PHYSICAL MODELS OF COGNITION

This paper presents and discusses physical models for simulating some aspects of neural intelligence, and, in particular, the process of cog nition. The main departure from the classical approach here is in util ization of a terminal version of classical dynamics introduced by the author earlier. Based upon violations of the Lipschitz condition at eq uilibrium points, terminal dynamics attains two new fundamental proper ties: it is spontaneous and nondeterministic. Special attention is foc used on terminal neurodynamics as a particular architecture of termina l dynamics possesses a well-organized probabilistic structure which ca n be analytically predicted, prescribed, and controlled, and therefore which presents a powerful tool for modeling real-life uncertainties. Two basic phenomena associated with random behavior of neurodynamic so lutions are exploited. The first one is a stochastic attractor - a sta ble stationary stochastic process to which random solutions of a close d system converge. As a model of the cognition process, a stochastic a ttractor can be viewed as a universal tool for generalization and form ation of classes of patterns. The concept of stochastic attractor is a pplied to model a collective brain paradigm explaining coordination be tween simple units of intelligence which perform a collective task wit hout direct exchange of information. The second fundamental phenomenon discussed is terminal chaos which occurs in open systems. Application s of terminal chaos to information fusion as well as to explanation an d modeling of coordination among neurons in biological systems are dis cussed. It should be emphasized that all the models of terminal neurod ynamics are implementable in analog devices, which means that all the cognition processes discussed in the paper are reducible to the laws o f Newtonian mechanics.