|
In pyramidal and fast spiking neurons of the rat somato-sensory cortex, firing rate adaptation in response to fluctuating, in vivo-like input occurs at multiple time scales (from a few milliseconds to many seconds). A simplified theory captures the adapted input-output curve of these neurons together with some elements of the firing rate dynamics. This allows one to build rate models whose parameters are directly related to the parameters of the original spiking model. The theory depends on mild assumptions usually fulfilled under cortical conditions and can be applied to any model of adapting spiking neuron whose response function is known. It also predicts some degree of independence between gain reduction by firing rate adaptation and sensitivity to input fluctuations. Under biophysically realistic conditions, an extension to the case of time-varying inputs is possible. The time-dependent model can predict the time-varying activity of a large population of adapting neurons also in the presence of fast input transients. The time constants governing the dynamics of the network are directly related to the time constant of excitatory and inhibitory post-synaptic current. |
|