Affordances in Human-Chatbot Interaction: A Review of the Literature.

Abstract

The present study advances our understanding of human-AI interactions, by identifying and analyzing chatbot affordances in prior research. The results of this review consolidate research findings on chatbots’ affordances, which must be taken into consideration when chatbot-based services are designed and deployed. Specifically, the review of state-of-the-art literature led to the identification of nine high level affordances: Human Like Conversing, Assistance Provision, Facilitation, Distilling Information, Enriching Information, Context Identification, Personalization, Fostering Familiarity and Ensuring Privacy. Our contribution is twofold. First, we map the chatbot affordances identified in prior research and group them in higher-level, overarching affordances through a thematic analysis. Furthermore, we identify areas for future research providing a foundation for researchers aiming to engage with the research area.