One tack being pursued by neuroscientists and engineers is the development of “smart” neural prostheses. These devices are intended to restore function, through electrical stimulation, to damaged motor neural circuits – the long, slender fibers that conduct neurochemical messages between nerve cells in the brain and spinal cord.
It is the rapid-fire transmission of messages between nerve cells that prompts the body’s movements, leading the hand to whisk away a fly, the leg to stretch, the head to turn. And it is disruption of these messages that leads to impairment, including paralysis, staggered gaits and other forms of motor dysfunction.
Simple forms of neural prostheses — some external, some implantable — have been developed over the last four decades to treat loss of hearing, bladder control and respiration. And recent advances have led to the development of some “smart” neural prostheses, which engage higher levels of brain function.
However, significant challenges remain in developing ever-more precise implanted neural interfaces that operate at the cellular level and that will provide even greater precision and fidelity in restoring function.
Harnessing the brain’s “plasticity”
To truly harness the capacity of neural prostheses to treat complex damage of the nervous system, the devices must be designed to exploit the brain’s “plasticity,” or capacity for change, says Michael Merzenich, PhD, UCSF Francis A. Sooy Professor of Otolaryngology and a member of the Keck Center for Integrative Neuroscience at UCSF.
Merzenich’s pioneering studies over three decades have revealed the capacity of the brain to rewire itself in response to new conditions, even during adulthood and aging. And in developing the first neural prosthesis – the cochlear implant, in the early 1980s — and software programs for language and learning disabilities in the mid 1990s – he has demonstrated that the brain has the capacity to actively engage in a remediation, or retraining, process.
“The brain is amazingly adaptive,” says Merzenich. “Our early studies developing the cochlear implant showed that the brain can take crude electrical inputs and interpret them and create new constructs,” he says.
“But our studies showed that the brain wants to receive this information in certain forms. Information delivered from the interface of a device has to be adequate for the brain to extract enough information to reestablish control.”
As neural prosthetics involve extracting neurological information from the higher levels of the brain, and transmitting it back to a critical nerve center in an unfamiliar form, he says, they should engage the brain in this process.
”The success with any complicated prosthetic device relates as much to how the brain adjusts to it, accepts it and controls its use as it does to the device itself. If we can figure out how to engage the brain to do its part it can make a merely adequate neural prosthetic device work marvelously.”
Merzenich will present a talk, “The role of plasticity in the nervous system in neural prosthetics,” at the AAAS symposium “Smart prosthetics: Interfaces to the nervous system help restore independence” (8:30-11:30 a.m., Friday, Feb. 16, 2007).
Neural prostheses can be “smart” in various ways, says Merzenich. They can:
be smart in and of themselves, by operating “intelligently” adapt to the brain tissue environment in which they are introduced be designed to grow in their utility as the brain is trained to take advantage of them
In all cases, he says, devices should be organized to engage the brain in ways that “enable plasticity and promote plasticity,” such as by:
delivering plasticity-enabling chemicals providing a body/brain/device interface that maximizes the potential for plastic adaptation applying stimuli in forms that effectively induce plastic change enabling the implementation of an intensive training program that makes the most out of the device
Alternative forms of plasticity-based training
Notably, Merzenich’s own current research focuses not on developing neural prosthetics, but rather on developing intensive plasticity-based mental and physical training programs. His targets are schizophrenia, bipolar disorder, functional losses in normal aging, mild cognitive impairment, Alzheimer’s disease, acquired movement disorders, autism, and learning, language and reading impairments in children.
“We are trying to see how far we can drive the brain in corrective directions by intensive training without a device,” he says.
In these cases, the neural circuits at play are those that receive sensory inputs – smell, touch, taste, sound and sight – support memory and cognition, and orchestrate behaviors.
Merzenich’s ongoing studies involving the use of software to accelerate the speed at which children with language and learning disabilities process sound suggest he’s on track. (His patented findings led to his founding in 1996, with Paula Tallal of Rutgers University, Scientific Learning, a therapeutic software company in Oakland, California.)
And numerous clinical trials targeting the other neurological conditions are producing encouraging results. A clinical trial for schizophrenia, underway at UCSF and Yale, aims to drive misdirected neural circuitry in a normal direction, though cognitive therapy, perceptual training, movement control, response control.
The results of this trial (supported by a second therapeutic software company that he has co-founded, Posit Science, in San Francisco) are “outstanding,” he says, far better than those produced by the standard medication for the disease, but at this early stage in the development of the strategy the regimen requires a burdensome 100 hours of work.
Other clinical trials under way at UCSF involve normal and infirm aging populations, including mild cognitive impairment and Alzheimer’s patients.
The studies on autism are the least developed, he says. “We’ve trained thousands of autistics with our child training programs, but our training tools and their effective applications are still very limited. We know that we can provide much better help for these individuals.”
Merzenich is not currently collaborating with neural stem cell scientists, but he talks with them, and thinks about their work. With the establishment of new neurons in the brain, he says, “brain plasticity will have to be a substantial and necessary part of recovery.”
“These are interesting stories,” he reflects. “They do not involve trying to substitute, compensate or work around a problem. In each case, the work involves trying to correct the processing in the machinery with the machinery being substantially intact.”