Emotional stimuli have two opposite effects on memory. On one hand, emotional stimuli improve memory; on the other hand, they induce false memories as well. Much research examined the effects of emotional valence and arousal on false memory, but no consensus has been reached yet. Some researchers found that emotional arousal impacted false memory to a larger degree while others preferred emotional valence. The present study seeks to address this conflict and look into different influences that emotional stimuli would cause on false memory compared to true memory. To the authors' knowledge, the controversy of emotional arousal and valence might result from a previous neglect in analyzing temporal components of false memory. The present study hypothesized that emotional stimuli would have different impacts on false memory at different points along its time course. Meanwhile, to understand false memory thoroughly, it's necessary to compare it with true memory from both behavioral and neuroscience perspectives. The present study used a within-subjects design and altogether 19 undergraduate students were recruited to participate. The study selected pictures from the International Affective Pictures System (IAPS) as emotional stimuli and used the DRM paradigm. There were 4 kinds of pictures, they were positive-high arousal, positive-low arousal, negative-high arousal and negative-low arousal. Each kind composed 150 pictures and they were divided into 3 groups, such as repeated(both show in study and recall stage), similar lures(only show in study stage) and new ones(only show in recall stage). Similar lures were left/right mirror reversals of studied pictures. Another 50 pictures were showed only in study stage. All participants first study pictures ,then do 20 addition tasks, after that recall pictures to decide whether each picture showed in study stage or not. To provide a comprehensive analysis, the present study collected behavioral data, eye movement data and ERP data. ERP data were recorded by ESI-64 system of Neuro Scan and analysised by Scan 4.3.1. Eye movement data were recorded by Tobii 1750, sampling rate were setted as 50 Hz. Results revealed that false memory, as expected, indeed changed across its whole time course. The present study involved altogether four kinds of false memory (i.e., negative-low, negative-high, positive-low, positive-high) and ERP data showed that although with similar starts, they ended up with different patterns, which were distributed mainly over the anterior regions of the brain. In addition, the differences between false memory and true memory proved salient. Regarding behavioral data and eye movement data, participants took longer to respond to false memory compared to true memory, and also displayed a larger amount of both fixation duration and count. In conclusion, it was found that both emotional arousal and valence could promote false memory but arousal took effect at an earlier time. Moreover, the differences between false memory and true memory were illustrated at the neural level. More specific, false memory was processed more deeply than true memory at earlier time points, showing that false memory required deeper processing to recognize stimuli; this indicated the “unfamiliar” effect of false memory in eye movement. In addition, false memory was processed to a greater degree than true memory at later points, impling that false memory took a larger amount of mental processing(e.g., information searching and comparing) to complete a final decision-making. Logically, false memory and true memory proved distinct temporally: false memory compared to true memory took longer to finish its formation at earlier time points and then its recall at later points.