595398.fig.001a
(a)
595398.fig.001b
(b)
595398.fig.001c
(c)
595398.fig.001d
(d)
595398.fig.001e
(e)
595398.fig.001f
(f)
595398.fig.001g
(g)
595398.fig.001h
(h)
Figure 1: Analysis of AP clusters and estimation of AP time precision in the CA3 neuron. The effect of the channel noise. (a) Overlapped output spike trains generated by identical input presented to neuron times. (b) Extracted and overlapped shapes of APs-forming AP clusters. Values depicted on the top of each cluster show the —the measure of AP time precision calculated as a standard deviation of AP times found in the AP cluster. (c) Shapes of -AP cluster’s mean voltage course ending with triggering an AP. (d) of consecutive AP clusters for 3 s duration stimulation. (e) Visualization of spike train groups. Based on the Hamming distance between spike trains elicited by repeating the same input, the trains were sorted according to their firing pattern into different spike train groups ( is the number of spike trains in each group). In this panel three groups are depicted by three lines differing in color. The ordinate shows the order of APs and the abscissa the firing time of AP. Despite the fact that each group contains different numbers of fired APs (some APs failed to be fired in individual trials), firing times in-between groups are well aligned. This indicates an interesting preference of the spiking mechanism to follow some pattern more frequently than others. (f) Aligned courses of of different AP clusters. (g) Correlations between voltages of all AP clusters at given times , and all clusters’ were calculated under the channel noise effect. Correlation curves are delimited with their 0.95 confidence intervals (indicated in red), obtained from 40 correlation curves obtained over 40 various discharge patterns of presynaptic neurons repeated times. (h) The histogram demonstrates the common distribution of ISIs in our model.