Would You Buy on Impulse During E-Commerce Digital Human Avatar Live Shopping? DHA Characteristics and Consumer...

Digital human avatars (DHAs) are emerging as a new type of intelligent shopping agent in livestream e-commerce; however, existing research has not fully explained how their characteristics influence impulse buying through consumer experience mechanisms. Drawing on the Stimulus-Organism-Response (S-O-R) framework, this study conceptualizes the interactivity, empathy, immersion, and vividness of e-commerce digital human avatars (EC-DHAs) as external stimuli and visual appeal, consumer-perceived arousal, and consumer engagement as internal experiential mechanisms, and examines their relationships with consumer impulse buying intentions. A sequential explanatory mixed-methods design was adopted. In the quantitative phase, purposive sampling was employed, and all participants were exposed to the same standardized EC-DHA livestream stimulus. Based on 467 valid questionnaires, partial least squares structural equation modeling (PLS-SEM) was used to examine the structural relationships, followed by artificial neural network (ANN) analysis to assess the relative predictive importance of key factors. In the qualitative phase, purposive sampling was used to recruit 21 participants for semi-structured interviews, and a grounded theory-informed coding procedure was applied to contextualize and further interpret the quantitative findings. The results show that interactivity and vividness are significantly and directly associated with consumer impulse buying intentions, whereas the direct relationships of empathy and immersion with impulse buying intention are not significant. Instead, empathy and immersion are primarily indirectly associated with impulse buying intention through experiential mechanisms, including visual appeal, consumer-perceived arousal, and consumer engagement, underscoring the important role of consumer experience in linking EC-DHA characteristics to impulse buying intention. The ANN results further identify visual appeal, vividness, and interactivity as key predictors of impulse buying intention. Overall, different EC-DHA characteristics exhibit distinct direct and indirect pathways to consumer impulse buying intentions. These findings clarify the role of consumer experience mechanisms in these relationships and provide practical implications for the visual presentation, interactive feedback, emotional expression, and consumer guidance of EC-DHAs in livestream e-commerce.