Restoration of vision, using a neuroprosthesis, depends upon providing the cortex with a well-controlled
temporospatial electrical stimulation pattern that mimics the pattern of neural activity normally associated
with vision, or uses the natural tuning properties of the visual system, to provide the cortex with meaningful
sensory input. Restoring vision presents a technical challenge because it is likely that a large number, at
least hundreds, of parallel channels of stimulation are required. This project seeks to understand the stimulation
of the primary visual cortex (area V1) using a large number of intracortical microelectrodes.
Attempts to stimulate vision by electrically stimulating the cerebral cortex go
back at least to 1918, when Lowenstein and Borchardt accidentally stimulated the
striate (visual) cortex of a man undergoing surgery for a bullet wound to the
head. The patient reported a twinkling sensation. Subsequent experiments
throughout the century confirmed this finding and showed that the stimulated
perceptual flashes, now called phosphenes, always reoccur in the same part of
the visual field as long as the same part of the cortex is stimulated. This
mapping, called retinotopy, was one of the first major observations concerning
the functionality of striate cortex; still, and perhaps unfortunately, it
remains the most influential idea in current thinking about the design and
implementation of visual prostheses.
Figure 1 shows the concept. Electrodes implanted in the visual cortex connect to a
fully implanted stimulator/receiver module. The stimulator is powered and
controlled by a wireless link. It is assumed that a camera/computer system
will register an image, digitize it, and transfer it to the cortex, bit by bit,
in order to create a perception of the image. Each point in the image is thus
represented in the cortex by means of an electrode that stimulates the
appropriate location, creating a perceptual spot of light known as a Phosphene.
That this approach could work is suggested by the retinotopy of striate cortex
and by a number of human and animal experiments demonstrating the feasibility of
chronic electrode implants. However, this “bitmap” approach may have serious
pitfalls.
Figure 1. Idealized prosthesis design. It is a common misconception that the brain works with "bitmapped" images, like computer screen. It represents such things as orientation, textures and color instead.
Figure 2 depicts the mental image we might induce using the
bitmap method to represent a face. It is a common misconception that the mental
image would look something like panel B; that is, high resolution (1000 pixels)
and an even gray distribution. Our current best would probably look something
like panel C, which uses black and white—reflecting the fact that we only know
how to turn neurons on and off—and an optimistic resolution of 256
regularly-spaced and constant-sized pixels.
A B C
Figure 2. Construction of a mental image. The real image is on the left. The middle Panel shows what the mental image might under extremely optimistic conditions; the right image is what we might be able to achieve in near future using only bitmap approach.
In
trying to achieve panel B, or ultimately, panel A, the problem is not just that
it requires resolution far beyond our current technical capability to implant
large numbers of electrodes. It is likely that there is a deeper flaw; simply,
that the visual system does not work this way. Our visual cortex does not
compute images in terms of spots, or pixels; rather, images are represented in
terms of edges, textures, colors, depths, and motion. Attempting to stimulate
vision through bitmapping is akin to playing the piano with one’s elbows; it
simply is the wrong interface. No one has ever found a striate neuron that
encodes simple spots. On the other hand, most striate neurons are sensitive to
orientation, spatial frequency (texture), binocular disparity (depth), color,
movement direction, and speed. These are the dimensions of vision, so it is
logical that a visual prosthesis should introduce these types of information
into the brain. This is the goal of our Intra-cortical visual prosthesis
project: to understand and optimize how an array of electrodes can communicate
visual information to the brain.
Inspiring experiments by Brindley, and others, in the 1970’s
studied the effects of visual cortical stimulation with relatively large
electrodes placed on the pia-arachnoid surface, and resulted in multiple
non-contiguous phosphenes, with uncomfortably high stimulus currents, that
produced occasional elicitation of pain due to meningeal or scalp
stimulation. By implanting microelectrodes within the visual cortex, with exposed tip sizes of the same order of magnitude
as the neurons to be excited, much more selective stimulation, at lower stimulus currents, can, in principle,
be achieved, resulting in more precise control of neuronal function. Studies of intracortical stimulation were
initiated at Huntington Medical Research Institute (HMRI), in 1979, in which the feasibility of safe, chronic,
intracortical stimulation of the cortex was established. Starting in 1983, work by
Brummer, and subsequent work
by Robblee, Rose, Cogan, and others at EIC Laboratories eventually resulted in microelectrodes made from activated
iridium. Implantation of 38 micro-electrodes in a human volunteer, at NIH, in 1994, provided the motivation for
the development of a fully implantable 1024-channel transcutaneous cortical stimulation system by the
Illinois Institute of Technology (IIT).
