From Cultural Knowledge to Critical Verification: Differential Development of Intercultural Competence in AI-Supported Independent EFL Learning

Abstract

This quasi-experimental study examines how an integrated AI-supported independentlearning methodology affected four dimensions of intercultural competence (IC) among university EFL students. Participants were 316 students from three higher education institutions in Uzbekistan, with 158 students in the experimental condition and 158 in the control condition. IC was operationalised through cognitive-cultural, affectiveattitudinal, behavioural-communicative, and critical-reflective dimensions, each assessed by five performance indicators. Pre-test scores did not differ significantly between groups (60.5 vs. 60.2; t = 0.29, p = .773). At post-test, the experimental group achieved a mean of 78.6 compared with 64.9 in the control group, t = 14.48, p < .001, Cohen’s d = 1.63. Large between-group effects were observed across all four dimensions, with the strongest effect in critical-reflective performance (d = 1.31), followed by cognitive-cultural (d = 1.17), behavioural-communicative (d = 1.05), and affective-attitudinal (d = 1.03). The proportion of experimental-group students at sufficient or high IC levels increased from 36.7% to 76.6%, compared with 35.4% to 44.9% in the control group. The findings suggest that intercultural development in AI-supported environments is strongest when cultural inquiry includes source verification, stereotype recognition, reflective revision, contextualised communication, and learner-first reasoning. Because the intervention combined task transformation, self-regulation, teacher facilitation, and AI-supported scaffolding, the results should be interpreted as effects of the integrated methodology rather than of AI in isolation.