Designing Anthropomorphic Conversational Agents: JAIS accepted research paper

The paper entitled “Texting with Human-like Conversational Agents: Designing for Anthropomorphism“ has been accepted for publication in the Journal of the Association of Information Systems (AIS Basket of Eight; VHB-JOURQUAL A).

Anna-Maria Seeger, Jella Pfeiffer and Armin Heinzl have successfully published their research paper on the design of anthropomorphic conversational agents in the „Journal of the Association of Information Systems“. The paper develops a theoretical design framework that explains how perceptions of anthropomorphism are stimulated in interactions with conversational agents (also named chatbots). Anthropomorphism refers to the attribution of human qualities, including consciousness, intentions, and emotions, to nonhuman agents. Anthropomorphism has various positive and negative consequences in terms of user perception and behavior. It may increase trust and connectedness but it may also reinforce social desirable and discriminative behaviors. The increasing popularity of conversational agents in domains as different as healthcare, e-commerce, and business, makes it very important to understand how to design and control anthropomorphism in interactions with conversational agents.  The theoretical model identifies three groups of factors to stimulate anthropomorphism: technology design-related, task-related, and individual factors. Findings from an online-experiment with 496 participants support the derived framework but also reveal novel, yet counterintuitive, insights. In particular, the paper demonstrates that not all combinations of anthropomorphic technology design cues increase perceived anthropomorphism, but that visual and communicative elements have to complement each other. This work advances our understanding of anthropomorphism in text-based human-computer-interactions and advises on the direction research and practice should take to find the right spot in anthropomorphic CA design.