In the journal Frontiers in Computational Neuroscience, an article appeared, the authors of which declare that they made an important step to understanding how the work of neurons in the human brain – the most complex structure known to us – determines the functions of this organ. A team of scientists from Switzerland, France and the United Kingdom participating in the Blue Brain project claims that our brain, in the process of processing information, creates multidimensional objects from neurons and connections between them and these objects represent a “world that we could not have imagined before”.
Blue Brain Project
Blue Brain is a project founded by the Institute for the Study of the Brain and Reason of the Federal Polytechnic School of Lausanne (Switzerland) in 2005. The goal of the initiative is to create a detailed digital version of the mammalian brain (first rat, and ultimately human) to understand the fundamental principles of his work and to find out how to use this knowledge in medicine.
To measure all the properties of the brain at all levels (in molecules, cells, individual regions of the cortex and their aggregate), and the interaction between them can be impossible. For this reason, scientists using biological data, the results of experiments and a supercomputer try to create a simulation that will give a detailed “map” of the brain. Such a model must respond to different stimuli in the same way as a real brain, so that scientists can see which processes and how go in it. In 2015, the Blue Brain team presented the first digital version of the neocortex fragment – 31,000 rat neurons responded to external stimuli with waves of coordinated electrical activity. Neocortex is the last area of the cerebral cortex that emerged during the evolution, responsible for higher nervous functions – sensory perception, conscious thinking, speech of humans.
Why it is so difficult to deal with the brain?
According to modern ideas, the human brain contains about 86 billion neurons – cells that store and transmit information. They are in constant interaction with each other, forming a neural network that allows us to act, think, recognize ourselves, make decisions, and so on.
A system consisting of such a huge number of neurons is too complicated to be studied in detail by modern methods. Because of this, we still do not know how not only the human brain functions, but also the brain of less complex mammals. Blue Brain expects that the mathematical model will help to understand the processes occurring in the head.
Processing of experiment.
Researchers studied neural connections using the method of algebraic topology – a system that describes the properties of objects regardless of their changing geometric shape. “It’s like a telescope and a microscope at the same time,” explains mathematician Catherine Hess, who worked on the project. “You can look at networks in detail to find hidden structures – like trees in a forest – and at the same time see voids and clearings.” Applied to the brain, this means that algebraic topology provides a tool for studying the properties of both individual neurons and the structure of the brain as a whole.
In the described experiment, algebraic topology was used to study virtual brain tissues, and then rechecked results on the real rat brain to make sure that the simulation worked correctly. Scientists have found that when the brain is presented with a certain stimulus, the neurons assemble into groups (“cliques”) that line up in sequence and form complex geometric objects with space (“cavity”) inside. The more neurons in the “click”, the more complex this group is arranged and the more dimensions have the resulting figures.
It turned out that tens of millions of figures of neurons with up to seven measurements can be formed in each individual part of the brain, and up to 11 measurements could be detected in some structures. According to the project participants, earlier scientists had never seen anything like this in either computer or biological models.
Multidimensional objects in the brain arise when it needs to process information; after the completion of this process they disintegrate. According to the scientists, neurons respond to the stimulus in a very organized way, creating figures from simple to complex. “The brain is building, and then destroying, towers from multidimensional fragments, starting with sticks (1D), then planes (2D), cubes (3D) and then more complex shapes with four, five or more dimensions,” says mathematician Ren Levy from Aberdeen University (Scotland). – Activity in the brain resembles a multidimensional sand castle, which materializes, and then completely crumbles. ”
So, what does it mean?
Scientists are not sure yet. They saw that in the functioning brain, the cells and signals between them do not behave chaotically. In the process of processing information, neurons build “cliques” and “cavities,” the existence of which had not previously been known. The figure of two interconnected neurons looks like a segment. Three is a triangle, four is a pyramid. The more neurons and connections between them, the more complex and multidimensional shapes are obtained, and although our brain can not imagine them, mathematicians can describe them.
According to Catherine Hess, the finding of the team from Blue Brain helps explain why it was so difficult for science to understand the relationship between brain structure and its functions. We are used to perceiving the world in three dimensions and we usually examine the activity of the brain with the help of appropriate tools that allow us to see only the “shadow” of what is really happening, not the full picture.
Now, researchers have to understand whether our ability to perform complex tasks depends on the creation of these figures inside the brain and their complexity. Also, it remains unknown what exactly causes neurons to be built in such an intricate way.