A brain computer interface (BCI) is a device that allows people with motor or cognitive disabilities to control computers or other devices. BCIs translate brain signals with a variety sensors and recording techniques, as well as a signal processing algorithm and machine learning. They can operate a wide range of devices, from cursors on computers to robotic arms and wheelchairs.
Most BCIs measure activity from the scalp, such as electroencephalographic (EEG) or functional near-infrared imaging (fNIRS) signaling. The signals are then recorded by a sensor that detects the presence of brain activity and software transforms them into operating commands.
In many BCI systems, the user must go through an iterative course of instruction to produce signals that the system recognizes. For instance, in a BCI that is specifically designed to type letters, the user is required to imagine moving their right or left hand.
The most exciting BCI breakthroughs involve implantable or invasible devices that record directly on neural tissue, instead of scalp. These devices are more accurate than noninvasive BCIs however they require surgery and are not without risk.
BCIs that are invasive remain in development and patients must be aware of the risks and advantages. BCIs can read neural signals containing sensitive health and behavioral information. Privacy and data security is an additional concern. Some people are opposed to BCIs because they fear that hackers can gain access to their minds and manage them.