Exploring Cognitive Synergy Framework on Computing Students Epistemic Agency in Robotics Education
Generative AI can support robotics programming, but students must still evaluate suggestions and justify design decisions. This study examined changes in grit-related learning dispositions and patterns of epistemic activity during instruction using the Cognitive Synergy Framework (CSF). In a mixed-methods study across six West African universities, we randomly allocated 410 undergraduates to CSF-supported or comparison instruction. A categorical-time mixed-effects model analysed 848 Grit in Robotics Programming scores from 351 students across Weeks 11, 13, and 15. Epistemic network analysis examined coded actions from 30 collaborative groups. The reported GRP condition-by-time interaction was significant, Wald χ 2 (2) = 15.02, p < .001. Relative to Week 11, estimated gains were 0.167 points greater under CSF at Week 13 (95% CI [0.059, 0.274]) and 0.213 points greater at Week 15 (95% CI [0.093, 0.333]). With a 0.20-point downward adjustment to imputed CSF scores at both follow-ups, the joint test yielded p = .103. Group-level network trajectories differed by condition (β = 0.09, 95% CI [0.03, 0.15], p = .004). Gains in independent epistemic agency remained untested. Future studies should match instructional support across conditions and assess individual reasoning, evaluative judgement, and knowledge construction on tasks completed without scaffolding.
Authors
- Elias Oupa Mashile (ORCID: https://orcid.org/0000-0001-7234-9031)
- Christian Basil Omeh (ORCID: https://orcid.org/0000-0002-1673-2578)
Institutions
- University of South Africa (ZA)
Publication Details
- Journal
- Journal of Educational Computing Research
- Published
- 2026-11-21
- DOI
- https://doi.org/10.1177/07356331261511181
- Primary Topic
- Teaching and Learning Programming
- Type
- article
- Field-Weighted Citation Impact
- 0.00