Presentation
Search Abstracts | Symposia | Slide Sessions | Poster Sessions | Lightning Talks
Neuroanatomical contributions to oral pseudoword 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 A125 in Poster Session A, Tuesday, October 24, 10:15 am - 12:00 pm CEST, Espace Vieux-Port
Sachi Paul1, Corrine Durisko1, Julie A. Fiez1; 1Learning Research & Development Center, University of Pittsburgh
Phonological dyslexia is an acquired reading disorder that preserves reading of real words but results in an inability to read pronounceable pseudowords. Despite substantial behavioral evidence of this phenomenon, attempts to identify its neuroanatomical basis have yielded inconsistent results throughout the perisylvian cortex. Notably, much of this prior work has involved a single case study approach, though modern voxel-based lesion-symptom mapping (VLSM) approaches aggregate across cases and may provide more reliable information about selective impairment. The current study focuses on two recent VLSM studies of phonological dyslexia that reported significant clusters associated with selective deficits in pseudoword reading. Ripamonti et al. (2014) identified the left insula (MNI=-38,-4,8) and left inferior frontal gyrus (IFG; average MNI=-40,8,20) as significant regions in a larger network associated with phonological dyslexia, while Dickens et al. (2019) reported a significant cluster centered in the left ventral precentral gyrus (lvPCG; MNI=-38,-2,16) associated with decreased pseudoword reading performance. The current study uses a region-of-interest (ROI) approach to assess replicability of these prior studies. Fifty-four participants with single, unilateral left hemisphere lesions were recruited irrespective of aphasia status (age: 62.80±10.91; education: 15.28±3.46; WAB-AQ: 92.86±9.87; 29 females). All participants completed an MRI scan and oral reading tasks consisting of 128 monosyllabic English words matched for frequency and regularity and 40 monosyllabic English pseudowords. Overlapping voxels between each of the three ROI spheres and each participant’s traced lesion were identified. Because Ripamonti et al. used a “syndrome-based approach,” with participants grouped by acquired reading disorder, and Dickens et al. used a “process-based approach,” with participants not recruited for having alexia or grouped by disorder, we explored both methods for each ROI. Fisher’s exact tests were performed for syndrome-based analyses. The results showed no significant associations between phonological dyslexia diagnosis and lesions to the left insula (p=0.49), left IFG (p=0.28), or lvPCG (p=0.31). Linear regression analyses were conducted for process-based analyses, with lesion volume and real word reading performance as covariates. The results showed no significant associations between pseudoword reading performance and lesion overlap percentage for each of the three ROIs when relevant covariates were included (left insula: b=-0.06, t(50)=-0.63, p=0.53; left IFG: b=-0.08, t(50)=-0.64, p=0.52; lvPCG: b=-0.009, t(50)=-0.08, p=0.93)). Overall, counter to our expectations, we did not find evidence for a selective impairment in pseudoword reading in any of three ROIs. This replication failure may be due to one or more methodological differences between studies. For instance, Ripamonti et al.’s study was conducted in Italian, where criteria for phonological dyslexia are different than in English because of syllable stress. Additionally, Dickens et al. used participants recruited for a study investigating aphasia recovery, while only 37% of our participants meet criteria for an aphasia diagnosis. These differences in underlying patterns of lesion location and neuropsychological profile may affect the variance structure of data and, therefore, decrease ROI replicability. Finally, both prior studies used VLSM, which allows for broader investigation of lesioned area. Our study is ongoing, and increased statistical power will allow for parallel VLSM analyses that may yield more convergent results.
Topic Areas: Reading, Disorders: Acquired