Emotion regulation involves deliberate, voluntary regulation and automatic, involuntary regulation. Automatic emotion regulation has been characterized as an unconscious goal-driven modulation of emotions, without paying attention to the process of regulating or engaging in deliberate control. Previous studies showed that when goals or norms referring to emotion regulation were primed, goal-relevant response or behavior can be activated without awareness. However, it remains unclear about the neural mechanism underlying the unconscious emotion regulation goal. Previous studies found that subliminal goal priming could change goal related behaviors. Meanwhile, attention allocation was a basic processing during emotion regulation. Furthermore, behavioral research found that automatic emotion regulation could decrease attention allocation to negative stimulus. Thus the present study aimed to examine whether subliminal priming of emotion regulation goals could change attention allocation in response to threatening information, and whether the amplitude of dot-locked P3 in the dot probe task would be modulated. A 2 (priming condition: regulation group vs. control group) × 2 (the congruency of cue and target: congruent vs. incongruent) mixed experiment design was carried out. Forty college students aged 17-25 years with normal or corrected-to-normal vision were recruited. Participants were randomly assigned to a regulation group (13 female, 21.75±1.92 years, 19-24 years) and a control group (14 female, M=20.35±2.13 years, 17-25 years). Participants in the regulation group were primed subliminally by emotion regulation goals. Participants in the control group were primed subliminally by the unrelated priming words. In the modified dot probe task, each trial started with a fixation cross in the middle of a computer screen for 500ms. Next, a priming word (describing an emotion regulation goal for the regulation group or an unrelated word for the control group) was presented for 20ms, preceded by a 100ms pre-mask stimulus and followed by a 100ms post-mask stimulus. Subsequently, a snake and a mushroom were presented simultaneously on either side of a fixation point for 500ms. Then, the stimuli disappeared from the screen and a dot target appeared in the location of one or the other picture. Participants were asked to respond to the dot target of a white point as quickly and accurately as possible by pressing F key = left and J key = right. The dot disappeared when the response was made. Then a 1000ms blank appeared after the response being finished. A total 144 experimental trials and 12 practice trials were used. EEG data were recorded and analyzed by NeuroScan system, we focused on the mean amplitudes of target-locked waveforms, including P1 during the time window of 80-140ms at the electrode of Oz, N2 during the time window of 260-340ms at the electrode of Fz, P3 during the time window of 300-500ms at the electrode of Pz. Priming (Between-subject factor, emotion regulation priming/ control priming group) × Congruency (Within-subject factor, congruent/ incongruent) mixed ANOVAs were conducted on reaction times and the mean amplitudes of P1, N2 and P3. Behavioral results showed that the interaction of group and congruency was significant. Reaction times of the incongruent condition were shorter than those of the congruent condition only in regulation priming condition. ERPs results showed that for the regulation priming group, the mean amplitude of P3 was larger in incongruent condition than in congruent condition. However, for the control priming group, the mean amplitude of P3 was smaller in the incongruent than the congruent condition. In addition, we found that the mean amplitude of N2 was larger in the incongruent than the congruent condition. The result of P1 showed no significant difference. These results suggest that subliminal priming of emotion regulation goal may attention withdrawal from fearful stimuli; further, increased posterior P3 in incongruent condition than in congruent condition may reflect the neural correlates of the subliminal regulation goal priming on the stage of attention.