Abstract Prospective memory represents the ability of remembering to carry out an intended action in the future. In a typical prospective memory experiment, participants are required to engage in two tasks: the ongoing task and prospective task. According to the nature of the prospective memory cues, there are two types of prospective memory: time-based and event-based prospective memory. There are two kinds of processes that can support prospective memory: preparatory attentional processes and memory processes theory (PAM) and multiple processes theory. The PAM assumes that prospective memory retrieval requires resource-demanding preparatory attentional processes, whereas the multiple process theory assumes that retrieval can also occur spontaneously. Aftereffects of prospective memory is a phenomenon that when prospective memory is finished or canceled, people still execute the prospective memory tasks repeatedly. The aftereffects of prospective memory can affect the quality of our lives. Lots of studies shows that aftereffect of prospective memory does exist in various conditions, but the processing mechanism of the aftereffect is controversial. The viewpoint of spontaneous extraction assumes that when prospective memory is completed, people are still able to spontaneously extract the original prospective memory task. The view of inhibitory states that when the cognitive resources used to suppress the original prospective memory are insufficient, people will make commission errors. The view of monitoring hold that people will continue to monitor the appearance of cues of prospective memory after their completion of tasks. Therefore, they are easy to make mistake when meeting the original clues. This study is to explore that which point is more reasonable. Experiment 1 adopted a single factor design. A total of 133 college students (26 baseline condition, 24 nonsalient/low frequency, 29 salient/low frequency, 27 nonsalient/high frequency, 27 salient/high frequency) participated the experiment. Experiment 2 adopted 2（cue saliency：nonsalient、salient）×2（cognitive loading：high、low） design. A total of 107 college students (24 nonsalient/low loading, 29 salient/low loading, 26 nonsalient/high loading, 28 salient/high loading) participated the experiment. Participants were asked to respond by pressing the key (F or J),but when meeting the prospective memory cues they should press the specified key (SPACE). During the course of experiments, Ongoing task and PM task were done simultaneously by participants. Participants’ behavioral data was recorded by computer automatically. The results showed that participants would make more commission errors in cue salient condition. Besides, in the four experimental conditions (nonsalient/low frequency, salient/low frequency, nonsalient/high frequency, salient/high frequency) the speed of ongoing tasks was slower than baseline condition after completion of prospective memory. In addition, no matter whether the clue was significant or not, participants would make more commission errors in high cognitive loading. Beyond these points, participants made the slowest reaction to the original prospective memory cues. The results shows that the appearance of aftereffect of prospective memory will occupy additional cognitive resources. What’s more, the cognitive resources are used to restrain primary prospective memory tasks. The process of inhibition is through the whole process of the experiment.