Exploring the origin of oscillation dynamics of subthalamic nucleus and globus pallidus in basal ganglia plays an important role in understanding the pathogenesis of Parkinson′s disease.In this paper,a more complete neural network model for Parkinson′s disease is established.The model contains the cortex with excitatory and inhibitory neurons,basal ganglia composed of striatum,subthalamic nucleus,external and internal segment of globus pallidus and thalamus.First,the dynamic results obtained through numerical simulation of the model are consistent with the biological experimental results related to the oscillations in the subthalamic nucleus and the globus pallidus.Secondly,the effects of synaptic connection weights associated with subthalamic nucleus and globus pallidus on their oscillation dynamics were discussed.Our results show that the oscillation in the subthalamic nucleus will disappear for stronger excitatory connection weights from the striatum and weaker inhibitory connection weights from external segment of globus pallidus.Also,oscillation in the globus pallidus will appear for both stronger excitatory and inhibitory connection weights from the subthalamic nucleus the striatum,respectively.It is hoped that the results of this study will play a guiding role in the treatment of Parkinson′s disease.