Excessive synchronous oscillation activities appear in the brain is a key pathological feature of Parkinson’s disease, the mechanism of which is still unclear. Although some previous studies indicated that \(\beta\) oscillation (13–30 Hz) may directly originate in the network composed of the subthalamic nucleus (STN) and external globus pallidus (GPe) neurons, specific onset mechanisms of which are unclear, especially theoretical evidences in numerical simulation are still little. In this paper, we employ a STN–GPe mean-field model to explore the onset mechanism of Parkinson’s oscillation. In addition to \(\beta\) oscillation, we find that some other common oscillation frequency bands can appear in this network, such as the \(\alpha\) oscillation band (8–12 Hz), the \(\theta\) oscillation band (4–7 Hz) and \(\delta\) oscillation band (1–3 Hz). In addition to the coupling weight between the STN and GPe, the delay is also a critical factor to affect oscillatory activities, which can not be neglected compared to other parameters. Through simulation and analysis, we propose two possible theories may induce the system to transfer from the stable state to the oscillatory state in this model: (1). The oscillation activity can be induced when the firing activation level of the population increases to large enough; (2). In some special cases, the population may stay in the high firing rate stable state and the mean discharge rate of which is too large to induce oscillations, then oscillation activities may be induced as the MD decreases to moderate value. In most situations, the change trends of the MD and oscillation dominant frequency are contrary, which may be an important physiological phenomenon shown in this model. The delays and firing rates were obtained by simulating, which may be verified in the experiment in the future.

大脑出现过度同步振荡活动是帕金森病的一个关键病理特征，其机制尚不清楚。尽管先前的一些研究表明\（\beta\）振荡（13-30 Hz）可能直接起源于由丘脑底核（STN）和外苍白球（GPe）神经元组成的网络，其具体的发病机制尚不清楚，尤其是数值模拟方面的理论依据还很少。在本文中，我们采用 STN-GPe 平均场模型来探索帕金森振荡的发生机制。除了\(\beta\)振荡，我们发现这个网络中还可以出现一些其他常见的振荡频带，例如\(\alpha\)振荡频带 (8-12 Hz)、\(\theta\)振荡频带 (4-7 Hz) 和\(\delta\)振荡频带（1-3 Hz）。除了 STN 和 GPe 之间的耦合权重外，延迟也是影响振荡活动的关键因素，与其他参数相比是不可忽视的。通过仿真和分析，我们提出了两种可能的理论可以诱导系统在该模型中从稳定状态转变为振荡状态：（1）。当种群的激发激活水平增加到足够大时，可以诱导振荡活动；(2)。在某些特殊情况下，种群可能停留在高放电率稳定状态，其平均放电率太大而无法诱发振荡，然后随着 MD 降低到中等值，可能会诱发振荡活动。在大多数情况下，MD和振荡主频的变化趋势是相反的，这可能是该模型中显示的重要生理现象。延迟和发射率是通过模拟得到的，未来可能会在实验中得到验证。