Randomized trial demonstrates significant gains in pragmatic competence using an AI platform in undergraduates, suggesting potential for improved academic communication.
Key Points
This study aims to enhance pragmatic competence in multilingual higher education through an AI-mediated multimodal platform.
Quasi-experimental mixed-methods design with 240 undergraduates (120 experimental, 120 control) over 5 weeks.
Utilization of pre- and post-intervention assessments, including tests on pragmatic competence and discourse completion tasks.
Integration of features like automated text analysis, multimedia video analysis, conversation simulations, and an AI consultation assistant.
The experimental group showed a significant improvement in pragmatic test scores (M = 15.0, p < .001) compared to the control group (M = 5.2, p < .001).
Improvement on discourse completion tasks was notable (experimental = +14.1 vs. control = +4.8, both p < .001).
Regression analysis identified engagement with conversation simulations (β = 0.58, p < .001) and AI consultation (β = 0.29, p = .012) as key predictors of learning.