Human Brain Project: preparations are in full swing
Europeans will create a computer model of the brain
Daniel Schrader, Deutsche Welle
The European Commission allocates more than a billion euros to create a model of the human brain. If successful, the project will become an important contribution to both medicine and digital technologies.
The European Union is preparing a scientific breakthrough. More than 1.2 billion euros will be spent in the next ten years to create a computer model that should combine all scientific knowledge about the human brain. This scale of funding puts the new Human Brain Project on a par with such programs as the flight to the Moon or the famous Human Genome Project. The similarity of the names is clearly not accidental. Ten years ago, American scientists presented to the world the sequence of nucleotides that make up human DNA, and European experts hope to create the most complete computer model of the brain in 10 years.
Like CERN (the European Center for Nuclear Research, another ultra-expensive European project), the Human Brain project will be managed from Switzerland. Here, in Lausanne, it is planned to create a huge scientific center called Neuropolis. However, this is just the tip of the iceberg. About 80 groups of scientists from all over Europe, as well as in Israel and the USA will work on modeling the human brain.
Since the main goal is computer modeling of the brain, the project is destined to combine the knowledge accumulated in neuroscience with the latest achievements of computer technology. Scientists expect to create a 3D model with a variable scale that would reflect both individual nerve cells and entire regions of the brain. This will require bringing together a huge body of knowledge about the structure of the brain. It is necessary to reproduce the pathways of nerve signals from neuron to neuron and from one area of the brain to another. To do this, scientists intend to use knowledge about the mechanism of transmission of certain signals, but they hope to understand their functions already in the process of creating a model.
A thinking supercomputer"I imagine that in ten years we may be able to understand why certain parts of the brain are arranged this way and not otherwise," formulates the goal of the project neuroscientist Katrin Amunts, an employee of the German Research Center in Julich and a professor at the University of Dusseldorf.
– Why does the area controlling motor skills have such a structure, and not another? And how does it differ from the area responsible for vision? What are the mechanisms behind this? What biological code triggers a cognitive or emotional response?"
Modeling the brain is not a goal, but a means to better understand how it functions, emphasizes Professor Amunts. The acquired knowledge will certainly be in demand in medicine, for example, in the therapy of Parkinsonism, Alzheimer's disease, epilepsy, and so on.
The research center in Yuliha is the largest of all involved in this project. There will be eight working groups working here – more than in London, Paris, Berlin or Lausanne. Here, the brain model is destined to settle physically. In a huge machine room, in 28 cabinets, each the size of a telephone booth, Juqueen is located – the most powerful supercomputer in Europe today. It came into operation in February of this year, its computing power approximately corresponds to the total computing power of one hundred thousand ordinary home computers. While Yukvin is busy with calculations in the field of plasma physics, molecular biology and climatology, but in ten years it is in him, or rather, in his even more powerful descendants, that a brain model capable of processing information, responding to external factors, and possibly thinking and even feeling, should settle…
Is it time for brain modeling?Despite the enthusiasm of the European Commission and the project participants, many European scientists are skeptical about the Human Brain project.
Professor of Neurology Christian Elger, head of the epileptology clinic in Bonn, is pleased with the increase in funding for this area of research, but believes that the time for brain modeling has not yet come. "In my opinion, this project is being started one or even two decades ahead of time," the scientist believes. – And maybe it's impossible in principle. Because the human brain, unlike a computer, is characterized by incredible complexity and variability."
Other concerns are related to the technological side of the project. After all, computer modeling assumes the presence of special software, and neuroscientists have long looked with distrust at the results of the work of computer scientists in the field of creating artificial intelligence. Too many research projects that tried to reproduce brain activity by computer means ended in failures in the 80s and 90s of the last century, emphasizes Professor Stefan Rotter, head of the Bernstein Center for Neuroinformatics at the Faculty of Biology at the University of Freiburg.
The scientist also believes that the modeling of mental activity is a matter of the distant future, but still expects interesting results from the project concerning the functions of certain subsystems of the brain. However, in his opinion, the bet should be made not on modeling, but on experimental work. "It's important to understand a real brain, not a technically complex model," explains Professor Rotter, "otherwise you'll get two problems later: a brain that you still don't understand, and a model that you don't understand either."
Knowledge about the brain and digital technologies of the futureBut the leaders of the Human Brain project – and, above all, its main initiator Henry Markram – emphasize that the first steps need to be taken now.
At the same time, they distance themselves from classical approaches of computer science. If artificial intelligence was supposed to imitate the work of the brain using relatively primitive programs, then a completely different approach was chosen in the new project. The entire nervous system, every neuron and every signal should be reflected in the model. All biochemistry, for example, the synthesis of neurotransmitters, is supposed to be omitted in this model. But the key role will be played by electrical signals exchanged by neurons. There are only two possible states – "yes" or "no", which corresponds to the binary code used in computer science – "1" or "0". This greatly simplifies the task: such signals can easily be transferred into the bowels of a computer system by creating an exact copy of them.
This project is designed to provide Europe with one of the leading places both in the field of brain research and in information technology. Perhaps the skepticism of many neuroscientists is due precisely to the fact that the computer in this project is given no less attention than the brain. The United States, for example, has taken a different path, having recently allocated more than three billion dollars for the development of molecular and cellular neurobiology. The Europeans do not want to compete with the Americans, but to complement them. Knowledge about the brain will help create the computer technology of tomorrow.
"We have to be ambitious," explains Carl–Christian Buhr, an expert of the European Commission on the development of digital technologies. He is sure that even if it is not possible to create a working model of the human brain in ten years, the knowledge gained in the course of research will pay off the costs a hundredfold. Apparently, it will take much more time than ten years to create a real thinking machine. But Europeans are starting to learn from nature, which has created such a perfect creation as the human brain, today.
Portal "Eternal youth" http://vechnayamolodost.ru27.05.2013