September 2010

Active Stance: Sensory substitution enhances O&P rehab

By Michael T. Wilson, CPO/LP

Rehabilitation science and medicine professionals are realizing that extracorporeal orthotics and prosthetics are sensory as well as functional substitution devices, and that these substitutions are equally important and mutually beneficial. Researchers have identified a high correlation co-efficient between normal body imagery and acquired sensory perception skills with orthotic-prosthetic control and manipulation skills.

Michael Merzenich, MD, a neuroscientist at the University of California San Francisco, believes we can make smarter prostheses when we’re smarter about integrating neuroscience with engineering and medical science. Merzenich believes that researchers cannot overestimate the capacity of the human brain to restore function, to be trained, and to make up what’s been lost in extraordinary ways; with the help of prosthetic devices, sensory information can continue to flow into the brain from the peripheral nervous system.1

Medical and prosthetic technology has advanced to the point where lost arms and legs can be replaced with artificial ones. Electrical and mechanical engineering, combined with suitable aesthetics, has the potential to create an artificial substitution device (prosthesis) with the same range of basic or rudimentary function as the limb it replaces, as long as the replacement has a suitable interface with the remainder of the user’s body. Until recently, however, not much attention has been paid to the other half of this circuit, namely feedback of sensation from these limbs to the user. Even though an amputee may regain the use of lost limbs, sensory impressions of those limbs remain elusive.

Sensory substitution

Orthotists and prosthetists often ask themselves why some clients experience so much difficulty with orthotic and prosthetic restoration while other clients breeze through the process. All things being equal, the problems are not always directly related to physiology, anatomy, histology, biomechanics, or psychology. It is the premise of this paper that some of the problems our clientele experience are often associated with the neuroscience, in particular the neuropsychological aspects, of O&P restoration. Gaining an understanding of how information from natural sensors is integrated into the activation of muscle systems is only part of the bigger picture of sensory substitution, however. The other is the mental construct that comprises the sense impressions, perceptions and ideas about the dynamic organization of one’s own body and its relationship to other objects.

The body is represented in the human brain in various ways, and such representations are key to the perception of static and moving bodily parts and in the understanding and imitation of motor acts. Within the context of a sensorimotor approach to understanding the nature of sensory experience, a main concern lies in studying the process by which subjects attain mastery of sensory perception from a substitution device. A series of five learning stages has been postulated by researchers at the Université Paris.2

1. Contact involves learning the sensorimotor skill necessary to maintain and control perceptual contact with a stimulus.

2. Exteriorization involves experiencing the stimulus as no longer located at or in the sensor that conveys it (eye, ear, skin, residuum), but as corresponding to an outside physical entity, such as a prosthesis or neuropathic limb.

3. Spatialization involves attribution of a spatial location for the experienced entity, with coherent understanding of its relation to the body.

4. Comprehension involves being able to not just spatially locate but also recognize the entity as a perceptual object among possible alternate objects.

5. Immersion is the state where the subject possesses all these abilities and feels he or she is physically immersed in an environment populated by objects that can be perceived through the substituted sensory modality.

Artificial vs. natural sensors

Instead of using artificial sensors in the prosthesis, it is possible to use the body’s own natural sensors. These come pre-installed, no assembly required, do not require battery power, are not prone to mechanical or electrical failure, and have been optimized through millions of years of natural evolution. Natural sensors provide cognitive feedback to the user that replicates communication with the brain more accurately than artificial sensors.3 Using natural sensors already present in the body is an attractive approach  because it avoids the need to strap artificial sensory devices onto the body or the prosthesis, which can get in the way of manipulating the prosthesis and have little cosmetic appeal.

More importantly, natural sensory communication with the brain is particularly effective for exteriorizing and spatializing sensory perception. The most advanced artificial receptor can process 32 simultaneous signals. In contrast, the fingers of the human hand have an estimated 17,000 touch sensing receptors, or 200-300 touch sensors cm2. With natural sensors, the sensorimotor loop is completed in approximately 70 ms.3

Sensory substitution theory

For the brain to correctly interpret information from a substitution device, it is not necessary that the information be presented in the same manner or form as the original sensory information system. It is only necessary to present information from a substitution device in a form of energy that can stimulate receptors at the man-machine (prosthetic) interface; for the brain, through the sensorimotor system, to know the origin of the information. This information reaches the perceptual level for analysis and interpretation via the somatosensory pathways and structures.4 We do not see with our eyes; the optical image does not go beyond the retina, where it is turned into patterns of pulses along nerves.5 Those individual pulses are not any different from the pulses of the big toe. It is the brain that recreates the image from these patterns of pulses.

Tactile vision substitute systems (TVSS) deliver optical information to the brain via an array of stimulators in contact with the skin on one or several parts of the body. Optical images picked up by the TVSS camera are transduced into energy (vibratory or direct stimulation) that can be mediated by skin receptors. This transduced pulse information is conveyed to the perceptual level of the brain for analysis and interpretation. After training with the TVSS, blind subjects report receiving the images in space instead of as stimulation on the skin.6

Similarly, an experimental biomedical device has been designed to facilitate neural correlation in the neuropathic and transtibial lower limb. The correlation between imagery and natural sensory substitution is referred to as exteriorized psychogenic proprioception (EPsP), otherwise known as “phantom limb sensation.” When using this device, the user first must imagine normality regardless of his physical state of being and degree of desensitization. Next the user develops an ability to anticipate specific events, in this case kinesthetic activity (awareness or perception of motion) based on sensory substitution. Finally, the user learns to associate sensory substitution with the image of normality in such a way not to be expected on the basis of chance alone.

Neuropsychological mechanisms

Within the skill-based or sensorimotor approach to understanding sensory awareness, sensation is a matter of the perceiver knowing that he is currently exercising his implicit knowledge of the way his body actions influence incoming sensory information.7 Why does seeing provide us with a qualitatively different sensory experience than hearing, taste or touch?  Indeed, why does sensory input provoke a sensory experience at all and why does our sensory experience differ in so many respects from other conscious mental phenomenon? The answers to these questions lie in the neuromechanisms involved.

Though knowledge is rapidly accumulating regarding the neuromechanisms involved, sensory consciousness can be explained within the context of sensorimotor function. Implicit knowledge of bodily actions is referred to as “corporality,” and is manifested and measured by the body’s response to an interaction between somatosensory and sensorimotor function.2 An illustration is provided by the sensation of softness one might expect in holding a sponge. Experiencing a sensation of softness consists of being aware that one can exercise certain practical skills with respect to the sponge. For example, one can press it, and it will yield under pressure. The anticipated experience or awareness or sensation of softness of the sponge is characterized by a variety of such possible patterns of bodily interactions with the sponge. Thus, the conscious experience of softness is easily characterized by the skill base or sensorimotor approach because it resides in, and is constituted by, the exploratory skill involved. It is impossible to imagine or anticipate all the exploratory patterns of softness while experiencing hardness. When a perceiver knows in an implicit and practical way, that at a given moment he is exercising sensorimotor skills associated with softness, then he is in the process of experiencing softness, becoming aware and conscious of the sensation of softness.

Laws that describe these sensorimotor skills or interactions are referred to as sensorimotor contingencies. These interactions can be explained in terms of corporality and alerting capacity.7 Corporality is further defined as the extent to which activation of a sensory receptor systematically depends on movement of the body. The alerting capacity of sensory input is the extent to which the sensation can cause automatic orienting behavior that peremptorily captures the organism’s cognitive processing skill. Within the context of sensory substitution, both corporality and alerting capacity remain implicitly perceptual experiences until the perceiver learns to predict or anticipate their mutual and reciprocal interaction, at which time the perception becomes explicit – a fully formulated, developed, and accurate kinesthetic event. In other words, proprioception is the neural input that signals mechanical displacement of the muscles and joints, but this positional input in and of itself does not have an experienced sensory quality, a condition similar to the autonomic nervous system. Rather, it is the anticipated effect of extensively processed positional input and its perceived relationship to alerting capacity that constitutes sensory awareness.

Until recently, no effort has been undertaken to analyze the laws of sensorimotor contingency related to a sensory substitution O&P device. The similarity between these sensorimotor contingencies and those that govern unimpaired limbs will determine the extent to which users will actually feel sensation in (corresponding to) the neuropathic limb via an orthotic device or in the prosthesis itself.

Optimizing O&P

Sensation is conscious when a person is poised to cognitively make use of the sensation in his or her judgments, decisions and rational behavior: that is, when the person has cognitive access to the sensation.7 The different types of sensation and the corresponding experienced characteristics can all be accounted for in terms of the differences between the sensorimotor contingency skills, and in terms of the way the neural channels are tuned to the environment, namely, by the properties of corporality and alerting capacity.

In optimizing orthotic and prosthetic restoration, these cognitive processes become so closely associated, so intricately intertwined, so mutually and reciprocally interactive that they become, in effect, indistinguishable. They, in fact, become neural correlates.

Empathy training

Understanding and clinical implementation of neural correlation modalities contributes to a more successful outcome of extracorporeal orthotic and prosthetic restoration by facilitating acquisition of sensorimotor contingency skills and by effecting an enhanced sense of wholeness, normality and well being while connected to and operating an O&P device. Consequently, clinical orthotists and prosthetists should be familiar with these concepts and practice them on a routine basis.

Walking on prosthetic feet attached to post-acute fracture braces will provide the wearer with a simple and practical demonstration of EPsP. When wearing and walking with these devices, familiar somatosensory and sensorimotor function will be compromised (momentarily disrupted) because the wearer is standing on top of articulated prosthetic feet rather than on the ground and because the anatomical ankles are immobilized. This will lead to a precarious, if not impossible, balancing situation (hence a safety belt).

Now let’s apply some neuropsychology principles. Instead of concentrating on the feet, the patient should concentrate on the floor. In other words, they should imagine stimulus experience as no longer coming from the feet, but coming from the floor. It is important that the patient not to try to maintain balance by moving the ankles and feet. Instead, they should completely relax the ankles and feet, and concentrate on the floor.  Again, imagination is the first critical step in facilitating EPsP. Imagine normality, and this image must extend to the floor and include the prosthetic feet.

Now, while the patient walks, they should anticipate what that image of normality is actually doing (kinesthetic activity), and at the same time, anticipate how they will perceive the unique sensory input related to walking in this particular circumstance (natural sensory substitution). Discernable correlation of the image of normality and sensory substitution will begin immediately because the sensorimotor and imagery skills are basically intact. If the patient has an ablated or neuropathic lower limb, correlation would still take place, but at a slower rate because proprioception and somesthesia have been compromised; the greater the compromise, the greater effort and a longer period of time will be necessary for correlation.

Clinical implementation and assessment of neural correlation modalities are also helpful in orthotic and prosthetic restoration because they are the most revealing methods of determining whether or not your client is safe when using an O&P device. Ask the patient to rehearse or choreograph a finite set of kinesthetic events relating to O&P device utilization, such as walking down a hallway, making a 180° left turn, walking back and making another 180° left turn and then coming to a standing stop. Remember to ask the client to anticipate everything they will feel and do throughout the entire sequence.

After they have completed this specific and finite sequence, ask if their imitation of sensorimotor skills (reenactment of their rehearsal and choreographic skills) was predictable, consistent and accurate. If the client reports their imitation skills as being 100% accurate, they can be deemed safe while utilizing the O&P device for that specific activity. If they report an inaccurate imitation (or any unexpected sensory or motor event or episode during the sequence), the client is unsafe and should not be allowed to independently continue that particular activity without receiving further training and supervision.

Anecdotal assessment of neural correlation skills relating to safe operation of prosthetic and orthotic devices will have to suffice until more reproducible scientific methods are introduced into the O&P profession that will allow the practitioner to physically measure sematosensory capacity and acquired sensorimotor contingency skills.


In a 2004 lecture, Hugh Herr, PhD, director of biomechatronics at the MIT Media Lab, identified “distributed sensing and intelligence” as a key area for the future of prosthetics research.8

“Advances in muscle-like actuators,  neuroprostheses, and biomimetic control strategies are necessary to increase the merging of body and machine to create an intimacy between the human body and prosthesis. It’s our thesis that such intimacy will create a paradigm shift in this area of rehabilitation,” he said. “To really push this area of medicine, we need to merge body with machine to create an intimacy between the human body and the prosthetic device.”

Practitioners are encouraged to develop an interest in and commitment to rehabilitation science and orthotic/prosthetic restoration to the fullest extent possible.

In his introduction to a 2002 special issue of the Journal of Head Trauma Rehabilitation focused on neuropsychological technologies, Douglas Chute, PhD, wrote: “The transportability of technology should allow the bridging of research protocols to clinical practice. There is no intrinsic reason why the neuropsychologist or rehabilitation specialist cannot fully engage with the new range of neuropsychological technologies appropriate for their patients in rehabilitation.”9

An understanding of sensory substitution can provide common ground between the clinical orthotist, prosthetist and neuroscientist, and forge a pathway for further communication of ideas and exchange of technologies between these two professions. Practitioners can also benefit from simple and practical methods of helping their clientele regain a more complete and personal image and impression of sensory as well as functional restoration.

Michael Wilson, CPO/LP, is an owner/operator of a prosthetic clinic in Missouri City, TX.


1. Merzenich MM. Smart prosthetics. Presented at National Academies Keck Futures Initiative: Smart Prosthetics: Exploring Assistive Devices for the Body & Mind, Irvine, CA, November 2006.

2. Botvinick M, Cohen J. Rubber hands “feel” touch that eyes see. Nature 1998;391(6669):756.

3. Haugland M, Sinkjaer T. Interfacing the body’s own sensing receptors into neural prosthesis devices. Technol Health Care 1999;7(6):393-399.

4. Bach-y-Rita P. Theoretical aspects of sensory substitution and neurotransmission related to spinal cord injury. Spinal Cord 1999;37(7):465-474.

5. Bach-y-Rita P. Brain Mechanism in Sensory Substitution. New York: Academy Press, 1972

6. White BW, Saunders FA, Scadden L, et al. Seeing with the skin. Percept Psychophys 1970;7(1):23-27.

7.  O’Regan JK, Noe A. Acting out our sensory experience. Behav Brain Sci 2001;24(5):1011-1021.

8. Herr HM. Multidisciplinary approaches to limb loss: A chain of events leading to a single step. Presented at Emerging Technologies in Support of the New Freedom Initiative: Promoting Opportunities for People with Disabilities, Washington, DC, October 2004.

9. Chute DL. Neuropsychological technologies in rehabilitation. J Head Trauma Rehabil 2002;17(5):396-377.

Leave a Reply

Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.