With an increasing number of technologies supporting transactions over distance and replacing traditional forms of interaction, designing for trust has become a core concern for researchers in both HCI and CMC. While much research focuses on increasing trust in mediated interactions, this paper takes a systemic view to identify the factors that support trustworthy behavior. In a second step, we analyze how the presence of these factors can be signaled to allow the formation of well-placed trust. For our analysis we draw on relevant research from sociology, economics, and psychology, as well as empirical findings in HCI and CMC research. The key factors that warrant trust in another actor are contextual properties (temporal, social, and institutional embeddedness) and the trusted actor's intrinsic properties (ability and motivation). In first interactions, trust is mainly warranted by contextual properties, as they provide external incentives and threat of punishment. As interactions are repeated over time and trust grows, intrinsic properties become more important. To increase the level of well-placed trust, researchers and designers need to identify signals for the presence of such trust-warranting properties that are reliable and easy to interpret. At the same time, they must be cheap to emit for actors whose actions are governed by them but costly to mimic for untrustworthy actors. Our analysis provides a frame of reference for the design of studies on trust in technology-mediated exchanges, as well as a guide for identifying trust requirements in design processes. We demonstrate application of the model in three scenarios: ecommerce, voice-enabled online gaming, and ambient technologies.

The aim of this paper is to establish a methodological foundation for Human-Computer Interaction (HCI) researchers aiming to assess trust between people interacting via computer-mediated communication (CMC) technology. The most popular experimental paradigm currently employed by HCI researchers are social dilemma games based on the Prisoner's Dilemma (PD), a technique originating from economics. HCI researchers employing this experimental paradigm currently interpret the rate of cooperation-measured in the form of collective pay-off-as the level of trust the technology allows its users to develop. We argue that this interpretation is problematic, since the game's synchronous nature models only very specific trust situations. Furthermore, experiments that are based on PD games cannot model the complexity of how trust is formed in the real world, since they neglect factors such as ability and benevolence. In conclusion, we recommend (a) means of improving social dilemma experiments by using asynchronous Trust Games, (b) collecting a broader range (in particular qualitative) data, and (c) increasing use of longitudinal studies.

Existing Web design guidelines give conflicting advice on the best position for the navigation menu. One set of guidelines is based on user expectation of layout, the other on results from user testing with alternative layouts. To resolve this conflict we test whether placing the menu in an unexpected position has a negative impact on search performance. The results show that users rapidly adapt to an unexpected screen layout. We conclude that designers should not be inhibited in applying design recommendations that violate layout conventions as long as consistency is maintained within a site.

Use of staff photographs is frequently advocated as a means of increasing customer confidence in an e-shop. However, these claims are not conceptually or empirically grounded. In this paper we describe a qualitative study, which elicited customer reactions towards an e-commerce site that displayed staff photographs and links to richer media. The results suggest that employing social and affective cues, particularly in the form of photos, can be a risky strategy. To be effective they should be combined with functionality and targeted specifically at the user types we identified.

In this paper, we investigate how interpersonal cues of expertise affect trust in different media representations. Based on a review of previous research, richer representations could lead either to a positive media bias (P1) or increased sensitivity for cues of expertise (P2). In a laboratory study, we presented 160 participants with two advisors-one represented by text-only; the other represented by one of four alternate formats: video, audio, avatar, or photo+text. Unknown to the participants, one was an expert (i.e. trained) and the other was a non-expert (i.e. untrained). We observed participants' advice seeking and advice uptake to infer their sensitivity to correct advice in a situation of financial risk. We found that most participants preferred seeking advice from the expert, but we also found a tendency for seeking audio and in particular video advice. Users' self-reports indicate that they believed that video in particular would give them the most detailed insight into expertise. Data for advice uptake, however, showed that all media representation, including text-only, resulted in good sensitivity to correct advice.