Review article · 2026 · Open access

To model human linguistic prediction, make LLMs less superhuman

Byung-Doh Oh, Tal Linzen

Trends in Cognitive Sciences · Published 1 June 2026 · DOI 10.1016/j.tics.2026.05.008

Open full textCite this paper

In brief

“To model human linguistic prediction, make LLMs less superhuman” is a 2026 review article by Byung-Doh Oh and Tal Linzen, published in Trends in Cognitive Sciences. OpenAlex counts 29 citations as of 17 November 2026. It is open access under a CC BY licence: the full text is free to read at Trends in Cognitive Sciences.

Citations
29
Most cited in
2026 (28)
Field-weighted impact
20.28
References
97
Authors
2

Abstract

When we read, we make predictions about upcoming words; these predictions influence our reading behavior. The success of large language models (LLMs), which, like humans, make predictions about upcoming words, has motivated their use as models of human linguistic prediction. Surprisingly, in the last few years, as LLMs' ability to predict the next word has improved, their ability to explain reading behavior has declined. We argue this is because current LLMs can predict upcoming words much better than human readers can. This 'superhumanness' is driven by LLMs' extensive training data, stronger long-term memory of training examples, and stronger short-term memory. We advocate for LLMs with human-like memory and for new experiments to measure the alignment between humans and LLMs, and we outline directions toward achieving these goals.

Cite this paper

Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.

APA 7

Oh, B.-D., & Linzen, T. (2026). To model human linguistic prediction, make LLMs less superhuman. Trends in Cognitive Sciences. https://doi.org/10.1016/j.tics.2026.05.008

MLA 9

Oh, Byung-Doh, and Tal Linzen. “To model human linguistic prediction, make LLMs less superhuman.” Trends in Cognitive Sciences, 2026. https://doi.org/10.1016/j.tics.2026.05.008.

Chicago

Oh, Byung-Doh and Tal Linzen. 2026. “To model human linguistic prediction, make LLMs less superhuman.” Trends in Cognitive Sciences. https://doi.org/10.1016/j.tics.2026.05.008.

Harvard

Oh, B.-D. and Linzen, T. (2026) ‘To model human linguistic prediction, make LLMs less superhuman’, Trends in Cognitive Sciences. doi:10.1016/j.tics.2026.05.008.

IEEE

B.-D. Oh and T. Linzen, “To model human linguistic prediction, make LLMs less superhuman,” Trends in Cognitive Sciences, 2026, doi: 10.1016/j.tics.2026.05.008.

BibTeX

@article{oh2026model,
  title     = {{To model human linguistic prediction, make LLMs less superhuman}},
  author    = {Oh, Byung-Doh and Linzen, Tal},
  journal   = {Trends in Cognitive Sciences},
  year      = {2026},
  doi       = {10.1016/j.tics.2026.05.008},
  url       = {https://doi.org/10.1016/j.tics.2026.05.008}
}

Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)

Topic and keywords

Social Sciences › Psychology › Developmental and Educational Psychology › Reading and Literacy Development

  • linguistic prediction
  • large language models
  • reading comprehension
  • next-word prediction
  • human-LLM alignment
  • training data size
  • short-term memory
  • long-term memory

Work with this paper

Most-cited papers in Psychology

All highly cited Psychology papers

Questions about this paper

How do I cite this paper in APA style?

Oh, B.-D., & Linzen, T. (2026). To model human linguistic prediction, make LLMs less superhuman. Trends in Cognitive Sciences. https://doi.org/10.1016/j.tics.2026.05.008

How many times has this paper been cited?

29 times, according to OpenAlex on 17 November 2026. Google Scholar and Scopus count different sets of sources, so their numbers differ.

Is this paper free to read?

Yes. It is open access at Trends in Cognitive Sciences: sign in and open the full text from this page.

Who wrote this paper?

Byung-Doh Oh, Tal Linzen.

What is the DOI of this paper?

10.1016/j.tics.2026.05.008, permanently at https://doi.org/10.1016/j.tics.2026.05.008. Sign in and open it from this page.

Bibliographic data from OpenAlex (CC0), last checked 17 November 2026. ResearchGuru does not host this paper; once you are signed in, the links open it at the publisher or the repository that does.