Numerous learning rules have been devised to carry out computational tasks in various neural network models. However, the rules for determining how a neuron integrates thousands of synaptic inputs on the dendritic arbors of a realistic neuronal model are still largely unknown. In this study, we investigated the properties of integration of excitatory and inhibitory postsynaptic potentials in a reconstructed pyramidal neuron in the CA1 region of the hippocampus. We found that the integration followed a nonlinear subtraction rule (the Cross-Shunting Rule, or CS rule). Furthermore, the shunting effect is dependent on the spatial location of inhibitory synapses, but not that of excitatory synapses. The shunting effect of inhibitory inputs was also found to promote the synchronization of neuronal firing when the CS rule was applied to a small scale neural network. © Springer-Verlag Berlin Heidelberg 2006.
CITATION STYLE
Wang, X. D., Hao, J., Poo, M. M., & Zhang, X. H. (2006). How does a neuron perform subtraction? - Arithmetic rules of synaptic integration of excitation and inhibition. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 3971 LNCS, pp. 7–14). Springer Verlag. https://doi.org/10.1007/11759966_2
Mendeley helps you to discover research relevant for your work.