AI-Driven Vocabulary Acquisition in EFL Higher Education: Interdisciplinary Insights into Technological Innovation, Ethical Challenges, and Equitable Access
AI-Driven Vocabulary Acquisition in EFL Higher Education: Interdisciplinary Insights into Technological Innovation, Ethical Challenges, and Equitable Access
Omer Elsheikh Hago Elmahdi,Asjad Ahmed Saeed Balla,2 Autores,Awwad Othman Abdelaziz Ahmed
TLDR
Adaptive Contextualized Learning is introduced, a novel pedagogical model emphasizing real-world context embedding, dynamic scaffolding, and cultural resonance that improves proficiency benchmarks by 35% compared to static AI systems, addressing gaps in temporally adaptive and culturally sustaining AI education.
Resumen
This study investigates the efficacy, cultural relevance, and ethical implications of AI-driven vocabulary learning tools through a mixed-methods approach combining a PRISMA-guided systematic review of 58 studies and controlled experiments across six global contexts. Results demonstrate that AI tools significantly outperform traditional methods, with a pooled Cohen’s d of 0.61 (95% CI: 0.52–0.70) for retention gains. However, efficacy varies by region: tools tailored to local cultural contexts (e.g., dialect-aware chatbots in Nigeria) achieved effect sizes up to d = 0.85, while culturally generic systems lagged (d = 0.38). The study introduces the Adaptive Contextualized Learning (ACL) framework, a novel pedagogical model emphasizing real-world context embedding, dynamic scaffolding, and cultural resonance. ACL-driven interventions improved proficiency benchmarks by 35% compared to static AI systems, addressing gaps in temporally adaptive and culturally sustaining AI education. Ethical risks, including algorithmic bias (e.g., 23% accuracy drops for non-native accents in speech recognition), were mitigated through the F.A.I.R. Implementation Framework, which prioritizes feedback loops with educators, federated learning for data privacy, and community co-design. Practical guidelines urge educators to integrate AI as supplemental tools, policymakers to fund offline-capable solutions, and developers to adopt modular designs for localization. Limitations include urban-skewed samples and confounding factors such as variable internet access. By bridging AI innovation with equity-centered pedagogy, this study advances theoretical discourse on culturally responsive edtech while offering actionable strategies for ethical AI deployment in diverse educational settings. Future research must prioritize rural adaptations and longitudinal cohorts to ensure inclusive scalability.