Neuroprosthetic devices are artificial computer chips, electrode arrays, biochips and other devices grafted straight onto the brainstem, or onto the surface of the brain itself. Their concept is to augment the function of the brain by adding additional capacity, or bypassing brain damaged areas, allowing patients to make a full recovery.
Neuroprosthetics offer great promise for the disabled, and for those seeking escape alike. Prosthetic limbs and organs connected to the body can be controlled by the brain as if you were born with them - after a little training, without conscious thought.
The ability to completely bypass a paraplegic body, or repair a hippocampus or cerebellum would bring new quality of life to those who have very little.
Conversely, letting the mind interface directly with VR, for the ultimate immersive experience allows a VR based avatar to truly become your sensory form.
These and many other applications are the goal of neuroprosthetics, and great strides have so far been made. However, there are several fundamental problems with such devices, which need to be addressed before they can ever become mainstream:
The implantation of modern neuroprosthetics is best described as dicey. A section of skull is opened up such that the brain is exposed. The neuroprosthetic is attached to the brain in the required area, and a hull drilled in the piece of skull that was removed. Wires are fed through the hole in the skull plate, and it is reattached and stitched into place. On the surface of the brain, the neuroprosthetic's electrodes individually seek out the strongest neuron signals they can, and bind with them.
The wires sticking out of the skull are then connected to a control microprocessor outside the body.
This implantation procedure has several immediate problems:
The Spinal Cord Barrier
Whenever the spinal cord is damaged, or a neurodegenerative disease takes hold, neural circuits between the brain and the body break down, and communication between body and brain fails. This can be a problem, as in order to interact with the body for virtual limbs, a neuroprosthetic must most likely be spliced onto the upper brainstem - part of the spinal cord.
The problem with damage to the spinal cord, is it is such a complex packed mass of spinal ganglion cells, each upwards of a metre long, and packed several thousand deep. Recreating even one of these spinal ganglia has so far been beyond our ability.
For reasons as yet unknown, the signal strength of the neurons the electrode arrays of a neuroprosthetic bond to, attenuates over time. Eventually, this renders the neuroprosthetic device near useless, and an additional surgical procedure is required to re-site the device in an area of increased signal strength.
Neuroprosthetic implants are still too large for the task they perform. At the time of writing, the most powerful, complex neuroprosthetic implant undergoing laboratory trials, fits 12,000 electrodes on a square centimetre implant. In the same area of the brain, there are several million neurons. This places the interface ratio of electrode to neuron at an extremely low value, and hampers the feedback fidelity produced.