Pattern Separation in the Model of Mammalian Visual System

Large artificial neural networks are examined. The structures discussed in this article simulate the cortex of the mammalian visual system. We use a modular architecture of the cortex divided into sub-regions. The modular structure of visual system is appropriate task for cluster computations. Our models consist of 16 to 65 thousands Hodgkin-Huxley neural cells. Such number of simulated neurons allowed us to observe liquid computing phenomena. Separation ability of groups of neural microcircuits was also observed. We show that such property depends on electrical parameters of particular cells. The separation ability as a function of the membrane capacitance is discussed in some detail