Predictive tools for analysis of verbal model suggestive potential

Context and relevance. Prognostic technologies of language phenomena research are in demand in different fields of activity: business communication, media planning, forensic science, psychological practice, suggestive linguistics, political and public activities. The theoretical basis is the concept of latent suggestive resources of verbal models of any complexity, which can be identified and analyzed with specially designed predictive tools. Purpose. The main objective of the study is to develop new and to adapt already available in other scientific paradigms technologies to explore verbal model suggestive potential, as well as to create a pool of tools that can be used to solve specific problems. Hypothesis. The working hypothesis is the assumption of causal relationship between qualitative and quantitative indicators of deep latent characteristics of any verbal model (brain rhythms patterns, associative color, informational redundancy properties, categorical-statistical assessments, emotional state status, rhythmic codes of complex models and many others). Methods and materials. The results of the research are based on a number of methods and technologies, among which experiments, mathematical statistics and scaling methods, correlation analysis, content-analysis, instrumental analysis of brain slow electrical activity, ranking using classifiers of different types (probabilistic, neural networks and sets of logical rules), creation of specialized software and development of digital models on the basis of the obtained experimental data, emotional states decoding and digital processing of visual information and others. In the selection of material and methods, the author is guided by the requirement of verifiability of results, the validity of which can be checked by other researchers. Results. The laws of action and impact of verbal models of different levels of complexity (sound-letter, word, text) have been established in the analysis. Five computer programs for automated data analysis in Russian, English, German, Tatar and Bashkir have been created. In addition, associative color matrices of the five languages sound-letters were calculated. Associative chromaticity of the languages was presented in the form of polychrome pictures. New analysis technology of text suggestive potential was described. A mathematical model for estimation of text information redundancy was build. Algorithms for estimating the suggestive potential of verbal models were proposed. Currently, work is ongoing to establish an author’s constant as an identifier of individual language. Conclusions. Latent impact potential of verbal models can be quantified and described by quality indicators of suggestive resources (from maximum negative to maximum positive). The synkrisis procedure developed by the author allows to describe the implicit conditionality of various parameters of the latent potential and on the basis of the established regularities calculate the duration of the impact.