PURPOSE To investigate the modifications of resting state fMRI(rfMRI) in complex partial seizures (CPS) epilepsy patients employing regional homogeneity(ReHo)、the amplitude of low frequency fluctuation (ALFF) and the functional connectivity(FC) techniques. METHOD AND MATERIALS Thirty-seven right-handed CPS patients were recruited and investigated in the comparison with 37 matched in age 、gender and education background controls. All subjects underwent MR scanning on a 3.0 Tesla scanner (GE-Discovery 750, Milwaukee, US.). Functional MRI scanning was performed using GRE-EPI sequences (FOV 24 cm×24cm, 64×64 matrix, flip angle 90°, TR 2000 ms,TE 30 ms , whole brain coverage,38 oblique axial ,4mm slices thickness and 0mm inter-slice space). Resting-state scanning lasted for 512s, producing 256 brain volume data sets. The first 10 images were excluded due to T1 equilibrium effects. T1-weighted 3DBRAVO-sequence images (FOV 24 cm×24 cm, 256×256 matrix, whole brain coverage, flip angle 13°, TR 7.8 ms, TE 3.0 ms) were achieved to assess the anatomical images for the co-registration of fMRI data with standard space coordinates. ReHo, ALFF and FC were processed using REST and SPM8 software to compare the resting state function in the whole brain between two groups. RESULTS Medial temporal lobe and surrounded brain regions were observed to participate the interictal epileptiform discharges (IEDs), and the default mode network (DMN).The cerebellum of CPS patients was the most commonly damaged region resulted from the abnormal neural electrical activities. CONCLUSION ReHo,ALFF and FC can detect interictal epileptiform abnormality, and can be possibly applied as an additional non-invasive tool for the detection of epileptogenic foci. Furthermore,the alterations in amplitude play a central role in epileptogenesis. CLINICAL RELEVANCE/APPLICATION ReHo,ALFF and FC can detect abnormal BOLD signals, localize the epileptic zones, and may be applied to evaluate the pathophysiological mechanisms of epilepsy.