Ambitious Billion-Euro Human Brain Project Kicks Off in Switzerland

19 25 Loading

brain-model
It’s mind-blowing how little we know about the human brain. The old saw is that if our brain were simple enough to understand, we wouldn’t be smart enough to understand it.

But that’s really just a cute way of saying that research on the brain is limited by the same complexity and centrality that make the organ so interesting to us. Animal models can’t shed light on the intricacies of our elaborate command centers. Even when performing brain surgery, doctors can’t see how the seemingly humble lump of grey goo works.

neuron-on-screenAs computers have grown increasingly powerful and genetics more sophisticated, they have opened up new avenues for exploring how the brain’s 85 billion interconnected neurons work. The Human Brain Project, which just kicked off with an initial round of meetings in Lausanne, Switzerland, has promised to bring together the scientific literature on mouse and human brains and build a functional computer model of the human brain to expand scientific understanding of the all-important organ.

It will be no small task.

Co-headed by neuroscientist Henry Markram, a South African native who also leads the Blue Brain Project, and Karlheinz Meier, a physicist with expertise in computer modeling, the 10-year project is mainly European but has recruited some specialized teams from Israel, China, Canada and the United States.

“What we are proposing is to establish a radically new foundation to explore and understand the brain and its diseases, and to use that knowledge to build new computing technologies,” Markram said in a video about the project.

The brain is so complex, however, that even today’s computers are not powerful enough to do that work. The work will require an exascale supercomputer, which won’t become available until roughly 2020.

“The brain has huge numbers of interacting parts: even if we had all the data we would need, simulating the relevant interactions would still be very difficult, especially at the molecular level. However, if we start work on the software now, and steer manufacturers to design machines that meet our requirements, supercomputers will soon give us enough computing power to at least begin molecular-level simulations,” the organizers explain.

The project’s modeling ambitions have stirred some controversy, David Kleinfeld, a U.C. San Diego physicist who is contributing research on the vascular structure of mouse brains, told Singularity Hub.

“The good news is that, at least in the U.S., people always approach neuroscience from an illness perspective, and they’re approaching it from an engineering perspective: This is nature’s way of organizing matter so that it absorbs knowledge, it’s pretty incredible really,” he said.

“The downside is that it’s a pretty immature field, despite the number of people involved, and it’s not clear what a central organization is going to do, if centralizing things is good or bad at the end for creativity,” Kleinfeld said.

Because the field is so new, Kleinfeld said, there has been no systematic audit of previous research. That work, which comes first in the project’s to-do list, may be the least glamorous but is arguably the most necessary. The results will subsequently be used to seed the computer models.

Images courtesy Human Brain Project

Discussion — 25 Responses

  • Calum October 15, 2013 on 9:29 am

    This looks like an undeclared whole brain emulation project. Which could be great, but what if it leads to an intelligence explosion? We haven’t got a plan yet for making sure that a conscious machine is good news for us meat-brain humans. A roadmap is sketched at bit.ly/170uDbM

  • IlyaRoy October 15, 2013 on 1:21 pm

    Looks to me like EU is trying to solve a very hard problem the same way they manage the continent – via centralized socialist means. I would predict the results will match the super-computing leap Japan launched in the 80’s.

  • why06 October 15, 2013 on 3:30 pm

    I believe initial research is good, but I don’t think you can force these things. A lack of resources often spurs more creative solutions. When all the parts are assembled, it will happen. Until then I think we just wait patiently and aim for the low hanging fruit.

  • Mike Huang October 16, 2013 on 12:03 am

    It’s mindblowing how much we know about the brain, but yet understand so little of it.

    • Calum Mike Huang October 16, 2013 on 1:39 am

      One of the interesting things about this approach to brain-building is that – theoretically, at least – you don’t really need to understand how the brain works in order to make a working copy. Imagine making an exact copy of a car, accurate down to the last millimeter. If you got it exactly right, it would work as well as the original even if you knew nothing at all about the physics of combustion or motion.

      And in part, that is the whole point. They are building a copy of a brain in order to help find out how ours work. The worry is that they may end up creating something very different instead.

  • seemsArtless October 16, 2013 on 5:55 am

    Interesting to see how this maps with the existing “Virtual Brain” project at http://www.thevirtualbrain.org/

  • Tober October 17, 2013 on 1:13 pm

    It’s just a waste of time and money. In such situations, you simply have to know the good solution before the starting.
    We have that the greatest threat to mankind turn into a solution for the biggest disease of modern society – a lack of humanness. Socialization – moral evaluation of data – is the main task because it is easy to make AI ,​​of course if you look from the right angle.

  • Nelson Cabrera November 3, 2013 on 8:01 am

    So if we are successful in creating a working model of a human brain on a supercomputer, what are the ethics involved if consciousness emerges? Do we have the right to experiment on what, in essence, is a human mind?

    • Calum Nelson Cabrera November 3, 2013 on 10:44 am

      Good question. A lot of ethical questions are going to be thrown up by this technology – if we survive the event. If the model generates a consciousness, we should treat it with the same respect we’d treat a natural-born, free-range meat human. There’ll be a strong temptation not to, though.

    • Tober Nelson Cabrera November 3, 2013 on 10:52 am

      Fortunately, there is no good working model. Big boys are just playing with big toys and very creative falsify falsification.
      Unfortunately, there are no ethics in the plans. But, ethics is the essence and goal.
      By the nature, we have to do that. Our task as individuals is to raise and educate children-next generation, and as a species to create just 1 supercomputer.
      Human mind is figuratively speaking – Great Pyramid of Giza. But just another tombstone.