Reliability and stochastic synchronization in type I vs. type II neural oscillators

Neural reliability and stochastic synchronization are remarkable features of real neurons with important consequences for neural computation. Both phenomena are general properties of any device with a resetting threshold, such as neurons. However, certain characteristics of the single neuron dynamics can notably improve neural reliability and stochastic synchronization. Here we show that neural resonators synchronize more reliably and more robustly than neural integrators. This suggests that neurons conveying sensory information in a spike-timing code are likely to be resonators, as supported by our recent studies on reliability and stochastic synchronization in the olfactory bulb.