Presentation
Search Abstracts | Symposia | Slide Sessions | Poster Sessions | Poster Slams
Left prefrontal regions mediate the influence of executive functioning on sentence processing in primary progressive aphasia
Poster B22 in Poster Session B and Reception, Thursday, October 6, 6:30 - 8:30 pm EDT, Millennium Hall
Andrea Gajardo-Vidal1,2, Diego L. Lorca-Puls1, Maria Luisa Mandelli1, Abigail E. Licata1, Buddhika Ratnasiri1, Rian Bogley1, Zachary Miller1, Bruce Miller1, Maya Henry3, Jessica de Leon1, Maria Luisa Gorno-Tempini1; 1Memory and Aging Center, Department of Neurology, University of California, San Francisco, 2Faculty of Health Sciences, Universidad del Desarrollo, Chile, 3Department of Speech, Language, and Hearing Sciences, University of Texas, Austin
Background: Previous neuropsychological studies have shown that executive functioning (EF) contributes to performance on language tasks that pose high cognitive demand. Furthermore, functional neuroimaging studies have provided neurobiological evidence that language tasks involve an interaction between language-related and domain-general executive control regions. Finally, it has been suggested that enhanced activity (or increased connectivity) in executive control networks may support language recovery in stroke patients. In the current study, we sought to investigate cross-sectionally and longitudinally the relationship between EF and language, and the neural substrates of that relationship (if any), in patients with primary progressive aphasia (a neurodegenerative syndrome). Methods: A total of 197 patients with a diagnosis of primary progressive aphasia (PPA) were selected from the UCSF Memory and Aging Center database. We performed four analyses involving behavioral and imaging data. (1) A principal component analysis across all EF tasks with no or minimal verbal output yielded one orthogonal component. Based on this component, we created a composite score of EF tasks, and run a series of correlational analyses between this composite score and representative language tasks. (2) A whole-brain voxel-wise analysis using grey matter images was implemented in the Mediation Toolbox to identify which brain regions mediate the relationship between EF and language. (3) A resting-state fMRI (rs-fMRI) analysis of data from 132 healthy controls was carried out to examine the functional connectivity of the regions identified in our mediation analysis. (4) A mixed ANOVA of longitudinal data from a subset of 83 PPA patients who were tested at two different time points was conducted to evaluate the interaction between EF performance and language decline. Results: (1) Across all language tasks, the sentence comprehension (SentComp, rho = 0.474, p<0.001) and the WAB spontaneous speech total score (SentProd, rho = 0.369, p<0.001) were the most strongly correlated tasks with our EF composite score compared to, for example, object naming (rho = -0.147, p = 0.041) or word reading (rho = 0.181, p = 0.031). (2) Two clusters located within the left middle frontal gyrus (LMFG) and left superior frontal gyrus (LSFG) mediated the relationship between: (i) EF and SentComp and (ii) EF and SentProd. (3) These LMFG and LSFG regions were part of the same fronto-parietal network that previous studies have associated with executive control and shown to be critical for more demanding tasks. (4) The SentComp and SentProd scores of the patients with better EF compared to those with worse EF were significantly better across the two time points and declined significantly less over time, even after controlling for potential confounds such as severity and education. Conclusions: Our results suggest that: (1) EF supports sentence processing in PPA patients; (2) two left prefrontal regions (LMFG and LSFG) mediate the relationship between EF and sentence processing; (3) these regions are part of the fronto-parietal executive control network rather than the classic perisylvian language network; and (4) EF might potentially play a “compensatory” role as language declines.
Topic Areas: Disorders: Acquired, Control, Selection, and Executive Processes