Spiking phase control in synaptically coupled Hodgkin–Huxley neurons

The spiking phase, defined as the time instant of action potential relative to a rhythmic signal, is a critical characteristic of information processing in brain circuits. Realistic neuron network models must, therefore, be able to reproduce dynamics in which the neuronal relative spiking phase could be sustained around a desired value in a broad interval through an appropriate neural plasticity mechanism. Many short- and long-term synaptic plasticity models have been proposed and used extensively, but the relative spiking phase dynamics have been largely overlooked in mainstream conventional plasticity models. In this work, we examined the classical biophysically relevant model of Hodgkin–Huxley neurons equipped with an excitatory synapse capable of realizing different types of plasticity, including short-term and long-term plasticity. The focus was to understand the impact of the various known forms of plasticity on the relative spiking phase dynamics. We found that two classical and most popular plasticity mechanisms, namely the short-term synaptic plasticity (STP) and long-term plasticity in the form of spike-timing-dependent plasticity (STDP), cannot control the relative spiking phase. STP, whilst achieving phase-locking around some value of the relative spiking phase when depression is significant, does not allow a flexible choice of this value. STDP, when considered independently from STP, fails to achieve any phase-locking. In contrast to these standard plasticity mechanisms, we found that STDP accounting for other non-neuronal biological processes (associated with glia and neuronal matrix) can effectively control and sustain the relative spiking phase in a broad range of values. This allows the generation of spiking patterns within the spiking neuron networks with a desired space–time distribution of spikes, which is essential for motor control tasks in cerebellar neural networks.