| Abstract [eng] |
Although chatbot anthropomorphism and its impact on consumer response has recently received a lot of attention from researchers, there is no consensus in the research on whether chatbot anthropomorphism positively or negatively influences consumers' perceptions and intentions (Blut et al. 2021). Different aspects of anthropomorphism, such as visual cues (Stroessner, Benitez, 2019; Go, Sundar, 2019; Youn, Cho, 2023; Pavone et al., 2023), psychological or emotional factors (Lee et al., 2023; Zhang et al., 2021), can positively influence the interaction between the AI and the user. However, in justifying the negative impact of conversational anthropomorphism on user response, researchers have focused more on the emotional response of users to AI. Many scholars refer to the Uncanny Valley Theory (Mori, 1970), which inherently has a negative perception of anthropomorphism. Anthropomorphised robots in research evoke negative feelings in consumers, such as feelings of creepiness (Rajaobelina et al., 2021), eeriness (Mende et al., 2019) and strangeness (Fukawa et al., 2023; Kim et al., 2019), and reduce trust, loyalty (Wang, 2023) and satisfaction (Crolic et al., 2022). Although the Uncanny Valley Theory (Mori, 1970) is based on the negative effect of anthropomorphism on consumer response, the theory states that increasing human similarity (anthropomorphism) leads to a more positive response up to a point where the similarity is excessive and causes feelings of discomfort or eeriness. As this work focuses on chatbots that have not yet reached a high level of anthropomorphism technologically and are not particularly human-like (anthropomorphism is expressed through text-based verbal communication and avatar), the Uncanny Valley Theory can be used to justify the positive effect of chatbot anthropomorphism on consumer response. Consumer response is used in the project to describe the constructs of consumer satisfaction, emotional response, attitudes towards chatbots and purchase intention. The results of the study showed that the chatbot anthropomorphism has a positive impact on consumer satisfaction, emotional response and attitudes towards chatbots. It also found that the relationship between chatbot anthropomorphism and consumer response is not moderated by the emotional tone of the situation. However, it was found that the emotional tone of the situation influences as a separate independent variable consumer satisfaction, emotional response, attitudes and purchase intentions. The findings also suggest that perceptions of chatbot anthropomorphism vary across countries. Moreover, satisfaction varies by country and age, emotional response and attitude by financial situation, and purchase intention by gender, country and age variables. |
