|
Spike-frequency adaptation (SFA) is a phenomenon present in many excitatory cortical neurons. While it may or may not have a dedicated functional role, the fast adaptation of the current frequency response certainly affects dynamics in cortical networks. Therefore, here we investigate the role of SFA for cortical dynamics and focus on the primary visual cortex (V1) as a model for a sensory system. More specifically, we investigate an extension of our network model of contextual effects in V1 [1,2], where the local connections between excitatory and inhibitory neurons are strong (compared to feedforward and other intra-cortical inputs). It has been shown that in such a ''balanced'' regime the external modulations (in our case from neurons in the receptive field surround) can have paradoxical effects [3,4] in the sense that an increased input to the local inhibitory neurons leads to a reduction of both the locally coupled excitatory and inhibitory neurons. Based on such responses one may want to infer the operating regime and local connectivity within V1. We investigate how these effects depend on the strength of SFA and hence may differ between initial transient and later sustained responses. In a simplified model, we take the gain of linear excitatory neurons as the key parameter affected by SFA and find that even small changes in the gain can switch between qualitatively different responses changes (paradoxical vs. non-paradoxical). Numerical simulations of recurrent networks with non-linear firing rate functions and explicit models of SFA confirm these predictions. Besides emphasizing that inferences about local network connectivity based on such response changes should be done with caution, our modeling study also suggests that in states with reduced SFA (as in the presence of increased levels of acetylcholine in states of sustained attention and alertness) the signaling of salient image locations in V1 is boosted. [1] Schwabe et al., J Neurosci, 2006. [2] Schwabe et al., Neuroimage, 2010. [3] Tsodyks et al., J Neurosci, 1997. [4] Ozeki et al., Neuron, 2009. |
|