Fig. 3
From: Responses of Leaky Integrate-and-Fire Neurons to a Plurality of Stimuli in Their Receptive Fields
Four example ISI probability density functions, \(g(t)\), calculated with four methods using different grid sizes. The column panels show the four different ISIs, with the spike history indicated in the top (with different times axes) of each column, and the sinusoidal stimulus for the corresponding time periods. The panels in the four lower rows show solutions of the different PDEs and IEs using increasing grid sizes in each row. In the three lower rows, the density function from the panels above using the second Volterra method with high accuracy is plotted as the reference line. As expected, the solutions become less accurate as the grid size increases. The second row from the bottom, indicated with a star in the upper right corner, shows the grid size used for estimation in the main analysis, which leads to decent approximations for all four methods