Work

Effects of AI-Generated High-Vocabulary-Density Texts on L2 Vocabulary Acquisition

Year: 2025

Type: conference-paper

Study designs: Randomized Controlled Trial, Clinical Trial

Abstract: This study examines the differential efficacy of AI-generated reading materials featuring high density of target words compared to traditional incidental reading approaches in second language (L2) vocabulary acquisition. Utilizing a randomized controlled trial design, participants were assigned to either an experimental (high-density) or control (incidental reading) condition. Following a five-day intervention, vocabulary learning outcomes were assessed through a tripartite framework encompassing: (a) word form recognition, (b) connotative understanding, and (c) contextual application accuracy. Three significant findings emerged: First, the experimental group demonstrated statistically superior performance across all assessment dimensions (p < .01), confirming that high vocabulary density enhances acquisition efficiency. Second, High-density reading showed significant inter-learner variability (SD = 1.2), suggesting proficiency-modulated effects. Third, a significant interaction emerged between vocabulary density and text length (η² = .18), with maximal benefits occurring when high density was implemented in shorter passages.

Source: International Conference on Education, Knowledge and Information Management (ICEKIM)

DOI: 10.1145/3756580.3756642

Authors: Caihong Cao, Zijian Xu

Institutions: Beijing Language and Culture University, University of California, Los Angeles

Language: English

Topic: Second Language Acquisition and Learning

Subfield: Developmental and Educational Psychology

Field: Psychology

Domain: Social Sciences

SDG: Quality education

Open Access status: hybrid

Funders: Beijing Language and Culture University, Fundamental Research Funds for the Central Universities