The market of modern neurointerfaces, despite its active development, unfortunately, can offer users only a number of existing prototypes that have a relatively low accuracy and identification reliability of the human operator control effects. In addition, any neurointerface on the market must be individually tailored to each operator, which makes it difficult to objectify its accuracy, precision and reliability. The first step in solving the above problems is to conduct a comparative analysis of different price segments of the market of existing neurointerface technologies, as presented in this article. The market research revealed that despite the disadvantages of electroencephalography, it is one of the most accessible non-invasive methods of recording biological signals in neurointerface systems. To facilitate future research, the main advantages and disadvantages of known models and methods of signal analysis in neurointerfaces have been considered and analyzed. In particular, in the context of signal pre-processing, advantages and disadvantages of such methods as Common Average Referencing, Independent Component Analysis, Common Spatial Patterns, Surface Laplacian, Common Spatio-Spatial Patterns and Adaptive Filtering are considered. At the stage of evaluating the informative characteristics of the signal, the analysis of models and methods based on the models of adaptive parameters of autoregression, bilinear autoregression, multidimensional autoregression, fast Fourier transform, wavelet transformation, wave packet decomposition is performed. Besides, a comparative analysis of the most common methods of identification (recognition) of control effects of the human neurointerface operator, namely, the method of discriminant analysis, the method of reference vectors, nonlinear Bayesian classifiers, classifiers of nearest neighbors, artificial neural networks is carried out. The study of neurointerface technologies provides researchers with additional grounds for a sound choice of mathematical, software and hardware of neurointerface systems, as well as contributes to the development of new versions with increased accuracy, reliability and reliability.