Artificial Intelligence-Assisted English Language Learning: A Systematic Analysis of Learning Outcomes, Motivation, and Student Engagement

AbstractThe rapid integration of Artificial Intelligence (AI) into English Language Teaching (ELT) hastransformed Computer-Assisted Language Learning (CALL) from passive instructional deliveryto adaptive, highly personalized learning environments. This study presents a systematicempirical investigation into the impact of AI-assisted language learning tools on objectiveEnglish language learning outcomes, psychological motivational shifts, and multi-dimensionalstudent engagement. Utilizing a mixed-methods empirical design, quantitative data werecollected from a stratified sample of 350 undergraduate English as a Foreign Language (EFL)students across public and private universities, supplemented by pre- and post-interventionlanguage proficiency tests. The empirical findings indicate statistically significant gains inoverall linguistic proficiency following AI-mediated instruction, with the highest mean scoreenhancements observed in speaking fluency ($t = 14.82, p < 0.001$) and writing mechanics ($t =12.65, p < 0.001$). Furthermore, regression analysis confirms that reduced foreign languageanxiety ($\beta = 0.384, p < 0.001$) and elevated cognitive engagement ($\beta = 0.312, p <0.001$) serve as primary predictors of proficiency acquisition. While behavioural and affectiveengagement remain consistently high, cognitive engagement requires scaffolded pedagogicaldesign to prevent passive over-reliance on AI text generation. Based on these findings, anintegrated pedagogical model is proposed, offering concrete strategies for curriculum integration,teacher training, and institutional policy implementation.