Researchers at Johns Hopkins University have developed a new technique that allows a partially paralyzed person to eat with robotic arms connected via a brain-machine interface, reports Engadget. The study participant only needed to make small movements with his fists at certain prompts (for example, “choose the incision location”) so that the robotic arms, equipped with forks and knives, cut the food and brought it to his mouth. According to researchers, a partially paralyzed person could get a piece of dessert in 90 seconds.

The new method centers on a shared control system that minimizes the amount of mental input required to complete a task. He could map his four-degree freedom of movement (two for each hand) to as many as 12 degrees of freedom for controlling the robot arms. The limbs’ prompt-based intelligent responses also reduced the workload.

The technology is still quite young. Scientists want to add touch-like feedback instead of relying solely on visual images. They also hope to increase accuracy and efficiency by reducing the need for visual confirmation. However, in the long run, the team sees that such robotic arms restore complex movements and provide greater independence for people with special needs.