Innovation in Language Learning

In Italian higher education, the teaching of pragmatics remains underdeveloped and is often addressed through traditional, form-focused approaches [1]. Pragmatics, however, is highly context-dependent and socially situated, which makes it particularly difficult to model in conversational agents, typically better at handling linguistic form than pragmatic use. The University of Piemonte Orientale Language Centre (CLUPO) is developing PragmAI, an oral task-based language learning (TBLL) chatbot designed to support the development of Italian L2 pragmatic competence in task-based interactions. The chatbot structures task-based activities by guiding learners through service-encounter scenarios and acting as a conversational partner that provides immediate, context-sensitive feedback. PragmAI will be integrated into the existing task-based units on Moodle (DIR) as part of the e-learning component of CLUPO’s blended programme. As an LLM-based task-oriented dialogue (TOD) [2] chatbot, PragmAI leverages contextual adaptability to sustain learner engagement beyond classroom time, while simulating real-world interactions. In designing these service-encounter tasks, the project refers to CEFR descriptors [3], CALL foundations [4], and TBLT principles [5] to strengthen pedagogical validity. A small-scale pilot will examine learners’ perception, refine tasks’ design, and revise training procedures, in order to explore the potential of AI-supported TBLL to enhance pragmatic competence in higher-education language programmes, and to inform scalable models for other university language centres.

Keywords: pragmatics; chatbot; oral interaction; TBLL; CALL; e-learning

REFERENCES