Аннотация
The purpose of this study is to evaluate the impact of systematically including physics Olympiad problems on the development of research competencies in prospective physics teachers. A quasi-experimental design was implemented between 2020 and 2024 with experimental and control groups, supported by a 2023 survey of 141 participants. The research emphasized advanced tasks such as extremum problems, applications of conservation laws, and graphical analysis, with particular attention to the depth of reasoning. Ethical standards were fully observed. Findings show that students in the experimental group demonstrated a steady increase in success rates compared to the control group, producing more well-reasoned solutions and making fewer methodological errors. Qualitative improvements were also observed: participants more often formulated hypotheses, selected models deliberately, and provided grounded interpretations. Survey data confirmed enhanced intrinsic motivation for research-oriented tasks. The novelty of the study lies in clarifying methodological foundations for applying Olympiad problems to research skills formation, and in justifying a gradual shift from standard to non-standard tasks, strengthened by interdisciplinary links and reflection. Practical significance is expressed in integrating this methodology into courses and electives, with guidance for task selection, organization, and assessment. Limitations relate to design, sample, and time frame. Future directions include larger samples, wider testing across physics domains, refined diagnostics, and digital or distance-learning adaptations.