Slide Slam N2
Resting-state connectivity and functional activation during reading in persons with alexia
Rachael Harrington1, Candace N. Arrington1, Lisa C. Krishnamurthy123, Bruce Crosson123, Robin Morris1; 1Georgia State University, 2Emory University, 3Atlanta Department of Veterans Affairs
Reading is essential for independence and activities of daily living. After stroke, between 68-80% of persons with aphasia have an acquired reading disability, or alexia (Benjamin et al., 2018, Brookshire et al., 2013). While there is much functional imaging research examining neural circuitry in persons with aphasia, very little research, outside of case studies, has been performed to understand those neural mechanisms used by persons with acquired reading deficits after stroke. In this project, we aimed to understand the altered or compensatory neural networks underlying reading in adults with aphasia. Here we present pilot data of 4 persons with alexia (M/F: 2/2, Age: 65 ± 16) who underwent resting state and task-based imaging in a Siemens 3T Prisma MRI Scanner. 3 participants had mild anomic aphasia and 1 participant had moderate conduction aphasia (WAB: 82 ± 8.3). All participants had mild to moderate reading impairment: Test of Word Reading Efficiency-2 (74 ± 11.4). Participants underwent imaging at rest and also performed the Fast fMRI Localizer Task (Fast-Loc) in the scanner (Malins et al., 2016). This task has been shown to be highly effective at identifying brain regions that are sensitive to the linguistic properties of reading, specifically lexicality, spelling-sound consistency, and semantic similarity. We performed standard preprocessing and processing with in-house pipelines using AFNI and FSL. Resting state data were analyzed with seeds in right and left hemisphere supramarginal gyrus. Resulting connectivity (p<1-12, cluster size>50) were analyzed within subject. Patients with more severe reading impairment had greater connectivity within reading and language areas and homologous areas within the left and the right hemisphere. Participants with less severe reading impairment had less connectivity between left hemisphere reading areas and right hemisphere SMG. Left hemisphere connectivity was greatest in patients with less severe reading impairment while participants with more severe reading impairment had decreased connectivity in left hemisphere. Further analysis of task-based activation in two of these participants were performed with t-tests of semantically related words and phonologically related words against a false font condition (p<0.001, cluster size > 50). While reading semantically related words, the participant with more mild alexia showed a pattern of dorsal stream activation and involvement of right hemisphere ventral stream areas and the participant with more severe alexia showed a pattern of dorsal stream activation as well as right hemisphere activation in superior and middle temporal gyrus. While reading phonologically related words, both participants showed a pattern of incomplete left hemisphere dorsal stream activation with decreased or no activation in the right hemisphere. The relationship between resting state connectivity in reading areas and task-based activation during reading in persons with alexia warrants further exploration. The increased RH connectivity in participants with more severe alexia could be important for understanding why some persons with alexia have greater reading impairment in the chronic stage post-stroke. Further understanding this relationship is important for determining target sites for non-invasive brain stimulation to improve reading abilities for persons post-stroke.