In order to explore the affective priming effect of emotion‐label words and emotion‐laden words, the current study used unmasked (Experiment 1) and masked (Experiment 2) priming paradigm by including emotion‐label words (e.g., sadness, anger) and emotion‐laden words (e.g., death, gift) as primes and examined how the two kinds of words acted upon the processing of the target words (all emotion‐laden words). Participants were instructed to decide the valence of target words, and their electroencephalogram was recorded at the same time. The behavioral and event-related potential (ERP) results showed that positive words produced a priming effect whereas negative words inhibited target word processing (Experiment 1). In Experiment 2, the inhibition effect of negative emotion‐label words on emotion word recognition was found in both behavioral and ERP results, suggesting that modulation of emotion word type on emotion word processing could be observed even in the masked priming paradigm. The two experiments further supported the necessity of defining emotion words under an emotion word type perspective. The implications of the findings are proffered. Specifically, a clear understanding of emotion‐label words and emotion-laden words can improve the effectiveness of emotional communications in clinical settings. Theo-retically, the emotion word type perspective awaits further explorations and is still at its infancy.