Sun, Y., Pan, Z., & Shen, L. (2008). Understanding the third-person perception: Evidence from a meta-analysis. Journal of Communication, 58, 280–300.
Sun, Y., Shen, L., & Pan, Z. (2008). On the behavioral component of the third-person effect. Communication Research, 35, 257–278.
Sussman, S., Sun, P., & Dent, C. W. (2006). A meta-analysis of teen cigarette smoking cessation. Health Psychology, 25, 549–557.
Tal-Or, N., Cohen, J., Tsfati, Y., & Gunther, A. C. (2010). Testing causal direction in the influence of presumed media influence. Communication Research, 37(6), 801–824.
Tal-Or, N., Tsfati, Y., & Gunther, A. C. (2009). The influence of presumed media influence: Origins and implications of the third-person perception. In R. L. Nabi & M. B. Oliver (Eds.), The Sage handbook of media processes and effects (pp. 99–112). Thousand Oaks, CA: Sage.
Tewksbury, D., Moy, P., Weis, D. S. (2004). Preparations for Y2K: Revisiting the behavioral component of the third-person effect. Journal of Communication, 54, 138–155.
Tsfati, Y. (2007). Hostile media perceptions, presumed media influence, and minority alienation: The case of Arabs in Israel. Journal of Communication, 57, 632–651.
Tsfati, Y., & Cohen, J. (2003). On the effect of the “third-person effect”: Perceived influence of media coverage and residential mobility intentions. Journal of Communication, 53(4), 711–727.
Tsfati, Y., & Cohen, J. (2005). The influence of presumed media influence on democratic legitimacy: The case of Gaza settlers. Communication Research, 32, 794–821.
Tsfati, Y., Cohen, J., & Gunther, A. C. (2011). The influence of presumed media influence on news about science and scientists. Science Communication, 33(2), 143–166.
Wechsler, H., Nelson, T., Lee, J. E., Seiberg, M., Lewis, C., & Keeling, R. (2003). Perception and reality: A national evaluation of social norms marketing interventions to reduce college students’ heavy alcohol use. Quarterly Journal of Studies on Alcohol, 64, 484–494.
Xu, J., & Gonzenbach, W. (2008). Does a perceptual discrepancy lead to action? A meta-analysis of the behavioral component of the third-person effect. International Journal of Public Opinion Research, 20(3), 375–385.
CHAPTER 23
How Does Technology Persuade?
Theoretical Mechanisms for Persuasive Technologies
S. Shyam Sundar, Jeeyun Oh, Hyunjin Kang, and Akshaya Sreenivasan
Historically, media and communication technologies have been seen as amplifiers of persuasive communications. From microphones to smartphones, communication technologies have served to boost the signal strength of persuasive messages. Mass-media technologies, such as the printing press, radio, and television, have boosted reception strength by expanding the reach and frequency of these messages. In fact, rich areas of persuasion research, such as propaganda, are premised on harnessing the vast dissemination potential of media and communication technologies.
Yet, theoretical attention to the role played by technology in persuasion is surprisingly scarce. The general tendency has been to view technology either as a channel for conveying persuasive messages (Fogg, Lee, & Marshall, 2002) or as a bundled environment with inherent and immutable constraints (Holbert, 2002). In assuming that a given media technology is a constant, both these approaches call for adapting message and psychological variables to suit the exigencies of the medium.
This is probably because classic definitions of “persuasion” (Miller, 1980, see chapter 5) focus on message and psychological variables by emphasizing (1) the use of symbols and (2) the social nature of the phenomenon. Highlighting the importance of communication, Dillard (2010) defines persuasion as one social actor using symbols to change the opinion or behavior of another social actor. However, the label of “social actor” need not be restricted to humans, but can indeed be extended to technologies, as demonstrated by numerous studies in the CASA (Computers as Social Actors) literature, which show that individuals tend to apply social rules of human-human interaction when interacting with technologies, even though they agree that computers do not have intentions (Reeves & Nass, 1996).
Likewise, the notion of “symbols” need not be restricted to the message content of communication. With the arrival of digital media, there is a growing realization that even nonlinguistic technological features can serve as symbols with persuasive appeal (Sundar, 2008a). In fact, the traditional separation of source and/or message from the technology that delivers the message is no longer conceptually defensible given that new media technologies are erasing the boundaries between source and receiver, as well as those between message and medium. For example, customization technologies make the receiver the source of messages, and interactive interfaces transform the nature of the message so significantly that the sheer existence of interactive features can serve as a persuasive message.
It must be noted, however, that technological features such as customization and interactivity are anything but fixed. They have become highly variable, given the “app culture” of modern media such as websites and tablets. The variable use and deployment of applications on websites makes it less useful for us to think and theorize about the Web as a whole media form or as a uniform “symbol system” (Salomon, 1979). Even specific genres of websites, such as social networking sites, cannot be treated as distinct, coherent media because they have several applications that afford dynamic changes in form and functionality. Therefore, no two examples of online social-networking platforms are the same. Not only are new applications developed all the time and continually change the technology of a medium, they are also increasingly available across media (e.g., same app available for tablets as well as smartphones), thereby diminishing the differences between media forms.
Given this lack of distinctiveness and the absence of uniformity within any given medium, the role of technology in persuasion cannot be meaningfully assessed by comparing different media (e.g., computers vs. robots), but by investigating the contribution made by specific features (e.g., morphology) or affordances (e.g., interactivity) of media technologies. Toward this end, we adopt a “variable-centered,” rather than “object-centered,” approach to the study of technology (Nass & Mason, 1990) and examine structural features that underlie interface design, characteristics that facilitate specific actions and thereby affect both the nature and effects of communication (Sundar, 2009).
Studies in the persuasive computing literature do not specify which aspects of a computer (e.g., Fogg, 2002), game (Bang, Torstensson, & Katzeff, 2006), or a household appliance (McCalley, Kaiser, Midden, Keser, & Teunissen, 2006) affect credibility and other user attitudes. We still do not know how and why persuasive technologies work. In the sections that follow, we attempt to provide some answers by viewing persuasive technologies not as specific tools or objects, but as variables related to technological affordances, such as interactivity and navigability, that may influence persuasion via five theoretical mechanisms.
How Technology Persuades
* * *
Theory and research suggest that technology can persuade individuals by (1) triggering cognitive heuristics about the nature of content, (2) enabling the receiver to be the source, (3) creating greater user engagement, (4) constructing alternative realities with enhanced vividness, self-representation, self-presence, spatial presence, and transportation, and (5) affording easier access to information.
Cognitive Heuristics
While cognitive heuristics triggered by message content are well-documented in the dual-process literature, those stimulated by interface affordances are only now beginning to be studied. Even if users do not actively engage interactive tools on an interface, the mere presence of interactivity can sometimes cue a series of cognitive heuristics that dictate their evaluation of the interface as well as its content (Sundar, 2008a). For instance, a website with a plethora of interactive tools can give users an impression that this is a participatory forum, open and democratic in nature, with visitors afforded a voice. The MAIN model (Sundar, 200
8a) proposes that affordances related to modality, agency, interactivity, and navigability manifest themselves in the form of interface cues that trigger mental shortcuts (i.e., cognitive heuristics) for judging the quality and credibility of the content delivered via the interface. Sundar, Xu, and Dou (2012) identify 20 such heuristics that could play a role in shaping consumers’ attitudes and behaviors in the context of online advertising and marketing.
Interface affordances do not exist as structural or ontological characteristics alone, but also possess cues that trigger perceptions in the form of quick evaluations (i.e., heuristics) about the perceived consequences of their use (Sundar & Bellur, 2010). In other words, the perception of a certain action possibility in the interface can directly contribute to positive or negative judgments of the credibility of content conveyed by the interface even without using it. For instance, the appearance of dialogue boxes during the course of browsing a website can enhance the feeling of constant interaction with the system, and thereby invoke the interaction heuristic. Similarly, control heuristic can be triggered by a device highlighting its ability to afford user control over the nature and flow of information (Sundar & Bellur, 2010). In fact, Sundar and Bellur (2010) offer a detailed list of heuristics for each type of interactivity: interaction and responsiveness heuristics for modality interactivity; activity, control, choice and ownness heuristics for source interactivity; and contingency, telepresence, and flow heuristics for message interactivity.
These heuristics are suggestive of theoretical mechanisms by which interactivity affordances influence the perceived value of the information and the medium, and thereby affect credibility, but they do not always have to occur via heuristic processing. In fact, the rules of thumb invoked by affordances could serve as important analytical tools for aiding systematic processing of underlying information. If the user is willfully applying the heuristic to arrive at credibility judgments of content, the processing is said to be conscious or controlled, rather than automatic. Therefore, interactivity can affect persuasion via both heuristic and systematic processes. The following three heuristics are quite reflective of this theoretical approach to understanding the effects of persuasive technologies.
Old Media Heuristic
Given the multimodal nature of modern media interfaces, the use of modalities resembling those used in older media can trigger mental shortcuts that lead to credibility judgments of content. Sundar (2000) found that providing audio downloads significantly lowered the perceived journalistic quality of news stories, especially when pictures were included in the stories. In contrast, text-only and text-plus-picture modalities elicited more positive evaluations. Newspapers have been traditionally seen as more credible sources of information, associated with stringent gatekeeping standards, whereas broadcast media outlets that use audio and/or video modalities tend to be perceived as less credible sources.
Thus, a website resembling a newspaper can serve to invoke a “newspaper schema,” leading to positive credibility evaluations. This rule-of-thumb is labeled old media heuristic by the MAIN model (Sundar, 2008a). As a result, the same message could be seen as more persuasive when presented via text rather than via audio and pictures. Depending on how the invoked heuristic is processed, this effect could follow one of two possible mechanisms, as would be predicted by dual-process models in social cognition: If it is processed heuristically, then the mere old-media look of the site can directly boost positive ratings. But, if it is processed systematically, the old-media look will prompt users to engage in controlled processing of the information, as opposed to the more automatic and passive processing of news content typically associated with electronic media. And, assuming that the content is strong, this effortful route is also likely to promote persuasiveness.
Machine Heuristic
When technology is the attributed source of communication, it can be quite persuasive. Sundar and Nass (2001) found that participants rated identical news stories as being higher in quality when they thought that the computer terminal, rather than news editors, chose them. This may be due to the operation of the machine heuristic—if a machine chose the news story, then it must be objective in its selection and free from ideological bias (Sundar, 2008a). A good example is our tendency to trust the order of search results by an automated engine, such as Google, and automatically assign higher importance to those results that appear at the top without critically assessing the intentionality behind the rank-ordering of the output (Pan et al., 2007). Therefore, persuasion is likely to be higher and less subject to counterargumentation when the message is delivered by an interface that is machine-like in its appearance as well as operation.
Bandwagon Heuristic
Aside from imbuing agency to the machine, modern communication technologies provide agency to receivers themselves, both collectively and individually. Social media have made it possible for users, as a collective, to weigh in on virtually everything online, from voting on top news stories to fact-checking on health information in a bulletin-board to reviewing products on e-commerce sites. Even without actively opining, users can send a collective message by simply visiting certain online venues. Their actions are compiled by any number of algorithms and applications to produce metrics such as number of hits, number of visitors, and so on. The number of Diggs on a social-bookmarking site and the list of most viewed news stories are just two examples of audience-as-source, which may ultimately influence what users choose to read or believe online.
Psychologically, audience-as-source activates the bandwagon heuristic, which has been shown to positively influence intention to purchase products from an e-commerce site (Sundar, Oeldorf-Hirsch, & Xu, 2008). This heuristic is triggered by any interface cue that signifies the popularity of certain content (e.g., the number of views of a YouTube video clip) or products (e.g., product review and star-ratings of a product listed in Amazon.com) and is shown to be stronger than the authority heuristic triggered by the presence of expert opinion (Sundar, Xu, & Oeldorf-Hirsch, 2009). Users seem to believe that if many other people think that something is good, then they must too. Elements of consensus (Chaiken, 1987) and/or endorsement (Metzger, Flanagin, & Medders, 2010) constitute this bandwagon effect, with important implications for persuasion using tools of social media.
Bandwagon, machine, and old-media heuristics are three examples of a wide range of mental shortcuts identified by the MAIN Model (Sundar, 2008a) as being triggered by interface features rather than content attributes. Yet, these heuristics can play an important role in determining the outcomes of persuasive communications (Sundar et al., 2012), both by their sheer presence and by offering unique functionality that was absent in older media.
User as Source
Self-agency is a powerful contributor of persuasive outcomes, as evidenced by the rise and success of technologies that afford customization by individual users. By enabling users to influence the nature and process of an interaction, these affordances make each individual user feel like they are a relevant actor in the interaction, and thereby aid persuasion. For instance, in marketing studies, products designed by the consumers generate significantly higher acceptance (Franke, Schreier, & Kaiser, 2009), and consumers in financial portals show higher willingness to provide personal information when they are provided with customizable web interfaces (Coner, 2003).
The agency model of customization (Sundar, 2008b) posits that “self-as-source” is the fulcrum of psychological benefits derived from customization. According to this model, technological affordances imbue a higher sense of agency by allowing the user to serve as a source of his or her information, and thereby become the center of his or her interaction universe. This translates to positive cognitive, affective, and behavioral responses toward both the interface and content of customizable media. Theoretically, two classes of mechanisms govern these persuasive effects. One pertains to the sheer affordance of the user acting as a source (e.g., the user’s ability to perform the tailoring on their own or digital
ly publish content that they create, as in social media). The other pertains to the content that results from the process of the user acting as a source.
Self-determination theory (Ryan & Deci, 2000) belongs in the former category, given its emphasis on user autonomy. Easy-to-use tools of customization and social media not only provide autonomy, but also a sense of competence or self-efficacy (Bandura, 1997) to operate them. In addition, they afford endorsement of content by others, which can nurture psychological bonding with both the process and components of communication, and thereby have a positive impact on content perception. Studies in community psychology (McMillan, 1996; McMillan & Chavis, 1986) have shown that feelings of membership, sense of belonging, and trust enhance sense of community, which can generate positive persuasive outcomes. For instance, Richardson et al. (2010) found that adding virtual community features to an online walking program helped participants stay engaged in the program. Another study showed that online spaces that foster a sense of community positively affected attitudes and behaviors toward the community (Firpo, Kasemvilas, Ractham, & Zhang, 2009). Kim and Sundar (2011) found that sense of community is indeed a significant mediator of the relationship between the perceived number of times a thread is shared in a discussion board and users’ attitudes toward posting. Together, these factors of autonomy, competence, and relatedness serve to enhance the degree of self-determination (Ryan & Deci, 2000) and thereby the intrinsic motivation to engage with the interface and its contents.
A related construct is that of perceived control afforded by the self acting as the source (Sundar, 2008b). Power users of technology are especially likely to feel a higher sense of control when using customizable features, in part because they seek personal control over their privacy (Sundar & Marathe, 2010), and a greater sense of empowerment (Weissman, 1988). When they customize, users are also known to feel a higher sense of accomplishment (Norton, Mochon, & Ariely, 2011) and ownership (Pierce, Kostova, & Dirks, 2003), both of which positively predict user attitudes toward the object of customization.
The SAGE Handbook of Persuasion Page 79
