• Abstract and Figures
• Public Full-text

13.7 News Feat Mind Machine MH 11/7/06 9:45 AM Page 125 NATURE|Vol 442|13 July 2006 NEWS FEATURE IN SEARCH OF THE SIXTH SENSE Implants in the brain could one day help paralysed people move robotic arms and legs. But first, scientists need to work out how our brains know where our limbs are, says Alison Abbott. hut your eyes. Now, touch your nose. to work in a much more life-like way. What’s but he can also tune in to the tug of a jacket Chances are you can do this without more, they hope to gain a deeper understand- sleeve to work out the direction his arm is even thinking about it. For this you can ing of how proprioception works, and how moving. Or to the cool air on his armpit when thank your sense of proprioception, they might be able to manipulate it. he raises his arm in a loose shirt. Neuropros- S R. MASSEY/GETTY which is so much a part of us that most of us Some months after his virus attacked, thetic engineers are realizing that many are unaware that it exists. This ‘sixth sense’ lets Waterman, only 19 years old, was lying in bed sensory feedback signals could be similarly our brain know the relative positions in space applying all his mental energy to the fight for harnessed. of different parts of our bodies. Without it, our control of his body. He tensed his stomach A neuroprosthetic is more accurately called brains are lost. muscles, lifted his head and stared down at the a brain–machine interface. Hundreds of elec- Ian Waterman knows how that loss feels. limbs that seemed no longer to belong to him. trodes, fixed into tiny arrays, are placed in or More than 30 years ago he lost this sense He willed himself to sit up. on the surface of the cortex, the thin, folded almost overnight, when a flu-like virus dam- outer surface of the brain that controls com- aged the required sensory nerves. His muscles Concentrated effort plex functions including the organization of worked perfectly, but he could not control Later, he realized that it was the visual feed- movement. The electrodes record the electri- them. “I lost ownership of my body,” he says. back that allowed his body to unexpectedly cal signals from the cortex’s neurons and these He could no longer stand, or even sit up by obey the mental instruction. “But the eupho- are translated by a computer algorithm and himself, and doctors said he would never be ria of the moment made me lose concentration used to drive specific actions — the movement able to do so again. Waterman’s condition arose and I nearly fell out of bed,” he remembers. of a cursor on a computer screen, for example, from a disease called acute sensory neuropathy From then on he learnt to compensate for or of an artificial limb. and is so rare that only a dozen or so similar his deficit in proprioception with other forms In this issue of Nature, two papers1–3 demon- cases are known to the medical literature. of sensory feedback to help him understand strate dramatic progress in the area. A team Some neuroscientists are taking a cue from where his limbs are, and thus control them. It consisting of John Donoghue’s group, based at Waterman’s experiences and starting to inves- requires constant, intense concentration, but Brown University in Rhode Island and Cyber- tigate whether robotic devices controlled by now, despite his profound impairments, he can netics Neurotechnology Systems in Foxbor- thought alone could be integrated with an arti- manage fairly normal movements. Most of the ough, Massachusetts, implanted 96 electrodes ficial sense of proprioception. If so, they rea- input that he relies on is visual — standing up into Matt Nagle’s motor cortex, the brain son, these ‘neuroprosthetics’ could be made with his eyes closed is still nearly impossible — region that processes information about move- 125 © 2006 Nature Publishing Group 13.7 News Feat Mind Machine MH 11/7/06 9:45 AM Page 126 NEWS FEATURE NATURE|Vol 442|13 July 2006 and redirected to the appropriate part of the cortex, or the sensory cortex itself. The ‘how’ refers to the design of the electri- R. FRIEDMAN cal signals to be fed into the cortex. These could mimic the sensory system’s natural nerve impulses, based on parameters such as frequency and amplitude. Or they could involve creating artificial signals that the sen- sory cortex is able to distinguish, in the hope that the brain can be trained to associate par- ticular signals with particular parameters. Once scientists have worked out how best to encode the signals, the idea would be to place sensors on artificial limbs to generate signals representing proprioceptive information such as angle of joint, vibration, force of grip — and other sensory information that Waterman has found helpful, such as temperature. Trained brain Most in the field have a hunch that the signal will not have to mimic neural activity per- fectly. The brain can, after all, cope with the Mind control: Matt Nagle’s neuroprosthetic lets him move a cursor using thought alone. very unphysiological signals generated by the most successful brain–machine interface to ment. Nagle is a quadriplegic patient and the A handful of researchers is starting to try to date: the cochlear transplant. Already, some first human volunteer to reach this advanced work out where and how to stimulate the sen- 110,000 profoundly deaf people have been stage of testing (see picture, above). Hooked up sory nervous system to reproduce the sorts of implanted with the device, according to the US to computers and attended by a team of techni- information that a limb might send to the sen- National Institutes of Health. The implant sits cians, Nagle could move a cursor to issue dif- sory cortex. It is early days: none of their work in the inner ear and interfaces with the audi- ferent instructions — for example, to open is published. And as so little is known about tory nerve. Its signals are totally artificial, and, e-mails or turn down the television. the system, there is no obvious place to start. at first, recipients can make nothing of the Krishna Shenoy’s group at Stanford Univer- Theoretically, the ‘where’ could be the noise. But the auditory cortex, it turns out, is sity, California, has done similar work in a nerves running from the limb into the spinal highly adaptable. With appropriate training, it non-paralysed monkey’s premotor cortex, the cord, or the spinal cord itself (see graphic). Or can quickly learn to associate particular codes area of brain where the animal’s movement- it could be higher — in the brain’s thalamus, with particular sounds, so that transplant related ‘intentions’ are generated. Using a new where incoming sensory signals are integrated recipients can learn to follow conversations algorithm, the team’s brain–computer inter- with ease. face produced results four times faster and HOW SOME OF THE BODY’S SENSORY “When the concept of stimulating the audi- more accurate than previously seen. INPUTS WIRE INTO THE BRAIN tory nerve emerged in the 1970s, people said it would be impossible to generate the right elec- Closing the loop Thalamus trical signal to the brain,” says Shenoy. “But it The two papers show how closely neuropros- turned out that you don’t have to get it perfect, thetics are approaching medical reality. But just close enough for the brain to do its own although moving a computer cursor by fine-tuning.” On the other hand, one does not thought alone may be dazzling, scientists have want to burden patients with having to learn long-term ambitions to make neuroprosthet- too much, says John Chapin a physiologist at ics reproduce more complex functions. Could the State University of New York Health patients direct a robotic arm to pick up a cof- Science Center in Brooklyn, and a pioneer in fee cup, for example? “For this, the devices using neural activity to control robots. “Ideally CLINICAL NEUROANATOMY AND RELATED NEUROSCIENCE AND RELATED NEUROANATOMY CLINICAL need to deliver feedback to the brain — we we should aim to mimic the natural signal as need to close the loop,” says Daofen Chen, closely as possible,” he says (see ‘Voyagers in director of the neural prosthesis programme at Face the cortex’). Somatic the US National Institute of Neurological Dis- sensory For now, whatever works will be good. “We orders and Stroke in Bethesda, Maryland. cortex don’t know if it will turn out to be possible to The brain’s sensory cortex receives signals — incorporate sensory information but we are proprioception, touch, pain and so on — from Arm going to try,” says neuroscientist Andrew the body (see graphic), and in response con- Schwartz of the University of Pittsburgh, an stantly modifies its movement-related com- expert in brain–computer interface technol- mands. The current generation of output-only of information Flow ogy for the control of robotic arm movement. neuroprosthetics are open-loop systems — Schwartz is working with Douglas Weber, with more limitations than Ian Waterman, who Trunk Nerve a bioengineer at Pittsburgh who is developing can at least use visual, temperature and tactile Spinal cord a model for studying sensory input. This feedback. “Brain–machine interfaces will have involves using electrodes to stimulate the FOLAN-CURRAN & J. FITZGERALD T. M. J. FROM: ADAPTED to become interactive,” says Chen. “But now sensory nerves from the limbs of an anaes- that we would like to exploit it, we realize we thetized cat at the point just before they enter Leg know next to nothing about sensory input.” the spinal cord, and simultaneously recording 126 © 2006 Nature Publishing Group 13.7 News Feat Mind Machine MH 11/7/06 9:45 AM Page 127 NATURE|Vol 442|13 July 2006 NEWS FEATURE Voyagers in the cortex If paralysed people can now use studies may have failed because rising and falling tones whether animal experiments on patients electronic implants to move a the cortex is too complicated to they are presented as sound, willing to take part.

Load more

  3

Copy Link