Deep brain stimulation is a treatment for Parkinson's disease that uses electrical stimulation to modulate neural activity in order to reduce motor symptoms associated with the disease. The design of the electrical stimulation signal used is strongly linked to the efficacy of such a treatment. We present computational models of the brain structures impacted by Parkinson's disease which are modulated by electrical current injections from chronically implanted electrodes as a part of deep brain stimulation. This work demonstrates that a high frequency signal with small random deviations in pulse timing can simultaneously induce regularized activity associated with improved motor symptoms and permit greater variation in neural responses, which are necessary to encode neural information.