HINDMARSH–ROSE NEURONAL NETWORK WITH SPIKE-TIMING-DEPENDENT PLASTICITY DEMONSTRATES COORDINATED RESET NEUROMODULATION

Pathological rhythms in Parkinson's disease are characterized by excessive neuronal synchronization in the subthalamic nucleus (STN), where synaptic strengths evolve following the spike-timing-dependent plasticity (STDP) rule. The coordinated reset (CR) stimulation technique has been shown, both theoretically and clinically, to induce long-lasting desynchronization while avoiding side effects of conventional deep brain stimulation. However, the influence of STDP model parameters on the predicted efficacy of CR stimulation remains unclear. We used the Hindmarsh–Rose (HR) neuronal model to simulate synchronized spiking activity in the STN. Our 100-neuron model incorporates excitatory chemical synapses whose strengths are governed by an STDP rule. Inputs from unmodelled neurons are considered using either uniformly distributed white noise or Poisson noise; the former resulted in a single stable state of synchronized activity while the latter resulted in two stable states (bistability), one synchronized and one desynchronized. We applied CR stimulation with a rapidly varying sequence (RVS CR) to our model to explore how stimulation frequency and the number of stimulation sites affect its efficacy at desynchronizing the network. Specifically, the neuronal population was divided into subpopulations representing distinct physical sites of stimulation and phase-shifted stimuli were delivered to each subpopulation in a random sequence. With careful tuning of the depression-to-potentiation ratio in the STDP rule, our model can reproduce the behavior of larger networks used in the literature, thereby providing similar analytical utility at lower computational cost. Our findings highlight the critical role of STDP parameter tuning in reconciling computational predictions with experimental observations.