Objective: Cervical disease, a type of chronic pain, can greatly impact quality of life. Traditional Chinese tuina, a form of therapeutic massage and manipulation, has been shown to be effective in relieving pain and other symptoms in patients with chronic neck pain. This study applied functional magnetic resonance imaging (fMRI) to explore the features of the resting state network of patients with chronic neck pain caused by cervical radiculopathy, and how tuina affects the causality between intrinsic brain networks. Methods: Using Granger causality analysis, effective connectivity of brain networks of 10 patients with chronic neck pain was compared with 10 healthy control subjects. Resting state fMRI data were using magnetic resonance scanning. Cervical spondylosis symptom scores were evaluated before and after 4 weeks of tuina therapy. Independent component analysis was applied to extract the specific networks related to sensation, execution, and cognition, including sensorimotor network (SMN), visual network (VN), auditory network (AN), anterior and posterior default mode network (aDMN, pDMN), left frontoparietal network and right frontoparietal network. Results: Compared with the control group, data from the treatment group revealed two major findings: before tuina therapy, SMN had a profound influence on aDMN and AN greatly affected pDMN; however, after 4 weeks of tuina therapy, aDMN and SMN showed reversed causality. Conclusion: Chronic neck pain caused by cervical radiculopathy may influence the DMN, which plays an important role in emotion, cognition, and memory, by stimulating the sensory afferent network. Tuina not only significantly relieves pain and discomfort, but also reverses the causality between aDMN and SMN.