Presentation

Search Abstracts | Symposia | Slide Sessions | Poster Sessions | Poster Slams

Concurrent Transcranial Direct Current Stimulation and task-fMRI to Discern Auditory Lexical Decision Processing in Chronic Aphasics

Poster B54 in Poster Session B and Reception, Thursday, October 6, 6:30 - 8:30 pm EDT, Millennium Hall
Also presenting in Poster Slam B, Thursday, October 6, 6:15 - 6:30 pm EDT, Regency Ballroom

Serena E. Song^1,2, Lisa C. Krishnamurthy^1,3, Simone Roberts^1,4, Amy D. Rodriguez^1,5, Bruce A. Crosson^1,5, Marcus Meinzer⁶, Venkatagiri Krishnamurthy^1,5,7; ¹Center for Visual and Neurocognitive Rehabilitation, Atlanta VA-Health Care System, ²Neuroscience and Behavioral Biology, Emory University, ³Physics and Astronomy, Georgia State University, ⁴Psychology, Georgia State University, ⁵Neurology, Emory University, ⁶Neurology, University of Greifswald, ⁷Medicine, Emory University

Transcranial direct current stimulation (tDCS) is a form of non-invasive brain stimulation used to improve language functions after stroke. Despite the promising effects of tDCS in stroke rehabilitation, it remains unclear how tDCS modulates brain systems during language tasks. This study aims to characterize the acute tDCS effects on language-related neural activity using a unique paradigm involving concurrent in-scanner tDCS and a lexical decision (LD) task-fMRI on chronic persons with aphasia (PWA). The tDCS influence on language-network brain activity and behavioral outcomes is assessed by comparing active and placebo sham-tDCS. This pilot study recruited three male English-speaking right-handed PWA (~23±14 months post-ischemic stroke, ~60±5 years old) with left hemisphere lesions including the inferior frontal gyrus (IFG) for subject-1 (S1), the precentral gyrus for subject-2 (S2), and the superior temporal gyrus for subject-3 (S3). All participants underwent an auditory LD task involving word and nonword recognition during two blinded sessions of either sham or active-tDCS. The tDCS electrodes were bilaterally placed on the IFG (F7/F8, 1mA for 20min). Functional images were processed independently for active and sham conditions to obtain a task-specific hemodynamic response function that uniquely characterizes LD processing during active and sham stimulation. General linear model tests contrasting active versus sham conditions were thresholded by a T-statistic of 3.296 (p<0.001, FWE cluster-wise corrected range of 2.4-4 cc), and balanced for sensitivity across sessions for each participant. The accuracy and reaction time (RT) of in-scanner LD behavioral performance was quantified by calculating the percent difference between active and sham sessions. Results showed that active-tDCS increased activation of the right inferior temporal gyrus (R-ITG) for S1 resulted in a 58.0% increase in accuracy and 19.3% faster RT for real word recognition compared to the sham condition. Active-tDCS also increased activation of the left angular gyrus (L-AG) during nonword recognition. Contrastingly, active-tDCS did not facilitate language performance for S2. Instead, sham-tDCS showed greater activation of the right postcentral gyrus resulting in an 11.5% increase in accuracy and 3.4% faster RT for word recognition. Lastly, active-tDCS increased activation of the left supramarginal gyrus for S3 without facilitating language performance. Instead, sham-tDCS resulted in a 20.8% increase in word recognition accuracy and a 21.7% increase in accuracy and 7.8% faster RT for nonword recognition. Collectively, the results suggest that immediate facilitation of language function occurred when active-tDCS was applied directly over the lesional/peri-lesional cortices as observed in S1. This direct stimulation facilitated the activation of language areas such as the R-ITG and L-AG. Stimulation of intact bilateral IFG, such as in S2 and S3, elevated neural recruitment of complementary task-related functions involving motor control and semantic decoding without facilitating improved lexical decision making. Our preliminary neuroimaging results and behavioral analysis demonstrate that tDCS-induced neural recruitment is influenced by lesion location and that lesion-specific stimulation most effectively recruits language areas. This provides support for individualized treatment planning for optimal tDCS rehabilitation in PWA. Future work will involve analyzing resting-state functional networks for LD processing and quantifying the hemodynamic parameters for a more comprehensive understanding.

Topic Areas: Methods, Disorders: Acquired