Patterns of Motor Cortical Activity Described by the First Latent Dimension Predict Presence of an Object, Poster 7

Brain computer interface (BCI)-controlled robotic neuroprosthetics can restore upper limb function to individuals with motor impairment in a laboratory setting. Presently, humans can learn to control a robotic arm using... [ view full abstract ]

Brain computer interface (BCI)-controlled robotic neuroprosthetics can restore upper limb function to individuals with motor impairment in a laboratory setting. Presently, humans can learn to control a robotic arm using signals recorded from microelectrode arrays implanted in the motor cortex. However, past work has shown deterioration of motor control when interacting with objects vs. performing movements in open space. Here we analyze population-level sources of variation during reaching and grasping movements with and without an object in the workspace. Factor analysis was used to identify the dominant patterns of co-modulation measured in the recorded population of neurons for these two conditions. We found a robust separation between the object and no-object neural trajectories in the first latent dimension, suggesting that the presence of an object is responsible for the greatest amount of variance in neural activity. The co-modulation of trajectories in the other latent dimensions suggests that variation captured here may be related to movement parameters, agnostic to the presence of an object. 

Acknowledgements: This material is based upon work supported by the Defense Advanced Research Projects Agency (DARPA) and Space and Naval Warfare Systems Center Pacific (SSC Pacific) under Contract No. N66001-16-C-4051 and the Revolutionizing Prosthetics program (contract number N66001-10-C-4056). Any opinions, findings and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of DARPA or SSC Pacific. Funding was also provided by the Undergraduate Program in Neural Computation at Carnegie Mellon University and the University of Pittsburgh.