Over the past year, there have been many developments in brain-computer interfaces for controlling prosthetics. One of the most recent involves a device that aims to control a prosthetic arm using memories of movement within the amputee’s brain.

A Science Daily report details a study at the Centre for Research and Advanced Studies (CINVESTAV), where scientists are developing a prosthetic arm that the user can control simply by imagining movements. In order for this to work, the researchers have to find out if there is a specific memory pattern in the user’s brain that remembers the movement. If so, the electrical signals can be used to manipulate a prosthetic arm.

According to Muñoz Guerrero, the scientist leading the research, one of the obstacles in this project stems from the fact that the brain registers a broad range of activities throughout the body, and a single movement pattern must be drawn from this complex registry. The first stage is to record the brain’s patterns of the amputee’s memory of how to move the arm using an EEG. The sensitivity of these signals to external stimuli, such as blinking or light, then has to be measured.

In order for the prosthetic to work, the device will need both electronic and mechanical components to activate it, as well as a system that can translate signals from the brain into actual movement. The team is currently finalizing the material, which will be used to develop the prosthetic arm. This is crucial as it must be the same weight as a real arm; between 2-3 kilograms.

This research marks breaking ground in the field of neuroprosthetics because it permits direct communication between the brain and prosthetic device connected to it.

Image credit: Science Daily


TechEmergence conducts direct interviews and consensus analysis with leading experts in machine learning and artificial intelligence. Stay ahead with of the industry with charts, figures, and insights from our unparalleled network, including executives from Facebook, Google, Baidu, Yahoo!, MIT, Stanford and beyond: