Abstract

Virtual influencers (VIs) have become a dominant force in digital marketing due to their flexibility in visual design and ability to embody diverse brand meanings. Unlike human influencers, VIs can be intentionally crafted to convey abstract attributes through metaphorical design, such as futurism, sustainability, or innovation, enabling brands to communicate complex ideas in more vivid ways. In practice, brands widely rely on such metaphorical representations, where the VI serves as a symbolic embodiment of product attributes. Despite their widespread use, little is known about whether different types of metaphorical design elicit different audience responses, particularly in video-based content where meaning unfolds dynamically. Prior research shows that metaphors enhance emotional engagement and persuasion (Van Mulken et al., 2014). But it primarily focuses on static imagery. Drawing on Conceptual Metaphor Theory and Visual Rhetorical Theory, we conceptualize metaphorical design as a mapping between the virtual influencer (source) and the product (target) (Phillips & McQuarrie, 2004). We propose that different visual structures (e.g., juxtaposition, fusion, replacement) systematically vary in their effects on emotional and behavioral responses due to differences in interpretability, processing fluency, and symbolic intensity. VI–product alignment further enhances these effects by increasing conceptual coherence. We plan to adopt a multimethod approach combining observational and experimental evidence. First, we analyze YouTube virtual influencer videos using computer vision and text analysis to identify metaphor types and measure emotional engagement, providing real-world evidence of their prevalence and impact. To establish causality, we propose a between-subjects experiment manipulating metaphor type and VI–product alignment. Participants will view standardized video stimuli in which all elements are held constant except the metaphor structure, and then report emotional responses, attitudes, and engagement intentions. This design allows us to directly test whether different metaphor types lead to distinct viewer responses and whether alignment strengthens these effects. This study contributes to IS literature in three ways. First, it advances research on influencer marketing by highlighting metaphorical design as a central mechanism through which virtual influencers communicate meaning and engage audiences. Second, it extends metaphor theory into dynamic, video-based contexts, demonstrating that different metaphor structures produce systematically different emotional and behavioral outcomes. Third, by combining observational data with experimental validation, it provides robust evidence of both real-world relevance and causal effects, offering actionable insights for designing effective virtual influencer content.

Download

Share

COinS