How Influencer Types and Message Types in Green Advertising Impact Consumer Attitudes: The Moderating Role of Consumer Attributes

This study explores how the match between influencer type and message type affects consumer responses to green advertising, with a focus on how consumption attributes serve as boundary conditions for this effect. Given the rapid rise of virtual influencers in sustainability marketing, existing research has largely considered influencer type and message framing as independent predictors of pro-environmental behavior, leaving important gaps in understanding how these two factors interact to generate persuasion and under what contextual conditions such interactions occur. Drawing on the Match-up Hypothesis (MUH), and the Stereotype Content Model (SCM), we propose that virtual influencers, perceived as high in competence but low in warmth, are best paired with competence messages, whereas human influencers, perceived as high in warmth, are better suited to warmth messages. This study tests five hypotheses through a preliminary survey and four experiments:The four experiments are organized as follows. Experiments 1A and 1B examine the basic interaction effect (H1) in a product context and a service context, respectively. Experiment 2 disentangles the two mediating pathways (H2 and H3). Experiment 3 tests the moderating effects of the utilitarian and hedonic attributes (H4 and H5). H1: There is an interaction effect between influencer type and advertising message type on consumer attitudes. H1a: For the virtual influencer, conveying competence message (vs. warmth) in green advertising elicits consumers’ more positive attitudes. H1b: For the human influencer, conveying warmth message (vs. competence) in green advertising elicits consumers’ more positive attitudes. H2: As virtual influencer matching with competence message (vs. warmth message), task self-efficacy mediates the positive effect on consumer attitudes. H3: As the human influencer matching warmth message (vs. competence message), social self-efficacy mediates the positive effect on consumer attitudes. H4: Utilitarian attribute moderates the matching effect of virtual influencer and competence message on task self-efficacy (a) and consumer attitudes (b). H5: Hedonic attribute moderates the matching effect of human influencer and warmth message on social self-efficacy (a) and consumer attitudes (b). The expected results are as follows: an interaction between influencer type and message type, such that a virtual influencer paired with a competence message and a human influencer paired with a warmth message elicit more positive consumer attitudes; mediation of the matching effect by task self-efficacy and social self-efficacy; and moderation by consumption attributes, such that utilitarian attributes favor the virtual influencer–competence message match, whereas hedonic attributes favor the human influencer–warmth message match. This study advances the theoretical integration of persuasion models in AI-mediated green communication and provides actionable guidance for marketers to optimize influencer selection and message design based on product characteristics.