“…Reports on DBR studies often do not achieve satisfactory coherence among theoretical perspectives, instructional design, and a posteriori analysis (Abrahamson & Wilensky, 2007;Artigue, Cerulli, Haspekian, & Maracci, 2009;Kelly, 2004;Puntambekar & Sandoval, 2009;Ruthven, Laborde, Leach, & Tiberghien, 2009). We attempt to achieve such coherence as follows: (a) our post-hoc analytic attention to the multimodality of discourse in our empirical data resonates with the epistemological commitments underlying our embodied-design framework (Abrahamson, 2009d); and (b) probability is a suitable content domain for the study of multimodal reasoning, because the challenges of expressing randomness numerically may impel students to resort to alternative discursive genres such as metaphor (Rubin & Hammerman, 2007).…”