---
title: "Lexical and Discourse Semantics in a Reading-time Corpus of"
description: "Reading-time corpora help psycholinguists test processing theories on natural text. 30 references."
authors: ["Jakub Dotlacil", "Laia Colina Fortuny", "Li Kloostra", "Johan Bos"]
year: 2026
venue: "Proceedings of the Fifteenth Language Resources and Evaluation Conference"
source: "https://aclanthology.org/2026.lrec-1.754/"
doi: "10.63317/5hcb799tf3uq"
licence: "CC BY-NC 4.0 (ELRA, LREC proceedings)"
licence_url: "https://creativecommons.org/licenses/by-nc/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/dotlacil-2026-lexical-discourse-semantics-reading-time/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/dotlacil-2026-lexical-discourse-semantics-reading-time/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2026](https://md.research.elamur.ai/free/en/papers/2026/page/34/) / Lexical and Discourse Semantics in a Reading-time Corpus of English

# Lexical and Discourse Semantics in a Reading-time Corpus of English

**Authors:** Jakub Dotlacil, Laia Colina Fortuny, Li Kloostra and [Johan Bos](https://md.research.elamur.ai/free/en/authors/johan-bos/) · **Year:** 2026 · **Venue:** [Proceedings of the Fifteenth Language Resources and Evaluation Conference](https://md.research.elamur.ai/free/en/venues/lrec/) ([LREC 2026](https://md.research.elamur.ai/free/en/venues/lrec/2026/)) · **Publisher:** [European Language Resources Association (ELRA)](https://md.research.elamur.ai/free/en/publishers/european-language-resources-association-elra/) · **Licence:** [CC BY-NC 4.0 (ELRA, LREC proceedings)](https://creativecommons.org/licenses/by-nc/4.0/)

**Links:** [Publisher's page](https://aclanthology.org/2026.lrec-1.754/) · [PDF](https://aclanthology.org/2026.lrec-1.754.pdf) · [DOI 10.63317/5hcb799tf3uq](https://doi.org/10.63317/5hcb799tf3uq) · [OpenAlex](https://openalex.org/W7160165267)

**Citations:** 0 citations (OpenAlex, 2026-10-14)

## Summary
Reading-time corpora help psycholinguists test processing theories on natural text. Until now, their linguistic layers have stopped at morphology and syntax, so semantic and discourse-level theories had no comparable test bed. The authors took the UCL corpus and annotated it with the full Parallel Meaning Bank (PMB) stack. The UCL corpus has 361 standalone British English sentences with self-paced reading, eye-tracking and EEG measures. The PMB layers are tokenization, semantic tags, symbolization, CCG derivations, WordNet senses, and Discourse Representation Structures built by Boxer, including VerbNet roles, anaphora and scope. The sentences were first annotated automatically, then corrected by hand over several months. Corrections were made to Boxer's inputs rather than to its output. The merged data is released through OSF and the PMB explorer. Two Bayesian case studies show the corpus is usable: one on interference from gender similarity during encoding, one on the cost of integrating semantic roles. Both produced patterns that agree with established theories or earlier findings. This makes the corpus the first reading-time resource with lexical and discourse semantics.

## Key points
- **Source corpus:** UCL corpus, 361 sentences averaging 13.7 tokens, drawn from unpublished online novels. Proper names were altered so that none occurs more than twice.
- **Reading measures:**
  - Self-paced reading: 117 participants, each reading a random subset of the sentences.
  - Eye-tracking: 43 participants, on the 205 sentences that fit on one line.
  - EEG: 24 participants, on the same 205 sentences.
- **PMB pipeline:** six steps.
  - Semantic tagging uses 80 fine-grained semtags grouped into 13 coarse classes.
  - Automatic tools were elephant (tokenizer), TNT (semtagger), EasyCCG (parser) and Boxer (DRS construction).
- **Notation:** meaning is given in a token-aligned sequence notation with three columns: WordNet synsets/constants, VerbNet roles with relative position indices, and the words themselves.
- **Statistical setup:**
  - Case studies used brms/Stan with principled priors and maximal random effects for subjects and word types.
  - Sampling ran 4 chains of 2,500 iterations; all Rhat values were below 1.02.
- **Availability:** data is at doi 10.17605/OSF.IO/PES63, with annotations also browsable on the PMB explorer.

## Interesting ideas
- Annotators almost never edited the final DRS; they fixed the tags, symbols or CCG parse feeding Boxer, so the corrected meaning stays derivable from corrected inputs.
- Semantic roles are written as signed offsets from the role-assigning token, so a whole meaning graph fits in token-aligned columns and joins easily with per-word data.
- A separate symbolization step normalizes lemmas into canonical meaning symbols, for example marking a pronoun's gender or mapping a modal to a logical operator.
- Annotation was tiered: each annotator fixed only cases they were confident about, doubtful items went to group discussion, and two people did a final consistency sweep.
- The resource was validated extrinsically, by checking that its annotations reproduce known processing effects rather than only reporting annotation quality.

## For Mova

How these ideas could serve Mova, our small open language model — its files, stages and tests:

- **I1** (learn): Repair rules and teacher corrections could be applied to the UD tree and tags, with frames regenerated afterwards, so every fix remains explainable through the normal pipeline.
- **I2** (frames): A column format with relative-offset role links could make frame outputs diffable against gold data and simple to align with tokens.
- **I3** (tokens): A symbol layer between lemmas and frames would give downstream stages consistent keys for pronoun features and modal force.
- **I4** (learn): This maps onto escalation in LLM teacher tiers: cheap witnesses settle confident cases, disagreements escalate, and a final pass checks global consistency.
- **I5** (learn): Mova's frame and world layers could also be evaluated by whether they recover known phenomena, alongside negative controls.

*Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.*

📚 Shelves (5): [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 16,039 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 15,891 works · [💭 Psychology](https://md.research.elamur.ai/free/en/shelves/psychology/) · 1,552 works · [💭 Developmental and Educational Psychology](https://md.research.elamur.ai/free/en/shelves/psychology/developmental-and-educational-psychology/) · 499 works · [📚 Literature and Literary Theory](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/literature-and-literary-theory/) · 519 works

🧭 Topics (3): [English](https://md.research.elamur.ai/free/en/topics/english/) · 2,263 works · [🧵 Discourse analysis](https://md.research.elamur.ai/free/en/topics/discourse-analysis/) · 466 works · [🧒 Psycholinguistics & language acquisition](https://md.research.elamur.ai/free/en/topics/psycholinguistics/) · 102 works

## Cite

Cite as: Jakub Dotlacil, Laia Colina Fortuny, Li Kloostra and Johan Bos. 2026. Lexical and Discourse Semantics in a Reading-time Corpus of English. Proceedings of the Fifteenth Language Resources and Evaluation Conference.

The paper itself, not this page. BibTeX from the library's record of the work.

```bibtex
@inproceedings{dotlacil2026lexical,
  title = {{Lexical and Discourse Semantics in a Reading-time Corpus of English}},
  author = {Jakub Dotlacil, Laia Colina Fortuny, Li Kloostra and Johan Bos},
  booktitle = {Proceedings of the Fifteenth Language Resources and Evaluation Conference},
  year = {2026},
  doi = {10.63317/5hcb799tf3uq},
  url = {https://doi.org/10.63317/5hcb799tf3uq}
}
```

- [← previous Leveraging Unannotated Sign Language Data via a Robust Data Augmentation Method for Contrastive Representation Learning](https://md.research.elamur.ai/free/en/papers/madjoukeng-2026-leveraging-unannotated-sign-language-data/)
- [next → Lexical and Syntactic Diversity: Still Lost in Machine Translation?](https://md.research.elamur.ai/free/en/papers/volkart-2026-lexical-syntactic-diversity-still-lost/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/34/)

## References

30 references, 3 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Lasha Abzianidze, Johannes Bjerva, Kilian Evang, Hessel Haagsma, Rik Van Noord, Pierre Ludmann, Duc-Duy Nguyen, and Johan Bos. 2017. [The parallel meaning bank: Towards a multilingual corpus of translations annotated with compositional meaning representations](https://md.research.elamur.ai/free/en/papers/abzianidze-2017-parallel-meaning-bank-multilingual-corpus/). In Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics, pages 242–247, Valencia, Spain.
2. Lasha Abzianidze and Johan Bos. 2017. Towards universal semantic tagging. In Proceedings of the 12th International Conference on Computational Semantics (IWCS 2017) – Short Papers, pages 1–6, Montpellier, France. [↗](https://doi.org/10.48550/arxiv.1709.10381)
3. Maria Barrett, Zeljko Agic, and Anders Søgaard. 2015. The dundee treebank. In The 14th international workshop on treebanks and linguistic theories (TLT 14), pages 242–248. [find](https://scholar.google.com/scholar?q=Maria+Barrett%2C+Zeljko+Agic%2C+and+Anders+S%C3%B8gaard%2E+2015%2E+The+dundee+treebank%2E+In+The+14th+international+workshop+on+treebanks+and+linguistic+theories+%28TLT+14%29%2C+pages+242%E2%80%93248%2E)
4. Johan Bos. 2015. Open-domain semantic parsing with boxer. In Proceedings of the 20th Nordic Conference of Computational Linguistics (NODALIDA 2015), pages 301–304. [find](https://scholar.google.com/scholar?q=Johan+Bos%2E+2015%2E+Open%2Ddomain+semantic+parsing+with+boxer%2E+In+Proceedings+of+the+20th+Nordic+Conference+of+Computational+Linguistics+%28NODALIDA+2015%29%2C+pages+301%E2%80%93304%2E)
5. Johan Bos. 2023. The sequence notation: Catching complex meanings in simple graphs. In 15th International Conference on Computational Semantics, pages 195–208. Association for Computational Linguistics (ACL). [find](https://scholar.google.com/scholar?q=Johan+Bos%2E+2023%2E+The+sequence+notation%3A+Catching+complex+meanings+in+simple+graphs%2E+In+15th+International+Conference+on+Computational+Semantics%2C+pages+195%E2%80%93208%2E+Association+for+Computational+Linguistic)
6. Johan Bos, Stephen Clark, Mark Steedman, James R Curran, and Julia Hockenmaier. 2004. Wide-coverage semantic representations from a ccg parser. In COLING 2004: Proceedings of the 20th international conference on computational linguistics, pages 1240–1246. [find](https://scholar.google.com/scholar?q=Johan+Bos%2C+Stephen+Clark%2C+Mark+Steedman%2C+James+R+Curran%2C+and+Julia+Hockenmaier%2E+2004%2E+Wide%2Dcoverage+semantic+representations+from+a+ccg+parser%2E+In+COLING+2004%3A+Proceedings+of+the+20th+international+co)
7. Marisa Ferrara Boston, John Hale, Reinhold Kliegl, Umesh Patil, and Shravan Vasishth. 2008. Parsing costs as predictors of reading difficulty: An evaluation using the potsdam sentence corpus. Journal of Eye Movement Research, 2:1–12. [↗](https://doi.org/10.16910/jemr.2.1.1)
8. Thorsten Brants. 2000. [TnT – a statistical part-of-speech tagger](https://md.research.elamur.ai/free/en/papers/brants-2000-tnt-statistical-part-speech-tagger/). In Sixth Applied Natural Language Processing Conference, pages 224–231, Seattle, Washington, USA. Association for Computational Linguistics. [↗](https://doi.org/10.48550/arxiv.cs/0003055)
9. Paul-Christian Bürkner. 2021. Bayesian item response modeling in R with brms and Stan. Journal of Statistical Software, 100(5):1–54. [↗](https://doi.org/10.18637/jss.v100.i05)
10. Vera Demberg and Frank Keller. 2008. Data from eye-tracking corpora as evidence for theories of syntactic processing complexity. Cognition, 109(2):193–210. [↗](https://doi.org/10.1016/j.cognition.2008.07.008)
11. Jakub Dotlačil. 2021. Parsing as a cue-based retrieval model. Cognitive science, 45:e13020. [↗](https://doi.org/10.1111/cogs.13020)
12. Kilian Evang, Valerio Basile, Grzegorz Chrupała, and Johan Bos. 2013. Elephant: Sequence labeling for word and sentence segmentation. In EMNLP 2013, pages 1422–1426. [find](https://scholar.google.com/scholar?q=Kilian+Evang%2C+Valerio+Basile%2C+Grzegorz+Chrupa%C5%82a%2C+and+Johan+Bos%2E+2013%2E+Elephant%3A+Sequence+labeling+for+word+and+sentence+segmentation%2E+In+EMNLP+2013%2C+pages+1422%E2%80%931426%2E)
13. Christiane Fellbaum. 1998. WordNet: An electronic lexical database. MIT press. [find](https://scholar.google.com/scholar?q=Christiane+Fellbaum%2E+1998%2E+WordNet%3A+An+electronic+lexical+database%2E+MIT+press%2E)
14. Edward Gibson. 2000. The dependency locality theory: A distance-based theory of linguistic complexity. In Image, language, brain, pages 95–126. MIT Press. [↗](https://doi.org/10.7551/mitpress/3654.003.0008)
15. Shinnosuke Isono. 2024. Category locality theory: A unified account of locality effects in sentence comprehension. Cognition, 247:105766. [↗](https://doi.org/10.1016/j.cognition.2024.105766)
16. Hans Kamp and Uwe Reyle. 1993. From Discourse to Logic. Introduction to Model theoretic Semantics of Natural Language, Formal Logic and Discourse Representation Theory. Kluwer, Dordrecht. [↗](https://doi.org/10.1007/978-94-011-2066-1)
17. Karin Kipper, Anna Korhonen, Neville Ryant, and Martha Palmer. 2008. A large-scale classification of english verbs. Language Resources and Evaluation, 42(1):21–40. [↗](https://doi.org/10.1007/s10579-007-9048-2)
18. Mike Lewis and Mark Steedman. 2014. [A* CCG parsing with a supertag-factored model](https://md.research.elamur.ai/free/en/papers/lewis-2014-ccg-parsing-supertag-factored-model/). In 2014 Conference on Empirical Methods in Natural Language Processing, pages 990–1000. Association for Computational Linguistics.
19. Bruno Nicenboim, Daniel J Schad, and Shravan Vasishth. 2025. Introduction to Bayesian data analysis for cognitive science. CRC Press. [↗](https://doi.org/10.1201/9780429342646)
20. Klaus Oberauer and Reinhold Kliegl. 2006. A formal model of capacity limits in working memory. Journal of Memory and Language, 55(4):601– 626. [↗](https://doi.org/10.1016/j.jml.2006.08.009)
21. Cory Shain, Idan Asher Blank, Evelina Fedorenko, Edward Gibson, and William Schuler. 2022. Robust effects of working memory demand during naturalistic language comprehension in language-selective cortex. Journal of Neuroscience, 42(39):7412–7430. [↗](https://doi.org/10.1523/jneurosci.1894-21.2022)
22. Nathaniel J Smith and Roger Levy. 2013. The effect of word predictability on reading time is logarithmic. Cognition, 128(3):302–319. 8. [↗](https://doi.org/10.1016/j.cognition.2013.02.013)
23. Language Resource References Berzak, Yevgeni and Nakamura, Chie and Smith, Amelia and Weng, Emily and Katz, Boris and Flynn, Suzanne and Levy, Roger. 2022. CELER: A 365-participant corpus of eye movements in L1 and L2 English reading. MIT Press One Broadway, 12th Floor, Cambridge, Massachusetts 02142, USA. PID https://doi.org/10.1162/opmi_a_00054. [↗](https://doi.org/10.1162/opmi_a_00054)
24. Cop, Uschi and Dirix, Nicolas and Drieghe, Denis and Duyck, Wouter. 2017. Presenting GECO: An eyetracking corpus of monolingual and bilingual sentence reading. Springer. PID https://doi.org/10.3758/s13428-016-0734-0. [↗](https://doi.org/10.3758/s13428-016-0734-0)
25. Frank, Stefan L. and Monsalve, Irene Fernandez and Thompson, Robin L. and Vigliocco, Gabriella. 2013. Reading time data for evaluating broad-coverage models of English sentence processing. PID https://doi.org/10.3758/s13428-012-0313-y. [↗](https://doi.org/10.3758/s13428-012-0313-y)
26. Frank, Stefan L and Otten, Leun J and Galli, Giulia and Vigliocco, Gabriella. 2015. The ERP response to the amount of information conveyed by words in sentences. Elsevier. PID https://doi.org/10.1016/j.bandl.2014.10.006. [↗](https://doi.org/10.1016/j.bandl.2014.10.006)
27. Futrell, Richard and Gibson, Edward and Tily, Harry J. and Blank, Idan and Vishnevetsky, Anastasia and Piantadosi, Steven T. and Fedorenko, Evelina. 2016. The Natural Stories Corpus. PID [http://github.com/languageMIT/naturalstories](http://github.com/languageMIT/naturalstories). [↗](http://github.com/languageMIT/naturalstories)
28. Hollenstein, Nora and Rotsztejn, Jonathan and Troendle, Marius and Pedroni, Andreas and Zhang, Ce and Langer, Nicolas. 2018. ZuCo, a simultaneous EEG and eye-tracking resource for natural sentence reading. Nature Publishing Group. PID https://doi.org/10.1038/sdata.2018.291. [↗](https://doi.org/10.1038/sdata.2018.291)
29. Kennedy, Alan. 2003. The Dundee Corpus. The University of Dundee, Psychology Department. CD-ROM. [find](https://scholar.google.com/scholar?q=Kennedy%2C+Alan%2E+2003%2E+The+Dundee+Corpus%2E+The+University+of+Dundee%2C+Psychology+Department%2E+CD%2DROM%2E)
30. Siegelman, Noam and Schroeder, Sascha and Acartürk, Cengiz and Ahn, Hee-Don and Alexeeva, Svetlana and Amenta, Simona and Bertram, Raymond and Bonandrini, Rolando and Brysbaert, Marc and Chernova, Daria and others. 2022. Expanding horizons of cross-linguistic research on reading: The Multilingual Eye-movement Corpus (MECO). Springer. PID https://doi.org/10.3758/s13428-021-01772-6.9. Optional Supplementary Materials: Appendices, Software, and Data 9.1. [↗](https://doi.org/10.3758/s13428-021-01772-6.9)

*Updated 1 November 2026*