Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Compensatory Reading Networks in Dyslexia: Revisiting the Dorsal-to-Ventral Shift Hypothesis
Poster Session E, Sunday, September 14, 11:00 am - 12:30 pm, Field House
Delaney E. Kelemen1,2, Camden Burnsworth2, Tracy M. Centanni1,2; 1University of Florida, 2Texas Christian University
Dyslexia is a common neurodevelopmental disorder affecting reading speed and fluency. Some individuals with dyslexia exhibit increased fluency reading inverted text, though the reason for this advantage remains unknown. We hypothesize this advantage may reflect altered reading network development in dyslexia, preserving reliance on parietal regions involved in mental rotation. Typically, as reading skill increases, activation shifts from dorsal regions, such as inferior parietal lobule (IPL), to ventral regions such as the visual word form area (VWFA) on the fusiform gyrus. Given that VWFA is hypoactive in dyslexia, continued reliance on dorsal structures in young adults with dyslexia may support an inverted text advantage. This study examined connectivity between dorsal and ventral structures during upright and inverted text reading using high-density electroencephalography (EEG) and behavioral measures in young adults with and without dyslexia. Young adults (18-35 years) with (DYS, N = 22), and without (TD, N = 24) dyslexia completed a modified version of the Gray Oral Reading Test (GORT-5), reading aloud with half the stories presented normally and half upside down. Average text deviations, self-corrections, time, and percentage of comprehension questions correct were recorded. High-density EEG recordings consisted of objects, words, and sentences presented at 0°, 90°, and 180° rotation. Connectivity analyses between bilateral fusiform gyrus, IPL, and inferior frontal gyrus (IFG) were conducted using Granger causality. GORT-5 performance was used to calculate inverted text advantage by calculating the absolute difference between average errors on inverted and upright stories (“inversion effect”). Participants with an inversion effect less than one standard deviation from the mean (0.38) were classified as “good inverters” (N = 5 in DYS). Correlations between Granger causality values and task performance revealed a trend toward a negative association between right IPL and right fusiform connectivity during inverted sentence reading and magnitude of the inversion effect (r = –0.46, p = 0.062) in the DYS group. During inverted word reading, a significant negative correlation was found in DYS good inverters between right IPL and right fusiform connectivity and magnitude of the inversion effect (r = –0.97, p = 0.006). Paired t-tests revealed that during word reading, the DYS group exhibited greater connectivity between left IPL to left fusiform for upright words (p = 0.012), but greater left IPL to right fusiform connectivity for inverted words (p = 0.010). In DYS good inverters, both left IPL to left IFG and right IPL to right fusiform were more active during upright word reading (ps ≤ 0.040). As opposed to continued dorsal pathway reliance, individuals in the DYS group may recruit a right hemisphere ventral pathway for inverted text reading and differ in reading efficiency and reliance on this pathway. The DYS good inverters appear to utilize a more efficient right hemisphere ventral pathway, which may explain their reduced inversion effect. In contrast, the DYS group exhibits a shift from a canonical left ventral pathway for upright words to a bilateral ventral pathway for inverted words. These results provide support for the development of compensatory right hemisphere reading networks in dyslexia.
Topic Areas: Reading, Disorders: Developmental