Presentation
Search Abstracts | Symposia | Slide Sessions | Poster Sessions | Poster Slams
No change in lateralization of language activity or connectivity in older adults
Poster E28 in Poster Session E, Saturday, October 8, 3:15 - 5:00 pm EDT, Millennium Hall
Christopher Grisham1, Andrew T. DeMarco1, Sachi Paul1, Elizabeth Dvorak1, Kelly C. Martin1, Peter E. Turkeltaub1,2; 1Center for Brain Plasticity and Recovery, Georgetown University, 2MedStar National Rehabilitation Hospital
Language processing in a large majority of adults is functionally dominated by left-hemisphere regions. The hemispheric asymmetry reduction in older adults (HAROLD) model states brain activity tends to be less lateralized in older adults compared to younger adults in a variety of cognitive functions (Cabeza, 2002). Although this model predicts that language lateralization might become reduced in older age, this hasn't been well tested. A confound in comparing fMRI activation across the lifespan is that younger versus older adults may exert disparate effort during the task. Here, we test the HAROLD model with respect to language processing. Because much of HAROLD findings come from task-related fMRI, we begin by examining task-activation in a language task that adapts for difficulty. Because language ability relies on interactions between gray matter regions, we then test for age-related changes in lateralization of both structural and functional connections between the areas activated by the fMRI task. Participants included 61 neurotypical adults (27 male, 34 female; ranged 31-83 years with mean age 61). Each individual completed an adaptive fMRI semantic decision paradigm to elicit language-related activation, a movie-watching resting state scan for functional connectivity, and a diffusion-weighted scan for structural connectivity. We then intersected regions activated in >60% of participants with a structural atlas, and identified contralateral homotopes to each atlas region to generate 23 left-right hemisphere paired nodes from which we generated structural and functional connectomes. For the three imaging metrics, we used Spearman correlations to examine the relationship between participant age and laterality index (LI) at edge-, node-, lobe-, and hemisphere-wise levels. We found no significant age-related changes to activation LI at either node-, lobe-, or hemisphere-wise levels. The functional connectome revealed weakly negative correlations with age throughout the whole network. However, functional connectivity LI showed no statistically significant changes with age. Interrogating our structural connectomes, we saw inter-hemispheric edges whose connectivity significantly declined with age. The nodes whose connections produced statistically significant decline include the left and right temporoparietal junctions, the left dorsomedial prefrontal cortex, the right medial frontal gyrus, and the right angular gyrus. However, we also observed no statistically significant changes to structural connectivity LI with age. Our analyses revealed no age-related changes in lateralization of language across any of our metrics, evaluated at either large- or small-scales. Declines in inter-hemispheric structural connectivity are consistent with previous studies demonstrating white matter structural degeneration with age. Our failure to find changes with LI over age for all of our metrics suggests that the HAROLD hypothesis may not be broadly applicable to all kinds of language processing demands. In prior studies that did not account for task performance confounds, reduced lateralization with advancing age may have related to increased effort required for fMRI task performance in older adults, rather than a fundamental change in the organization of brain networks with increasing age. Our results add to the body of literature evaluating changes in language processing with increasing age and underscore the importance of further studies to better understand the picture of neurocognitive aging.
Topic Areas: Language Production, Development