Cyborg Laboratory Breakthrough
Brain implants are intended to help paraplegics regain independent movement. A new generation of brain-computer interfaces is coming remarkably close to that goal.
Every time I think about it, I get goosebumps: Erik Sorto, who has been completely paralyzed since the age of 21 due to a gunshot wound, is sitting in my laboratory drinking a beer by himself for the first time in ten years. A year earlier, my team and I at the California Institute of Technology in Pasadena had implanted electrodes in his cerebral cortex to register the brain's movement signals and transmit them to an electromechanical arm. Now he grabs the beer bottle and puts it on Sorto's lips. We watch with fascination as our test subject masters this supposedly simple, but in reality highly complex task.
Day after day we move our limbs without thinking about it. But how can a mechanical prosthesis be controlled by the power of thought? There are two main classes of such so-called brain-computer interfaces (BCI for short). In one, a device transmits electrical signals to nerve cells in the brain. The technology forms the basis for numerous implants already used by patients, such as the cochlear implant. The inner ear prosthesis stimulates the auditory nerve of deaf people, allowing them to perceive sounds.
Deep brain stimulation represents another application of the first class of BCIs. Targeted electrical impulses in the basal ganglia can, for example, alleviate the usually very stressful motor disorders of Parkinson's patients. Devices for stimulating the retina for blind people based on a similar principle are currently being clinically tested.
The second group of brain-computer interfaces is still under development. These devices record neural activity from the brain and use it to control prostheses, for example. Electroencephalography (EEG) and functional magnetic resonance imaging (fMRI) are fairly rough measurement methods. The former measures the average brain tissue activity from the skull surface. It only registers the summed up signals from many millions of nerve cells.
fMRI allows a slightly better spatial resolution, but only records brain activity indirectly via the blood flow in the brain tissue. This leads to very strong time delays, which is why the method is not suitable for registering rapid changes. …
