Presentation
Search Abstracts | Symposia | Slide Sessions | Poster Sessions | Lightning Talks
Voxelwise and connectome lesion-symptom mapping reveals distinct organization of sublexical and lexical reading
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 A124 in Poster Session A, Tuesday, October 24, 10:15 am - 12:00 pm CEST, Espace Vieux-Port
Sara Dyslin1, J. Vivian Dickens1,2, Andrew DeMarco1,2, Sarah Snider2, Rhonda Friedman2, Peter Turkeltaub1,2,3; 1Center for Brain Plasticity and Recovery, Georgetown University, 2Center for Aphasia Research and Rehabilitation, Georgetown University, 3MedStar National Rehabilitation Hospital
Oral word reading relies on knowledge of both sublexical and lexico-semantic properties of words, whereas reading of novel words (e.g., “dofe”) relies primarily on knowledge of sublexical orthography-to-phonology mappings since they have no lexical/semantic representation. Neurocognitive models of reading delineate two distinct routes for sublexical vs. lexical reading, with sublexical reading relying on dorsal stream temporoparietal and inferior frontal regions, and lexical reading relying on ventral stream temporal and parietal regions (Taylor, Rastle, & Davis, 2013). Individuals with post-stroke alexia and specific behavioral deficits (e.g., a lexicality effect) provide an avenue for refining our understanding of the cognitive and neural architecture for reading in the brain. Recent research has sharpened our knowledge of the dorsal stream, finding that different lesion locations produce distinct phonological reading impairments related to deficits in sensory-motor integration vs. motor-phonological processing (Dickens et al., 2021). Here, we replicate and extend this prior work, examining how lesion location and/or structural disconnections explain reading deficits of a new set of pseudowords manipulated on orthographic body-to-phonology mappings. Participants included 64 adults with history of chronic left-hemisphere stroke (Age=62.27(13.97)); Gender (29F/35M); Months since stroke=49.61(58.37)), and 68 age-matched controls. All participants completed an MRI scan and oral reading of 200 real words and 60 pseudowords. Three types of pseudowords were presented, differentiated based on the number of orthography-to-phonology body mappings that exist in English: zero mappings (ZM e.g., “dofe”), one mapping (OM e.g., “yoon”), or multiple mappings (MM e.g., “chead”). Lesions were manually traced and warped to MNI space. Whole brain connectomes were derived from diffusion data. Multivariate Lesion-Symptom Mapping was used for parallel analyses of voxelwise lesion data and connectomes. Age, education level, and lesion volume were regressed out and corrected thresholds were applied based on permutation analyses. We conducted five pairs of analyses: 1) all pseudowords controlling for words, 2) words controlling for pseudowords, 3) ZM accuracy, controlling for OM and MM accuracy 4) OM accuracy, controlling for ZM accuracy, and 5) MM accuracy, controlling for OM accuracy. Lesions to the supramarginal gyrus and disconnections to frontal and parietal regions within the dorsal stream resulted in relative reduction of pseudoword accuracy. Lesions to extensive areas of the superior temporal sulcus (y=-2 to y=-71) resulted in relative reduction of real word accuracy. There were no significant findings for the contrast of ZM accuracy controlling for OM and MM accuracy. However, lesions and disconnections to frontal regions including the middle frontal gyrus and premotor areas resulted in relative reduction of MM and OM compared to ZM accuracy. Disconnections to the premotor regions and parietal areas resulted in relative reduction of MM compared to OM accuracy. These results show discrete organization of sublexical and lexical reading along dorsal and ventral regions respectively. Significant findings within the sublexical contrasts indicate that existing phonology-to-orthography mappings may aid in the reading of pseudowords and may additionally rely on frontal regions and their connections within the dorsal route. Further research will be needed to confirm and clarify these subprocesses and their implications within the reading network.
Topic Areas: Reading, Disorders: Acquired