Application of graph theory to both resting-state and task-based fMRI data to uncover brain-behavior relationships related to therapy outcomes in aphasia

Project Summary Although many treatment paradigms exist for aphasia rehabilitation with behaviorally successful outcomes, there is still much debate regarding the ?best? pattern of neural reorganization during recovery. This lack of consensus may be due, in part, to incomplete exploration of the link between neural reorganization and behavior. The proposed project will apply graph theoretical measures to both task-based and resting-state fMRI (rs-fMRI) data to examine changes in functional connectivity related to treatment outcomes in persons with aphasia. Graph theory and rs-fMRI are currently underutilized in clinical aphasiology. Graph theory is gaining popularity in other areas of cognitive science as a meaningful technique for the characterization of brain network dynamics. Resting-state fMRI has been gaining popularity in other areas of clinical research as an easily acquired and informative measure of general cognitive functioning. Twenty participants with aphasia (PWA) will be scanned using fMRI at four time points. Between scan 1 and scan 2, no therapy will be provided, establishing a baseline control for each PWA. Between scan 2 and scan 3, ten weeks of word finding therapy will be provided. Between scan 3 and scan 4, no therapy will be provided, allowing for measurement of maintenance effects. The treatment used is abstract word retrieval training. This treatment was chosen because it has been shown to be not only successful at improving abstract word retrieval, but also improving retrieval of related concrete words that are not trained (i.e., generalization) (Kiran, Sandberg, & Abbott, 2009; Sandberg & Kiran, 2014). This makes the treatment more effective and a good candidate for promoting positive neural changes. Furthermore, abstract and concrete words have been shown to have differing neural activation patterns. This allows for the systematic examination of direct training and generalization effects of treatment on brain reorganization. For the graph theory analysis, regions of interest (ROIs) will be defined using a data-driven spatially-constrained approach developed by Drs. Hillary and Molenaar. Once correlations matrices are defined for each time point, graph theoretical measures will be calculated and compared at each time point to systematically examine changes in functional connectivity related to therapy outcomes. The methodological approach in this proposal is innovative in the use of two baseline scans before treatment and two maintenance scans after treatment and the application of graph theory and rs-fMRI to examine treatment-related brain reorganization in aphasia. The successful completion of this project is expected to help inform current theories of optimal reorganization patterns after stroke and provide groundwork for establishing informative and appropriate methods for measuring brain reorganization in aphasia.