Achieving educational equity in Social-Emotional Learning (SEL) requires Differentiated Instruction (DI) that tailors pathways to diverse learner needs, moving beyond mere technological access. This study proposes a “Dynamic Matching Model” integrating TPACK with Cognitive Load Theory to align AI scaffolding and VR immersion with learner readiness in empathy education. Seventy-seven university students were randomly assigned to an AI-based summarization tool or a desktop VR simulation to engage with a migration narrative. ANCOVA results revealed a significant Expertise Reversal Effect: AI significantly enhanced Empathic Concern (EC) for novices (low prior knowledge) by reducing cognitive load (mean difference = 0.169, p = 0.003), while VR induced overload in this group. Conversely, VR boosted EC for high-prior-knowledge learners, particularly females (mean difference = 0.155, p = 0.031; simple effect for female experts, p = 0.004), enabling deep embodied engagement via existing schemas. Qualitative data corroborated AI’s cognitive offloading for novices and VR’s embodied advantages for experts. These findings extend TPACK by positioning learner readiness as a critical “Context” moderator. The model advocates a precision-oriented strategy: AI for accessible entry points across all learners and VR for advanced emotional simulation among the prepared. This dual-path approach promotes equitable empathy development in technology-enhanced SEL.
Yangchun Xiong (2026) studied this question.