Comprehensibility of AI-generated and human simplified texts for L2 learners

There is currently limited investigation of readers’ comprehension of AI simplified text from the perspective of educators, but such research can help to more effectively address the specific needs and perspectives of language teachers and learners regarding the comprehensibility of AI simplified text. Therefore, the purpose of this study was to compare second language readers’ comprehension of an original reading passage, a version that was simplified by human experts, and a version that was automatically simplified by AI. Results showed that the AI and human simplified texts were not any easier to comprehend than the original version of the text when readers’ topic familiarity was taken into account in the analysis. These findings suggest that AI-generated simplified text does not yet improve L2 learners' comprehension. Thus, further study is needed on the effect of AI simplifications using learner-centered qualitative assessments like think aloud to assess the efficacy of simplified texts. These findings also serve as a reminder to researchers and teachers that they must carefully evaluate AI tools before attempting to apply them to instructional practice.

Authors

Institutions

Publication Details

Journal
Language learning & technology
Published
2025-05-05
DOI
https://doi.org/10.64152/10125/73627
Citations
2
Primary Topic
Topic Modeling
Type
article
Field-Weighted Citation Impact
4.96
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Comprehensibility of AI-generated and human simplified texts for L2 learners

Dennis Murphy Odo
2 citations
Language learning & technology
Topic Modeling
4.96
article

Comprehensibility of AI-generated and human simplified texts for L2 learners

Dennis Murphy Odo
article en
2 citations

Abstract

There is currently limited investigation of readers’ comprehension of AI simplified text from the perspective of educators, but such research can help to more effectively address the specific needs and perspectives of language teachers and learners regarding the comprehensibility of AI simplified text. Therefore, the purpose of this study was to compare second language readers’ comprehension of an original reading passage, a version that was simplified by human experts, and a version that was automatically simplified by AI. Results showed that the AI and human simplified texts were not any easier to comprehend than the original version of the text when readers’ topic familiarity was taken into account in the analysis. These findings suggest that AI-generated simplified text does not yet improve L2 learners' comprehension. Thus, further study is needed on the effect of AI simplifications using learner-centered qualitative assessments like think aloud to assess the efficacy of simplified texts. These findings also serve as a reminder to researchers and teachers that they must carefully evaluate AI tools before attempting to apply them to instructional practice.

Language learning & technologyVol. 29(1)
Pusan National University (KR)
Openalex Percentile: Top 5%
Topic Modeling
4.96
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.