Behavior-Aware Argumentation: Integrating Toulmin’s Model for Adaptive Learning in Digital Environments

Abstract

This study investigates an adaptive learning mechanism within Viat‑map, a learning tool based on Toulmin’s Argument Model, to address engagement challenges among struggling and gaming students. In addition to scoring correctness, the system monitors two key behavioral indicators: the number of steps taken and the time spent on tasks. These indicators are used to tailor the number of multiple‑choice options, reducing them for overloaded learners and increasing them for those who may be guessing strategically. In a controlled experiment with sophomore EFL students, pre‑and post‑tests, ANCOVA, and effect size analyses showed significant gains for struggling learners, while high performers benefited less. Findings highlight the value of behavior‑aware analytics in optimizing educational technologies, and offer insight into refining personalized learning algorithms.