Full length articleReal-life prosocial behavior decreases after being socially excluded by avatars, not agents
Introduction
The desire to belong to a social group and maintain positive relationships with others is a life-long fundamental human need (Baumeister & Leary, 1995). It may be satisfied in everyday face-to-face contacts or via the Internet on virtual platforms (i.e., Second Life), social network sites (i.e., Facebook) or even in online massively multiplayer role playing games (MMORPGs). Many users seek social support via the Internet (e.g., Cole and Griffiths, 2007, Dupuis and Ramsey, 2011), as online social interaction seems to allow for social integration particularly for shy and lonely individuals (i.e., in MMORPGS; Stetina, Kothgassner, Lehenbauer, & Kryspin-Exner, 2011). Although the Internet may have a beneficial effect on users (c.f., Best, Manktelow, & Taylor, 2014) and may even serve as a training ground for social skills training in socially awkward individuals (c.f., Lehenbauer, Kothgassner, Kryspin-Exner, & Stetina, 2013), negative online experiences are also common.
For instance, cyberbullying represents a frequently encountered negative online experience especially among those who are most vulnerable to it (i.e., children and youth; Tokunaga, 2010). Cyberbullying is characterized by a “behavior performed through electronic or digital media by individuals or groups that repeatedly communicates hostile or aggressive messages intended to inflict harm or discomfort on others” (Tokunaga, 2010, p. 278). Bullying via the Internet may take on many shapes, one of which is ostracizing or socially excluding another person from social network sites or from mutual guild activities in MMORPGs.
The concept of ostracism or social exclusion has a long history of empirical research (Carter-Sowell, Chen, & Williams, 2008) and has recently gained even more attention in the context of online social interactions (c.f., van Bommel, van Prooijen, Elffers, & VanLange, 2016). Overall, ostracism refers to a range of detrimental social actions such as ignoring, rejecting or excluding another person. Prior studies (e.g., Zadro, Williams, & Richardson, 2004) have shown that even a brief experience of face-to-face social exclusion may have a number or detrimental effects on the individual. According to the Temporal Need-Threat Model of ostracism (Williams, 2009) social exclusion triggers reactions on the following three levels: (1) the four fundamental human needs belonging, self-esteem, meaningful existence and control (c.f., Williams, 2001, Williams, 2007, Zadro et al., 2004) are threatened as a result of an immediate reflexive response to social exclusion. (2) This subsequently leads the excluded individual to increasingly reflect on the experience; coping mechanisms are activated and the threatened fundamental human needs are fortified. (3) However, if the experience of social exclusion is ongoing, the individual is no longer able to uphold coping mechanisms, and resignation as well as helplessness and depression set in. Prior research supports this model, as ostracized individuals have consistently been found to react to different social exclusion stressors with increased anger, aggression and depression as well as with lower self-esteem and heightened distress (Boyes and French, 2009, Kelly et al., 2012, Leary et al., 2003, Twenge et al., 2001, Warburton et al., 2006, Zöller et al., 2010).
While the above mentioned studies mostly examined short-term effects of social exclusion in face-to-face interactions (Williams & Sommer, 1997) or text-based communications (Smith & Williams, 2004), only two studies (Kassner et al., 2012, Kothgassner et al., 2014) have thus far explored social exclusion in a context which is more informative for online social interactions. They each used a virtual reality (VR) simulation in which the participants were excluded from a ball tossing game by a virtual entity representing another person (avatar) or a computer (agent). The results suggest that being ostracized by virtual others may equally threaten the four basic human needs and lead to a comparable increase in aggression as in face-to-face social exclusions. However, both studies only used all-female samples and thus, generalizations to larger populations are limited. Another drawback is that both studies only assessed the immediate reaction corresponding with level 1 of Williams' model (2009) and thus, do not allow for more far-fetched conclusions about the effect of virtual social exclusion on emotional states and subsequent behaviors (i.e., regarding level 2).
In light of these quite robust immediate effects of social exclusion, research increasingly shifts towards examining the next stage of reactions such as subsequent social interactions with others. This phase is particularly interesting since – according to the Temporal Need-Threat Model (Williams, 2009) – the socially excluded individual has by then had time to reflect on the experience and fortify his/her fundamental human needs (Carter-Sowell et al., 2008). Hence, two behavioral outcomes are plausible: Either the ostracized individual tries to regain control and recognition following the experience of social exclusion and thus, acts anti-socially or aggressively; or s/he chooses to appease others by mimicking their behaviors and engaging in more prosocial behaviors (Carter-Sowell et al., 2008). Prosocial behaviors include a range of actions which are intended to benefit others such as “helping, comforting, sharing, and cooperating” (Batson & Powell, 2003, p. 463). This may include working harder on a mutual task, being more charitable, seeking the other's company, or helping others during a mishap. These behaviors may generally be understood as rooted in a feeling of belonging to a social community (Twenge, Baumeister, DeWall, Ciarocco, & Bartels, 2007).
A small number of studies have thus far examined the impact of experimentally manipulated face-to-face or text-based social exclusion on prosocial behaviors. They found that social threats lead to a reduction in prosocial behavior towards others (Twenge and Baumeister, 2005, Twenge et al., 2007, Wentzel and McNamara, 1999, van Bommel et al., 2016). For instance, Twenge et al. (2007) found their excluded participants to donate less money, show less helpful and cooperative behaviors and volunteer less for further lab experiments. Overall, these results support the first hypothesized behavioral outcome of social exclusion, i.e., the attempt to regain control or more recognition via antisocial behaviors (c.f., Carter-Sowell et al., 2008).
These observed reactions seem to hold true also for online communities, as van Bommel et al. (2016) were able to show that being ostracized online diminished subsequent helping behaviors in an Internet forum. Yet, to our knowledge, there is no research examining the impact of online or virtual social exclusion (i.e., being socially excluded by a virtual entity) on subsequent real-life prosocial behaviors. This constitutes a considerable research gap, as online and offline social interactions are increasingly intertwined, and the transfer of online experiences to the offline context is gradually more relevant in everyday settings (i.e., transferring online friendships to offline friendships). At the same time, online social interaction is complicated by the fact that the identity of the online partner may not be known to the user; for instance, in an MMORPG the user may interact with either guild members who s/he has previously met face-to-face, or with strangers, or even – known or unbeknownst to the user – with the computer. In order to thoroughly assess the impact of virtual social exclusion on the individual it is, thus, essential to distinguish between two types of virtual social entities.
Generally, two virtual entities have been described in the literature: (1) avatars, constituting digital representations of human beings and (2) agents representing digital analogues of computer algorithms (Bailenson, Blascovich, Beall, & Loomis, 2003). Research has suggested that humans tend to respond to both entities in a social manner (Bailenson et al., 2005, Gratch et al., 2007). However, to date there is still an ongoing debate on whether the observed social response is identical for the two entities. A number of studies demonstrate that participants indeed show a comparable reaction to avatars and agents, even if they are aware of who or what controls the digital human (Kothgassner et al., 2014, Von der Pütten et al., 2010). A model dedicated to explaining this phenomenon is the so called Media Equation Concept (Nass and Moon, 2000, Reeves and Nass, 1996). It follows an evolutionary perspective which assumes that people are prepared to react to both, computer controlled entities and ‘real’ human beings in the same way as long as they contain sufficient social cues (i.e., gestures, mimics etc.). While both avatars and agents may trigger similar immediate social responses, it is, however, still unclear whether both also effectuate the same social behaviors (i.e., prosocial actions) in real-life.
A means of evaluating the social experience with a virtual entity is proposed by the concept of social presence (Biocca, Harms, & Burgoon, 2003). It entails the attribution of mental states (i.e., consciousness, feelings etc.) to a virtual entity which commonly results in treating this entity like another human being (c.f., Blascovich et al., 2002, Nowak and Biocca, 2003). Thus, higher levels of social presence are accompanied by a heightened sensual awareness of another person even though this person may be located miles away (Goffman, 1959), or they may lead to feeling (physically) close to a computer-controlled virtual entity and interacting with it as if it were real even if it is clear to the user that it is not.
Therefore, the main objective of the current study was to examine the impact of being socially excluded from a virtual ball tossing game by either a human-controlled avatar or a computer-controlled agent. According to Williams' model (2009), the effect of social exclusion was evaluated on two levels regarding (1) the immediate impact on fundamental human needs and (2) the influence on subsequent emotional states and real-life social behaviors (e.g., prosocial behavior vs. social distance). Hence, we were interested whether there would be differences in (1) fundamental human needs, (2) emotional states and (3) prosocial behaviors between excluded and included participants when considering agency (avatar vs. agent). Additionally, social presence was evaluated to provide more insight into interactions with avatars and agents.
Section snippets
Participants
Participants were recruited via University courses and volunteered to take part in the experiment in exchange for course credits. The final sample consisted of 22 male and 23 female students, ranging in age from 19 to 35 years (M = 25.71; SD = 3.929) with no group-differences regarding age (inclusionary state: t(43) = 0.961, p = n.s.; agency: t(43) = -0.378, p = n.s.) and regarding prior computer experience (inclusionary state: t(43) = 0.485, p = n.s.; agency: t(43) = -0.021, p = n.s.). Male
Results
Results are depicted in more detail below; all statistical analyses were carried out using SPSS Version 20 (Inc. Chicago, USA).
Discussion
In light of today's ubiquitous online communications, the effect of negative online experiences such as social exclusion has moved into the spotlight. Despite its relevance, however, only two studies (e.g., Kassner et al., 2012, Kothgassner et al., 2014) have thus far examined how virtual social exclusion may instantly affect the individual. Apart from the immediate impact, long-term effects are also in a great need to be examined more closely. Evaluating subsequent behavioral changes would
Acknowledgements
We would like to thank Marlene Kaufmann, Katharina Punz and Johanna Wörle for their support in the acquisition of parts of the data.
References (54)
- et al.
Online communication, social media and adolescent wellbeing: A systematic narrative review
Children and Youth Services Review
(2014) - et al.
Having a Cyberball: Using a ball-throwing game as an experimental social stressor to examine the relationship between neuroticism and coping
Personality and Individual Differences
(2009) - et al.
Reactions to acceptance and rejection: Effects of level and sequence of relational evaluation
Journal of Experimental Social Psychology
(2004) - et al.
State, not trait, neuroendocrine function predicts costly reactive aggression in men after social exclusion and inclusion
Biological Psychology
(2011) - et al.
All alone with sweaty palms—physiological arousal and ostracism
International Journal of Psychophysiology
(2012) - et al.
An online self-administered social skills training for young adults: Results from a pilot study
Computers & Education
(2013) - et al.
Beyond the fascination of online-games: Probing addictive behavior and depression in the world of online-gaming
Computers in Human Behavior
(2011) Following you home from school: A critical review and synthesis of research on cyberbullying victimization
Computers in human behavior
(2010)- et al.
“It doesn't matter what you are!” explaining social effects of agents and avatars
Computers in Human Behavior
(2010) - et al.
When ostracism leads to aggression: The moderating effects of control deprivation
Journal of Experimental Social Psychology
(2006)