Presentation
Search Abstracts | Symposia | Slide Sessions | Poster Sessions | Lightning Talks
Individual word representations dissociate from linguistic context along a cortical unimodal to heteromodal gradient
There is a Poster PDF for this presentation, but you must be a current member or registered to attend SNL 2023 to view it. Please go to your Account Home page to register.
Poster C119 in Poster Session C, Wednesday, October 25, 10:15 am - 12:00 pm CEST, Espace Vieux-Port
Susanne Eisenhauer1, Tirso Gonzalez Alam1, Piers Cornelissen2, Jonathan Smallwood3, Elizabeth Jefferies1; 1University of York, 2Northumbria University, 3Queen's University, Canada
Language comprehension involves multiple hierarchical processing stages across time, space, and levels of representation. When processing a word, the sensory input is transformed into increasingly abstract representations that are integrated with the linguistic context. While neuroimaging research has traditionally focused on mapping individual brain regions to the underlying subprocesses, recent studies suggest that whole-brain distributed patterns of cortical activation might be highly relevant for efficient cognition. Specifically, recent theories propose that the sensory- and memory-dependent processes underlying language comprehension are structured along a principal cortical gradient from unimodal to heteromodal brain regions. We investigated the gradient’s role in sentence reading, using multiple linear regression on open functional magnetic resonance imaging (fMRI) and magnetoencephalography (MEG) recordings from 102 participants. We found that the gradient is recruited differentially for the processing of individual word representations (visual, orthographic, and lexical) in contrast to properties reflecting a word’s relation to its linguistic context (semantic similarity and position within the sentence): Word representations gradually involved the unimodal end to a stronger extent, while contextual representations gradually increased towards the heteromodal end of the gradient. Individual word representations showed opposite effect directions on brain activation in fMRI and MEG, for example, reduced brain activation for more vs. less frequent words in fMRI, and the reverse pattern in MEG. Nevertheless, the association of individual word representations with the unimodal end of the gradient was consistent across both neuroimaging modalities. MEG revealed that the observed distinction along the gradient persisted through time, suggesting parallel processing across word representation levels and context. Our findings indicate the gradient captures the neural organization of language by providing a gradual dissociation between word vs. contextual representations. Furthermore, the gradient reveals convergent patterns across neuroimaging modalities (similar location along the gradient) in the presence of divergent responses (opposite effect directions).
Topic Areas: Reading, Meaning: Lexical Semantics