Wild Narratives: Exploring the Effects of Animal Chatbots on Empathy and Positive Attitudes toward Animals

TL;DR

This paper introduces animal-perspective chatbots and experimentally tests how verbal, nonverbal, and identity cues shape users’ perceptions and empathy toward animals. Using a mixed-methods 2x2x2 design with 240 participants and a horse-narrative, the study finds that verbal cues significantly boost affective and cognitive empathy and prosocial intentions, while nonverbal cues (animal-emoji) can reduce cognitive empathy due to perceived artificiality. Perceived animal-like (horse-like) identity also enhances empathy, attitudes, and willingness to act, suggesting authenticity in animal narratives is key. The findings offer design guidance for conservation-oriented chatbots and highlight the nuanced impact of nonverbal cues on empathy, informing future research on non-human agent design and public engagement with wildlife welfare.

Abstract

Rises in the number of animal abuse cases are reported around the world. While chatbots have been effective in influencing their users' perceptions and behaviors, little if any research has hitherto explored the design of chatbots that embody animal identities for the purpose of eliciting empathy toward animals. We therefore conducted a mixed-methods experiment to investigate how specific design cues in such chatbots can shape their users' perceptions of both the chatbots' identities and the type of animal they represent. Our findings indicate that such chatbots can significantly increase empathy, improve attitudes, and promote prosocial behavioral intentions toward animals, particularly when they incorporate emotional verbal expressions and authentic details of such animals' lives. These results expand our understanding of chatbots with non-human identities and highlight their potential for use in conservation initiatives, suggesting a promising avenue whereby technology could foster a more informed and empathetic society.

Figures (13)

• Figure 1: Example of emojis used as nonverbal cues
• Figure 2: Example screenshots of two versions of the chatbot interface. Left: Verbal, non-verbal, and identity cues enabled. Right: No cues enabled.
• Figure 3: Illustration of the study procedure. Our study includes eight experimental groups according to the combination of cues enabled (verbal cues, nonverbal cues, and identity cues).
• Figure 4: Results about the effects of chatbot design cues. The dots represent mean values, and the error bars show one standard error. The significance levels are labeled ($p<0.05$: *, $p<0.01$: **, $p<0.001$: ***). (a): Effect of verbal cues on affective empathy. (b): Effect of verbal cues on cognitive empathy. (c): Effect of non-verbal cues on cognitive empathy. (d): Effect of verbal cues on prosocial behavior intention.
• Figure 5: Results about the effects of perceived identity the chatbot. The dots represent mean values, and the error bars show one standard error. The significance levels are labeled ($p<0.05$: *, $p<0.01$: **, $p<0.001$: ***). (a): Effect on affective empathy. (b): Effect on cognitive empathy. (c): Effect on attitudes toward the animal. (d): Effect on prosocial behavior intention.