-
Categories
-
Pharmaceutical Intermediates
-
Active Pharmaceutical Ingredients
-
Food Additives
- Industrial Coatings
- Agrochemicals
- Dyes and Pigments
- Surfactant
- Flavors and Fragrances
- Chemical Reagents
- Catalyst and Auxiliary
- Natural Products
- Inorganic Chemistry
-
Organic Chemistry
-
Biochemical Engineering
- Analytical Chemistry
-
Cosmetic Ingredient
- Water Treatment Chemical
-
Pharmaceutical Intermediates
Promotion
ECHEMI Mall
Wholesale
Weekly Price
Exhibition
News
-
Trade Service
Written | nagashi editor | Wang Cong typesetting | Shuichengwen With the development of science and technology, artificial intelligence such as face recognition and image recognition has entered the daily life of the public.
Most of these image recognition software are based on deep learning algorithms and massive data Set, extract and analyze facial or object features in pictures and video sources to identify people's identities and object names.
It is very interesting that for artificial intelligence, the recognition of human faces or objects is a very complex task, but for higher mammals, it is a very simple problem, although the specific learning process may involve complex nerves.
Meta reconstruction and brain activity.
Regrettably, the specific formation mechanism of these learning processes has not been explained.
On April 21, 2021, researchers from the Max Planck Institute of Neurobiology published a research paper titled: Mouse prefrontal cortex represents learned rules for categorization in Nature.
The study shows that mice have amazing classification capabilities.
In the learning process, mouse prefrontal cortex neurons show different dynamics when acquiring category selectivity, and show different levels of participation in subsequent rule conversions.
This proves how abstract information behaves at the neuron level.
For example, a toddler may be able to easily recognize objects that have been seen in a new picture book, such as "chairs.
"
The child made the right decision, which seems not particularly noteworthy, but it must go through a complicated process of classification and identification: the child must associate the chairs depicted in the book with the chairs he already knows, even if they may have Different shapes or colors.
But the question is, how do children do it? The answer is classification, which is the basic element of thinking.
Sandra Reinert, the first author of the study, explained: “Whenever a child encounters a chair, he will store this experience.
Based on the similarity between the chairs, the child’s brain will abstract by forming the category of “chairs”.
The attributes and functions of the chair allow him to quickly associate the new chair with the categories and knowledge it contains.
"In fact, our brains constantly classify things, not only chairs in childhood, but also Is any information.
So, what benefits will this bring us? Pieter Goltstein, the corresponding author of the study, said: "Our brains are trying to find a way to simplify and describe the real world.
Without classification, we cannot interact with the environment as efficiently as we do now.
In other words, if we lose this ability , We must learn to recognize every newly encountered chair.
Therefore, classifying sensory input is vital to living beings, but this underlying process of the brain is largely unknown.
"Mouse With amazing classification ability.
In this research, Pieter Goltstein and his research team have been committed to studying how the brain stores abstract information, such as learning categories.
Since this is difficult to study in humans, the researchers tested whether mice are classified in a way similar to ours.
To do this, the researchers showed the mice pictures of different striped patterns and gave them a classification rule.
One group of animals must divide the pictures into two categories based on the thickness of the stripes, while the other group must classify them according to the orientation of the picture.
The researchers were surprised to find that these mice were able to learn their own rules and reliably classify pictures into the correct category.
What's even more surprising is that after the initial training phase, these mice can even classify striped patterns that they haven't seen before into the correct category—just like kids who get a new book.
Mice can inductively classify visual stimuli.
Not only that, when researchers switch the classification rules, the mice can ignore what they have learned before and reclassify pictures according to the new rules—this is exactly what we humans are learning new things.
Things that I’ve been doing all the time.
Therefore, this study proved for the first time the extent to which mice are classified, with what precision, and have the ability to abstract close to humans! Neurons gradually developed a category representation.
It is worth noting that in the process of classification learning, neurons showing category selectivity were found in several regions of the neocortex of the mammalian brain.
Among them, the prefrontal cortex seems to play a role in this aspect.
Important role.
Based on this, the research team further studied the basis of mouse brain classification.
They focused their research on the prefrontal cortex of the brain, which is also the brain area where humans participate in complex thinking processes.
The results of the study showed that when the mice classified the striped patterns, certain neurons in this area became very active.
What's more interesting is that different groups of neurons will react selectively to different categories.
A single neuron in the mouse prefrontal cortex produces a category-selective response, Tobias Bonhoeffer, director of the Max Planck Institute for Neurobiology, said: "The discovery of category-selective neurons in the mouse brain is a key point.
This makes us number one.
The activity of these neurons from the beginning to the end of category learning can be observed at a second time.
This shows that neurons do not gain selectivity immediately, but gradually develop during the learning process.
"Classification selection neurons are part of long-term memory.
Scientists believe that only When the acquired knowledge changes from short-term memory to long-term memory, the classification-selective neurons in the prefrontal cortex will play a role.
There, these cells store the classification as part of semantic memory, and semantic memory is the collection of all factual knowledge.
The research team repeatedly imaged individual cells in the mouse prefrontal cortex to study the formation of neuron types throughout the learning process.
They found that in the learning process, prefrontal cortex neurons showed different dynamics when acquiring category selectivity, and showed different levels of participation in subsequent rule conversions.
Some neurons selectively and uniquely respond to categories and reflect generalization behavior.
The prefrontal cortex of mice contains unique class-regulating neurons.
Therefore, the class representation in the prefrontal cortex of mice is obtained gradually during the learning process, rather than temporarily recruited.
At the same time, this gradual process shows that neurons in the prefrontal cortex are part of a specific semantic memory for visual categories.
It is worth noting that learning categories are a way for the brain to simplify the real world, but it also means that these categories may not correctly reflect reality.
Conclusion In a complex natural environment, the ability of animals to classify sensory stimuli is vital to their survival.
Learning to remember categories rather than just individual examples allows animals with classification capabilities to show better behavioral flexibility And computing advantages.
This study confirmed that mice also have a strong classification ability, which can efficiently recognize and classify striped patterns according to established rules.
This classification ability is gradually acquired in the learning process, and is closely related to the neuronal activity of the prefrontal cortex of the brain, which reveals the specific formation mechanism of the classification ability of higher mammals.
By studying the category learning of mice, the research adds important details to the neuronal basis of abstract thinking, and reminds us that complex thinking is not just the exclusive ability of humans.
Link to the paper: Open for reprinting This article is open for reprinting: Just leave a message in this article to inform
Most of these image recognition software are based on deep learning algorithms and massive data Set, extract and analyze facial or object features in pictures and video sources to identify people's identities and object names.
It is very interesting that for artificial intelligence, the recognition of human faces or objects is a very complex task, but for higher mammals, it is a very simple problem, although the specific learning process may involve complex nerves.
Meta reconstruction and brain activity.
Regrettably, the specific formation mechanism of these learning processes has not been explained.
On April 21, 2021, researchers from the Max Planck Institute of Neurobiology published a research paper titled: Mouse prefrontal cortex represents learned rules for categorization in Nature.
The study shows that mice have amazing classification capabilities.
In the learning process, mouse prefrontal cortex neurons show different dynamics when acquiring category selectivity, and show different levels of participation in subsequent rule conversions.
This proves how abstract information behaves at the neuron level.
For example, a toddler may be able to easily recognize objects that have been seen in a new picture book, such as "chairs.
"
The child made the right decision, which seems not particularly noteworthy, but it must go through a complicated process of classification and identification: the child must associate the chairs depicted in the book with the chairs he already knows, even if they may have Different shapes or colors.
But the question is, how do children do it? The answer is classification, which is the basic element of thinking.
Sandra Reinert, the first author of the study, explained: “Whenever a child encounters a chair, he will store this experience.
Based on the similarity between the chairs, the child’s brain will abstract by forming the category of “chairs”.
The attributes and functions of the chair allow him to quickly associate the new chair with the categories and knowledge it contains.
"In fact, our brains constantly classify things, not only chairs in childhood, but also Is any information.
So, what benefits will this bring us? Pieter Goltstein, the corresponding author of the study, said: "Our brains are trying to find a way to simplify and describe the real world.
Without classification, we cannot interact with the environment as efficiently as we do now.
In other words, if we lose this ability , We must learn to recognize every newly encountered chair.
Therefore, classifying sensory input is vital to living beings, but this underlying process of the brain is largely unknown.
"Mouse With amazing classification ability.
In this research, Pieter Goltstein and his research team have been committed to studying how the brain stores abstract information, such as learning categories.
Since this is difficult to study in humans, the researchers tested whether mice are classified in a way similar to ours.
To do this, the researchers showed the mice pictures of different striped patterns and gave them a classification rule.
One group of animals must divide the pictures into two categories based on the thickness of the stripes, while the other group must classify them according to the orientation of the picture.
The researchers were surprised to find that these mice were able to learn their own rules and reliably classify pictures into the correct category.
What's even more surprising is that after the initial training phase, these mice can even classify striped patterns that they haven't seen before into the correct category—just like kids who get a new book.
Mice can inductively classify visual stimuli.
Not only that, when researchers switch the classification rules, the mice can ignore what they have learned before and reclassify pictures according to the new rules—this is exactly what we humans are learning new things.
Things that I’ve been doing all the time.
Therefore, this study proved for the first time the extent to which mice are classified, with what precision, and have the ability to abstract close to humans! Neurons gradually developed a category representation.
It is worth noting that in the process of classification learning, neurons showing category selectivity were found in several regions of the neocortex of the mammalian brain.
Among them, the prefrontal cortex seems to play a role in this aspect.
Important role.
Based on this, the research team further studied the basis of mouse brain classification.
They focused their research on the prefrontal cortex of the brain, which is also the brain area where humans participate in complex thinking processes.
The results of the study showed that when the mice classified the striped patterns, certain neurons in this area became very active.
What's more interesting is that different groups of neurons will react selectively to different categories.
A single neuron in the mouse prefrontal cortex produces a category-selective response, Tobias Bonhoeffer, director of the Max Planck Institute for Neurobiology, said: "The discovery of category-selective neurons in the mouse brain is a key point.
This makes us number one.
The activity of these neurons from the beginning to the end of category learning can be observed at a second time.
This shows that neurons do not gain selectivity immediately, but gradually develop during the learning process.
"Classification selection neurons are part of long-term memory.
Scientists believe that only When the acquired knowledge changes from short-term memory to long-term memory, the classification-selective neurons in the prefrontal cortex will play a role.
There, these cells store the classification as part of semantic memory, and semantic memory is the collection of all factual knowledge.
The research team repeatedly imaged individual cells in the mouse prefrontal cortex to study the formation of neuron types throughout the learning process.
They found that in the learning process, prefrontal cortex neurons showed different dynamics when acquiring category selectivity, and showed different levels of participation in subsequent rule conversions.
Some neurons selectively and uniquely respond to categories and reflect generalization behavior.
The prefrontal cortex of mice contains unique class-regulating neurons.
Therefore, the class representation in the prefrontal cortex of mice is obtained gradually during the learning process, rather than temporarily recruited.
At the same time, this gradual process shows that neurons in the prefrontal cortex are part of a specific semantic memory for visual categories.
It is worth noting that learning categories are a way for the brain to simplify the real world, but it also means that these categories may not correctly reflect reality.
Conclusion In a complex natural environment, the ability of animals to classify sensory stimuli is vital to their survival.
Learning to remember categories rather than just individual examples allows animals with classification capabilities to show better behavioral flexibility And computing advantages.
This study confirmed that mice also have a strong classification ability, which can efficiently recognize and classify striped patterns according to established rules.
This classification ability is gradually acquired in the learning process, and is closely related to the neuronal activity of the prefrontal cortex of the brain, which reveals the specific formation mechanism of the classification ability of higher mammals.
By studying the category learning of mice, the research adds important details to the neuronal basis of abstract thinking, and reminds us that complex thinking is not just the exclusive ability of humans.
Link to the paper: Open for reprinting This article is open for reprinting: Just leave a message in this article to inform
