Acta Psychol (Amst). 2025 Oct 8;260:105649. doi: 10.1016/j.actpsy.2025.105649. Online ahead of print.
ABSTRACT
The proposed meta-analysis study examines the pedagogical role of artificial intelligence (AI)-based interventions in English as a Foreign Language (EFL) education through a synthesis of empirical evidence of 23 peer-reviewed experimental and quasi-experimental experiments published in 2019-2025. Scheduling a strict systematic review process in accordance with the PRISMA principles, the work investigates AI solutions, including chatbots, automated writing assessment platforms and virtual reality applications, assessing their effectiveness in relation to various demographic groups of learners, teaching environments, and language achievements. The effects sizes were calculated using the Comprehensive Meta-Analysis (CMA) software and g of Hedge on a random-effects model. The combined findings reported a statistically significant and large overall effect (g = 1.10, SE = 0.18, 95 % CI [0.75, 1.44]) that means that AI-ready pedagogies improve EFL learning outcomes significantly, especially regarding such aspects as the accuracy of writing, fluency in speaking, and motivation of the learners. In addition, other beneficial affective effects of using AI included the reduction of anxiety and motivation and enjoyment among the learners. Subgroup analyses also indicated that the kind of measurement tool had a pronounced moderating effect on effect size, with affective variables (e.g., motivation, engagement) having stronger increases than only quantitative variables (e.g., word count). Heterogeneity was high (I2 = 92.66) in order to highlight the role of contextual and methodological differences. The findings are part of the increasing literature on AI in language learning that provides empirical data on informing educational practices to guide further studies on the adoption of the concept of intelligent technologies in EFL teaching.
PMID:41066858 | DOI:10.1016/j.actpsy.2025.105649