Obsessive-Compulsive Disorder(OCD) is characterized by persistent and unwanted obsessions mostly accompanied by ritualistic compulsions. According to the World Health Organization, OCD is one of the top 10 causes of disability, manifesting its severe impact on quality of life. But the mechanism of OCD is still unclear. Some researchers suggested that the attention bias to threat-related stimuli may be the cause and maintenance of OCD. In the past few decades, many studies have investigated attention bias in OCD by using the emotional Stroop task, mostly relying on reaction time. However, those behavioral studies provide little information about the relative involvement of sensory versus cognitive processes in attention bias. Therefore, we use ERPs(Event Related Potentials), which can exquisite temporal resolution and sensitivity to emotional processing, to explore the time course of attention bias in high obsessive-compulsive tendency individuals (HOC). Meanwhile, there was an evidence that the component of attention bias in HOC is difficulty in disengaging attention away from threat. So, which ERPs component will be represented to reflect this information processing. In order to fulfill our curiosity in this two questions, we carried out this research. Accordingly, we obtained the behavioral and electroencephalogram (EEG) data from HOC using the emotional Stroop task. Brain electrical activity was recorded from 64 scalp sites using tin electrodes mounted in an elastic map(Brain Products GmbH, Gilchiing, Germany), with the reference electrodes placed on the left and right mastoids. The experiment was designed to discover the component of attention bias in HOC by manipulating stimulus types(OCD related negative words, generate negative words, neutral words) and different groups (HOC and LOC). Generate negative words and neural words were selected from previous studies(Moritz, et al., 2004; Thomas, Johnstone, & Gonsalvez, 2007), OCD related negative words were chosen by clinical psychologies or psychiatrists. Subjects were divided into two groups by PI (Padua Inventory) scores. The mean amplitude of nine electrodes were selected to analyzed, which was F3, Fz, F4, C3, Cz, C4, P3, Pz, P4. We analyzed the results using repeated-measures analyses of variance (rANOVA) with group as between-subjects variables, stimuli type as within-subjects variables. We observed that HOC shared the same reaction time with LOC. However, interaction effect is observed between group and stimulus type (F(2,29)=5.43, p=.007, ?p2=.15). Specifically, HOC displayed enhanced P2 amplitude, when processing OCD related negative words compared with neutral words(p=.046), while LOC showed enhanced P2 amplitude to general negative words competing with neural words. Moreover,compared with general negative words and neutral words, HOC manifested larger LPPs towards OCD related words. The results of the study support that HOC manifested attention bias to OCD related negative stimulus, and the component of attention bias is difficulty in disengagement away from threat, which occurs not only at sensory processing stage but also at strategic processing stage.