Effects of deafness and sign language fluency on white matter connectivity
Stephen McCullough1, Karen Emmorey1; 1SDSU
Previous literature has shown that auditory deprivation and sign language exposure alter the gray and white matter structure of the human brain; however, not much is known about the changes in the white matter (WM) connectivity between brain areas in deaf signers. Using diffusion tensor imaging (DTI) and quantitative anisotropy (QA) based connectometry (Yeh, 2016), we investigated the effects of deafness and fluency in American Sign Language (ASL) on the macroscale connectome of the human brain. A total of 27 participants, 14 deaf signers (8 female, mean age = 29.4, SD = 4.3) and 13 hearing controls (7 female, mean age = 25.3, SD = 5.8), underwent MRI. We used the extended (35 sentence) version of the ASL Sentence Reproduction Test (ASL-SRT; Supalla et al., 2014) to measure ASL fluency. The preprocessing and analyses of structural and diffusion MRI data were carried out using AFNI and DSI Studio. Diffusion MRI data for each participant was acquired using 25 diffusion sampling directions with 2 mm in-plane resolution and 3 mm slice thickness. The diffusion data were reconstructed in the MNI space using q-space diffeomorphic reconstruction to obtain the spin distribution function (Yeh et al., 2010; 2011). A T-score threshold of 3.0 and tract length threshold of 20 mm were used in the deterministic fiber tracking algorithm (Yeh., 2013) to find tracts. To estimate the false discovery rate (FDR), 4000 randomized permutations were applied to the data to obtain the null distribution of the track length. To examine the effect of deafness on WM connectivity, we performed a regression analysis on each groups’ normalized quantitative anisotropy (QA) values with sex and age variables partialled out. To determine the effect of ASL fluency on the WM connectivity in deaf signers, we calculated Pearson's correlation between normalized (QA) values with z-transformed ASL-SRT scores while partialling out the effects of age and sex. Group connectometry analysis results (thresholded at FDR ≤ .01) revealed increased QA in left and right inferior fronto-occipital fasciculus (IFOF), left inferior longitudinal fasciculus (IFL), left uncinate fasciculus, right corticospinal tract, corpus callosum forceps major, and right medial lemniscus for the deaf group. No tracts with increased QA were found for the hearing group. Higher QA for the deaf group in the IFOF, left IFL, and right medial lemniscus suggests more robust connections between visual areas, proprioceptive sensory, and language areas for the deaf group, possibly due to neural changes arising from congenital deafness and/or to the visuo-spatial and motoric processing demands of ASL. Correlational tractography results of sign language fluency and QA showed positive correlations (F ≤ .001) for the IFOF bilaterally, right medial lemniscus, right IFL, right corticospinal tract, and left corticospinal tract. In addition, the results showed negative correlations (F ≤ .001) for corpus callosum forceps minor, corpus callosum tapetum, corpus callosum forceps major. The tracts showing positive QA correlation with the ASL fluency, again, suggest increased connectivity changes specific to visuo-spatial language processing. Moreover, negative QA correlations suggest increased connectivity between hemispheres for less fluent ASL signers.
Topic Areas: Signed Language and Gesture, Development