Consumers' Emotional Responses to AI-Generated Versus Human-Generated Content: The Role of Perceived Agency, Affect and Gaze in Health Marketing

Combining the theory of planned behavior, anthropomorphism and affect, this study examines the role of perceived agency and affect in investigating the effects of AI-generated versus human-generated content in the context of the COVID-19 vaccine uptake. We conducted a 2 (agency: chatbot versus human) x 3 (affect: anger, embarrassment, neutral) between-subjects lab experiment. Eye-tracking was used to test attendance to the affect-inducing stimuli. Findings revealed that participants preferred human agents over chatbots to vent their anger. However, when embarrassment was evoked, they preferred chatbots to avoid being judged by others. Anger elicited significant indirect effects on vaccine intentions. The eye-tracking data confirmed the effectiveness of visual stimuli in affect elicitation. Findings contribute most directly to the emerging literature on the complexity of human-machine interactions in health marketing. Together, findings demonstrate that marketers and policymakers must consider context-matching of emotional appeals and perceived agency when designing strategic campaigns.