Can Virtual, CGI-Generated, Influencers Help Sell Products on Instagram? Effects of Perceived Realism and Disclosure on Brand-Related Attitudes and Behavioral Intentions

Joe Phua, Nathaniel J. Evans, Youngjee Ko (Ph.D. candidate), and Janice Lee (Ph.D. candidate) (Forthcoming) “Can Virtual, CGI-Generated, Influencers Help Sell Products on Instagram? Effects of Perceived Realism and Disclosure on Brand-Related Attitudes and Behavioral Intentions,” International Journal of Internet Marketing and Advertising.

Abstract: Virtual influencers, CGI-generated avatars with fabricated personalities, are increasingly being used by brands to advertise products on social media platforms. Applying theoretical frameworks including the match-up hypothesis (Kamins, 1990), parasocial identification (Horton and Wohl, 1951; Rubin and Perse, 1987), source credibility (Ohanian, 1990) and the persuasion knowledge model (Friestad and Wright, 1994), the current study assessed effects of perceived realism (high versus low) and disclosure (yes versus no) of a virtual influencer in an Instagram brand ad on perceived brand credibility, brand attitude, purchase intention and eWOM intention. Using a two (realism: high versus low) x two (disclosure: yes versus no) between-subjects experiment, results revealed that perceived realism exerted a significant main effect on the dependent measures, and also interacted with disclosure to influence brand-related outcomes. Further, perceived prosocial characteristics of the virtual influencer also mediated between realism and the dependent measures. Theoretical and practical implications are discussed.