Archives for Technical Sciences | 2026
Authors: Young C.S.; Mamadjanova M.; Vakhidova F.; Tursunova G.; Isroilova B.; Achilova O.; Shoyimova K.
DOI: 10.70102/afts.2026.1835.178
Journal: Archives for Technical Sciences
Year: 2026
Publisher: Technical institute of Bijeljina
Document Type: Article
Open Access: All Open Access; Gold Open Access
Cited by: 0
The cultivation of cross-cultural communicative competence (CCCC) among advanced language learners remains ongoing, particularly regarding exposure to authentic socio-pragmatic contexts and the lack of adaptive feedback in traditional learning. The research paper suggests a socio-pragmatic scaffolding system that uses large language models to generate socio-pragmatic dialogues and provide context-sensitive remediation, based on generative artificial intelligence Architectures like GPT-4 from OpenAI. A quasi-experimental design was used with 120 advanced learners, who were separated into an experimental (n=60) and a control group (n=60), with the intervention lasting 12 weeks. In both the experimental and control groups, AI-mediated role-play, discourse adaptation tasks, and pragmatic feedback cycles were undertaken, while traditional task-based instruction was delivered. Pragmatic appropriateness, discourse flexibility, and intercultural sensitivity were measured by pre-and post-tests based on validated rubrics and a standardized communicative competence scale. Findings show that there is a statistically significant benefit in the experimental group in terms of an overall increase of CCCC scores by 32.8 % (p < 0.01), as compared to 12.3 % in the control group. The accuracy of pragmatic appropriateness increased by 29.4%, the score of intercultural sensitivity indices increased by 88.2%. The impact of AI-driven scaffolding was strong, as evidenced by the effect size analysis (Cohen's d = 0.86). Moreover, the learner engagement metrics showed a 41% voluntary task completion and a 41.6% drop in pragmatic transfer errors. The result proves that socio-pragmatic scaffolding supported by generative AI can substantially enhance cross-cultural communicative competence by providing adaptive, context-rich, and iterative feedback. The research finds that an AI-based discourse simulation, when incorporated into a high-level language course, can help fill pragmatic gaps and promote the development of culturally responsive communication skills in a globalised learning setting. © 2026, Technical institute of Bijeljina. All rights reserved.
adaptive feedback systems; advanced language learning; AI-mediated Language Instruction; cross-cultural communicative competence; generative artificial intelligence; intercultural pragmatics; socio-pragmatic scaffolding