Abstract：The interaction between recognition memory and emotion has become a research hotspot in recent years. Dual process theory posits that familiarity and recollection are two separate processes contributing to recognition memory. It has been hypothesized that item retrieval is based on both familiarity and recollection, whereas source retrieval relies solely on recollection. Event related potentials(ERPs) studies on recognition memory have identified two old/new effects that refer to familiarity and recollection, respectively. The mid-frontal old/new effect, occurring at approximately 300~500ms over frontal electrode sites, is attributed to familiarity. The parietal old/new effect, approximately 500~800ms maximal over parietal electrode sites, reflects recollection. In source memory paradigm, when both the item and source are remembered accurately, it is inferred that the memory decision is based on recollection. When correct item recognition is accompanied with an incorrect source attribution, it is posited that the memory decision is based on familiarity. Recent studies have discussed the interaction between source memory and emotion under “emotional context.”, and many studies have supported the theory that source retrieval is mediated by valence of emotional context. In the present study, we used ERPs techniques to explore the cognitive and neurological mechanisms of source retrieval under different emotional contexts while applying a source memory multiple-task paradigm. Subjects in our experiment were seven men and ten women. In the study phase, participants were asked to study a series of Chinese characters (item) with emotional pictures of three valences (source) appearing as the background of the items. Meanwhile, participants were told to press a button when an animal name was shown. In the test phase, only the Chinese characters were presented. Participants were instructed to press one of four buttons on the response box to indicate whether the word was presented at encoding phase with a neutral background, a positive background, a negative background, or whether it was new. It was found that ERPs were more positive for old items of both successful and unsuccessful source retrieval than new items under all three emotional contexts between 300 and 500 ms after stimulus onset (old/new effect). The old/new effect of successful source retrieval did not differ from unsuccessful source retrieval under the neutral context, but it was larger for successful source retrieval compared to unsuccessful source retrieval under emotional context (positive and negative). At 500-650 ms, there were old/new effects for successful source retrieval under all three emotional contexts, and there were no significant differences between unsuccessful source retrieval and new items. Observations gather from similar scalp topographies of successful source retrieval and unsuccessful source retrieval revealed further supporting evidence. Moreover, emotion appeared to have a prominent effect on successful source retrieval during this time window. ERP Amplitudes of items presented with positive context were greater than those of items with negative and neutral context. In addition, results of intracranial source estimation analyses showed different brain regions related to retrieval of negative and neutral context. Our results show that item retrieval and source retrieval are based on two different cognitive processes: familiarity and recollection, respectively. Furthermore, our findings suggest that successful source retrieval and unsuccessful source retrieval may be supported by a common mechanism. Finally, there is evidence to indicate that both item retrieval and source retrieval are influenced by the emotional valence of context. Our results show that item retrieval and source retrieval are based on familiarity and recollection respectively, which are two different cognitive processes. Successful source retrieval and unsuccessful source retrieval might be supported by a common mechanism. It is also indicated that both item retrieval and source retrieval are influenced by the emotional valence of the context.