High CT students used the AICA for code understanding and debugging while low CT students used it for direct answer retrieval¶

Claim · All claims
Evidence · 1 study · associational r2 · q2
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? Lag Sequential Analysis of screen recordings showed distinct behavioral transitions: high CT students showed unique transitions such as RC → AAEC and AAEC → DP, while low CT students showed UT → AANQ, AACC → CPC and RC → PPC, indicating dependence on AICA-provided answers. → Shu Zhao 2026

Evidence¶

Shu Zhao 2026¶

Shu Zhao, Ziqi Wang, Tingting Chen, Chunchen Kang, Yu-Ju Lan. (2026). Exploring the effect of computational thinking levels on students' learning performance, cognition, and behavior when using AI coding assistants. Education Tech Research Dev. https://doi.org/10.1007/s11423-026-10663-6

q2 · i? · associational · r2

LSA of 52 screen recordings coded with a ten-category behavioral scheme (GESQ 5.1), each group showing 15 significant sequences. The article reports the low CT group's unique transitions and concludes "low CT students only sought answers by querying AICA during coding."

"the unique transitions for low CT group but not appearing in high CT group were UT → AANQ, AANQ → AANQ, RC → AACC, AACC → CPC and RC → PPC. It revealed that low CT students only sought answers by querying AICA during coding."

Discussion¶

Related Claims¶

• Behavioral trends across three tasks showed high CT students maintained stable independent coding while low CT students' AICA use fluctuated — related
• High and low CT groups showed comparable prior coding knowledge before the AICA intervention — related
• High CT students significantly exceeded low CT students in task strategies, elaboration, and self-assessment during AICA interactions — related
• Role design facilitates transitions among higher-order collaborative behaviors in human-GAI collaboration — related
• High CT students significantly outperformed low CT students on the post-test after AICA-assisted coding instruction — related
• 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 — related
• 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
• Lag-sequential Analysis reveals temporal ordering patterns in CSCL interaction that traditional coding-and-counting approaches hide — related
• ENA revealed high CT students had significantly more coherent SRL cognitive networks spanning planning, execution, and self-reflection than low CT students — related