In this chapter we take a broad view of the social psychology of intergroup relations and apply it to the internet. We start by addressing two popular propositions about the internet. The first of these is that the internet is a safe haven for racists who, using the cover of anonymity, engage in hostile flaming of members of other communities. We can call this first proposition the anonymous cyberhate view of the internet. The second popular proposition is the idea that the internet is a hotbed of social activism where people develop plans to change the world using a device that spans continental boundaries. We call this second proposition the global activist view of the internet.

Let us be clear that there is some truth to both propositions and it is easy to find evidence for both of them. Our contention, however, is that both of the propositions are gross oversimplifications that tend to lead commentators and observers to misunderstand the dynamics present in this medium.

No doubt, social interactions supported by modern technologies have become an important part of our personal, professional, and political lives. At least since the last decade social psychologists have become interested in studying interactions that are mediated by modern technologies such as the internet. Research topics have focused on how computer-mediated interactions differ from traditional face-to-face interactions. One aspect that has drawn a lot of attention is that computer-mediated interactions have greater potential for anonymity. As a consequence, many studies have focused on the effects of whether or not the actor is identifiable to the people she is interacting with, referencing the SIDE model (Spears et al., 1990, 2001). SIDE stands for Social Identity De-individuation Effects. The model proposes that social identity moderates the effects of anonymity. It proposes that deindividuation can accentuate social influence when one identifies strongly with an online group but reduce social influence in the absence of strong identity.

When studying social interactions that are supported by new technologies, there is an inclination to start with the premise that social behavior is in some or all ways fundamentally different as a result of the technology. A notable exception is the research on the SIDE model, which started with the position that deindividuation and social identity have been studied in social psychology at least over the last three decades (Diener, 1977, 1979), and that we could learn from applying what we know from these research traditions to advance our understanding of the role of anonymity in computer-mediated interactions.

Social interaction leads – online as much as offline – to mutual, more or less equal social influence of the interacting parties on each other. Internet users send out information, summaries of subjective experiences, and opinions using communication tools. Recipients acquire unconsciously new insights or strive to learn or form an opinion about a topic. Research studying the facilitating and hindering factors for such media-based social influence has a long-standing history (for a summary of the early work see Williams, 1977). Nonetheless, the interplay between personality characteristics and media attributes as preconditions of social influence has hardly received any attention yet. Therefore, the current chapter will apply interactionism (Endler and Magnusson, 1976) to online social influence. The combination of personal and situational variables is relevant in the context of new communication media as in any other context, because individuals may respond differently to the specific situational conditions provided by computer-mediated communication (CMC). (For initial evidence see Amichai-Hamburger et al., 2002; Cress, 2005; Guadagno and Cialdini, 2002; Hertel et al., 2008; Sassenberg et al., 2005.) Hence, considering individual differences for the prediction of media effects will allow more precise predictions about the impact of CMC on social influence.

In this chapter, different types of social influence will be introduced and the features of CMC that are relevant for social influence will be discussed. Afterwards, the research on these types of social influence in CMC will be reviewed. This review serves two purposes. First, it will show that almost all existing research in this domain has focused on the impact of situational characteristics that either (a) are inherent in the technology (e.g., few social context cues) or (b) result from the social setting in which CMC takes place (e.g., the impact of group norms on a person's attitudes). Second, it will outline how the moderation of social influence processes by communication media characteristics is best explained by the interactionist approach that includes inter-individual differences (e.g., personality traits). Before this approach is described, the few studies that take into account the moderating impact of inter-individual differences between users on the effects of situational characteristics (i.e., personality traits or cognitive styles) will be summarized. (For a framework model putting forward a similar argument for CMC research in general see Joinson, 2003.)

Contemporary web-based communication technologies often present a juxtaposition of multiple sources. In January of 2009, for instance, CNN.com and Facebook.com partnered so that internet users could see and hear President Barack Obama's inauguration in one window, and read and write comments about it with myriad other viewers in another window. YouTube.com presents not only videos but also the comments that viewers have posted about the video. Facebook.com supports individuals’ self-descriptive information online, but also “wall postings” by a profile owner's friends, which may enhance, modify, or contradict the self-presentations individuals attempt to make. Sellers on eBay advertise goods online, and other buyers’ ratings of those sellers are a click away. Travel sites offer hotel listings alongside ratings and reviews by previous guests. And in addition to the overt comments and statements left online by people, the computational systems supporting these interactions often provide interesting statistical coefficients summarizing the evaluations or even the electronic footprints users leave behind: How many people rated a product evaluation as useful? How many times has a viral video been viewed, or a journal article been downloaded? How many friends does an individual have on Facebook? In all of these cases, there is a central message, and there are socially generated responses. How do these mutual sources of influence affect users?

New communication technologies are changing the manner of reception by which individuals acquire information from institutional, interpersonal, and peer information sources. Technology changes the temporal and contiguous presentations of these sources, and may in fact change the information processing and social influence dynamics among these sources; that is, the sequence with which sources are sampled or the simultaneity with which they appear may have potent effects on the information processing filters and biases. Such developments in communication technologies are raising new questions and resurrecting old questions about the interplay of interpersonal and mass communication. New technologies blur the boundaries between interpersonal and mass communication events and/or the roles that communicators take using new systems. Arguments have been made that the “convergence” of old and new media demands new and unified perspectives on traditionally segregated processes. “Media convergence” is a term that has been used to connote several phenomena that are brought about by advancements in telecommunication technology that may change some aspect of the communication process. Sometimes the term refers to the blending of previously individuated mass media: one can watch movies on one's computer, for example. We wish to discuss another kind of convergence: the potential for simultaneous communication via computers of both mass and interpersonal channels. Additionally, new and salient sources of information are present, and potentially influential, via contemporary online information systems: peer communication and computational summaries of social behavior. How these forms and sources of information coalesce, confirm, or contradict one another asks for conceptual and analytic frameworks to address them. New technologies invite research that will advance understanding of how individuals conceptualize communication and interpret messages, and how their attitudes are changed as a result of new mediated message forms and content.

The internet is a defining feature of modern life. Individuals increasingly conduct social interactions online or use other forms of mediated communication, such as text messaging. Understanding how this communication medium affects interpersonal interactions and relationships is a key task for psychologists. Specifically, researchers have been interested in the extent to which individuals are willing to trust others that they encounter online, and the degree to which this trust is justified.

Developing trusting relationships is important if connections forged online are to move to deeper levels. Trusting relationships may be more likely to provide social support and other beneficial outcomes. Of course, just as in offline relationships, appropriate calibration of trust levels is also important. If individuals are too trusting in their online interactions, others may take advantage of them. An optimal level of trust is one in which the person is open to gaining deep and fulfilling relationships, but is not gullible enough to be cheated or taken advantage of (Rotter, 1980; Yamagishi, 2001).

This chapter presents a basic overview of ostracism, describing the reasons for its use, the situations in which it can occur, and the effects it has on its targets. Ostracism can happen in both face-to-face and electronic-based (e.g., internet, cell phones, and virtual reality) interpersonal interactions. The authors describe a variety of experimental research methods that have been used to study the effects of ostracism in both of these interpersonal contexts, comparing and contrasting the results of these different studies. Particular attention is paid to studies involving electronic-based interactions. Both immediate and long-term effects of ostracism on targets’ behaviors are discussed, focusing specifically on how people can be negatively impacted by being ostracized in an electronic-based interaction. The authors discuss how the nature of electronic-based interactions may affect people's reactions to ostracism in novel ways, providing greater understanding of the similarities and differences between face-to-face and electronic-based interactions. These findings are reexamined and explained in the context of virtual bravado – uninhibited behavior perhaps due to the anonymity provided by electronic-based communication. Finally, future directions of research are suggested for studying ostracism in electronic-based interactions, as well as people's general behavior in these types of interactions.

Introduction

There are several needs that are essential to our lives as human beings, and the desire to satisfy these needs influences our attitudes, behaviors, and interactions. Four of the most fundamental human social needs are the needs for belonging, self-esteem, control, and meaningful existence (Williams, 2001, 2009). Humans have a need to belong – they desire interpersonal relationships and meaningful interactions (Baumeister and Leary, 1995). Humans also have a desire to have reasonably high self-esteem (Leary et al., 1995; Tesser, 1988). Humans also have more action-focused needs; they desire some degree of control over their environment (Friedland et al., 1992), and knowledge that their existence matters to others (Greenberg et al., 1986; Solomon et al., 1991).

Introduction

The new communications technologies are developing at such a fast pace that it is difficult for research and theorizing to keep up. Although exploring the range of applications and instantiations of the latest forms of computer-mediated communication (CMC), texting, and video-based phone systems provides many useful insights, research and theorizing that lag behind the technological developments will run the risk of being phenomenon- and even technology-driven, making it difficult to anticipate new uses and consequences. In this chapter we therefore adopt a theory-focused approach to make some sense of the effects of the new technologies (as Kurt Lewin said, there is nothing so practical as a good theory), and a primarily experimental methodology to test this. We focus on a theoretical framework that we have developed over a number of years to gain insights into the effects of CMC in social and organizational settings: the SIDE model. We have found this model useful in helping to correct a tendency, in the literature on CMC in particular, to underestimate the role of social influences on and within these technologies, and an equal (and perhaps opposite) tendency to overestimate their capacity to counteract the impact of status and power.

In particular, we think this theoretical model has been useful in helping us to understand (and predict) some of the more counterintuitive findings of behavior found using computer-mediated communication. The idea that people actually conform when isolated from and anonymous to their group is a good example of such an effect explained by the SIDE model. Gender, which is a key focus in the present chapter, also forms an interesting case study in this respect. Much theorizing and research has proposed that women might become more assertive and less submissive when liberated by the anonymity of CMC. Our research suggests that this is not necessarily the case. The SIDE model helps to explicate when and why the technology helps disempowered groups to transcend inequalities of status and power, and when it leaves them more vulnerable to the power divide. Of course, people are not just passively exposed to the effects of technologies such as CMC – a key argument is that they provide strategic opportunities for people to “manage” their identities contra “less mediated” face-to-face communication. So, for example, when women are given the chance to conceal or deceive their gender identity, do they do this, and if so, with what effect? And are these strategies and effects similar for other groups and categories? After outlining our model and providing some evidence of empirical support, we concentrate on the “gender divide” as one important test case, in which the social and power dimensions of these communications technologies in particular can be examined in some detail. Finally, we consider some of the implications of these findings for gender, as well as relations between groups divided by power and status in general.

Communication technologies and the internet have created an explosion in the number and types of groups that exist in contemporary society – some of which could not have been imagined twenty years ago. The ubiquitous range of information and communication technologies (from hand-held devices to laptops to videoconference rooms to virtual reality immersion labs) provides a variety of channels (audio, video, text, tactile, and graphics in two and sometimes three dimensions) and functionality (communication, collaboration, file sharing, social networking, etc.) which has enabled groups to move beyond the conventional “same time, same place” model of face-to-face groups to an “anytime, anywhere” or even an “all the time, everywhere” model of the digital world (Hollings-head and Contractor, 2002). Some types of online groups, which were novel fifteen years ago, are now commonplace, such as geographically distributed work teams and internet-based discussion forums and online communities.

With the advent of Web 2.0 and Web 3.0 on the horizon, new group forms are continually emerging. Facebook, MySpace, and other social networking sites have enabled people with common acquaintances, similar interests, or complementary goals from across the country or around the world to connect and, in some cases, form discussion groups, friendship groups, political action groups, and even business ventures. Some multiplayer online games, such as “World of Warcraft,” are organized around “guilds” and players must work as a team to overcome obstacles and achieve goals. Virtual worlds, such as Second Life, are user-defined and user-created environments where “residents” can explore, meet other “residents,” socialize, participate in individual and group activities, and conduct business (Wikipedia, 2008).

When the lights go out in a kid's room, a familiar place can become many places – a strange land inhabited by frightening creatures or a playground of warm and fuzzy animals as playmates. The loss of visual cues permits the experience of a place to hinge on subtle cues – unfamiliar noises or the lingering fantasies of a bedtime story. The experience of social interaction is often shaped by place. When the sense of place is not well defined by a physical context, we, like the kid when the lights go out, can experience any one of many places, and our behaviors and the interpretations of others’ actions change depending on the place we construct. One of the emerging themes in this volume is that place is not well defined in computer-mediated interactions. In fact, the physical place is often irrelevant. The meaning of an e-mail exchange does not depend on whether the communicators are in their offices or at the local coffee shop when they send and receive the messages. Whether the exchange is viewed as a business negotiation or idle chit-chat depends less on where the communicators are located than on other cues, often subtle, associated with the interaction. Hence, physical place is often irrelevant to the definition of situation in the person x situation frame.

Another theme that threads its way through the chapters in this book is that the identity of the person is malleable. Actors’ identities are often partly or completely masked by the medium. Moreover, what one reveals about one's self is controllable and the opportunity for portraying a fraudulent self is great. I can become whomever my desires, imaginations, and ambitions dictate. Thus, anonymity has been and continues to be a variable of interest in the study of computer-mediated interaction. The emphasis in much of the early work was on the dark side of anonymous interactions – the bad things that happen when one escapes the constraints of norms and accountability. However, a sampling of the chapters in this volume reveals that the dark side is only part of the picture. For example, the social identity model of deindividuation effects (SIDE model; Spears et al., this volume) provides a much more nuanced and dynamic view of what can happen when actors’ identities are hidden or filtered by electronic media. Spears and his colleagues show how anonymity can lead to more or less normatively scripted behavior depending on the other cues in the context.

The use of computer-supported social interaction (CSI) has become a primary feature of communication among individuals, due in part to its structural features (e.g., freedom of time and geographical constraints) and psychological features (e.g., anonymity). As a consequence, many social scientists have investigated the social processes in computer-supported interactions, including online impression formation, relationship development, and group dynamics. Because individuals communicate via the use of computers in many personal, educational, and professional settings, it is important to continue and encourage the study of social processes in such environments. These studies have identified a number of influences on the behaviors (i.e., conformity, economic choices, etc.), thoughts (i.e., attitude change, impression formation, etc.), and, to a lesser degree, physiological/emotional states of people involved in computer-supported interaction.

The goal of this volume is to impose the global theoretical framework of the person-by-situation interaction (Snyder and Ickes, 1985) onto the study of computer-supported social interaction. This perspective recognizes that people are affected by the expectations and limitations of social situations, but to varying degrees. The extent that a person will respond or react to social forces has been found to depend on the levels of internal characteristics (e.g., self-esteem) that a person brings with them to the social situation. For example, social psychologists have observed that those individuals who express higher levels of self-esteem are more resistant to social pressure and as a result are less likely to conform to the influence of others (Baumeister, 1982), and are more likely to persist in the face of obstacles to their goals (Gist and Mitchell, 1992). Originally posited by Lewin (1935), reinvigorated by Mischel (1968), and then tested and validated in decades of social psychological research across a broad range of social phenomena, the explanatory rubric of social psychology can now be applied to the communication forums that have emerged in the relatively brief existence of internet technologies. Before discussing the details and benefits of the person-by-situation interaction perspective, we will review the existing theoretical frameworks that have been applied to computer-supported interactions.