Prospective memory (PM) refers to the ability to remember to perform a delayed intention. In a classical event-based PM paradigm, individuals engage in an ongoing task and simultaneously remember to accomplish a previously encoded intention when the PM cue (associated with that intention) occurs. Focal and non-focal PM tasks differ in the extent to which the ongoing task encourages the processing of the PM cue features. A lot of studies have focused on the influence of emotional cues on prospective memory. However, the results of these studies are contradictory and it is not clear when or how emotion influences prospective memory. By reviewing the previous studies, we found focality of PM task was likely to be a mediator. We supposed the performance of PM task depended on whether the processing patterns in focal and non-focal PM tasks were consistent with the ways in which individuals processed emotional information. In general, negative information associated with increased engagement of perceptual processes, and positive information leading to enhanced recruitment of conceptual processes. If the processing way of the emotional cues was consistent with that in PM task, emotional cues would facilitate the PM. If there was an inconsistency, emotional cues would impair the PM. We conducted two ERP experiments to verify this hypothesis. In both experiments, the ongoing task was a one-back working memory task using neutral, pleasant, and unpleasant pictures as stimuli. Each experiment was composed of three PM sessions, which varied for the emotional valence of the PM cue (pleasant, unpleasant, neutral). Experiment 1 was a focal task, and participants were asked to quickly press the PM key immediately after making the decision for the ongoing task, whenever a pre-specified picture occurred. Therefore, perceptual processing was required in Experiment 1. Experiment 2 was a non-focal task, and the conceptual processing was needed. Participants were asked to pressed the PM key, whenever a picture belonging to the same category as it in the encoding phase occurred. The behavioral results showed that focal task performance was significantly better than non-focal task in the accuracy, but there was no significant difference between the emotional cues and the neutral cues in the two tasks. When compared with the ERP elicited by pictures of ongoing task, the grand-averaged ERP elicited by PM cues were characterized by the typical modulations related to PM. More specifically, PM cues were found to elicit two positive components: the frontal positivity, which occurred between 220 and 450 ms after stimulus onset; and the parietal positivity, which occurred between 400 and 1000 ms. The early frontal positivity is considered to reflect the detection of the PM cue and is associated with more automatic processes. It was smaller for positive than neutral and negative cues in focal task. In contrast, the frontal positivity for positive cues was larger in non-focused task. The parietal positivity reflects the retrieval of the intention from memory and is associated with postretrieval monitoring processes and with coordination between PM and ongoing responses. For parietal positivity, it was smaller for negative than positive and neutral cues in focal task. In contrast, it was larger for negative cues in non-focal task. In conclusion, the results confirmed our hypothesis that the focality of PM task could modulate the effect of emotional cues on prospective memory. The influence of emotional cues on prospective memory may be critically tied to the way in which negative and positive information was processed.