Investigating the Impact of Artificial Intelligence-Mediated Interaction on Second Language Development in Academic and Digital Learning Contexts

Abstract

The current research examines the effect of a second language (L2) development under the influence of the interaction mediated by the use of artificial intelligence (AI) in academic and informal digital learning settings. The study is based on the interactionist, output-based, and sociocultural theories of SLA, which analyze the process of AI interaction as a method of meaning negotiation, feedback uptake, and the use of strategic language. The longitudinal mixed-methods design was used, which consisted of quantitative pre-, post-, and delayed assessments, and of the qualitative analysis of the AI-mediated interaction transcripts, the reflective journals, and the semi-structured interviews. A sample of 50 developed L2 English learners was purposely chosen in a Pakistani higher education environment. It was found that long-term AI-mediated communication improves negotiation, repair, and strategic use, which contribute to the growth of discourse and strategic competence. The skills developed in unstructuring digital environments are transferred successfully to academic communication and seem to be long-term, yet practicality nuances might have to be refined further. Evaluative literacy and learner agency became the key to effective engagement. AI-mediated practice combined with reflective prompts, genuine academic assignments, and scaffolding to be used as the most effective way to achieve optimal L2 growth and long-term communicative competence.