Abstract:To model behavioral and neural correlates of language comprehension in naturalistic environments researchers have turned to broad-coverage tools from natural-language processing and machine learning. Where syntactic structure is explicitly modeled, prior work has relied predominantly on context-free grammars (CFG), yet such formalisms are not sufficiently expressive for human languages. Combinatory Categorial Grammars (CCGs) are sufficiently expressive directly compositional models of grammar with flexible constituency that affords incremental interpretation. In this work we evaluate whether a more expressive CCG provides a better model than a CFG for human neural signals collected with fMRI while participants listen to an audiobook story. We further test between variants of CCG that differ in how they handle optional adjuncts. These evaluations are carried out against a baseline that includes estimates of next-word predictability from a Transformer neural network language model. Such a comparison reveals unique contributions of CCG structure-building predominantly in the left posterior temporal lobe: CCG-derived measures offer a superior fit to neural signals compared to those derived from a CFG. These effects are spatially distinct from bilateral superior temporal effects that are unique to predictability. Neural effects for structure-building are thus separable from predictability during naturalistic listening, and those effects are best characterized by a grammar whose expressive power is motivated on independent linguistic grounds.

Improved Combinatory Categorial Grammar Induction with Boundary Words and Bayesian Inference.

Taxonomy Induction from Chinese Encyclopedias by Combinatorial Optimization.

Neural Bi-Lexicalized PCFG Induction

An Empirical Study of Compound PCFGs

Parsing Combinatory Categorial Grammar with Answer Set Programming: Preliminary Report

Parsing TCT with Split Conjunction Categories.

Bayesian Grammar Induction for Language Modeling

Large-scale CCG Induction from the Groningen Meaning Bank

An Improved Approach for Semantic Graph Composition with CCG

Improved parsing with taxonomy of conjunctions

Category Locality Theory: A unified account of locality effects in sentence comprehension

Classifying Syntactic Categories in the Chinese Dependency Network.

Dependency Induction Through the Lens of Visual Perception

Controlling Complexity in Part-of-Speech Induction

Experimental Support for a Categorical Compositional Distributional Model of Meaning

AN EFFICIENT PARSING MODEL FOR UNIFICATION CATEGORIAL GRAMMAR WITH OBJECT-ORIENTED KNOWLEDGE REPRESENTATION AND SELECTION SETS

PCFGs Can Do Better: Inducing Probabilistic Context-Free Grammars with Many Symbols

Modeling structure-building in the brain with CCG parsing and large language models

Visually Grounded Compound PCFGs

Re-evaluating the Need for Multimodal Signals in Unsupervised Grammar Induction

Statistical properties of probabilistic context-sensitive grammars