Latent state-space models for neural decoding.

摘要: Ensembles of single-neurons in motor cortex can show strong low-dimensional collective dynamics. In this study, we explore an approach where neural decoding is applied to estimated dynamics instead the full recorded neuronal population. A latent state-space model (SSM) used estimate from measured spiking activity population neurons. second representation then decode kinematics, via a Kalman filter, The SSM-based illustrated on primary monkey performing naturalistic 3-D reach and grasp movements. Our analysis that performance based comparable