What Limits Our Capacity to Process Nested Long-Range Dependencies in Sentence Comprehension?
Open accessThe DOI matches its Crossref record; 15 of the 15 references it cites were found in a registry or an index.
Abstract
Sentence comprehension requires inferring, from a sequence of words, the structure of syntactic relationships that bind these words into a semantic representation. Our limited ability to build some specific syntactic structures, such as nested center-embedded clauses (e.g., “The dog that the cat that the mouse bit chased ran away”), suggests a striking capacity limitation of sentence processing, and thus offers a window to understand how the human brain processes sentences. Here, we review the main hypotheses proposed in psycholinguistics to explain such capacity limitation. We then introduce an alternative approach, derived from our recent work on artificial neural networks optimized for language modeling, and predict that capacity limitation derives from the emergence of sparse and feature-specific syntactic units. Unlike psycholinguistic theories, our neural network-based framework provides precise capacity-limit predictions without making any a priori assumptions about the form of the grammar or parser. Finally, we discuss how our framework may clarify the mechanistic underpinning of language processing and its limitations in the human brain.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2018) · arXiv · doi:10.48550/arxiv.1810.04805 · Cited by 73,173
Learning representations by back-propagating errors
Rumelhart, D. E., Hinton, G. E., & Williams, R. J. (1986) · Nature · doi:10.1038/323533a0 · Cited by 31,892
Connectionism and cognitive architecture: A critical analysis
Fodor, J. A., & Pylyshyn, Z. W. (1988) · Cognition · doi:10.1016/0010-0277(88)90031-5 · Cited by 4,022
Linguistic complexity: locality of syntactic dependencies
Gibson, E. (1998) · Cognition · doi:10.1016/s0010-0277(98)00034-1 · Cited by 2,537 · cited 3×
“…of a relative clause (RC) within a sentential complement (SC), for example: “The fact that the employee who the manager hired stole office supplies worried the executive (SC/RC)”, which was shown to be easier to process than the reversed embedding (RC/SC) and doubly center-embedded sentences [12].”
BackgroundInfluentialInvariant visual representation by single neurons in the human brain
Quiroga, R. Q., Reddy, L., Kreiman, G., Koch, C., & Fried, I. (2005) · Nature · doi:10.1038/nature03687 · Cited by 1,988
Natural speech reveals the semantic maps that tile human cerebral cortex
Huth, A. G., de Heer, W. A., Griffiths, T. L., Theunissen, F. E., & Gallant, J. L. (2016) · Nature · doi:10.1038/nature17637 · Cited by 1,504
A probabilistic earley parser as a psycholinguistic model
Hale, J. (2001) · Second meeting of the North American Chapter of the Association for Computational Linguistics on Language technologies 2001 - NAACL '01 · doi:10.3115/1073336.1073357 · Cited by 1,150 · cited 2×
“To quantify expectation, Hale [18] suggested the Shannon information content, or ‘surprisal’, as a measure for processing difficulty: surprisal ( w i + 1 ) = − log ( p ( w i + 1 | w 1 ... w i ) , where w i are words in the sentence (for illustration purpose, we assume that the sentence is presented…”
BackgroundSmolensky, P. (1990) · Artificial Intelligence · doi:10.1016/0004-3702(90)90007-m · Cited by 959
A probabilistic model of lexical and syntactic access and disambiguation
Jurafsky, D. (1996) · Cognitive Science · doi:10.1016/s0364-0213(99)80005-6 · Cited by 714
Holographic reduced representations
Plate, T. A. (1995) · IEEE Transactions on Neural Networks · doi:10.1109/72.377968 · Cited by 657 · cited 1×
“Subsequent proposals, such as Vector Symbolic Architectures [22], Binary Spatter-Coding [23], Holographic Reduced Representation [24], proposed hand-crafted solutions to efficiently approximate tensor products with a limited number of neurons.”
Method
Cited by
The neural architecture of language: Integrative modeling converges on predictive processing
Schrimpf, M., Blank, I. A., Tuckute, G., Kauf, C., Hosseini, E. A., Kanwisher, N., Tenenbaum, J. B., & Fedorenko, E. (2021) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2105646118 · Cited by 517
Syntactic Structure from Deep Learning
Linzen, T., & Baroni, M. (2021) · Annual Review of Linguistics · doi:10.1146/annurev-linguistics-032020-051035 · Cited by 202
The neural architecture of language: Integrative modeling converges on predictive processing
(2020) · doi:10.1101/2020.06.26.174482 · Cited by 129
Causal Transformers Perform Below Chance on Recursive Nested Constructions, Unlike Humans
Lakretz, Y., Desbordes, T., Hupkes, D., & Dehaene, S. (2021) · arXiv · doi:10.48550/arxiv.2110.07240 · Cited by 4
“And what does that tell us about the data that those models are trained on and about the potential processes that humans may use to process such constructions (Lakretz et al., 2020)?”
BackgroundSelf-citation
Metadata
- Authors
- Lakretz, Y., Dehaene, S., & King, J.-R.
- Venue
- Entropy
- Publisher
- MDPI AG
- Published
- 2020-04-16
- Volume
- 22
- Issue
- 4
- Pages
- 446
- ISSN
- 1099-4300 (ISSN-L)
- OpenAlex
- W3016825020
- Cited by
- 33
- References
- 15
- FWCI
- 2.38
- Type
- journal article
- Language
- en
- Open access
- gold
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.