Beyond the Algorithm: A Person-Centered Investigation of Engagement, Self-Efficacy, and Anxiety Among Chinese University EFL Learners in the Age of Large Language Models

Department of Interpreting and Translation, Wake Forest University, Winston-Salem, United States

^* Corresponding author: This email address is being protected from spambots. You need JavaScript enabled to view it.

Abstract

Implementation of large language models (LLMs) in higher education is a paradigm shift in the learning setting of its own right, but the psychological aspect of the change has yet to be theorized properly. This conceptual paper argues for a person-centered approach to understanding how Chinese university students—as English as a Foreign Language (EFL) learners—navigate the cognitive and affective implications of LLM-mediated learning. Moving beyond reductive input-output frameworks of human–machine interaction, this study synthesizes existing literature to examine the interplay among three core psychological constructs: learner engagement, self-efficacy, and foreign language anxiety within AI-mediated contexts. This paper will inform a more responsible, ethically aware, and humanly designed addition of the use of AI in second language learning by highlighting the less obvious, yet contrasting orientations of participation, self-trust, and apprehension that students use to respond to AI. This paper recommends the following based on such findings: 1. The process of pedagogy should be psychologically adjusted to the performance of students. 2. The self-awareness aspect of students plays an essential role in SLA.