The central nervous system is remarkably adaptive to changes in sensory input. Cortical plasticity following visual deprivation was first observed in kittens (Wiesel and Hubel, 1965a
)) and extensive reorganizations of cortical representational maps have been observed in humans after limb amputation (Flor et al., 1995
). This is manifested also in cases of increased sensory use in normal subjects such as musicians, who exhibit increased auditory cortical representations (Pantev et al., 1998
), and in professional string players, the cortical representation of the fingers of the left hand is greatly increased (Elbert et al., 1995
). Such findings illustrate how the brain may re-organize functionally, to adapt to changes in sensory demands.
Blindness provides a rare model of crossmodal neuroplasticity. Cortical structures normally specialized for visual processing may be used by blind individuals for auditory and tactile purposes (see Theoret et al., 2004
, for a review). Extensive neuroanatomical differences – even outside the occipital cortex – have been observed in those blind from an early age (early-onset blind or EB individuals). Elbert and colleagues (2002)
found that the area of the tonotopic region of the auditory cortex is almost twice the size of its counterpart in sighted individuals. Hippocampal volumes are abnormally enlarged in both EB and LB (late-onset blind) individuals, perhaps offering an anatomical substrate for their enhanced navigational skills (Fortin et al., 2007). White matter connectivity between primary somatosensory and visual areas is also increased in EB individuals; Wittenberg et al. (2004)
applied repetitive transcranial magnetic simulation (rTMS) over the primary somatosensory cortex and observed significant occipital cortex activity using positron emission tomography. This effect was noted only in EB, but not in sighted or LB individuals. Using voxel-based morphometry to measure anatomical integrity more directly, Noppeney and colleagues (2005)
found that EB individuals had decreased white matter volume in the optic radiation and sensorimotor system, and reduced gray and white matter volume in primary visual areas. The optic radiation showed no detectable deficit in a small group of LB individuals assessed with diffusion tensor imaging (Schoth et al., 2006
). Using diffusion tensor tractography (DTT), Shimony et al. (2006)
found atrophied geniculocalcarine tracts in the EB, while connections between visual cortex and the orbital frontal and temporal cortices were relatively preserved. Yu et al. (2007)
also observed increased fractional anisotropy of the corticospinal tract in early-blind men using DTT, perhaps reflecting increased myelination.
Here we examined whole-brain volumetric changes in both EB and LB compared to sighted individuals using tensor-based morphometry (TBM) with fast fluid registration. We aimed to create a 3D map of the level of voxel-wise volumetric gains and losses in EB and LB subjects. The purpose of the current study was two-fold. First, to explore, in blind individuals, new cerebral regions that have not specifically been examined in previous studies, as well as to confirm differences found by previous studies using other methods. And secondly, to offer a straightforward comparison between EB and LB individuals, which is often lacking in studies of cross-modal plasticity. We hypothesized that there would be deficits in primary visual cortices and the occipital lobes generally, but we predicted hypertrophy in extra-occipital brain regions, and in the corresponding callosal sectors carrying interhemispheric fibers, due to compensation from other senses. We also hypothesized that the pattern of gains and losses might be accentuated in the early-blind, as plasticity is expected to be greater while the occipital lobes are still rapidly developing, shedding light on the time-courses and possible substrates of the changes (e.g., late myelination).