Rapid neocortical network modifications via dendritic plateau potential induced plasticity

Abstract

Learning in brains is associated with changes in neuronal network activity thought to be driven by synaptic plasticity. While recent work in the hippocampus has revealed some of the mechanisms involved there, less is known about how neocortical circuits adapt, especially during behavior. Here to determine if neocortical areas possess rapid plasticity mechanisms that could support online adaptations we used optical imaging and intracellular membrane potential (Vm) recordings to examine the activity of layer V neurons in a higher visual area of mice learning a task. The introduction of a novel rewarded stimulus resulted in a rapid modification of population activity that featured abrupt alterations in single neuron selectivity. Vm recordings revealed that both naturally occurring and experimentally-induced dendritic calcium plateau potentials (plateaus) rapidly alter the action potential (APs) output and Vm dynamics of neurons over many seconds of time around the plateau, in some cases from one trial to the next. Trains of high frequency APs had no effect. Finally, experimental inhibition during learning of the distal dendritic region responsible for initiating plateaus reduced the rate of population level adaptation. Our findings suggest that the deep layers of higher order visual cortex possess a rapid learning mechanism mediated by plateau-induced synaptic plasticity.