This exploratory study examines whether chatbot-mediated written interaction supports adult B2 English learners’ performance in online interaction, pluricultural competence, and mediation in Official Language Schools (OLS) in Spain. The intervention was built around a fictional-culture scenario in which learners had to resolve a cultural misunderstanding between a Spanish visitor and a host from an invented culture. In the experimental condition, students interacted with a chatbot previously configured with information about the fictional culture; in the control condition, students worked in pairs in a chatroom, with one peer acting as the cultural expert. Interaction texts were independently rated by two researchers using a Common European Framework of Reference for Languages (CEFR) Companion Volume-informed rubric. The dataset comprised 16 learners in the control group and 24 in the experimental group, each rated by two evaluators. Inter-rater reliability reached acceptable levels for all aggregated dimensions, with ICC(2,1) values above 0.70. Mann–Whitney U tests showed no significant between-group differences in online interaction or pluricultural competence, whereas the chatbot-supported condition, which included sustained-questioning scaffolding, was associated with significantly higher mediation scores. The findings suggest that chatbot use may be pedagogically promising for mediation-oriented writing tasks, although the evidence should be interpreted cautiously because the study is exploratory, the sample is small, and the scenario relied on a fictional cultural frame.
Cores-Bilbao et al. (Wed,) studied this question.