Abstract
In blindness, the visual cortex supports novel non-visual functions through mechanisms of cross-modal plasticity. Several fMRI studies suggest that cross-modal activation is related to semantic representation. It is unknown if these cross-modal responses to language processing reflect particular semantic categories. Further, it is unknown the extent to which these responses relate to the structural and metabolic alterations of visual cortex that accompany blindness. We studied congenitally blind subjects and sighted controls using BOLD fMRI while they performed a semantic decision task. During scanning, subjects judged the plausibility of spoken sentences drawn from three semantic categories: visual (‘the door is cracked’); auditory (‘the bee is buzzing’); and tactile (‘the cloth is velvety’). Control stimuli consisted of sentences played backward (lacking semantic content) and white noise. In addition to these functional data, resting ASL-based perfusion and diffusion tensor imaging data were collected. We have obtained preliminary data from 3 blind subjects and 2 controls. All three blind subjects demonstrated activation of occipital cortex, including the calcarine sulcus, for the contrast of forward- and backward-sentences vs. white noise, and further for forward-sentences vs. backward sentences. Visual cortex activity was not seen in control subjects. There was greater activation within visual areas in the left hemisphere compared to the right (% signal change L 1.34±0.64 vs. R 0.78±0.58; t = 3.6, p = 0.07). There were no significant differences in activation between semantic categories in any brain areas in these data. With additional subjects this comparison will be revisited. Also, extent of cross-modal activation will be related to the resting perfusion of occipital gray matter and the fractional anisotropy of occipital white matter.