UC Irvine-led Study Achieves Brain-Controlled Walking with Artificial Sensory Feedback

UC Irvine researchers (from left) Dr. An Do, associate professor of neurology; Payam Heydari, professor of electrical engineering and computer science; and Zoran Nenadic, professor of biomedical engineering, recently participated in a study that demonstrated a brain-computer interface technology that enables spinal cord injury patients to walk with a robotic exoskeleton and feel lifelike sensory responses, a key factor in safe and realistic mobility. Debbie Morales / UC Irvine
 

• An interdisciplinary, multi-institutional team is the first to use a bidirectional brain-computer interface to control the gait of a walking exoskeleton.
• The system enables patients suffering spinal cord impairments to regain both the ability to walk and the associated sensation, which results in safer, more realistic movement.
• The project was funded by the National Science Foundation.

Irvine, Calif., April 16, 2026 — Researchers at the University of California, Irvine, Caltech and Keck School of Medicine of USC have developed a bidirectional brain-computer interface that allows a person to control a robotic walking exoskeleton using brain signals and receive artificial leg sensation through direct electrical stimulation of the sensory cortex.

The project is the first implementation of a BDBCI for walking that incorporates bilateral interhemispheric leg sensorimotor brain areas and represents a critical step toward restoring full ambulatory function in individuals living with spinal cord injury and paraplegia. The team’s results are reported in a paper published recently in the journal Brain Stimulation.

“Millions of people worldwide suffer from paralysis from spinal cord injury, with loss of lower-extremity motor and sensory function leading to wheelchair dependence and increased risk of serious secondary conditions including heart disease, osteoporosis and pressure ulcers,” said co-author Dr. An Do, UC Irvine associate professor of neurology. “Recovering the ability to walk ranks among the highest rehabilitation priorities for paralyzed individuals.”

He said that while robotic gait exoskeletons have emerged as a promising technology for restoring walking ability, existing systems rely on manual control and provide no sensory feedback, a significant limitation as impairment of the sensation of taking steps is known to reduce gait speed and increase the risk of falls.

The team’s BDBCI apparatus directly addresses both these challenges. By decoding motor intent from electrocorticography signals recorded from the leg motor cortex and delivering artificial leg sensation through targeted electrical stimulation of the somatosensory cortex, it creates a closed-loop, brain-driven experience of walking.

“This work demonstrates that it’s feasible to restore both the motor and sensory dimensions of walking using a single, compact, embedded brain-computer interface system,” Do said. “We believe this lays a critical foundation for the development of fully implantable systems that could one day give paraplegic patients a meaningful and natural sense of movement.”

Learn more about the study in the full press release in UC Irvine News.