Effect of anthropomorphism and perceived intelligence in chatbot avatars of visual design on user experience: Accounting for perceived empathy and trust

Abstract

Introduction: The rapid advancement of intelligent chatbots has transformed human-AI interaction, offering novel opportunities to enhance user experience (UX) through psychological and design interventions. However, the mechanisms by which chatbot design features influence UX remain understudied, particularly regarding the roles of emotional and cognitive mediators.

Methods: This study employed a 2 × 2 within-subjects experimental design with 160 participants to investigate the effects of anthropomorphism (high vs. low) and perceived intelligence (high vs. low) in chatbot avatars on UX. Structural equation modeling (SEM) was utilized to analyze the mediating roles of perceived empathy and trust in this relationship.

Results: Direct effects of anthropomorphism and perceived intelligence on UX were nonsignificant. However, their combined influence was significantly mediated by perceived empathy and trust (β = 0.48, *p* < 0.01). Specifically, highly anthropomorphic avatars correlated with elevated empathy (β = 0.32) and trust (β = 0.27), which in turn improved UX.

Discussion: These findings underscore the importance of emotional engagement over mere intelligence in designing effective chatbots. This research contributes unique insights into the complex mechanisms governing user interactions with intelligent chatbots, emphasizing the need for design strategies that prioritize emotional connections and cognitive ease.