Bottom-up and Top-down Inputs Drive the Variability of Cortical Neurons

Summary

Date: 
February 27, 2015 - 11:00am
Location: 
NW 353
About the Speaker
Name: 
Camille Gomez-Laberge (Born/Kreiman Labs)

Neurons in the cerebral cortex respond inconsistently to a repeated sensory stimulus. The origin and nature of this variability remain poorly understood yet have important implications for the neural code. We found that reversibly inactivating sources of either bottom-up or top-down input to cortical visual areas in the alert primate reduced both spike train irregularity and trial-to-trial variability of single cortical neurons. This reduction in variability can be described by a simple model where a fraction of the pre-synaptic inputs are silenced, provided that there exist temporal correlations primarily within, but not between, excitatory and inhibitory pools of inputs. A large component of the variability of cortical neurons can therefore be ascribed to synchronous input produced by signals arriving from multiple sources.