International Journal of
Information and Education Technology

Abstract

Geometric concepts and spatial reasoning are fundamental components of early mathematics education. However, young learners often experience difficulties in understanding abstract spatial and shape relationships through conventional instruction alone. Although Augmented Reality (AR) and game-based learning have been increasingly applied in mathematics education, relatively limited attention has been paid to integrating AR, image recognition, and real-world object exploration into game-based learning activities for lower-grade elementary students. To address this gap, this study developed a game-based AR learning system with image recognition functions to support first-grade students in learning geometric concepts. The system enables learners to identify geometric shapes from everyday objects, visualize corresponding three-dimensional models through AR, and apply the recognized shapes in interactive game-based matching tasks. A one-group pretest-posttest study was conducted with 31 first-grade students from an elementary school in Taiwan. Data were collected through a mathematics achievement test, a learning motivation questionnaire, and a flow experience questionnaire. Paired-samples t-tests, independent-samples t-tests, one-sample t-tests, and Analysis of Covariance (ANCOVA) were used to examine students’ learning achievement, motivation, and flow experience. The results indicated that the system supported students’ learning of geometric concepts, particularly among students with low prior knowledge after controlling for prior achievement. Students also reported a significantly high level of flow experience during the learning activity. However, no significant difference was found between pretest and posttest learning motivation, possibly because students’ initial motivation was already high. These findings suggest that the proposed game-based AR system may serve as a promising instructional support tool for elementary geometric concept learning. Nevertheless, because this study was conducted with a small sample and without a control group, the findings should be interpreted with caution. Future studies should include larger and more diverse samples, delayed posttests, and comparison groups to further examine the long-term learning effects of AR- and image recognition-supported game-based learning.