This study investigates how university-level English language learners across three learning contexts (China, the United Arab Emirates, and a United States program enrolling international graduate students) perceive, identify, and critique cultural bias in AI-generated content. Guided by critical AI literacy and intercultural communicative competence frameworks, the study employed a qualitative comparative case study design in which 50 participants across the three sites analysed an identical 400-word AI-generated essay on climate change and completed a structured reflection protocol. Reflexive thematic analysis identified six cross-cutting themes: initial appreciation of textual clarity alongside recognition of limited analytical depth; detection of Western-centric and globalised framing; critique of an emphasis on individual responsibility that minimised systemic accountability; recognition of underrepresented perspectives from developing regions and vulnerable populations; context-specific mismatches between proposed solutions and local realities; and emerging awareness of AI outputs as reproductions of dominant discourses shaped by the data and assumptions embedded in model development. Across all contexts, participants demonstrated evaluative engagement of varying depth, supported by a scaffolded protocol. Notably, some participants identified problematic assumptions without explicitly labelling them as 'bias', suggesting that critical AI literacy may be enacted in practice before it is articulated through formal analytical terminology. The explicitness of such articulation appeared to vary according to educational level, disciplinary background, and cultural context. Contextual differences aligned with participants' sociopolitical orientations: Chinese students foregrounded collective and state-led action, students at the UAE site emphasised feasibility in relation to climate and cultural constraints, and U.S.-based international students drew on transnational perspectives to critique structural inequality. The findings inform TESOL pedagogy by supporting the integration of critical AI literacy instruction that scaffolds learners' abilities to interrogate AI-generated text as culturally situated rather than neutral.
This article examines why the Language Awareness (LA) movement, despite gaining significant momentum in UK schools during the 1980s, struggled to sustain itself through the following decade. Drawing on interviews with eight educators - including teachers, teacher trainers, and researchers - most of whom were directly involved in the movement, the study investigates both the educational significance they attributed to the Language Awareness Approach (LAA) and the factors they perceived as driving its decline. Findings show that participants experienced the LAA as professionally reorienting and enriching: it repositioned learners as active linguistic explorers, opened space for creative and cross-curricular pedagogy, and fostered collaborative communities that extended beyond individual classrooms. Yet these gains proved fragile. The introduction of the National Curriculum (NC) narrowed professional autonomy; local advisory networks and material-sharing infrastructures gradually eroded; linguistic preparation in teacher education remained insufficient; and assessment frameworks within the NC left little room for LA outcomes. Together, these pressures rendered individual teacher commitment unsustainable. The study suggests that LA's marginalisation reflected not a failure of the approach itself, but of the institutional conditions required to sustain it. These findings have the potential to inform the introduction of LA in other worldwide contexts.
Self-efficacy and metacognitive strategies play important roles in determining English achievement. However, how they interact with each other remains understudied. Moreover, an emerging theory called the Island Ridge Curve (IRC) contends that the effects of cognitive and motivational factors on language performance fluctuate with the increase in language proficiency. The current study explored the interplay between self-efficacy and metacognitive strategies in determining English achievement and the variations of these relationships across low-, medium-, and high-proficiency students. We recruited 1,184 undergraduates (females = 48%, Mage = 18.95, SD = 0.76) from two Chinese universities. Results of multi-group structural equation modelling showed that: (1) both self-efficacy and metacognitive strategy use were positively associated with English achievement, and these associations fluctuated in the nonsignificant-significant-nonsignificant pattern; (2) metacognitive strategy use positively mediated between self-efficacy and English achievement, and the mediation fluctuated in the nonsignificant-significant-nonsignificant pattern. These results demonstrate the utility of the IRC in examining nonlinearity in language research. (sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic).(sic)(sic), (sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic).(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(Island Ridge Curve; IRC)(sic)(sic)(sic)(sic)(sic)(sic), (sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic).(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic), (sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic),(sic),(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic).(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)1,184(sic)(sic)(sic)(sic)((sic)(sic)(sic)(sic)48%, (sic)(sic)(sic)(sic)18.95(sic), (sic)(sic)(sic)0.76).(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic): (1)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic), (sic)(sic)(sic)(sic)(sic)(sic)(sic)"(sic)(sic)(sic)-(sic)(sic)-(sic)(sic)(sic)"(sic)(sic)(sic)(sic)(sic);(2)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic), (sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)"(sic)(sic)(sic)-(sic)(sic)-(sic)(sic)(sic)"(sic)(sic)(sic)(sic)(sic).(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic)(sic).
This study examines the role of language awareness, affect, and linguistic justice in the practices of educators engaged in the revitalisation of Basque across formal and non-formal educational contexts. Adopting a qualitative and interpretative approach, the study draws on twenty-four semi-structured interviews with teachers, social educators, and leisure-time educators. The analysis explores how professional language awareness, understood as educators' situated sensitivity to language use and its social implications, is constructed across contexts and relates to motivation, linguistic attachment, and everyday decisions regarding the use of Basque in multilingual environments shaped by sociolinguistic inequality and the hegemonic presence of Spanish. The findings identify context-specific configurations of language awareness. Formal education is characterised by more declarative orientations, socio-educational and residential contexts by relational but constrained practices, and organised educational leisure by a more integrated approach. From a decolonial and intersectional perspective, the study concludes that language revitalisation requires socially just and affectively grounded educational conditions that strengthen legitimacy, emotional security, and everyday opportunities for language use, while highlighting the need for context-sensitive professional development in language awareness and multilingual pedagogies.
