Characteristic of signal response in coupled inferior olive neurons with Velarde-Llinás model

In the cerebellum, the function for motor control is obtained in the cerebellar learning process, where the synaptic weights between the parallel fibers and the Purkinje cells are adjusted by long term depression (LTD) according to the error signal from inferior olive (IO) neurons. Schweighofer et al. have demonstrated that chaotic irregular firing can be produced by moderate electrical couplings in the IO network, and rich error signals for efficient cerebellar learning may be allowed by this chaotic effect in the computer simulation for their Hodgkin-Huxley type compartmental model. In this paper, focusing on chaotic behavior in the IO network consisting of Velarde-Llinás IO neuron model, we investigate the signal response in both cases of chaotic resonance (CR) and stochastic resonance (SR) under the conditions for real IO neuron, i.e., sustaining low firing frequency, asynchronous firing and sub-threshold oscillation. It is confirmed that CR can meet with these conditions, but cannot almost all SR.