<![CDATA[The goal of this study is to ascertain how successfully the Geoboard-assisted Discovery Learning approach affects fifth-grade elementary school students' problem-solving abilities in the field of planar figures. This study used a nonequivalent control group design and a quasi-experimental methodology. 21 students from SDN Kembangarum 2 made up the experimental group, and 29 students from SDN Kembangarum 1 made up the control group. The experimental group received instruction using the Geoboard-assisted Discovery Learning paradigm, while the control group received conventional training. Mathematical problem-solving tasks were used to gather data, and the Shapiro-Wilk normality test, Levene's homogeneity test, Independent Sample T-Test, and N-Gain Score were used for analysis. According to the findings, there was a 25.41-point difference between the experimental group's mean posttest score (70.62) and the control group's (45.21). With t = 5.245 and Sig. (2-tailed) = 0.000 < 0.05, H0 was rejected, according to the Independent Sample T-Test. Furthermore, there was a ±17.29% difference between the experimental group's N-Gain Score of 44.79% (medium category) and the control group's of 27.50% (low category).. These findings demonstrate that the Geoboard-assisted Discovery Learning model significantly and positively influences students' problem-solving abilities and is more effective than conventional instruction. The use of the Geoboard as a concrete manipulative medium enables students to experience Bruner's three stages of representation—enactive, iconic, and symbolic—thereby fostering deeper conceptual understanding and more optimal development of problem-solving skills.]]>
