CAN-5e: Chinese affective norms for five discrete emotion categories

The human affective space is predominantly described by two theoretical accounts: the dimensional perspective and the discrete-emotion approach. While numerous dimensional datasets (e.g., valence and arousal norms) have been published, there is a dearth of resources grounded in the discrete-emotion approach, particularly for the Chinese language. To address this gap, the present study expands on the initial norms of Yao et al. (2017) to establish CAN-5e, a new normative dataset that provides affective ratings of 1,100 Chinese words along five discrete emotions (happiness, anger, fear, disgust, and sadness). A sample of 525 native Chinese speakers volunteered for the rating tasks, with each participant evaluating 220 words for one discrete emotion using a five-point Likert scale. The results revealed significant relationships between discrete emotions, affective dimensions (valence and arousal), and key psycholinguistic variables (familiarity, concreteness, imageability). Furthermore, although ratings for all discrete emotions were strongly correlated between male and female participants, the two groups showed differential susceptibility to specific emotions. Then, distributional analysis distinguished between “pure” words (strongly tied to one emotion) and “mixed” words (connected to more than one), suggesting that words associated with negative emotions often overlap across categories, whereas happiness-related words demonstrate the strongest category specificity. Finally, linear regression analyses showed that discrete emotions could not predict lexical decision performance (i.e., response time, accuracy). Collectively, the CAN-5e dataset complementarily extends existing Chinese affective corpora and allows for designing new experiments examining the role of discrete emotions in language processing within the discrete-emotion framework, thereby contributing to a deep understanding of the relationship between language and emotion.