International Journal of
Information and Education Technology

Abstract—This paper explores how an adaptive Artificial Intelligence (AI)-based mathematics platform is experienced by a group of learners who are often overlooked—students who learn at a slower pace. Using a qualitative usability approach, the study does not aim to produce broad generalizations. Instead, it provides an “existence proof”, showing how certain design choices can unintentionally create barriers to learning. Rather than presenting sweeping conclusions, the study offers a detailed look at where and how interactions with the platform break down, giving a nuanced picture of the challenges slow learners may face. By placing learners at the center of our study and using think-aloud protocols alongside participant observation, we were able to uncover key design flaws that directly hinder the learning process. Our findings point to two main patterns of failure. The first is what we call a “Frustration–Disengagement Loop”. Here, the platform’s punitive approach to gamification pushes students away from genuine learning. Instead of staying focused on understanding the material, they shift their goals toward simply trying to outmaneuver or bypass the system. The second failure pattern we observed is what we term a “Representational Divide”. While the platform makes effective use of visual aids, it falls short in helping learners make the crucial connection to symbolic understanding. This gap often leaves students with only a surface-level sense of competence, reinforced by procedural hints that mask deeper misunderstandings. To address these issues, we propose a set of evidence-based design heuristics, along with visual redesigns, intended to support more inclusive, effective, and ethically responsible educational technologies.

Keywords—Human-Computer Interaction (HCI), inclusive design, educational technology, cognitive walkthrough, slow learners

[PDF]

Cite: Topanus Tulak, Andi Kaharuddin, and Harmelia Tulak, "The Representational Divide: A Qualitative Usability Analysis of an Adaptive AI for Atypical Learners," International Journal of Information and Education Technology, vol. 16, no. 3, pp. 611-616, 2026.

Copyright © 2026 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).