Abstract
Engineering education faces a structural tension: the need to develop students’ systems-level reasoning about complex technical objects conflicts with the wide variance in students’ prior knowledge, cognitive readiness, and disciplinary background. Standard fixed-sequence curricula address neither the breadth of the hierarchy students must traverse nor the individual variation in how they traverse it.
AI-assisted diagnosis tools - including knowledge space theory-based assessment, Bayesian student modelling, and natural language interaction through large language model tutors - are proposed as the mechanism for adaptive pathway assignment and transition. The paper analyses the theoretical foundations of each adaptive mechanism, maps them to specific hierarchical instructional stages, and provides curriculum design guidelines for engineering faculty.
References
This work is licensed under a Creative Commons Attribution 4.0 International License.