Graduate student

Joint Graduate Group in Bioengineering

UC Berkeley & UCSF Graduate Program Bioengineering (San Francisco, CA)

Jose Carmena

Exploring the role of stiffness control in movement accuracy and stability, and the related neural correlates in M1 and PMd, with the ultimate goal of incorporating dynamics control into brain-machine interfaces.

The central nervous system (CNS) produces dexterous, skilled movements despite inherent unpredictability in the motor system and external environment. Recent studies show that limb impedance (inertia, damping, and stiffness) may be one mechanism used to overcome these challenges. Limb stiffness, in particular, shows task-specific modulation in a variety of tasks, and even appears to be optimized to task requirements (e.g. Burdet et al 2001, Wong et al 2009).These exciting findings suggest that limb stiffness is manipulated by the CNS. How does the CNS achieve task-relevant control of limb stiffness? Previous work reveals that co-contraction of antagonist muscle pairs, tuning of spinal reflex loop gains, and posture selection are key mechanisms for stiffness modulation. But how these different schemes are controlled and combined to create task-level control is still unclear. Related questions about how limb stiffness control is learned and represented by the CNS also remain open. Cortical areas, such as primary motor cortex (M1) and dorsal premotor cortex (PMd) likely contribute, but activity in these areas has not been studied in the context of limb stiffness. M1 shows correlations with both movement dynamics and muscle activity, as well as abstract movement parameters. This makes the motor cortex a probable candidate for impedance representations, and also begs the question of what form such representations might take. Cortical involvement, moreover, raises the exciting possibility of incorporating stiffness control into neuroprostheses. Current brain-machine interfaces (BMIs) have focused on decoding kinematic movement parameters, which alone are insufficient for dexterous control and interaction with external environments. Given limb stiffness’ strong role in controlling movement accuracy and stability, BMIs that incorporate modulation of upper-limb stiffness would provide more reliable performance in real world scenarios.

2008 Mathematics in Brain Imaging Summer Course, UCLA Institute for Pure and Applied Mathematics

2008 Neuroinformatics Summer Course, Marine Biological Laboratory

2008-present NSF Graduate Research Fellow