A Direct Sensorimotor Connection with the Spared Neural Code of Movement to Regain Motor Function

Project Information

GRASPAGAIN

Grant agreement ID: 101118089

DOI

10.3030/101118089

EC signature date 17 October 2023

Start date 1 January 2024

End date 31 December 2028

Funded under

• European Research Council (ERC)

Total cost

€ 1 495 271,00

EU contribution

€ 1 495 271,00

Coordinated by

FRIEDRICH-ALEXANDER-UNIVERSITAET ERLANGEN-NUERNBERG
Germany

Periodic Reporting for period 1 - GRASPAGAIN (A Direct Sensorimotor Connection with the Spared Neural Code of Movement to Regain Motor Function)

Reporting period: 2024-01-01 to 2026-06-30

Summary of the context and overall objectives of the project

Spinal cord injury (SCI) affects millions of people worldwide, and among the most devastating consequences is the loss of hand function. For individuals with cervical SCI, the inability to grasp, hold, and manipulate objects profoundly limits independence and quality of life. Restoring even basic hand function is consistently rated as the top priority by people living with paralysis.

Current approaches to restoring hand function after SCI face significant limitations. Invasive brain-computer interfaces require neurosurgery and carry associated risks. Functional electrical stimulation provides only coarse, pre-programmed movements. Conventional prosthetics and orthoses lack intuitive control, leading to high abandonment rates.

GRASPAGAIN takes a fundamentally different approach. Rather than bypassing the injured spinal cord through brain implants, the project investigates the motor neurons below the injury that once controlled the hand. Recent evidence suggests that even after clinically "complete" spinal cord injury, some motor neurons survive and may retain the ability to respond to voluntary commands — even when no visible movement occurs. If these surviving neurons can be detected and decoded, they could serve as a natural control signal for assistive devices, using the same neural commands the person used before their injury.

The project pursues three integrated objectives. First, it aims to understand how motor neurons controlling the hand are organized and how they behave after spinal cord injury (neurophysiology). Second, it develops non-invasive technology to detect and decode the activity of these surviving neurons in real time (neural interfacing). Third, it creates a training framework that allows individuals with paralysis to learn to use their residual neural signals to control assistive devices such as robotic hands and exoskeletons (sensorimotor training).

The expected impact is both scientific and clinical. Scientifically, the project advances fundamental understanding of how the nervous system controls the hand and what changes after injury. Clinically, it aims to demonstrate that non-invasive, intuitive hand function restoration is achievable — offering a realistic pathway to assistive technology that could improve daily independence for people living with paralysis.

Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far

During the first two years of the project, substantial progress has been made across all three objectives.

In neurophysiology, the team recorded the electrical activity of individual motor neurons in the hand muscles of healthy volunteers using advanced electrode arrays inserted into the muscle and placed on the skin surface. A key discovery was that nearly half of the motor neurons studied change their activation order depending on what task the hand performs — for example, spreading the fingers versus bending them. This challenges a long-standing principle in neuroscience that assumed motor neurons always activate in a fixed order. In a separate study of dynamic hand movements, the team found that motor neurons are highly specialized for specific finger movements, with very little overlap between those controlling single-digit and multi-digit tasks.

In neural interfacing, the project recruited eight individuals with motor complete cervical SCI through partnerships with five rehabilitation centers across Germany. Despite complete hand paralysis, the team successfully detected voluntary motor neuron activity in the paralyzed forearm muscles and decoded more than ten different intended hand movements using artificial intelligence. Three participants with over eight years of complete paralysis were then able to use these decoded signals to control a wearable robotic finger, performing everyday tasks such as unscrewing bottle caps — using the same neural commands they used before their injury. This represents the first demonstration of intuitive, proportional control of a physical robotic device by individuals with motor complete SCI using a fully non-invasive interface.

In the sensorimotor training framework, the team developed a real-time platform that allows individuals with SCI to see and control the activity of their own surviving motor neurons on screen. This system was also applied successfully in a six-year-old child with complete hand paralysis following brain surgery, who was able to play a computer game and grasp objects using a robotic orthosis for the first time. The team also released an open-source software framework enabling rapid adaptation of neural decoders to individual users across different neurological conditions.

Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)

The project has produced several results that advance the field significantly beyond what was previously known or achievable.

The most impactful finding is that individuals with clinically complete spinal cord injury retain meaningful voluntary control over motor neurons that once moved their hands. This challenges the prevailing clinical view that motor complete injury means permanent, total loss of voluntary motor function below the lesion. By showing that these surviving neurons can be detected non-invasively and used to control robotic assistive devices within minutes, the project opens a realistic pathway toward wearable technology that restores hand function without surgery. This approach is fundamentally more accessible than invasive alternatives requiring brain implants.

In fundamental neuroscience, the discovery that motor neuron recruitment order is flexible and task-dependent represents a significant departure from a principle that has guided the field for over fifty years. This finding has implications not only for understanding healthy motor control but also for designing more effective rehabilitation strategies and neural interfaces.

On the technology side, the project developed and publicly released several open-source tools that are already being adopted by other research groups. These include a framework for rapid neural decoder adaptation across different neurological conditions, a biophysical modelling platform for studying motor neuron behavior computationally, and a real-time motor unit decomposition system capable of processing hundreds of electrode channels simultaneously.

For further uptake and clinical translation, the key needs include larger-scale clinical validation studies, regulatory approval pathways for wearable neural interface devices, and partnerships with medical device manufacturers. A first step toward commercialization has been taken through a dedicated proof-of-concept grant for pediatric applications. The technology has been demonstrated to international medical delegations and at major medical technology fairs, generating interest from clinicians and industry partners. Continued engagement with rehabilitation centers and patient advocacy groups will be essential to ensure the technology meets real-world needs.