Fig. 8: The sparse-wave network regime boosts spike inputs while the dense-wave network shunts.

a A 0.2 × 0.2 mm² pool in the sparse-wave network model received a 20 Hz Poisson spike train input for 10 ms aligned either to a period of depolarization (blue shaded region) or hyperpolarization (red shaded region) as defined by the spike-LFP phase relationship. The dark blue and red lines are the mean evoked firing rate after receiving the spiking input in either the depolarized or hyperpolarized phase respectively (light blue and red lines represent N = 40 individual trials). The black line is the firing rate of the neuron pool when no input was given. b Same as (a), but the inputs were delivered to the dense-wave network. The evoked responses were much weaker as the network shunted the currents evoked by the incoming spikes. c The response gain between the distributions of spontaneous and evoked activity across N = 40 presentations of spiking input. In the sparse-wave network (left bars), inputs during the depolarized state had larger relative responses as compared to inputs during the hyperpolarized state (3.09 ± 0.09 compared to 2.11 ± 0.05 mean ± standard deviation; p = 3.57 × 10⁻⁸, two-tailed Wilcoxon’s signed-rank test). In contrast, the dense-wave network (right bars) responses did not differ in their response gain during the hyperpolarized and depolarized states (1.10 ± 0.04 and 1.11 ± 0.05, respectively; p = 0.20).