The cognitive load module encodes cross-domain learning burden into student state representations as a low-to-high load gradient¶

Claim · All claims
Evidence · 1 study · design r2 · q2
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person · recheck by 2029: tool and model accuracy changes quickly

Subclaims¶

q2 i? UMAP visualization of test-interaction representations shows LT-MKT produces a clearer low-to-high cognitive-load gradient than the variant without cognitive load modeling. → Haotian Zhang 2026

Evidence¶

Haotian Zhang 2026¶

Haotian Zhang, Shucun Wang, Jinze Wu, Liang Ding, Shuochen Liu, Zhenya Huang, Jing Sha, Shijin Wang, and Qi Liu. (2026). Incorporating Cognitive Load and Knowledge Transfer for Multi-Domain Knowledge Tracing. arXiv:2608.24005. https://arxiv.org/abs/2608.24005 doi:10.48550/arxiv.2608.24005

q2 · i? · design · r2

Representation-level analysis on test interactions grouped into low-, medium- and high-load tertiles by a cognitive load index built from question difficulty, domain transition and domain coverage; representations visualized with UMAP (Figure 3).

"In contrast, LT-MKT produces a clearer and more continuous low-to-high load gradient in the representation space. This pattern indicates that the proposed cognitive load module helps encode cross-domain learning burden into the student state representation"

Discussion¶

Related Claims¶

• A qualitative case study shows LT-MKT jointly captures beneficial knowledge transfer and load-induced learning friction — related
• Removing the cognitive load module causes the largest ablation performance drop, while state fusion has a smaller effect — related
• Modeling knowledge transfer within and across domains alleviates the cold-start problem on unseen concepts — related