The Effects of Visual Deprivation: Implications for Sensory Prostheses - Artificial Sight: Basic Research, Biomedical Engineering, and Clinical Advances - page 48

Classical studies by Recanzone et al. have demonstrated that removal of a finger
leads to striking reorganization within primary somatosensory cortex [1, 2].
Similarly, extensive practice on an auditory discrimination task results in an
expansion of the region of primary auditory cortex responsive to the trained
frequency [3]. The remarkable plasticity of auditory cortex is thought to be a
major contributing factor to the success of cochlear implants: language compre-
hension tends to be quite poor immediately after implantation, and then improves
over many months [4].
In the field of visual prostheses the hope is, of course, that visual cortex
will show a similar capacity for adult plasticity. However, there is a signif-
icant amount of data suggesting that early visual cortex may show much less
adult plasticity than early auditory or somatosensory areas. It is surprisingly
difficult to find perceptual learning effects that can be definitively attributed to
primary visual areas. While some electrophysiology studies in monkeys have
found neural changes within primary visual area V1, these tend to be both
restricted in magnitude and task dependent [5, 6]. In humans, V1 changes in
responsivity have been found as a result of training in an orientation discrim-
ination task using functional magnetic resonance imaging (fMRI), but these
changes may be partially mediated by attentional feedback from higher visual
areas [7].
One type of cortical plasticity that has been extensively studied is the “filling
in” of retinal lesions (a process in which cortical neurons that subserve a retinal
scotoma begin to respond to nearby intact retina). While rapid filling in has
been found in cat [8, 9], reorganization in response to scotomas seems to occur
extremely slowly in both the monkey and the humans. A recent study in macaque
found no evidence for remapping of regions of cortex that represented a retinal
scotoma over several months [10]. However, in the case of human patients, there
is one study examining cortical mapping in patients with well-established (several
years) foveal scotomas due to macular degeneration which does demonstrate
remapping [11]. Perhaps if adult retinotopic reorganization occurs as a result of
restricted visual loss, it happens extremely slowly.
One possible explanation for why early visual areas show so little plasticity
is that visual cortex is already performing a demanding host of visual functions.
Reorganization of visual cortex in response to auditory deprivation could poten-
tially undermine the ability of visual cortex to perform basic visual functions.
As described below, the changes that occur in somatosensory cortex consequent
on learning to read Braille with three fingers result in worse performance in
discriminating which finger had been stimulated. In the case of early visual
cortex, the need to perform normal visual functions may limit the scope for
changes consequent on the demands of deafness.
By contrast, one might expect more substantial reorganization in higher-level
visual areas that tend to be more experience dependent. Neuronal changes as a
function of learning do seem to be larger in higher visual areas [12], though no
direct electrophysiological comparisons between low- and high-level plasticity
have been made to date. One reason for assuming that higher levels of cortex are