On March 26, 1945, shrapnel from an enemy shell struck the right arm of David Cumming Simpson, a commissioned soldier from Scotland. It left him temporarily unable to use the arm and got Simpson sent home to Scotland, where he had planned to pursue a career as an accountant. He shifted his focus to medical research, received his PhD from Edinburgh University and began to develop medical instruments, including one of the first successful fetal heart monitors.

A turning point in his career came more than a decade later. Chemie Grünenthal, a German pharmaceutical company, had been distributing a new sedative and anti-hypnotic drug called thalidomide; soon, they realized it was also an effective treatment for morning sickness. Thalidomide was sold worldwide, though it was never approved by the Food and Drug Administration for use in the U.S. Just one dose of the drug taken during a critical developmental period of a pregnancy (days 27-40 after fertilization) was enough to cause fetal limb abnormalities. By the time it was taken off the market in 1961-62, more than 10,000 babies–mostly in Europe, Australia, Canada and Japan–had been born with missing arms, legs, digits and organ deformities. Up to 40% of these small victims died before their first birthday.

Simpson was asked to design new upper limb prostheses for the children who survived infancy. He started working with one of the best devices available at the time: pneumatic limbs called “pat-a-cake” arms. Children moved their shoulders forward and backward to make the prosthetic hands clap together, allowing the children to hold objects. Simpson got to work on designing additional useful features, such as a powered elbow joint, and the ability to rotate the hand inward and outward.

But the more movements Simpson added, the more difficulty the children had controlling them. The prostheses demanded all the children’s focus, and presented “a considerable challenge both to the intellect and to the powers of concentration,” as Simpson later wrote. Using a properly designed prosthesis, he felt, should feel unconscious and automatic.

So in 1965—50 years ago, and 20 years after Simpson’s own wartime injury — he began testing a major advance in prosthetics. His “position-servo system” provided the child with feedback about where the end of the prosthesis was. Doing so would, in theory, allow it to take more full advantage of the body’s natural sense of proprioception, an awareness of where parts of the body are in space. Proprioception enables the nearly effortless coordination of body parts, such as kicking a soccer ball without looking, by means of a complex interplay of sensory neurons in the inner ear, receptors in muscles and tendons, and the central nervous system.

The position-servo system would closely link the movement of the intact shoulder to the prosthetic arm, so the position, force and acceleration of the prosthetic arm fed back to the shoulder. Simpson theorized that, if the movement in an artificial joint corresponded proportionately to the movement of the natural shoulder joints, the information received by the shoulder would travel to the central nervous system and confer a more intuitive sense of the artificial arm’s position. By 1968, after two and a half years’ testing, Simpson wrote that children using the limb achieved a “considerable degree of unconscious control and position awareness of the limb.”

In 1974, in a chapter of a book on upper limb prostheses, Simpson coined a term describing this fundamental idea: extended physiological proprioception (EPP). The principle states that brain can map an artificial limb to itself, essentially making it part of the body, if given the right feedback channels. This was a new idea for prosthetics development, and one that helped thalidomide children navigate their world in a much more natural way.

Scaling EPP up to ever-more-complex prostheses has proven more difficult. Only one arm (the Boston Digital Arm) currently employs Simpson’s positional servo technology. But a new case study, published in Science Translational Medicine in February 2014, explores Simpson’s concept of EPP on a new frontier.

Researchers in Italy and Switzerland have created an artificial hand that contains sensors, which are then connected to electrodes implanted in the wearer’s median and ulnar nerve fascicles.  When the prosthetic hand touches an object, the sensors send a signal to the wearer’s nerves, inducing the sense of touch and strength of grasp.  Using sensory feedback, the subject in the study was able to produce three different levels of grasping force by modulating his grasp accordingly.

Prosthetics with ever-more sophisticated feedback mechanisms may offer a new testing ground for EPP, and the elusively natural feel in an artificial limb.