims that if there are enough neurons (fast processors), then when the processing will be faster, much more parallel, the processing stream will think it is conscious.
    I think otherwise. In my opinion, we have largely understood the imitation of the senses and their perception but are not aware of them, therefore more kilos of those neurons will not necessarily make a difference. I believe that it can be understood in terms of exact science, someday, even if
    There is God and even if there is not. I do not include the assumption of its existence in the discussion.
    Apart from that, if we assume for a moment, and I believe so, that this is how consciousness is created, as some have said. From the moment it was created, it is a new level of abstraction.
    She is aware and there are processes that can be discussed from the perspective of psychology and philosophy. There is no need to go down to the language of the machine and it does not contribute to understanding either. If as a result the robots (we) produce a culture, it is appropriately spiritual.

    Now in my private opinion and I am a minority on this site, I believe in the existence of God - infinite consciousness. I do not force my opinion on anyone, and I do not believe that it is required for discussion between us or should prevent it. I believe that the same incomprehensible process can be understood with a scientific breakthrough.

  3. Before using explosive words such as consciousness or God, their meaning must be explained
    Because they have no collective meaning, and each person has a different meaning for the same words.

    The difference between a computer and a brain comes down to the fact that a computer cannot experience pain, heat, cold, taste and more...
    The computer can differentiate between heat and cold or taste but in the form of an output and not an experience.

  4. Are you willing to be moral as if there is a God even in the absence of God?
    Above the hardware layer, if a software layer is created, does the actions the software does have no meaning?
    That is, having a value that goes beyond the hardware that supports it. For philosophy above the layer of physics and biology.

  5. Joseph
    When a computer will consist of 100 billion processors, each processor connected to tens of thousands of other processors - then there is a chance for a machine with consciousness at the level of a human. But - these processors have to be extremely small, because otherwise there is an effect on the transit time of the signals. We are very far from that.
    But - consciousness is a property of a complex system like the brain. I mean - there is nothing outside of neurons. There is no soul, and there is nothing spiritual.

  6. Noah Thanks, I'm in. I wasn't convinced. In my opinion there is consciousness, we don't necessarily need God to explain it even though I believe in his existence. We do not understand her. Annie is convinced that Watson the computer (somewhat deep blue) has a consciousness as a result of its functionality that rivals ours. Maybe I'm wrong. I believe, and I may very well be wrong, that machine learning gives functionality to tools. I believe that there is consciousness, which arises from a higher level process, from a much more continuous flow of such processes. Something we don't know yet. What is here in the article is a program counter of a chip, or a stack pointer. not consciousness. I may be wrong. The research is not useless. It shows the creation of connections that are more complex or different from the basic ones.

  7. Every supposedly rediscovered machine learning topic does not explain the creation of consciousness. In my opinion, it is not created given a large enough pool of neurons, but there is a mechanism that has not been revealed yet. It can be said that we understand the tool used by consciousness.
    The movie Watson, the smartest computer in the world, describes the impressive development. However, I don't believe that consciousness simply joins the wonderful abilities that the IBM computer discovered to play the Jeffery game.

    Someone has not yet deciphered how consciousness is created to my understanding unfortunately. All the voices seem to be close to deciphering, and all the models of the human brain, mouse brain - will reach the functionality of the brain, but in the absence of an understanding of consciousness, and if there is consciousness, and it is not a by-product of a certain amount of neurons, we are far from there.

  8. There is a big difference between the human brain and the computer:
    The human mind examines reality and selects the words that suit it in order to talk about this reality.
    No, not the computer.
    The computer has no connection to reality other than a limited system of input and output.
    If they didn't tell him he wouldn't know if it was day or night. If it's black or white, if it's good or bad.
    The structure of programming languages ​​is also fundamentally different from the structure of human language.
    A programming language has a very limited vocabulary. No duplicates. There are no synonyms. And there are not many ways to link the words in order to build sentences.

Leave a Reply

Email will not be published. Required fields are marked *

This site uses Akismat to prevent spam messages. Click here to learn how your response data is processed.