EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

A Corpus Linguistics Analysis of Metadiscourse in AI-Generated and Human-Written Literature Reviews

Aasia Majeed, Sammy Q. Dolba, Sehrish Aslam

Journal of Arts and Linguistics Studies · 07.04.2026 · 4 (4) · 31-73

DOI: 10.71281/jals.v4i4.710 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Arts and Linguistics Studies, online erschienen am 07.04.2026. Verfasst von Aasia Majeed und 2 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The increasing use of artificial intelligence (AI) in academic writing has raised important questions regarding the rhetorical quality of AI-generated texts compared with human-written academic discourse. This study investigates the use of metadiscourse in AI-generated and human-written literature reviews using a corpus linguistics approach based on Hyland’s (2005) interpersonal model of metadiscourse. Two comparable corpora consisting of AI-generated and human-written literature reviews were analysed quantitatively and qualitatively using AntConc. The analysis examined the frequency, distribution, and contextual functions of both interactive markers (transitions, frame markers, endophoric markers, evidentials, and code glosses) and interactional markers (hedges, boosters, attitude markers, self-mentions, and engagement markers). The findings reveal that both AI-generated and human-written literature reviews employ a wide range of metadiscourse resources to organise information, express authorial stance, and engage readers. However, AI-generated texts demonstrate a greater reliance on transitions, evidentials, and engagement markers, producing highly structured and reader-oriented discourse. In contrast, human-written literature reviews exhibit stronger authorial presence through more frequent self-mentions and greater rhetorical flexibility, resulting in a more authentic and nuanced academic voice. The study concludes that while AI can effectively reproduce many structural features of academic writing, it remains less successful in replicating the interpersonal and rhetorical sophistication characteristic of human authors. These findings contribute to corpus linguistics, metadiscourse research, and the growing field of AI-assisted academic writing by providing empirical evidence on the similarities and differences between AI-generated and human-written literature reviews.

Zitieren

APA

Majeed, A., Dolba, S. Q., & Aslam, S. (2026). A Corpus Linguistics Analysis of Metadiscourse in AI-Generated and Human-Written Literature Reviews. Journal of Arts and Linguistics Studies, 4(4), 31-73. https://doi.org/10.71281/jals.v4i4.710

Vancouver

Majeed A, Dolba SQ, Aslam S. A Corpus Linguistics Analysis of Metadiscourse in AI-Generated and Human-Written Literature Reviews. Journal of Arts and Linguistics Studies. 2026;4(4):31-73. doi:10.71281/jals.v4i4.710

BibTeX
@article{Majeed2026,
  author = {Majeed, Aasia and Dolba, Sammy Q. and Aslam, Sehrish},
  title = {A Corpus Linguistics Analysis of Metadiscourse in AI-Generated and Human-Written Literature Reviews},
  journal = {Journal of Arts and Linguistics Studies},
  year = {2026},
  volume = {4},
  number = {4},
  pages = {31--73},
  doi = {10.71281/jals.v4i4.710}
}
Kostenlos starten →