Scilight Press

This paper examines the grain of content in junior high mathematics digital textbooks from People’s Education Press (PEP) using Cognitive Load Theory (CLT) in a sequential explanatory mixed-methods design: (1) bibliometric analysis of 2008-2023 142 publications found substantial gaps in cognitive-aligned pedagogical design; then, (2) large-scale surveys of 231 teachers and 102 students found critical gaps in navigation intuitiveness (71.3%), interactive affordance deficiency (68.9%), and personal pathway rigidity (76.5%). (3) Interviews with 6 teachers and 3 developers further revealed these deficiencies lay in: (1) content fragmentation serving procedural skills at the expense of conceptual integration; (2) sequence disruption violating CLT’s intrinsic load tenets; and (3) passive multimodal serving static text/images (82% of resources) limiting germane processing. We thus innovated a CLT-driven framework to reduce intrinsic load by animating schema builders chunking complex concepts, minimize extraneous load by Gestalt-principled UI redesign serving spatial consistency, and enhance germane load by adaptive analytics serving personal pathways. Empirical results showed 34% more knowledge retention (p<0.01, d=1.87) and 28% less perceived cognitive load (NASA-TLX) relative to conventional textbooks. Our work contributed both a theoretically grounded resource optimization model and an advancement of CLT in technology-enhanced mathematics instruction.