Circuit Dynamics and Computational Neuroscience I.1.a Single-cell modeling Monday AM + Wednesday AM

2457 - Dissecting dynamic gain reveals differential contributions of subthreshold impedance and spike generation dynamics


Abstract Body

The dynamic gain function (DGF) quantifies the input-output-function of neurons driven by background fluctuations to an in-vivo-like operating point. Recent studies suggest that a wide bandwidth of the DGF, which reflects precise spike-timing, has functional importance, might be under evolutionary pressure (Lazarov et al. 2018) and may even be related to cognitive performance. The DGF shape is influenced by the nano-physiology of the axon initial segment, dendrite size (Eyal et al. 2014) and axonal resistance (Brette et al. 2013). Relative contributions of these factors are difficult to disentangle, as DGF entails signal transformations from input current to membrane voltage to firing probability.

Here, we decompose the DGF into its spike-generator component and a subthreshold impedance in neurons with various morphologies and firing patterns at different developmental stages and with perturbed axonal ion channel density. We find that the transformation from input current to membrane voltage is shaped by neuronal morphology and insensitive to correlations in the background fluctuations. Transformation from somatic voltage to firing probability has strong high-pass characteristics with a sharp cut-off, whose position is relatively insensitive to morphology, but affected by axonal perturbations. DGFs of all neurons showed improved encoding of high frequencies when background fluctuations was dominated by low frequencies. This ‘Brunel-effect’ (Brunel et al. 2001) manifests mainly in the supra-threshold spike-generation gain. The subthreshold impedance contributes by attenuating low frequencies.

Our results suggest that the active conductances controlling spike initiation play a crucial role for the DGF bandwidth and boosting of high frequencies.