In the late 1950s, a colleague and I installed a biochemistry lab for the purpose of doing research in brain biochemistry in a building devoted to the teaching of psychiatry. In the same space, another graduate student was building an electronic model of a brain according to a design published (in 1952) by W. Ross Ashby, an English cyberneticist. (Cybernetics is the study of complex systems including computers and brains.) This early electronic “brain” would change its response to a particular input from the operator if the operator “punished” it by activating a specific switch. In this way the output of the machine could be “shaped” (within the limits of its abilities) by how the operator behaved toward it. The machine thus crudely mimicked the formation of a conditional response seen in animals such as Pavlov’s dog which learned to salivate at the sound of a bell.
Today, we know far more about brains. In response to this new knowledge Dr. Henry Markram began a project entitled the “Blue Brain Project” in 2005. Using IBM’s Blue Gene supercomputer he modeled a small, well known area of the rat’s brain. His ultimate goal is to incorporate into his computer-based model everything that is currently known about how our brain is connected and how these connections are altered over time (as a result of experience). The enormity of this task can be appreciated when one realizes that there are between 80 and 120 billion neurons in each of our brains and each of these has hundreds to thousands of synapses (connections) to other neurons. Furthermore, about 60,000 research papers are published each year in the fields relevant to brain functioning, many of which present us with new findings about how the brain functions. Dr. Markram’s desire is to be able to model the brain sufficiently well to better understand disorders of the brain and the interactions of drugs upon the brain in the hope that this will lead to more effective treatment of brain diseases.
This project has been controversial and Dr. Markram has been accused of spending a lot of money on a project that may prove impossibly complex. However, there is a chance of a sizable prize for him. The European Commission has selected his project as one of six competing for two Future and Emerging Technologies Flagship Awards to be announced this year and then funded with $1.4 billion over the next ten years.
There is an old cybernetic axiom that no machine can understand another machine as, or more, complex than itself. If true, it would mean that we humans could not expect to ever fully understand our own brains. However, Dr. Markram’s goal appears to lie short of this and his electronic model may well enable us to glean some new and valuable insights into the workings of our minds in health and disease. W. Ross Ashby created many pithy aphorisms relevant to this effort to make brain-like devices, of which the following is an example: “Some say the first requirement is that it shall weigh [only] 45 ounces,” (the weight of a human brain).
Questions and suggestions from readers are welcomed and will be responded to in future editions of this column. Contact me at firstname.lastname@example.org