Lexico-Grammatical Patterning in Human and AI-Generated Abstracts: A Comparative Corpus Study
The Critical Review of Social Sciences Studies · 05.03.2026 · 4 (1) · 2657-2664
DOI: 10.59075/dh07ga80 ↗Kurz gefasst
Fachartikel in der Zeitschrift The Critical Review of Social Sciences Studies, online erschienen am 05.03.2026. Verfasst von Afaq Ahmed Khan und 2 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
The growing use of large language models in academic writing necessitates systematic investigation of their linguistic output. This study presents a comparative corpus analysis of lexico-grammatical patterning in human-written and AI-generated academic abstracts in linguistics. A parallel corpus of twenty abstracts that comprise ten human-authored (HWAC) and ten ChatGPT-generated (AGAC) on matched topics was analysed across four dimensions: lexical density, grammatical structure, epistemic stance, and rhetorical move realisation. Findings reveal that human-written abstracts exhibit greater lexical specificity, concise syntactic structures, and direct empirical reporting, whereas AI-generated abstracts display generalised academic vocabulary, elevated nominalisation, syntactic elaboration, and a conceptually explanatory orientation. Epistemic stance also diverges, with human authors employing assertive constructions and AI texts favouring hedged formulations. Although AI-generated abstracts successfully emulate academic discourse conventions, meaningful lexico-grammatical differences persist, with implications for genre analysis, authorship assessment, and AI-assisted scholarly communication.
Zitieren
APA
Khan, A. A., Rabail, & Shabbir, N. (2026). Lexico-Grammatical Patterning in Human and AI-Generated Abstracts: A Comparative Corpus Study. The Critical Review of Social Sciences Studies, 4(1), 2657-2664. https://doi.org/10.59075/dh07ga80
Vancouver
Khan AA, Rabail, Shabbir N. Lexico-Grammatical Patterning in Human and AI-Generated Abstracts: A Comparative Corpus Study. The Critical Review of Social Sciences Studies. 2026;4(1):2657-2664. doi:10.59075/dh07ga80
BibTeX
@article{Khan2026,
author = {Khan, Afaq Ahmed and Rabail and Shabbir, Nida},
title = {Lexico-Grammatical Patterning in Human and AI-Generated Abstracts: A Comparative Corpus Study},
journal = {The Critical Review of Social Sciences Studies},
year = {2026},
volume = {4},
number = {1},
pages = {2657--2664},
doi = {10.59075/dh07ga80}
}
EUVIMED