The Man-Machine Interface

Of all types of diseases and injuries, complications due to nerve damage tend to be the most terrifying. While we have potential ways to overcome cancer or heart disease, medicine is helpless in the face of neural injury, which is often permanent and debilitating. Whether it is natural damage to neurons from a disease such as Multiple Sclerosis, a spinal injury that causes paralysis, or an amputation of a limb, the common denominator that prevents treatment is that science has yet to find ways to regenerate lost neurons.

Although some cell types, such as skin cells or liver cells, can easily replicate and heal, neurons are so sophisticated and reliant on an organizational schema carefully set in place by the brain during development that they do not heal so easily, and may even be prevented from doing so by surrounding glial cells to prevent further harm, forming a neural “scar” of tissue and chemical factors that inhibit growth [1,2]. And unlike some species of salamanders or starfish, humans obviously do not have the stem cell capacity to regenerate lost limbs. In fact, some of the most promising theoretical methods for regenerating neurons comes from utilizing neuronal stem cells, although research in this field has only just begun, and other neuroscientists are researching methods for reducing the chemical “scar” in the hopes that neurons will naturally heal if unfettered by inhibitory signals [3].

So while we wait for stem cell and neuroscience research to advance to its full potential, what can we do to aid those who have lost limbs or have incurred spinal paralysis? The answer lies in the realm of neuroprosthetics, and the idea that we can use machinery and computers to restore limb functionality without actually healing the nervous system. And although this seems to be an idea taken out of science fiction, the last fifteen years of research in the field shows great strides have been taken in prosthesis technology.

Research has attempted to entirely bypass the peripheral nervous system of the body, which controls limb movement and is responsible for sending sensations of pain and touch to the brain, instead focusing on understanding how the brain controls these functions directly. Specifically, researchers look to the motor cortex of the brain, the region that sends signals to limbs to move, and examine it using electronic microelectrode arrays along the surface of the head or implanted beneath the skulls of both animals and humans. Because most movements generate a unique pattern of neuron activation in the brain, the microelectrodes can detect what these patterns are, and rapid analysis of brain patterns by computers in real-time when a subject is attempting to move can be used to program robotic limbs to mimic the intended actions [4]. Although these electrode arrays merely capture the electrical activity of a miniscule subset of the brain’s cells, they seem to provide more than enough information for computers to decode simple tasks from electrical data.

In this way, scientists have trained rats to push a lever with a robotic arm, and monkeys have been able to move a computer cursor, both using only their brains [5,6]. But this control schema, aptly known as a brain-machine interface, has recently been applied to more than just animals. Patients with tetraplegia, or full-body paralysis, have also been able to move computer cursors, control television sets, and even move prosthetic fingers after being implanted with a small microelectrode array next to their motor cortices [7]. One tetraplegic woman, who has had the electrode implant for over five years, was even able to control an entire robotic arm to sip from a bottle of coffee lifted from a table to her mouth.

Modern prosthetics take advantage of these tremendous advances in technology, both in terms of neural control and in actual robotics hardware, with some prostheses being articulated at over two dozen joints and have individually functioning fingers [8]. However, the next roadblock in achieving fluidity of movement closer to that of a true hand lies in the problem of sensation. No matter how articulate prostheses are constructed and how fine-tuned scientists can make the neural control, without the feeling of touch, users will have difficulty with the most basic of tasks. Understandably, it is difficult to grasp objects without knowing how hard the fingers are pressing, and this is a problem many users of prosthetics have reported encountering; whether it means breaking dishes because the hand grips too much, or dropping things meant to be picked up, there is a level of control not easily achievable without a sense of touch [7].


Although normal prosthetics attempt to give some measure of sensation, through special harnesses that squeeze other sections of the live body to denote resistance in the hands and feet, these feelings are unnatural and are often the ultimate cause for rejection of the prosthesis altogether [8]. Therefore, the next step in prosthetics research is in giving back sensation of touch to the user through the limbs, and there are a variety of ways proposed to do so. In the brain-machine interface mentioned above, electrodes are implanted next to the brain that are able to read brain signals; however, what would happen if a computer was able to do the reverse, to stimulate the region of the brain associated with feelings of touch?

One line of research, termed by the researchers at Duke University as the “brain-machine-brain interface,” aims to do just that, using a second set of electrodes to provide electrical shocks to the brain’s sensory cortex in response to activity by the motor cortex. When performed in monkeys, the animals’ sensory cortices were given varying levels of stimulation when in contact with objects on a computer to simulate different sensations of virtual “touch,” and were even able to tell identical objects on a computer simulation apart by the apparently different “feelings” associated with the shocks. The ability to distinguish the sensation of different virtual surfaces in monkeys gives hope that a similar method could be used to improve haptic feedback in human prosthetics [9].

In a less invasive method, scientists have come up with a more obvious solution: to route neuronal signals to the brain using the remains of the severed nerves still present in the arms, just as real arms would. Researchers at the University of Utah have attempted to thread stimulating electrodes onto the nerves that carry information about spatial location, or proprioception, as well as touch, in the remaining arms of patients. Computers were then programmed to stimulate nerves with precise shocks based on the location of the limb and proportional to the level of pressure in the fingers, just as nerves activated in a live forearm would to the nerves in the upper arms. The study subject was able to tell the difference between materials such as rubber or foam at a higher chance than random, hinting at the method’s effectiveness [10,11].

And in a perhaps more complicated approach, nerves can be rerouted to other sections of the body, where they can innervate the surrounding tissue and be useful for sensation again. When one patient underwent this “targeted innervation approach” at the Rehabilitation Institute of Chicago, nerves from his arms were moved to his chest. When he thought about moving his hands, instruments in the machines detected signals in his chest and he was able to easily control a prosthetic arm. However, in an unintended side effect, the sensory neurons from his arms were transferred to the chest as well, and so he began to feel areas of his chest as if they were his hands and arms. Scientists aim to take advantage of this phenomenon to use the patient’s own nerves to feel again, albeit in different portions of the body, and develop prosthetics that stimulate these new regions for a more authentic feeling of touch [12].

Although science has accomplished much in the field of neuroprosthetics, there is so much more that must be done to develop prostheses that can truly replace a living limb. The science behind haptic sensation is still in its early stages, and there is no telling if direct brain stimulation is causing feelings of pain or actual sensations of touch, or if electrode stimulation of nerves can result in nerve damage after prolonged use. And even if a more functional prosthesis is to be developed, in order to be implemented, we need lighter materials, better articulation, a long-lasting power supply; the list goes on and on.

However, in an age before we have decoded how neural damage works and stem cells can cure humanity’s wounds, neuroprosthetics research is the best hope we have for the many people suffering from paralysis or loss of limbs.


  1. Aguayo, Albert. Nerve regeneration revisited. Nature Reviews Neuroscience. 7, 601 (August 2006)
  2. Filbin, Marie T. Myelin-associated inhibitors of axonal regeneration in the adult mammalian CNS. Nature Reviews Neuroscience. 4, 703-713 (September 2003).
  3. Iwanami A et al. Transplantation of human neural stem cells for spinal cord injury in primates. J. Neurosci. Res., 80, 182-190. 2005.
  4. Yanagisawa T et al. Real-time control of a prosthetic hand using human electrocorticography signals. J Neurosurg. 114, 1715–1722, 2011.
  5. Chapin JK et al. Real-time control of a robot arm using simultaneously recorded neurons in the motor cortex. Nat Neurosci. 2(7):664-70. July 1999.
  6. Serruya MD et al. Instant neural control of a movement signal. Nature. 416(6877):141-2. March 14, 2002.
  7. Hochberg LR et al. Reach and grasp by people with tetraplegia using a neurally controlled robotic arm. Nature. 485, 372–375. 17 May 2012.
  8. Kwok, Roberta. Neuroprosthetics: Once more, with feeling. Nature. Nature News Feature. May 08, 2013
  9. O’Doherty JE. et al. Active tactile exploration using a brain–machine–brain interface. Nature. 479,  228–231. 10 November 2011.
  10. Dhillon, GS et al. Residual function in peripheral nerve stumps of amputees: implications for neural control of artificial limbs. Hand Surg. 29, 605–615. 2004.
  11. Horch, K. Object Discrimination With an Artificial Hand Using Electrical Stimulation of Peripheral Tactile and Proprioceptive Pathways With Intrafascicular Electrodes. IEEE Trans Neural Syst Rehabil Eng. 19(5):483-9. October 2011.
  12. Kuiken TA et al. Redirection of cutaneous sensation from the hand to the chest skin of human amputees with targeted reinnervation. PNAS. 104(50), 20061-20066. 2007.
  13. O’Doherty JE et al. Active tactile exploration using a brain–machine– brain interface. Nature. 479: 228-230. 2011.
  • Image credits: Nature News, Neuroprosthetics: Once more, with feeling.

Arthur Jurao is a graduate of Georgetown University, and is the current Executive Director of E-Publishing for The Triple Helix Online. Follow The Triple Helix Online on Twitter and join us on Facebook.