<p>Educators face persistent challenges in fostering deep understanding of abstract STEM concepts. Theories of embodied cognition propose that grounding abstract ideas in bodily action and perception can enhance learning. Teachers often intuitively use embodied pedagogies such as gesturing and hand-drawing in the classroom, both by performing these themselves or asking students to do these actions, yet the comparative effects of these different types of embodied pedagogies are not well understood. Furthermore, the effect of these pedagogies needs to be put into perspective of students’ varying levels of prior understanding. It is unlikely that a single pedagogy will suit all learners. Building on research examining how embodied pedagogies interact with prior knowledge, we contrasted gesture-based and hand-drawing-based instruction in a college-level introductory statistics and data science course. Students were randomly assigned to view instructional videos that enacted either drawing or gesture then reenacted the demonstrated actions. Results showed that prior knowledge significantly moderated instructional effectiveness: For low-prior-knowledge learners, gesture-based instruction outperformed drawing-based instruction, whereas no difference emerged for high-prior-knowledge learners. The findings of this study contribute to our theoretical understanding of how embodied pedagogies impact learning and offer practical insights for tailoring embodied pedagogies to learners’ prior knowledge in the classroom.</p>