The brain is a very complex system, composed of many (billions) specialized cells, called neurons, that are dynamically interacting with each other via so called synaptic connections. The neurons in the brain, accordingly, form a very large network that ultimately processes peripheral sensory signals and performs various functions, such as control of movements. The synaptic connections in this network play a very important role because their specific organization in a given brain area determines a specific function performed by that area. Researchers in the Brain Sciences Center develop large-scale mathematical models of the brain network and then explicitly simulate these models using high-performance supercomputers.
Large-scale computer simulations are complementary to experimental studies of the brain. Mathematical models and computer simulations are important for understanding how the organization of synaptic connections shapes the underlying brain function, and also allow to decipher specific mechanisms of the function implementation. In addition to stimulating the building of unified theoretical frameworks of brain function, explicit computer simulations of large-scale realistic models also help researches screen out a number of working hypotheses much faster and at a lesser expense than testing them out in experimental studies. Predictions of simulations, in turn, provide guidance for future experiments.
The broader impact of large-scale modeling studies carried out at the Brain Sciences Center is two-fold.
First, the success of the current research would stimulate studies of functional disorders caused by focal brain lesions, such as stroke, resulting in the loss of cells and/or connections between them. The large-scale neural network simulations are able to model explicitly spatio-structural aspects of the underlying brain structure and should provide, therefore, appropriate theoretical means for the investigation of the cell/connection loss-induced neurological disorders.
Second, the current research should be also important for developing new generation of prosthetics that are driven by brain signals to assist, for example, paralyzed patients or amputees. Particularly, simulations of large-scale brain models could be important for pin-pointing the relevant brain signals and their spatial localization.