Shijia Geng
Papers
6
Total Citations
196
H-Index
4
About
Shijia Geng is a pioneering researcher in neuroprosthetics and brain-machine interfaces (BMIs), with a particular focus on developing adaptive, autonomous control systems that can compensate for neural signal changes. Her most influential work, "Using Reinforcement Learning to Provide Stable Brain-Machine Interface Control Despite Neural Input Reorganization" (77 citations), introduced a groundbreaking approach that enables BMI systems to maintain stable performance even when the brain's neural representations shift—a critical challenge for real-world neuroprosthetic use. Geng's research uniquely combines reinforcement learning algorithms, such as actor-critic and Hebbian learning, with neural decoding to allow users to control robotic arms using only binary evaluative feedback, dramatically simplifying the user's cognitive load. She has also advanced the common marmoset as a key primate model for behavioral neuroscience (46 citations), enabling more naturalistic studies of motor control and body schema extension. Her work on extracting error-related signals from the striatum during robotic arm perturbations further demonstrates her commitment to creating closed-loop, self-correcting neuroprosthetic systems. With a total of nearly 200 citations across her core publications, Geng's contributions are shaping the next generation of adaptive, user-friendly neural prosthetics that can function reliably in daily life.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5
- 6