The Effect of the Degree of Multimodal Information Explanation by AI Streamers on Consumers’ Purchase Intention: The Moderating Role of Product Type

Abstract

With the increasing adoption of AI virtual streamers in livestream commerce, product presentations are undergoing a fundamental transformation—from linear, host-led explanations toward interactive, AI-driven formats in which intelligent agents integrate multimodal information and actively scaffold user understanding. While prior research on multimodal communication in livestreaming has largely emphasized presentation formats or the number of modalities employed, it has paid limited attention to the extent to which AI virtual streamers intervene in organizing and explaining information within multimodal environments, and how such explanatory intervention shapes consumer decision-making. Addressing this gap, the present study conceptualizes and examines the degree of multimodal information explanation provided by AI virtual streamers. Drawing on a 2 (product type: utilitarian vs. hedonic) × 3 (degree of multimodal information explanation: low, medium, high) within-subject experimental design, we systematically investigate its effects on consumers’ purchase intention and technology acceptance, as well as the moderating role of product type. The results reveal that higher levels of multimodal information explanation significantly enhance technology acceptance, while their effects on purchase intention are contingent upon product type. Specifically, product type moderates the relationship between explanatory depth and purchase intention, whereas no significant moderating effect is observed along the technology acceptance pathway. By shifting the analytical focus from modality configuration to AI-driven explanatory intervention, this study extends the theoretical framework of multimodal communication in livestream commerce and advances understanding of how explanation depth functions as a critical mechanism in facilitating user cognition and shaping consumption responses. These findings provide actionable implications for optimizing explanation strategies of AI virtual streamers and designing more effective multimodal content in livestreaming contexts.

Keywords: AI Streamers, Multimodal Information Explanation, E-commerce Live Streaming, Purchase Intention, Product Type

DOI: 10.54941/ahfe1007503