Power-law scaling of synchronization robustly reproduced in the hippocampal CA3 slice culture model with small-world topology

Abstract

The hippocampal CA3 is a recurrent network included small-world topology. The percentage of co-active neurons in CA3 slice cultures is approximated by power-law. We show that the power-law scaling of synchronization is reproduced in the CA3 slice culture model where synaptic weights are log-normally distributed and balanced excitation/inhibition regardless of network topologies. However, small-world topology improves the robustness of the reproduction of the power-law scaling in the culture model. Power-law scaling is known as a sign of optimization of a network for information processing. These results suggest that CA3 may be robustly optimized for information processing by excitation/inhibition balance, log-normally distributed synaptic weights and small-world topology. © 2012 Springer-Verlag.