Beyond tool use: how ecological coupling configures AI’s empowerment for language learning engagement

TL;DR

This study explores how AI can enhance language learning by interacting with teachers, peers, and the environment.

Contribution

The paper introduces a new model showing AI's impact on learning depends on its integration within the learning ecosystem.

Findings

AI's effect on engagement is mediated by perceived ecological coupling with teachers, peers, and environment.

Three learner profiles were identified based on coupling and engagement levels.

The model suggests a shift from tool use to strategic ecological governance for better learning outcomes.

Abstract

This study develops and tests an AI-empowerment Configural Model to explain how artificial intelligence (AI) empowers language learning engagement. Grounded in ecological systems theory (EST) and ecological affordance theory (EAT), the model theorizes AI as an interactive agent within the learning ecosystem. A mixed-methods study of 475 Chinese university language learners demonstrates that AI’S effect on engagement is significantly mediated by the perceived quality of its ecological coupling with teachers, peers, and the environment. Latent profile analysis (LPA) further identifies three distinct learner configurations: low coupling-low engagement, moderate coupling-moderate engagement and high coupling-high engagement, which systematically differ in their coupling of AI. The model ultimately shifts the paradigm from tool implementation to strategic ecological governance, providing a practical basis for designing learning environments that leverage synergistic human–AI coupling to foster deeper, sustained engagement.

Figures