Footnote:
Diese Datenquelle enthält auch Bestandsnachweise, die nicht zu einem Volltext führen.
Description:
To provide a basis for the validation of balanced spiking neuronal networks and their dynamics, we characterized resting-state (no task) and task-related spiking activity of arm/hand area of macaque (pre-)motor cortex.We recorded spiking activity using a 100-electrode Utah Array (Blackrock Microsystems) during resting-state (REST) and reach-to-grasp (R2G) behavior. In REST, we defined epochs of rest (RS), sleepiness (RSS) and spontaneous movements (M); in R2G: preparatory periods (PP) and task-related movements (TM). Single units were separated into putative excitatory and inhibitory.On the level of single units, we found that ~50% of all units changed their rates significantly with behavioral epochs. Next, we characterized the network activity based on a) the dimensionality which reveals the number of principal components needed to describe the parallel spiking activity, and b) the balance between putative excitatory and inhibitory population firing (absolute difference or instantaneous correlation). RS & PP show the highest dimensionality and the lowest instantaneous balance. Only R2G epochs show a prevalence of excitation (PP) or inhibition (TM), indicating superiority of REST for the validation of balanced network models.Support: DFG SPP1665 DE2175/2-1 & GR1753/4-2; DFG 368482240/GRK2416; EU Grants 720270 & 785907 (HBP).