Disentangling Syntax and Semantics in the Brain with Deep Networks
This provides a framework for isolating linguistic representations in brain activity, advancing neuroscience and computational linguistics, though it is incremental in applying existing methods to new data.
The study tackled the problem of understanding how linguistic classes like syntax and semantics are represented in the brain by decomposing GPT-2 activations and fMRI data from 345 subjects listening to narrated text, finding that compositional representations recruit a widespread cortical network and that syntax and semantics share a common neural substrate rather than being separated.
The activations of language transformers like GPT-2 have been shown to linearly map onto brain activity during speech comprehension. However, the nature of these activations remains largely unknown and presumably conflate distinct linguistic classes. Here, we propose a taxonomy to factorize the high-dimensional activations of language models into four combinatorial classes: lexical, compositional, syntactic, and semantic representations. We then introduce a statistical method to decompose, through the lens of GPT-2's activations, the brain activity of 345 subjects recorded with functional magnetic resonance imaging (fMRI) during the listening of ~4.6 hours of narrated text. The results highlight two findings. First, compositional representations recruit a more widespread cortical network than lexical ones, and encompass the bilateral temporal, parietal and prefrontal cortices. Second, contrary to previous claims, syntax and semantics are not associated with separated modules, but, instead, appear to share a common and distributed neural substrate. Overall, this study introduces a versatile framework to isolate, in the brain activity, the distributed representations of linguistic constructs.