Claim Missing Document
Check
Articles

Found 11 Documents
Search

Enhancing EFL Reading Comprehension via an AI-Chatbot-Guided Toulmin Mapping in Viat-Map Andoko, Banni Satria; Subhi, Dian Hanafudin; Wafiuddin, Naufal Falah; Wang, Xia; Hirashima, Tsukasa
Utamax : Journal of Ultimate Research and Trends in Education Vol. 7 No. 2 (2025): Utamax : Journal of Ultimate Research and Trends in Education (In-Press)
Publisher : LPPM Universitas Lancang Kuning. Pekanbaru. Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31849/9nwj8r52

Abstract

The growing demand for English as a Foreign Language (EFL) proficiency necessitates innovative approaches that go beyond conventional reading practices, which often emphasize translation and literal comprehension without fostering critical thinking. Although the Toulmin Argumentation Model has proven effective in enhancing logical structuring skills, it remains insufficient in supporting higher-order thinking without additional scaffolding. This study bridges that gap by embedding an AI-powered chatbot, driven by a Large Language Model (LLM), into the Viat-Map system to promote reflective engagement and argument construction during reading tasks. Using a quantitative pre-test and post-test design, 16 university students engaged with the AI-enhanced system over two weeks. Findings reveal a significant improvement in comprehension performance (p < 0.01), with the number of interaction steps emerging as the strongest predictor of learning gain (r = 0.482, p < 0.05). Time alone did not significantly affect outcomes, yet its combination with active interaction produced a moderate synergistic effect (p = 0.041, η² = 0.16). These results underscore that active learner engagement, rather than mere exposure time, is the critical factor for meaningful comprehension improvement. The study contributes to the literature by demonstrating how AI-driven Toulmin mapping effectively cultivates deeper comprehension and critical reasoning. In a larger perspective, this research signals the possibility of reimagining global language education through intelligent systems that foster autonomy, equity, and inclusivity in diverse learning environments.