Download Full PDF

Emma Rao

Ruohan Zhao

The Hong Kong University of Science and Technology, Hong Kong SAR, China

 

Abstract

From a sociocultural perspective, and based on an activity theoretical framework, this study examines the integration of ChatGPT into a structured, extracurricular program entitled the Lens on China workshop series, investigating how Chinese language learners use this tool to construct cultural context scripts. Accordingly, two research questions are posed: (1) How does learners’ perception of ChatGPT’s role influence their approach to scriptwriting within a cultural context? (2) What factors influence the creation of cultural scripts with ChatGPT? The results indicate that the perceived role of ChatGPT varies and influences prompting strategies: learners viewing it as an assistant prefer contextual prompts, while those who see it as a tool tend to use procedural prompts. Contextual prompts provide learners with appropriate culturally relevant content, while procedural prompts offer necessary structural elements. A sequential combination of contextual-first and subsequent procedural prompts proved most effective for achieving culturally appropriate and structurally sound scripts. Further analysis through Activity Theory identified several contradiction pairs. The contradiction between rules and community is identified as key to understanding the task and learning outcomes. This tension arose when ChatGPT generated culturally plausible-sounding but unauthentic content, leaving learners unsure of genuine norms. When this contradiction was effectively mediated by the instructors—who articulated explicit cultural and pragmatic rules— it transformed abstract cultural knowledge into performable linguistic expressions, thereby significantly enhancing intercultural communicative competence and speaking skills. Consequently, instructors’ role must shift from a resource provider to a mediator of contradictions, harnessing the gap between AI logic and language culture to deepen learners’ intercultural competence.

 

Keywords

ChatGPT, Activity Theory, cultural context script