We studied the learnability of English filler-gap dependencies and the “island” constraints on them by assessing the generalizations made by autoregressive (incremental) language models that use deep learning to predict the next word given preceding context. Using factorial tests inspired by experimental psycholinguistics, we found that models acquire not only the basic contingency between fillers and gaps, but also the unboundedness and hierarchical constraints implicated in the dependency. We evaluated a model’s acquisition of island constraints by demonstrating that its expectation for a filler-gap contingency is attenuated within an island environment. Our results provide empirical evidence against the argument from the poverty of the stimulus for this particular structure.
Skip Nav Destination
April 20 2023
Using Computational Models to Test Syntactic Learnability
In Special Collection: CogNet
Ethan Gotlieb Wilcox,
Ethan Gotlieb Wilcox
Online ISSN: 1530-9150
Print ISSN: 0024-3892
© 2022 by the Massachusetts Institute of Technology
Massachusetts Institute of Technology
Linguistic Inquiry 1–44.
Ethan Gotlieb Wilcox, Richard Futrell, Roger Levy; Using Computational Models to Test Syntactic Learnability. Linguistic Inquiry 2023; doi: https://doi.org/10.1162/ling_a_00491
Download citation file: