Presentation
Search Abstracts | Symposia | Slide Sessions | Poster Sessions | Lightning Talks
Gradients of brain connectivity modulate language processing with modality-specific mechanisms.
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 C74 in Poster Session C, Wednesday, October 25, 10:15 am - 12:00 pm CEST, Espace Vieux-Port
This poster is part of the Sandbox Series.
Lidon Marin Marin1,2, Susanne Eisenhauer1,2, Elizabeth Jefferies1,2; 1University of York, 2York Neuroimaging Centre
Auditory and visual language comprehension involves a series of processing steps in the brain, starting with modality-specific primary cortices, and eventually recruiting common heteromodal regions (such as the ventral anterior temporal cortex or “semantic hub”) responsible for decoding higher-level language features and integration with the evolving semantic context (1–3). Language processing at intermediate steps is also proposed to involve activation of modality-tuned brain regions, as different features of the stimuli are analysed (e.g., phonetic or orthographic characteristics) (1,2). Although previous investigations have identified the location of these mechanisms in the brain, recent studies suggest that whole-brain patterns of functional connectivity might also be crucial for language processing. The ‘principal gradient of intrinsic connectivity’ – i.e. the component of resting-state fMRI data that explains the greatest variability in whole-brain connectivity patterns – represents the gradual shift from input-driven processes to more abstract ones, since higher gradient values are located closer to heteromodal brain regions, and lower values coincide with primary sensory regions (4). Previous studies have started to explore the relationship between this gradient and heteromodal semantic cognition (5–8), while a recent study found that visual, orthographic and lexical properties of words are represented towards the sensory end of the gradient (9). Moreover, the gradient explaining the second-most variance in intrinsic connectivity, which represents the separation of visual from auditory and somatomotor regions (4), suggests that distinct dimensions of large-scale cortical organisation may capture the similarities versus the differences in language processing depending on modality. However, this hypothesis is largely unexplored. The main objective of our study is to investigate how these two gradients of human brain connectivity can capture the organisation of language processing in different modalities. To do so, we will use a publicly available fMRI dataset of 102 participants scanned during sentence listening (10) and will replicate the methodology of a previous study using the same dataset with visual stimuli (9). First, we will explore whether the principal gradient is related to the brain’s response to the psycholinguistic characteristics of auditory stimuli (GLM). These characteristics will include word-level (auditory duration, phonological distance and word frequency) and contextual measures (semantic similarity and word position in the sentence). We will also compare our results to our paper of reference (9), in order to explore modality differences. We expect to find word-level auditory characteristics to be mainly represented at the sensory end of the gradient, and contextual characteristics at the heteromodal end (significant correlation of brain’s effect of characteristics and gradient). We also expect word-level relationships to be localised in modality-tuned regions that are distinct from primary systems, and recruited in a dissociable fashion for spoken and written inputs (9). We will also explore whether the second gradient underlies the brain’s response to different modalities at different levels of the language processing hierarchy. Taken together, these findings will establish whether two gradients capturing the large-scale organisation of connectivity underlie similarities and differences in language processing across modalities. This will provide a framework for understanding language activation in diverse task contexts.
Topic Areas: Speech Perception, Meaning: Lexical Semantics