Student-agent interaction featured a significant Query-Evaluation-Query verification loop (RM-QR-EA), distinct from the linear QR-UN path of student-instructor interaction¶

Claim · All claims
Evidence · 1 study (2 entries) · causal r2 · q2
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? Lag sequential analysis identified significant EA->QR, QR->EA, and RM->QR transitions in student-agent interaction, forming an offload-evaluate verification loop. → Li 2026 q2 i? Student-instructor interaction showed a different, simpler pattern with significant QR->UN transition and self-transitions in AN, AP, and UD. → Li 2026 (2)

Evidence¶

Li 2026¶

Li, X., Liu, Z., Jiang, S., Chen, J., & Chen, W. (2026). The impact of an LLM-based educational agent on learning achievement, cognitive dynamics, and student perceptions in computer science education. International Journal of STEM Education. https://doi.org/10.1186/s40594-026-00641-y

q2 · i? · causal · r2

Lag sequential analysis (GSEQ 5.0, adjusted residuals Z > 1.96) of student queries to DBagent shows significant cross-behavior transitions "EA-> QR , QR-> EA , and RM-> QR", which the authors interpret as an offload-evaluate cycle in which students critically evaluate agent output.

"Meanwhile, EA-> QR , QR-> EA , and RM-> QR transitions exhibit significant features that cross different behaviors."

Li 2026 (2)¶

Li, X., Liu, Z., Jiang, S., Chen, J., & Chen, W. (2026). The impact of an LLM-based educational agent on learning achievement, cognitive dynamics, and student perceptions in computer science education. International Journal of STEM Education. https://doi.org/10.1186/s40594-026-00641-y

q2 · i? · causal · r2

LSA of the control class's student-instructor interaction found significant QR->UN transition and self-transitions across AN, AP, and UD; the authors describe this interaction pattern's cognitive transformation as relatively straightforward with low interaction frequency.

"Table 6 shows that students' engagement is significant in QR-> UN and reveals the self-transition feature across three behaviors (AN, AP , and UD )."

Discussion¶

Related Claims¶

• Learners in agent interaction became locked in lower-order self-transition loops (e.g., Understand z = 49.08, Application z = 51.48) and rarely transitioned spontaneously to higher-order cognition — related
• Only the Collaborator group shows a complete significant sequential metacognitive chain from orientation through elaboration; the Replacement Tool group shows only self-loops — related
• Role design facilitates transitions among higher-order collaborative behaviors in human-GAI collaboration — related
• High CT students used the AICA for code understanding and debugging while low CT students used it for direct answer retrieval — related
• Evidence suggests over-reliance on generative AI may negatively affect students' self-regulated learning behaviors, potentially harming retention and transfer to novel contexts — related
• Cognitive offloading tendency is negatively associated with critical thinking, technical problem-solving, and metacognitive self-regulation — related
• Lag-sequential Analysis reveals temporal ordering patterns in CSCL interaction that traditional coding-and-counting approaches hide — related