Rubber hand feels real for amputees

One of the bigger challenges facing researchers who are developing artificial limbs is to create prostheses that not only act but also feel like real limbs. This is especially true for the hand, which is one of the most sensitive parts of the human body, and although advanced prosthetic hands with fully articulated digits which move independently of one another are now available, they would be far more useful if they provided the user with sensory feedback.

Last year, surgeons from the Rehabilitation Center of Chicago made some progress towards this goal: they fitted amputee Claudia Mitchell with a prosthetic arm, after re-routing the intact nerves from the stump to muscles in her chest. Now, when that part of her chest is pressed, she feels a crude sensation that the fingers of her missing limb are being touched. An alternative approach is to connect the prosthesis directly to the brain; this could be achieved, in theory at least, by attaching sensors in the device to electrodes which are implanted into the somatosensory cortex, or into the peripheral nerves which project up to it.

Both of these procedures involve surgery, on either the brain or the peripheral nerves, and so are extremely invasive. Now Swedish researchers show that a simple illusion can induce amputees to experience a rubber hand as their own, so that tactile stimuli directed to it produce sensations which are localized to their missing limbs. The study, which is published in an open access paper in the journal Brain, is therefore an important step towards the development of neuroprostheses which feel like real limbs. 

Earlier this month, cognitive neuroscientist Henrik Ehrsson and his colleagues at the Karolinska Institute in Stockholm described the body swap illusion. They first made their participants view a mannequin's body, or that of another person, from the first-person perspective, using video cameras and head-mounted visual displays. When the participants' bodies were stroked with a short rod, in the same way and at the same time as the body of the mannequin or the other person, they reported experiencing the other body as their own.

In the current study, a similar phenomenon, called the rubber hand illusion, was used. The study involved 18 participants, all of whom have had one of their arms amputated somewhere between the wrist and elbow. The participants sat with the stump of their amputated arm hidden from view, and with a life-sized rubber hand in full view on the table in front of them. An experimenter then simultaneously touched the stump and the index finger of the rubber hand with soft paintbrushes.


Under these conditions, the amputees experienced the rubber hand illusion, reporting strong sensations of touch from the rubber hand when it and their stump were stroked simultaaneously. Using the skin conductance response, the researchers confirmed these reports objectively. Stabbing the rubber hand with a needle caused the participants to sweat, leading to a measurable increase in skin conductance. This shows that stabbing the rubber hand was stressfull for the participants, because they experienced the prosthesis as being a part of their own bodies.

In normal volunteers, the rubber hand illusion can only induced when stimulation is applied to the corresponding part of the real hand. In amputees, this is of course impossible, because the exact location of stimulation cannot be matched, but stroking the amputees' stumps was nevertheless sufficient to evoke the illusion. Beforehand, the researchers had mapped the sensations evoked by touching the participants' stumps and stroked the area which had elicited a phantom sensation in the index finger. They suggest that the illusion was induced by a similar mechanism to that in normal volunteers: the brain receives tactile information from the real hand, but this information is overridden by the conflicting visual inputs, so that the sensation is "transferred" to the rubber hand.

Interestingly, the researchers note that the shorter the time since amputation, the stronger was the induced illusion. This is perhaps because there has been less time for the somatosensory cortex to reorganize itself. Following removal of a limb, this reorganization causes phantom limb sensations in the vast majority of amputees, but these sensations eventually fade. These sensations are therefore still present during the first few years after amputation, so the evoked illusion is vivid. If, on the other hand, a longer period of time has elapsed since amputation, the brain has become accustomed to the lack of sensory inputs from the missing limb, and the evoked illusion is weaker.

This study opens up new possibilities for the development of prosthetic limbs. It suggests that athe rubber hand illusion could be reproduced using an artificial limb equipped with tactile fingertip sensors connected to an array of stimulators on the stump. When an object is touched  with the prosthesis, the tactile stimulus would instantaneously be transmitted to the stump, thus tricking the brain into processing that information as coming from the artificial finger. The illusion would make the user feel that they "own" the prosthesis (i.e. that it is a part of their body), and  the sensory feedback from it would enable them to better control it 


Ehrsson, H. H. et al (2008). Upper limb amputees can be induced to experience a rubber hand as their own. Brain DOI: 10.1093/brain/awn297.

More like this

I saw an article on this very topic recently in the "optical illusion" issue of Discover magazine. They were showing how fooling the visual input of someone with a missing limb could cause the body to induce sensation as if the phantom limb was truly there still.

They were proving that they could then "relax" phantom limbs that were locked in painful positions by visually moving the optical illusion of the limb into a comfortable position thus relieving the pain by fooling the brain.

I think it is of no effect for blindman.