Designing Anthropomorphic Enterprise Conversational Agents

Stephan Diederich,Lutz M Kolbe,Alfred Benedikt Brendel

doi:10.1007/s12599-020-00639-y

Stephan Diederich, Lutz M Kolbe + Show 1 more

Open Access

https://doi.org/10.1007/s12599-020-00639-y

Copy DOI

Abstract

The increasing capabilities of conversational agents (CAs) offer manifold opportunities to assist users in a variety of tasks. In an organizational context, particularly their potential to simulate a human-like interaction via natural language currently attracts attention both at the customer interface as well as for internal purposes, often in the form of chatbots. Emerging experimental studies on CAs look into the impact of anthropomorphic design elements, so-called social cues, on user perception. However, while these studies provide valuable prescriptive knowledge of selected social cues, they neglect the potential detrimental influence of the limited responsiveness of present-day conversational agents. In practice, many CAs fail to continuously provide meaningful responses in a conversation due to the open nature of natural language interaction, which negatively influences user perception and often led to CAs being discontinued in the past. Thus, designing a CA that provides a human-like interaction experience while minimizing the risks associated with limited conversational capabilities represents a substantial design problem. This study addresses the aforementioned problem by proposing and evaluating a design for a CA that offers a human-like interaction experience while mitigating negative effects due to limited responsiveness. Through the presentation of the artifact and the synthesis of prescriptive knowledge in the form of a nascent design theory for anthropomorphic enterprise CAs, this research adds to the growing knowledge base for designing human-like assistants and supports practitioners seeking to introduce them into their organizations.

Highlights

Technological advances, in machine learning and natural language processing, continue to change the way in which we live, work, and interact with each other, thereby expanding the scope for innovation and automation of human activities (Brynjolfsson and McAfee 2016; Davenport and Kirby 2016)
Different studies highlight that social responses and associated perceptions of anthropomorphism can contribute to a positive user perception of conversational agents (CAs), for example with regard to service satisfaction (Gnewuch et al 2018; Diederich et al 2019c), enjoyment (Lee and Choi 2017) or trust (Araujo 2018)
How the human-like design of a CA can provide utility despite its limited capabilities has yet to be investigated. We address this problem and contribute to the knowledge base on anthropomorphic CA design with the following research question: How can a CA in a professional context be designed to offer a human-like interaction while mitigating feelings of uncanniness due to limited conversational abilities? we bring together prescriptive knowledge for CAs gained mostly in experiments and propose a design for a CA that offers a human-like interaction experience enabled through the combination of social cues with approaches to address the limited responsiveness of present-day CAs

Summary

Introduction

Technological advances, in machine learning and natural language processing, continue to change the way in which we live, work, and interact with each other, thereby expanding the scope for innovation and automation of human activities (Brynjolfsson and McAfee 2016; Davenport and Kirby 2016) One phenomenon in this wave are conversational agents (CAs), defined as software which users interact with through natural language (McTear et al 2016). Several studies at the same time indicate that human-like design may lead to undesired negative effects due to feelings of uncanniness (Wiese and Weis 2019), and that a ‘‘more is more’’ approach does not necessarily lead to increased user perception of anthropomorphism (Seeger et al 2018) Against this background, the Theory of Uncanny Valley (Mori 1970) posits a sharp drop in affinity for human-like artifacts where a user’s attention abruptly shifts from the human-like qualities to its inhuman imperfections (MacDorman et al 2009)

Objectives

Methods

Results

Conclusion