Active tactile exploration using a brain–machine–brain interface

Prosthetics get a touch better Brain–machine interfaces have evolved to a stage at which robotic arms can perform complex movements, such as reaching and grasping, in response to signals from the brain. Further progress, towards systems that allow dexterous control of an artificial limb, will require the development of somatosensory feedback, allowing touch to inform the use of the limb. A step towards that goal has been achieved with the demonstration of an interface that multiplexes neuronal signals from the motor cortex to control elements of a computer display with artificial tactile feedback through microstimulation of the somatosensory cortex. Monkeys learned to use this interface to move a computer cursor or hand image to explore visual targets. To discover which target would yield reward, the monkeys had to discriminate the microstimulation evoked from each target when the actuator touched the objects.