FUNDAMENAL RESEARCH OF THE SPEECH SEGMENTOLOGY PHENOMENON (CONSIDERING COGNITIVE REFLECTION AND COMMUNICATIVE ACTIVITIES APPROACH)


2018. № 3 (17), 225-237

 Moscow State Linguistic University, Lomonosov Moscow State University

Abstract:

This paper considers implications of the semantic decoding process of Russian spoken and written texts on the basis of media-segmentation and macrosegmentation. We suggest some experimental approaches to analysis of verbal utterances regarding monologues and dialogues especially for distant mediated discourses (telephone conversations). The experimental corpus included specially adapted texts (quasitexts) where punctuation marks, capital letters, gaps, paragraph marks were eliminated.

There were also two types of experiments: perceptual-auditory and perceptual-visual ones.

To determine the role of the auditory canal for the semantic interpretation, two experiments were conducted; two groups of subjects (native speakers) analyzed quasitexts in two forms: spoken language and written language reproduction of the same special quasitexts with and without any distortion.

On the basis of quantitative probabilistic-frequency analysis it is concluded that the establishment of boundaries for quasitext semantic segments is probabilistic in nature. Localization of boundaries and ways of marking are determined by the choice of a particular marker type and semantic structure of the text.

It is assumed that the process of perception, cognitive analysis, interpretation and semantic “reconstruction” of a quasitext on media- and macro-segmentation levels activates various verbal and cogitative processes. The degree of media-segmentation variability when reconstructing texts is significantly lower than the degree of macro-segmentation variability.

The study of the written texts visual perception with special disorders and auditory perception of two kinds of these quasitexts is very important to the understanding of human cognitive mechanisms and cogitative activity especially in the field of semantic recognition on the basis of segmentation phenomenon.