AI-MEDIATED LANGUAGE LEARNING: AN APPLIED LINGUISTICS ANALYSIS OF THE XEROPAN PLATFORM

Abstract

In this quantitative research it is to explore the efficiency of an AI supported language learning platform, Xeropan operating under applied linguistic grounds supported by Sociocultural Theory. A real experimental pre-test–post experiment design was used in secondary education students, including control group with traditional instruction and experimental group learning on the Xeropan platform. Statistical analysis included descriptive analyses and independent samples t-tests to compare group performance, as well Levene’s test for equality of variances. There was no significant pretest score difference between the ESMS and control groups (ESMS group M = 21.90; control group M = 26.52), providing support for initial group equivalence. Post-test analysis demonstrated that there was a statistically significant difference in favour of the experimental group with experimental group gaining much more (M = 44.00) than their counterparts in the control group (M = 25.75), t(38) = −17.93, p <. 001. The variance of scores in the experimental condition was also less, meaning more uniform learning outcomes. These results provide strong empirical evidence that AI-based learning with Xeropan notably increases learners’ language proficiency, especially in performance related to reading. “Artificially intelligent language tutors: How tutoring can support language learning” finds that AI-guided online sessions for developing both the spoken and written word can be used as educational tools in a scalable yet felt, human-like way.