Language, brain and computation: from semiotic asymmetry to recursive rules

This article reviews the comparative analysis of the most promising theoretical areas exploring the correlation principles of the structural / functional brain organization and cognitive characteristics of the language. A special role in the presented approaches is assigned to the principles of computability. The article discusses the advantages and disadvantages of the modular and holistic models in the context of semiotic asymmetry. It is important to emphasize that the language model as a set of hierarchically represented objects is consistent with only with the connectivism paradigm (brain as a network), in which the main element is the functional neuron, and it is not the biological neuron but its simplified information model. The informational simplification creates a number of contradictions. We have analyzed the methodological basis of neural network approach and principles of recursive organization. The main question is, do neuron structures determine the principles of mental representations and computational grammar models? The uniformity of syntax and logical structure of neurons creates metaphorical and a not completely correct hypothesis that brain is a network of functional units that computes grammar and syntax rules. One of the objectives of this work is to identify the methodological inconsistencies arising from the computational models of language structures and language activity.