Skip to main content

ETHNOS_APP

Home • Search • Journals • List 0

Form Without Function? Register Sensitivity in Cefr‐Conditioned LLM and Second‐Language Learner Writing

Bibliographic Data

ID22416398
AuthorsMichele Carlo (0009-0007-9011-6531, Graduate School of Foreign Language Education and Research Kansai University, Japan), Michele Del Carlo (0000-0003-0431-662X, Kansai University, corresponding author), Osamu Takeuchi (0000-0002-5195-992X, Graduate School of Foreign Language Education and Research Kansai University, Japan)
Year2026
Publication date2026-07-02
Peer ReviewedYes
Open AccessYes
TypeARTICLE
VenueInternational Journal of Applied Linguistics (JOURNAL)
Journal identifiersISSN: 0802-6106 • E-ISSN: 1473-4192
PublisherWiley (PUBLISHER • GB)
DOI10.1111/ijal.70284
OpenAlexW7167243873
LanguageEN
References cited23

Large language models (LLMs) can generate text that resembles L2 learner writing at requested proficiency levels, but whether such texts adapt to communicative contexts as human learners do remains unclear. This study compared LLM‐generated and human second‐language writing across three task‐defined categories (informal–personal, semi‐formal–expository, formal–argumentative) and four CEFR levels (A2–C1) using Biber's multi‐dimensional analysis (MDA). Six instruction‐tuned models responded to Write & Improve prompts ( n = 592 human; n = 4724 LLM), with EFCAMDAT as a human baseline. Under CEFR‐only prompting, LLM outputs shifted on MDA dimensions tied to informational density and explicitness, but showed weaker, differently directed, or less variable differentiation on dimensions tied to persuasion and elaboration, especially in formal–argumentative tasks. The largest observed gap was at C1, where CEFR descriptors most emphasize register flexibility. LLM outputs were more uniform than human writing and clustered near the center of the human distributional space, showing less variation than authentic learner texts. The broad pattern recurred across all six model families. CEFR‐conditioned generation can reproduce some structural signals of proficiency while showing limited human‐like register differentiation in MDA space, a pattern we interpret as form without function.

Human communication · Persuasion · Writing center · EFL/ESL Teaching and Learning · Innovative Teaching and Learning Methods · Second Language Acquisition and Learning

  • Register, Genre, and Style

    Open Access•Douglas Biber, Susan Conrad•Register, Genre, and Style•2019

  • Data Analysis Using Regression and Multilevel/Hierarchical Models

    Open Access•Andrew Gelman, Jennifer Hill•Data Analysis Using Regression…•2006

  • Variation across Speech and Writing

    Open Access•Douglas Biber•Variation Across Speech and Writing•1988

  • Discourse on the Move

    Douglas Biber, Ulla Connor et al.•Discourse on the move•2007

  • Emmeans

    Russell V Lenth, Julia Piaskowski•CRAN: Contributed Packages•2017

  • Stanza

    Open Access•Peng Qi, Yuhao Zhang et al.•Proceedings of the 58th Annual…•2020

  • Does writing development equal writing quality? A computational investigation of syntactic complexity in L2 learners

    Open Access•Scott A Crossley, D S Mcnamara•Journal of Second Language Writing•2014

  • Fitting Linear Mixed-Effects Models Using lme4

    Open Access•David M Bates, Douglas Bates et al.•Journal of Statistical Software•2015

  • LmerTest Package

    Open Access•Alexandra Kuznetsova, Per B Brockhoff et al.•Journal of Statistical Software•2017

  • Reflexivity in human-written and ChatGPT-generated English research article abstracts

    Open Access•Man Zhang, Jiawei Zhang•Applied Linguistics•2025

  • Does ChatGPT Write Like a Student? Engagement Markers in Argumentative Essays

    Open Access•Feng Jiang, K Hyland•Written Communication•2025

  • The Emergence of Complexity, Fluency, and Accuracy in the Oral and Written Production of Five Chinese Learners of English

    Diane Larsen‐freeman, D Larsen-Freeman•Applied Linguistics•2006

  • Theoretical Bases of Communicative Approaches to Second Language Teaching and Testing

    Michael Canale•Applied Linguistics•1980

Citation velocityhistorical
Highly citedNo

Tools

Open DOI
Ethnos_APP • Open Source Project • MIT License • Frontend v2.0.0 • Privacy and Cookies • API Documentation: api.ethnos.app/docs • API Source Code: GitHub • DOI: 10.5281/zenodo.17049435 • Frontend Source Code: GitHub • DOI: 10.5281/zenodo.17050053 • cruz.rio.br • Expectantes Misericordiae