Abstract
<jats:p>The article deals with the problem of system analysis of the concept "educational physical experiment" (UPE) in the methodology of teaching physics using artificial intelligence. The relevance of the research is due to the ambiguity of the interpretation of UPE in the scientific and methodological literature, the multiplicity of its definitions and the need for a comprehensive review of its structure, functions and psychological and pedagogical aspects. The purpose of the work is to analyze the key characteristics of UPE (goals, functions, types, stages of implementation, didactic requirements, pedagogical principles) using the neural network. In the course of the study, six types of definitions of UPE were obtained (methodological, pedagogical, psychological-pedagogical, structural-functional, through scientific analogy, extended), mental map reflecting the multidimensional structure of the concept UPE, including projection on the axis "conditions–result–analysis" were con-structed. Special attention is paid to the connection of UPE with other teaching methods (explanatory and illustrative, problem-based, research), as well as its role in solving theoretical and experimental problems. The key psychological and pedagogical aspects are identified: the influence on perception and thinking, the development of cognitive processes, the formation of internal motivation, and the implementation of an activity-based approach. The results demonstrate the potential of AI as a tool for systematizing pedagogical concepts and obtaining qualitatively new interpretations, which is of practical importance for improving the methodological training of future physics teachers and increasing the effectiveness of physical education.</jats:p>