Stochastic neural firing properties in neurons of a cerebellar control system

The cerebellar system for the voluntary control of arm-hand movements involves a large number of different neuron types. These neurons are located both inside the cerebellum but also in extracerebellar brain structures, which provide processed motor and sensory information to the cerebellum. The various different neuron types have different morphology, receive different types, number and patterns of synaptic inputs and have different firing rates and kinetics. However, a common trait for all neurons is that their firing properties have clear stochastic components, which is evident in recordings from the neurons recorded in brain preparations when all synaptic inputs is removed (in vitro). Here, we take advantage of a unique, comprehensive database of the various neuron types present within the cerebellar arm-hand control system, recorded from the brain in vivo, to provide a comparative description of their spike firing patterns. Although the inter-spike intervals for most cell types in this system can be described by a simple type of distribution characteristic for stochastic neurons, it is only for a few exceptional cases that the consecutive inter-spike intervals are independent of each other. We conclude that the spike patterns of these neurons may be the result of multi-factorial sources of variability that include the patterns in the various synaptic inputs that neurons receive in vivo and the inherent stochasticity of spike generation.