In selective attention tasks, the efficiency of processing a target stimulus in a given trial is often influenced by what happens in a previous trial. When a to-be-attended target on a current trial (the probe trial) matches the ignored or inhibited distractor on a previous trial (the prime trial), a response to the target is typically delayed compared with when the two stimuli are not associated with one another. This phenomenon is called the negative priming (NP) effect. Although there have been many studies demonstrating the factors that influence the manifestation of the NP effect, most of these experiments used the traditional paradigm, in which the target and distractor are presented simultaneously in both the prime and probe trials. These studies explore how target selection is achieved when the target is presented concurrently with one or multiple distractor stimuli in a display. The experiments reported in this thesis used a recently developed new paradigm, the NP paradigm under rapid serial visual presentation, or NP under RSVP (Wong, 2012), to explore target selection among temporally separated stimuli. In RSVP presentation, each stimulus is presented very briefly and at the same spatial location. In the present study, each trial couplet consisted of a prime and a probe stream. Each stream included a target, a unique coloured distractor, and several task-irrelevant neutral stimuli. Whereas the neutral stimuli were always letters, in different experiments, the target and distractor could be digits, number words in English, or number words in Chinese. The critical manipulation was the relationship between the probe target and the preceding prime distractor, which could be unrelated (the Control condition) or matched (the IR condition). When they were matched, they could be in the same language or in different languages. In four experiments, I explored the factors that might influence the manifestation of the NP effect under RSVP. In Experiment 1, a baseline experiment, the prime distractor was a digit, and it was identical in form to the probe target. A significant NP effect was found. In Experiment 2, the prime distractor was an English number word, but the probe target was a digit. A significant NP effect was found in accuracy, indicating that the NP under RSVP could be found even when the relationship between the prime distractor and probe target was conceptual rather than identical. In Experiment 3, the target and distractor were shown in two different representational forms. In the prime trial, the distractor was a Chinese number word, but the target was a digit. In the probe trial, the distractor was a digit, but the target was a Chinese number word. A significant NP effect was once again found, suggesting that participants could shift quickly from one representational form (digit) to another (Chinese number word) between the prime and probe trials and still show NP. Experiment 4 was a cross-language NP experiment with Chinese-English bilinguals. In half the trials (the L1-to-L2 trials), the prime distractor was a Chinese number word while the probe target was its English translation equivalent. In the other half of the trials (the L2-to-L1 trials), the prime distractor was an English number word while the probe target was its Chinese translation equivalent. A significant NP effect was found in the L1-to-L2 trials, but not in the L2-to-L1 trials. Taken together, these results extended the finding of previous research under RSVP. They also provided additional evidence for the revised distractor inhibition account of NP (Tipper, 2001).