Аннотация
In this study, we aimed to empirically evaluate the effectiveness of a specialized science, technology, engineering, arts, and mathematics (STEAM) program that purposefully utilizes 2D and 3D modeling tools to develop technical, logical, and spatial thinking in 1st-4th-grade students.To test the hypothesis of an association between the STEAM approach and the development of thinking, a quasi-experiment with pre- and post-testing in parallel groups was conducted (N = 172). The experimental group (EG) (n = 85) participated in project-based activities involving the creation of physical and digital models throughout the academic year, whereas the control group (CG) (n = 87) followed a standard curriculum for the same period.Adapted versions of the Bennett Mechanical Comprehension Test (technical thinking), Raven's Progressive Matrices (logical thinking), and the Yakimanskaya–Zarkhin–Kadayas Spatial Representations Test (spatial thinking) were used, with grade-level differentiation and score normalization to ensure cross-grade comparability of the results.The results revealed a statistically significant superiority of the EG over the CG in the post-test across all measured indicators (p < 0.001, Mann-Whitney U test) with large effect sizes (r = 0.72–0.76). The greatest improvement was recorded in spatial-thinking skills. Intra-group analysis (Wilcoxon signed-rank test) confirmed significant dynamics within the EG (r > 0.80) and no significant change in the CG. The results also suggested the potential importance of the teacher's role as a facilitator in implementing STEAM-based activities.The findings provide empirical support for the potential effectiveness of integrating a modeling-based STEAM approach into primary education to foster the development of technical, logical, and spatial thinking in children.