Designing Authentic Customer-Chatbot Interactions: A Necessary Condition Analysis of Emotional Intelligence and Anthropomorphic Features in Human-Computer Interaction
Md Irfanuzzaman Khan, Arun Kumar Tarofder, Sharmini Gopinathan, Ahasanul Haque
Abstract
This study offers a novel framework of perceived authenticity (PA) in chatbot-mediated service interactions by drawing on insights from Mind Perception Theory, Theory of Mind, the Authenticity Model of Computer-Mediated Communication, and Uncanny Valley Theory. The model integrates emotional and anthropomorphic cues by including perceived humanness, empathy, warmth, and humor. Survey data from 396 participants were analyzed using Partial Least Squares Structural Equation Modeling. The results indicate that empathy exerts the strongest influence on perceived authenticity, followed by perceived humanness and warmth, while humor plays a complementary role. PA significantly enhances trust, rapport, and satisfaction but has limited effect on frustration. Necessary Condition Analysis identifies minimum thresholds of key predictors needed to achieve high PA. Moderation analysis reveals that empathy and humanness are more effective for male-presenting chatbots, while humor enhances authenticity for female-presenting ones. The findings offer significant theoretical and practical implications in the domain of human -chatbot interaction.