Modeling Structure‐Building in the Brain With CCG Parsing and Large Language Models

Published on July 7, 2023

Imagine you’re trying to build a complex structure with Legos. You could use the standard basic blocks, but they might not be enough to bring your creation to life. That’s where the power of Combinatory Categorial Grammars (CCGs) comes in for modeling language comprehension in our brains. Just like how CCGs offer more flexibility and expressiveness for grammar, they also offer a better fit for understanding neural signals in our brain while we listen to a story. The left posterior temporal lobe seems to light up with excitement when CCG-based measures are used, showing that structure-building is separate from predictability in our brains. Combined with the insights gained from a transformer neural network language model, this research opens up new horizons for studying how our brains process language in naturalistic environments. Want to dive deeper into this fascinating study? Check out the full article!

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 (CFGs), 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 functional magnetic resonance imaging (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.

Read Full Article (External Site)

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes:

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>