Studying the economy of energy expenditure in a large balanced spiking neuron network

The link between neural activity and energy flows forms the basis of several forms of functional neuroimaging. Since the biophysics of neurovascular interactions is extremely complex, it would be worthwhile to investigate this question using simple computational models. Since neural networks are models of computation in the brain it would be interesting to study energy utilization in these models under various conditions of operation. In this paper, we study energy utilization in large, sparse spiking neuron network containing a mixture of excitatory and inhibitory neurons. In such a network, a balanced state, in which the total excitation and inhibition are designed to cancel out, has been considered to reflect the situation in real cortical networks. In our simulations, the network in balanced state is found to correspond to a state of minimum energy consumption very often. Such a state is also associated with low regularity of firing of individual neurons, and only moderate levels of synchrony across the network.