Bayesian Item Response Model with Condition-specific Parameters for Evaluating the Differential Effects of Perspective-taking on Emotional Sharing
- Keishi Nomura, Life Sciences/Graduate School of Arts and Sciences/Yotsumoto Lab, The University of Tokyo, Meguro-ku, Tokyo, Japan
- Aiko Murata, NTT Communication Science Laboratories, Nippon Telegraph and Telephone Corporation, Atsugi, Kanagawa, Japan
- Yuko Yotsumoto, Life Sciences/Graduate School of Arts and Sciences/Yotsumoto Lab, The University of Tokyo, Meguro-ku, Tokyo, Japan
- Shiro Kumano, NTT Communication Science Laboratories, Nippon Telegraph and Telephone Corporation, Atsugi, Kanagawa, Japan
AbstractIt is known that perspective-taking helps humans recognize another’s emotional state on an individual basis. Here, we investigated how perspectives influence emotional sharing, namely the act of understanding mood, or a relationship between other people in a multiparty conversation. In order to capture the effects of perspectives on sensitivity and bias in responses, we introduced condition-specific parameters in a Bayesian item response model. The model revealed that interlocutors are more sensitive and biased to emotional incongruency when they give ratings for a pair including themselves than that excluding them. This relationship holds for observers who did not participate in the conversation and took the respective perspectives. The findings support the assimilating effects of perspective-taking through which people can perceive mood as the target does.