Pavlov principle and brain reverse engineering

The general principle of the operation of neural systems is considered, which makes clear the efficiency of information-computational structures consisting of neural-like elements. At the heart of the formulated principle lies the highly anticipated hypothesis of I.P. Pavlov that the "correctly organized" modifications of the connections between the elements and structures of the nervous system is the basis of "higher nervous activity of man" (in the terminology proposed by Pavlov). The possibility of a clear formulation of the principle, which is naturally called the Pavlov Principle (PP), arose only recently, as a result of the victorious procession of the family of computer training algorithms such as Deep Learning, which led to extremely efficient schemes for using "fine-grained" computing devices (GPU, etc.) solving cognitively complex problems. The article briefly outlines the line of development of ideas about the work of the nervous system, starting with the discovery by I.P. Pavlov of systemic conditioned reflexes, then through the neural and synaptic schemes of E. Konorsky and D. Hebb, the perceptron of F. Rosenblatt, the connectionist works of the 1980s, to modern schemes of deep learning. For the proper functioning of PP based neural networks essential is (initial) randomness of the interneuronal connections. A computational example of the solution of the model problem based on the "direct" application of the Pavlov Principle is given.