<![CDATA[This study aims to determine the differences between PjBL-based GDL and TCL learning models on students' mathematical communication abilities, examine differences in FI and FD cognitive styles on students' mathematical communication abilities, and investigate the interaction between PjBL-based GDL and TCL learning models and cognitive styles on students' mathematical communication abilities. This study employed a quantitative approach with a quasi-experimental design. The research subjects consisted of fourth-grade students from two elementary schools in Cluster III, Depok District, Sleman, Yogyakarta. Data analysis was conducted through an independent samples t-test and two-way ANCOVA. The results revealed that the PjBL–GDL model had a significant effect on students’ mathematical communication skills (F = 116.75; p < 0.05; η² = 0.522). Significant differences were also found between FI and FD students (F = 46.29; p < 0.05; η² = 0.302). However, no significant interaction was observed between learning models and cognitive styles (p > 0.05), indicating independent effects. These findings imply that the PjBL–GDL model can be implemented inclusively without the need to differentiate instruction based on students’ cognitive styles, making it a practical approach for heterogeneous classrooms. The novelty of this study lies in integrating PjBL and GDL to enhance mathematical communication across diverse learners. It is concluded that PjBL–GDL is effective regardless of cognitive style. Teachers are recommended to apply this model flexibly, while future studies should examine its impact on other skills such as motivation and critical thinking.]]>
