Designing Interfaces with Social Presence: Using Vividness and Extraversion to Create Social Recommendation Agents*

Article excerpt


Interfaces now employ a variety of media-rich, social, and advanced decision-making components, including recommendation agents (RA) designed to assist users with their tasks. Social presence has been identified as a key consideration in website design to overcome the lack of warmth, social cues, and face-to-face interaction, but few studies have investigated the interface features that may increase social presence. Recent research on RAs has similarly acknowledged social presence as a key factor in the design of online RAs and in building trust in this technology, but there has been limited empirical work on the topic. In this study an experiment was conducted to explore how social technology cues, media capabilities, and individual differences influence social presence and trust in an RA. RA personality (extraversion), vividness (text, voice, and animation), and computer playfulness were found to influence social presence, with social presence serving in a mediating role and increasing user trust in the RA. Vividness also had a moderating effect on the relationship between RA extraversion and social presence such that increased levels of vividness strengthen this relationship.

Keywords: Recommendation agent, decision aid, social presence, vividness, personality, multimedia, computer playfulness, trust, decision making

1. Introduction

Interface design has advanced to support end-user decision making through interfaces that are media-rich and highly interactive. Interfaces now regularly employ sound, pictures, video, and animation. More recent enhancements include socializing or personalizing the interface (Hassanein and Head, 2007; Lee and Nass, 2005), in many cases through the incorporation of recommendation agents (RAs) or decision aids that provide assistance and advice to online consumers (Wang and Benbasat, 2005). While the sophistication of interfaces and RAs continues to grow, many questions remain about how social, multimedia interface characteristics influence user perceptions and behaviors with technology.

Social presence, the feeling of warmth and sociability conveyed through a medium, has recently been identified as a key variable in establishing a connection between a website and its visitors (Hassanein and Head, 2007; Kumar and Benbasat, 2006). Recent studies have shown that perceptions of social presence affect user trust in websites (Gefen and Straub, 2004; Hassanein and Head, 2007; Kumar and Benbasat, 2002) and serve as an enabler for trust-building cues (Gefen and Straub, 2003). Researchers have noted that "the potential importance of social presence raises the question of how a website can be designed to increase social presence" (Gefen and Straub, 2004 p. 418). Yet little empirical research has been conducted to determine what interface features can be used to increase the perceived social presence of websites (Hassanein and Head, 2005/6; Hassanein and Head, 2007).

Social presence has been similarly suggested as a key factor to consider in the design of online RAs, as the social cues and content conveyed can lead to increased trust, involvement, and satisfaction in the RA (Kumar and Benbasat, 2002; Qui and Benbasat, 2005a). But again, few studies have been conducted to identify ways to increase social presence (Kumar and Benbasat, 2006; Qui and Benbasat, 2005a). In an extensive review of RA research, Xiao and Benbasat (2007) developed a conceptual model of RA characteristics, use, and various outcomes. Social aspects of RAs, however, have received little attention in the literature and, thus, were not included in their model. Vividness, a media capability and potential determinant of social presence, was specifically excluded from the RA model due to a lack of empirical investigation and inconsistent results (Xiao and Benbasat, 2007). RAs are widely used in practice and are the subject of several recent studies on trust in technology (Wang and Benbasat, 2005; 2007; 2008). …