دانشگاه آزاد اسلامی واحد تهران جنوب

صفحه اصلی / نخستین همایش هوش مصنوعی و پژوهش های نوظهور: همگرایی انسان و سیستم های هوشمند

Enhancement of ELT Learners’ Socio-pragmatic Shortfalls through AI-Developed Modules for Interlanguage Pragmatics

نویسندگان :

Amir Mohammed Asadi¹ Esmaeil Arib²

1- دانشجوی دانشگاه آزاد اسلامی واحد تهران جنوب 2- هیئت علمی دانشگاه آزاد اسلامی واحد تهران جنوب

کلمات کلیدی :

Artificial Intelligence،Language Learning،Instructional Modules،Pragmatic Competence

چکیده :

This research investigates the efficacy of instruction gained by AI-authored instructional modules produced with Grok 3 for xAI for improving the socio-pragmatic awareness of 159 upper-level Tehran-based English as a Foreign Language students. Socio-pragmatic ability, requisite for contextually appropriate performance with language, is underdeveloped for EFL students, leading to pragmatic breakdowns in intercultural communication. Quantitative experimental design compared a treatment condition exposed to five AI-authored modules to a theoretical control condition of equal size, with pre- and post-tests (25 multiple-choice) administered online through Google Forms. Statistical tests, namely paired t-tests and descriptive statistics, saw the treatment condition post-test mean (17.84) as superior to that of the control condition (17.00; t = 2.32, df ≈ 316, p = 0.021), rejecting null expectations of no module effect upon speech act ability as well as socio-pragmatic awareness. Moderately within-subject gains (treatment: t = 0.102, p = 0.920) combined with a ceiling effect due to high baselines (71.2% treatment, 67.8% control) show a modest overall effect. Targeted improvement for particular speech acts, such as requests and apologies, shows the context-sensitive effectiveness of the modules, but inconsistencies point to a more sophisticated design for the modules.