On the Design of and Interaction with Conversational Agents: An Organizing and Assessing Review of Human-Computer Interaction Research

Conversational agents (CAs), described as software with which humans interact through natural language, have increasingly attracted interest in both academia and practice because of improved capabilities driven by advances in artificial intelligence and, specifically, natural language processing. CAs are used in contexts such as peoples private lives, education, and healthcare, as well as in organizations to innovate or automate tasks for example, in marketing, sales, or customer service. In addition to these application contexts, CAs take on different forms in terms of their embodiment, the communication mode, and their (often human-like) design. Despite their popularity, many CAs are unable to fulfill expectations, and fostering a positive user experience is challenging. To better understand how CAs can be designed to fulfill their intended purpose and how humans interact with them, a number of studies focusing on human-computer interaction have been carried out in recent years, which have contributed to our understanding of this technology. However, currently, a structured overview of this research is lacking, thus impeding the systematic identification of research gaps and knowledge on which future studies can build. To address this issue, we conducted an organizing and assessing review of 262 studies, applying a sociotechnical lens to analyze CA research regarding user interaction, context, agent design, as well as CA perceptions and outcomes. This study contributes an overview of the status quo of CA research, identifies four research streams through cluster analysis, and proposes a research agenda comprising six avenues and sixteen directions to move the field forward