Over the past decade, a number of studies have shown that brain signals can allow people with severe disabilities to communicate by simply thinking, i.e., without using conventional muscular communication strategies such as speech, facial expression, or hand gestures. These devices, which record signals from the brain and translate them into control of an output device, are called Brain-Computer Interfaces or BCIs. BCI systems provide the only communication option for people who are totally paralyzed and who cannot communicate in any way using conventional methods. While previous studies have demonstrated the tremendous potential of BCI systems, current techniques have serious drawbacks that have thus far delayed their widespread application outside the laboratory to help such severely disabled people. The BCIs developed to date require extensive user training and may work well in only some individuals, and/or they require that recording devices be implanted within the brain. These implants have risks such as infection and carry no guarantee of long-term success. A brain-controlled communication device that is easy to use and that does not depend on risky surgical implants within the brain could allow thousands of patients who have lost the ability to communicate to again share their wishes with the outside world. With such a device, these patients could use a word processor, surf the Internet, or even control a prosthesis. The exciting initial findings of this current study are that a technique called electrocorticographic (ECoG) recording, which uses electrodes that do not penetrate the brain, can be used to control a BCI communication device that is powerful and easy to use.
This novel technique might offer to people who are totally paralyzed (such as patients with ALS who are fully "locked in" to their bodies without being able to communicate in any way) the only possibility to again share their wishes with the outside world. In the future, it might also allow people with less severe disabilities, such as people with spinal cord injury, to control their environment directly from their brain. The method gives signals that can provide the user with excellent control with very little training. The recording electrodes that detect the brain signals are placed on the surface of the brain in a procedure already commonly used in epilepsy surgery. Other BCI methods require electrodes that penetrate into the brain and/or require extensive training.
These initial results move beyond a recent study that showed that humans can use ECoG signals to move a cursor in one dimension (up/down) on a computer screen (Leuthardt et al., 2004). While studies from other labs have passively recorded ECoG signals, our two studies are the very first reports that actually used these brain signals for communication. Our first study reported one-dimensional control of a computer cursor; the present study explores two-dimensional control, and thus greatly extends the potential practical applications.