#35: When Visuals Come First: Aligning Pre-Instruction and Embodied Pedagogies to Support Learning of Statistical Models

Introductory statistics courses are challenging in part because core concepts such as statistical models are highly abstract. Prior work shows that visual representations and embodied pedagogies can each support abstract learning, yet little is known about how they interact. In this study, 112 undergraduates were randomly assigned to pre-instruction with or without visualizations, followed by an embodied intervention (gesture or drawing). Students who received visual pre-instruction outperformed those without visual support on an immediate posttest (Posttest 1). However, pre-instruction type did not directly predict later performance (Posttest 2), nor did it moderate the effectiveness of gesture versus drawing. Exploratory analyses revealed a moderated mediation: the relationship between Posttest 1 and Posttest 2 performance varied by pre-instruction type and students’ reported use of visual information from the embodied intervention. These findings underscore the importance of representational coherency across learning phases, offering new insights for designing instructional sequences in statistics education.