Abstract
<jats:p>The article addresses the urgent issue of integrating generative artificial intelligence technologies into the process of teaching philological disciplines amidst the digital transformation of higher education. The paper substantiates the necessity of transitioning from reproductive reading methods to innovative models for visualizing the semantic structures of text. The relevance of the study is driven by the contradiction between the increasing volume of digital information and the declining capacity of students for "deep reading." The scientific novelty of the research lies in the development of an original "cognitive mapping" methodology that defines a sequence of stages for transforming textual meanings into visual models. Drawing on L. S. Vygotsky’s theory regarding the mediative role of signs, AI tools (Canva, D-ID) are considered for the first time not as design aids, but as "external scaffolds" for semantic analysis and text deconstruction: identifying macropoetics, character hierarchies, and metaphorical series. The novelty is further supported by the creation of an original typology of educational prompts for working with literary texts. During a pedagogical experiment conducted at the Zh. Dosmukhamedov Higher Pedagogical College, the effectiveness of this technology was verified using L. N. Tolstoy’s epic novel War and Peace. The obtained statistical data confirm a qualitative growth in analytical competencies: students' interpretative activity increased by 24%, and the systemic understanding of the text improved by 23%. The practical significance of the work consists in the creation of a ready-to-use methodological case, including step-by-step instructions for prompt engineering and evaluation criteria for students' analytical assignments. The results were tested through the supervision of student research projects (using the analysis of Jack London’s Martin Eden), which confirms the method's versatility across different literary genres.</jats:p>