Object, Me, Symbiote, Other: A social typology of player-avatar relationships
First Monday

Object, Me, Symbiote, Other: A social typology of player-avatar relationships by Jaime Banks

Despite how humans and non-human objects relate in very social ways, the relations between players and their online game avatars are most often examined through conspicuously parasocial lenses. That is, the player-avatar relationship (PAR) is generally seen as one-way and non-dialectical as players think, feel, and acts toward the avatar without consideration for the avatar’s role in that relationship. The present study examines the potential for PARs to be fully social, in which player and avatar both materially contribute to the relationship. Through interpretive thematic analysis of in-depth interviews with diverse World of Warcraft players, analysis revealed that PARs sometimes feature the fully social characteristics of self-differentiation, emotional intimacy, and shared agency. It is argued that these differences in sociality may be best understood according to a four-point PAR typology, ranging from non-social to fully social.


Perspectives on player-avatar connections
Player-avatar relationships as variably social
Discussion and conclusions




“Every year we, the denizens of Suncrown Village, would take part in a ritual to keep the different aspects of nature under our control. Dominion over the element of water was symbolized by the summoning and enslavement of an elemental named Aquantion. Through a wicked perversion of our magic, the scoundrel broke his bonds and enslaved us instead!” — Geranis Whitemorn, from the quest “Forgotten Rituals” in World of Warcraft.

At its simplest, an avatar is “an interactive, social representation of a user” [1] such as a textual username, a graphic profile photo, or a three-dimensional body. At perhaps its most complex, an avatar is an entity that haunts cyberspace as “the face of it-ness, who-ness, and what-ness, mediating community and unseating the subject’s Eigentlichkeit (self-possession, and ‘having’ of what is my own)” [2]. Derived from the Sanskrit word avatara (a god in earthly form), the avatar is generally understood to be ”a delegate, a tool or instrument allowing an agency to transmit signification to a parallel world” [3], ranging in digital games from a simple shape (such as the paddle in Pong or the circular character in PacMan) to complex, three-dimensional, evolving entities (such as the assemblage in the Katamari series or the humanoid characters in World of Warcraft).

In many massively multiplayer online games (MMOs), players create a three-dimensional avatar — usually humanoid — choosing its appearance, functions, traits, and names according to the themes of each game. For example, one might have a dainty Elven hunter with a scar across her left eye or a burly, broken-horned Minotaur who shuns his combat duties to instead pick flowers. From the moment of its creation, the avatar is generally understood to be functioning on behalf of the player. The player uses a keyboard, mouse, or other controller to indicate what the avatar should do in the gameworld — move, jump, gesture, speak, interact with characters, cast spells — and the avatar enacts those intentions. This perspective most certainly accounts for the ways that the player acts upon the avatar, but what of the ways the avatar acts independently and acts upon the player, and of the interaction between the two? In other words: what about the potential sociality of the player-avatar relationship?

Despite acknowledgements that humans engage non-human objects in highly social ways (Reeves and Nass, 1996), the connections between players and avatars in online games has been examined from conspicuously parasocial perspectives. That is, the player-avatar relationship is assumed to be one-way, non-dialectical, and existing only in the mind of the player (Lewis, et al., 2008; cf., Horton and Wohl, 1956). As such, there is not a clear picture of the potential for online game avatars to be engaged as fully social actors, even though the potential for social engagement is particularly strong for interactive technologies such as online games (Nowak and Rauh, 2005). This study is an exploration of online gamers’ stories about their avatars, analyzed for variations in how player-avatar relations (PARs) mirror human social relationships. Ultimately, analysis revealed that PARs are variably social, and this spectrum of sociality is best understood according to a four-point typology that links relational features to motivations for online gameplay.



Perspectives on player-avatar connections

In creating and engaging an MMO avatar, players enter into an odd scenario. They are, in a sense, responsible for two bodies — one made of physical atoms and one of code and pixels. In this scenario, embodiment and identity are complicated — the human and non-human, physical and digital, material and immaterial are broken down and rebuilt (Haraway, 1991) such that the demarcations between ‘self’ and ‘other’ vary based on the nature of the player-avatar relation. Although theorists (e.g., Boler, 2007; Stone, 1995; Turkle, 1995, 2007) often attend to such complexities of human-technology relations, empirical investigations of player-avatar connections focus primarily on the way the player acts on, thinks about, and feels toward the avatar as an object. As a result, much of our understanding of these connections is player-centric, resting on three intertwined phenomena: identification, attachment, and instrumentality. Identification is the degree to which players see avatar as themselves or not (Rehak, 2003; Taylor, 2002). Attachment is the way a player has affinity for an avatar as a character, through senses of likeness, physical control, responsibility, and suspension of disbelief (Lewis, et al., 2008). Instrumentality is the way that avatars are taken as tools — as means to particular, individual ends (Schultze and Leahy, 2009).

These player-centric perspectives draw on the notion that relationships with media characters are parasocial — they are one-way, non-dialectical attachments in the mind of the audience (Horton and Wohl, 1956). Importantly, parasocial connections with game avatars influence how players experience gameplay phenomena (Li, et al., 2013; Trepte and Reinecke, 2010; Van Looy, et al., 2010). However, this focus on how human factors exist, are enacted, or are augmented with respect to technological factors does not account for how avatars themselves actively participate. The parasocial perspective dismisses avatars’ potential agency — the potential to matter in a relationship.

Player-avatar relationships as potentially social

A social relationship is a valenced connection between two people through which each party influences the other (Berscheid and Peplau, 1983; Harvey and Pauwels, 2009). A requirement for social relationships is self-differentiation — the degree to which a person holds a concrete definition, understanding, and experience of himself or herself as distinct from another entity (Bowen, 1978). If an entity is seen as “other,” an association may emerge through relational knowing, interrelatedness, continuity, and endurance (Harvey and Pauwels, 2009; Snyder and Lopez, 2007), and each party influences the other according to transactional information sharing and how each sees that information as valuable or meaningful (Levinger, 1979; Thibaut and Kelley, 1959). That meaning is shared (Mead, 1934; Pearce, 2007) as people work to understand how they are seen in the eyes of others (Goffman, 1959). Sometimes meanings include physical, cognitive, and emotional intimacies that beget a perception of closeness: feelings of care and understanding, affirmation and value, and warmth and belonging (Sinclair and Dowdy, 2005).

Although it may seem that only humans have the ability to fulfill these relational criteria, further consideration reveals that avatars can fulfill them as well. As cybernetics theories posit (e.g., Haraway, 1991; Wiener, 1948), we engage in recursive, mutual information exchanges with our digital technologies, and meaning emerges through those transactions. Non-human objects hold agencies — including the ability to exchange information — that are different from human agencies (Bogost, 2012; Harman, 2005; Latour, 2005) but that are no less important in understanding human-technology relationships. Just as players influence avatars, avatars influence players through encoded features and abilities, such as occupying space and representing purpose in the gameworld, conveying personality in their autonomous in gait, voice, and postures, and even priming players to think and behave in accordance with given cues (Yee and Bailenson, 2007). Intersecting rules of physical and game environments govern these exchanges (Lessig, 1999) in the same way that social norms govern interpersonal human relations. Acts of play emerge in this feedback loop similar to acts of interpersonal communication, as information flows across the interface (see Burn, 2003; Gee, 2005).

Through graphical, textual, aural, and procedural information exchanges between player and avatar as active agents, the connection between player and avatar may feature characteristics of an interpersonal social relationship. As such, PARs should be examined for their potential to be two-way, dialectical relationships rather than as one-way, non-dialectical connections. To this end, this study examined players’ descriptions of their avatars and gameplay for characteristics of interpersonal relationships to answer the question: Do players and avatars have social relationships?




To explore the potential for PARs to be fully social relationships, this investigation focused on gamers’ phenomenal accounts in order to effectively examine the relationship from the “inside,” as experienced by one entity in the dyad. In particular, interviews were conducted with players of the MMO World of Warcraft (WoW). WoW is an ideal space in which to examine PARs because the game’s longevity, stability, and persistence affords players the opportunity to develop strong ties with their avatars as stable personalities over time (King, et al., 2010; Lee, 2007), with some players having been involved with their avatars for more than 10 years.

Research context

Part drama of interracial war and survival, part complex competition system, and part cartoonish social club, WoW is the most popular MMO on record, with nearly 12 million players at its peak (Blizzard Entertainment, 2010). The subscription-based game launched in 2004 has, in the last decade, spawned a rich culture of official lore and fan fictions, fan sites, online and off-line player communities, conventions, and commercial goods. Broadly, scholarly investigations of WoW characterize the gameworld as an immersive, social space emerging at the juncture of human and technological agencies (see Giddings, 2007; Taylor, 2009). Through this avatar-mediated interactivity, the gameworld and its culture are continuously reconstituted through social interaction (McArthur, 2008; Williams, et al., 2006), economic activity (Castronova, 2005), achievement and competition (e.g., Rettberg, 2008), and identity- and meaning-making (e.g., Bates, 2009; Bessière, et al., 2007; Hagström, 2008).

Variations in PARs may emerge in WoW in part because players engage their avatars according to what Mateas and Stern (2006) call “constrained freedom” — players materially contribute to their avatars’ appearance and behavior, but these contributions are governed and augmented by the game’s design and avatars’ inherent logic (Steinkuehler and Williams, 2006). As such, both the player and the avatar contribute to defining the relationship according to their unique agencies.

When beginning the game, for example, players must create and customize an avatar. However that customization is limited to 13 races (e.g., Dwarf, Orc, Goblin), to binary gender (male or female), to 11 classes (e.g., mage, warrior, hunter), to a limited number of features (e.g., one of nine skin colors, eight hairstyles, seven piercings), to only one audible voice per race/gender combination, and to unique names that adhere to content standards (e.g., no sexual references). Once an avatar is created, the player is free to use keyboard and mouse to move the avatar around the world, but only according to game rules and avatar physics — the avatar must travel on foot at first until it has gained enough power to ride a mount or fly, and it may pass through other avatars but not through walls or trees. The player may also advance in the game by fighting monsters through a combination of spells and abilities, but they are limited to the spells assigned to the avatar’s class. Similarly, players may use avatars to communicate with other players — typed text appears in a speech bubble over the avatar’s head or in a chat window — but this text is only readable by players in the same faction. Players may also choose among more than 200 “emotes” or gestures (such as /flirt, /chuckle, or /dance) to express themselves, however the way the avatar performs that gesture is predefined. For example, typing in /chuckle makes a female human perform a default flirty laugh, and a male dwarf defaults to a hearty guffaw.

A given player’s approach to this constrained freedom is understood to be driven by particular motivations for gameplay (see Billieux, et al., 2013), such as competing and earning achievements, socializing and forming relationships, and following the narrative and becoming immersed in the world (Yee, 2006). Those driven to compete may focus on game-defined goals and excelling in spite of avatar constraints, using the avatar to complete quests (e.g., killing a monster, collecting an object) in order to enhance the avatar’s skill and weapons. Players driven to socialize may not care as much about advancing their avatars, and instead focus on the avatar’s utility in building friendships and in helping their friends advance. Those motivated by gameworld immersion may develop unique backstories for their avatars and the role it plays in the world narrative, and use the avatar to explore interesting environments (see Yee, 2006). These variable motivations and associated gameplay style present an ideal opportunity to examine the experiential conditions under which different types of PARs may emerge.

Recruitment and sampling

Participants were recruited through a combination of convenience sampling from open calls in public forums, posts in Facebook groups, paid Facebook ads, and known players. Potential participants were invited to visit a Web site with study information and a survey addressing demographics, play habits, and open-ended questions about favorite avatars. Compensation for completing the survey was an entry into a drawing for one year of WoW game time ($180 value).

Among 404 complete survey responses, 335 players indicated they were interested in participating in interviews in exchange for one month of WoW game time ($15 value). From this frame, 30 cases were purposively sampled to include a range of player and avatar features, preferred types of play (e.g., roleplayers, raiders, casuals), and relationship intensity. Relationship intensity was determined through a preliminary heuristic evaluation of responses to open-ended questions about feelings and memories about the avatar: low-intensity (e.g., “It’s just pixels.”), moderate-intensity (e.g., “I spend a lot of time on it and I like it a lot.”), or high-intensity (e.g., “I LOVE my avatar — it’s like my child.”).

The purpose of this sampling approach was not to garner a representative sample, but to elicit participation from many different types of players in order to make inclusive observations about PARs broadly. Participants were sampled in tandem with iterative analysis, toward theoretical saturation (Glaser and Strauss, 1967). In completing the twenty-fifth through thirtieth interview sets and reviewing emerging patterns, no new concepts had been introduced and key themes were well-developed; as such data collection was halted. Five participants were ultimately excluded due to incomplete data; field notes and partial recordings suggested that concepts represented those cases were adequately represented in other cases.

This approach yielded a 25-participant sample with 12 male, nine female, and four genderqueer players. Of these, 10 players played avatars of the opposite sex. Among them, 10 had favorite avatars in the Alliance and 15 in the Horde (the game’s two opposing and culturally divergent factions), and every available avatar race (e.g., Night Elf, Goblin) and class (e.g., mage, warrior) was represented. Players varied in ethnic and sexual identities, and in gameplay habits.

Data collection

Two in-depth interviews were conducted with each participant. The first was a semi-structured interview conducted via Skype, following a loose question guide that addressed topics of avatars, social groups, environments, gameplay, and game cultures, beginning with broad questions such as “Please tell me a bit about <avatar name>. What is <avatar name>’s story?” and moving to more specific questions like “<avatar name> is a <race><class> — why did you choose that combination?” The second was a gameplay interview — an unstructured interview combining participant-led Skype conversations with participant-led in-game activities, such as working toward in-game achievements or hunting for rare game items. This combined-interview approach was designed to invite stories that might not have been salient outside of gameplay.

Interviews lasted between 45 minutes and two hours, depending on participants’ openness, depth of experience, and chosen play activity. Both sessions were recorded using a Skype audio-recording plug-in, and cooperative gameplay events were screen-captured. Following the interviews, session memos were written to document first impressions and emerging themes. Interview recordings were professionally transcribed, checked for accuracy, and names were masked with participant-chosen pseudonyms (also used to attribute data in this article). In total, approximately 70 hours of audio recordings resulted in 1,500 pages of transcribed text. These transcripts were the data analyzed.

Data analysis

To examine the potential for PARs to be social relationships, interview transcripts were subjected to thematic analysis. Thematic analysis is an iterative, interpretive, inductive process of identifying and integrating emergent patterns within and across data sets (Braun and Clarke, 2006; Guest, et al., 2012), where the units of observation are data segments constituting discrete thoughts (e.g., phrases or sentences). Analysis was conducted in five steps, as prescribed by Braun and Clarke, with interpretive memos written throughout (Glaser, 2012). Important to note is that, although data was collected from individual players, the unit of analysis in this study is not the player, but instead the player-avatar relationship.

First, transcripts and screen captures were engaged throughout the data collection, preparation, and pre-coding stages, through active, repeated readings (four-five iterations) before formal coding. Second, a set of initial codes were generated, identifying particular features of the data. Coding was loosely guided by the aforementioned definition of “relationship” (a valenced connection with mutual influence), remaining open to other characteristics and the possibility that social relationship features may not be present. Third, codes were evaluated for unifying patterns, within and across cases (Strauss and Corbin, 1990). The output of this step was a list of themes and sub-themes and corresponding data segments. See Table 1 for an example of a theme, subthemes, and data points.


Table 1: Example of coded themes, sub-themes, and corresponding data.
ThemeSubthemeSample transcript excerpts
“Realness”Having a true/real self“I think both the being in a world where stomping on everybody’s what you’re supposed to do, and having a character connected to a lore that was reminding me of who I really am.” ~Kayne
Death isn’t death“I used to kind of make me a little bit sad, but I stopped caring a while ago when it started happening a lot.” ~Perry
Lore/Fantasy“There are some points where you could easily break immersion if you’re supposed to be in a fantasy world and everyone’s calling you by your real name, even if they’re your real friends.” ~Penny
Evidence of in-game events“You know, the tabard is a trophy.” ~Roy
Digital as Real“It’s just it is really striking. It really is one of those things where it’s like you have a memory and you almost don’t want to violate the memory by seeing the reality, but no, it’s good. It is pretty, on an aesthetic level.” ~Heiko


In the final step, themes were reviewed for prevalence and for relevance in explaining similarities and differences among PARs. Here, prevalence is the degree to which a theme or sub-theme was common in some form across cases. Relevance is the degree to which a theme encapsulated a particular quality in the data that contributed to a deeper understanding of the PAR as social or non-social (Braun and Clarke, 2006). Relevance was considered both in how it stood on its own, and how it combined with other themes to present a clear picture of PAR features. Finally, those themes most prevalent and relevant, along with their originating data segments, were cross-referenced with existing literature and triangulated with screen captures to name, contextualize, and identify associations among them.




Analysis revealed that three characteristics of human social relationships were present in PARs: self-differentiation, emotional intimacy, and variations in perceived agency.


Relationships varied in the degree to which the player and avatar were experienced as distinct social agents. In some, avatars’ preferences, goals, or motivations were separate from players’, while in others avatars, they were extensions of players. Self-differentiation was revealed in adverb use — using “I” or “he” or even “we” to refer to avatars — and in stories of avatars’ existence as whole, independent personalities in the gameworld. For example, Heiko spoke of his avatar’s resistance to completing an infamous torture quest, despite his own interest as a gamer in seeing how it played out. “He didn’t do it,” he said. “I just could not do that quest with him and it made no sense.” Conversely, other relationships featured self-similitude, as players described avatars as objects they owned and used in the same way one might use a car or a golf club. In those cases, the avatar was a plaything — something rather than someone — and were most often referred to as “it” or “me.” Chris, for example, actively rejected that avatars are personalities: “I don’t try to put him on a pedestal. He’s nothing but an extension of my arm into that universe. He’s not his own entity, he’s me.”

These variations in avatars’ distinct social or functional agents represent different levels of relational self-differentiation (Bowen, 1978). These differences suggest that, in some PARs, the connection goes beyond a “user-tool” relation to something more akin to an interpersonal relationship — a “we.”

Emotional intimacy

Emotional intimacy emerged as an important PAR feature, varying from a distinct lack of sentiment to intense emotional ties. All players expressed some degree of emotional attachment to their avatars, however that attachment ranged from a strategic attachment to the avatar as an optimized gamepiece to a social attachment to the avatar as a unique, independent companion. Some players also suggested a sense of being cared for by the avatar. Emotional intimacy with avatars is manifested in three sub-themes: language of care, senses of sharing experiences, and appreciating the relationship as meaningful rather as offering competitive advantage in the game.

Language of emotion and care. The use of emotional language suggested a range of intimacies, from dismissive detachment to deep respect, love, and loyalty. For example, Dani noted, “I don’t think I will ever stop feeling a connection to the character ... All that history and all that emotion makes a really realistic character out of someone who started out as a joke.” Sometimes expressions of care manifested in statements of commitment, as in Chas’ contention that although he would consider re-customizing or even deleting most of his characters, his main avatar had “immunity” and would never be changed. In contrast, other players either used no emotional or caring language, or explicitly stated that their avatar was merely a thing used to play the game. As Synth explained, “If we were all just un-textured blocks that were running around, I would still be an un-textured block running around.”

Shared experience. In some PARs, players’ own motivations and abilities were the key drivers of gameplay experiences, as they focused on their own sense of control, skill, and problem-solving. In contrast, some PARs featured player and avatar sharing meaningful experiences, from killing difficult monsters “together” to jointly making decisions about what to do in the game. Often, these stories were accompanied by the avatar having played an important role in the player’s life more generally. For Kayne, his avatar “was the cornerstone of [his] social life for a couple of years.” Dani explained that she had been in an abusive relationship and that she found strength in her avatar’s personality, noting “I just don’t think I’ll ever shake the feeling that he saved my life.” In these ways, the perception of shared experience was a way that avatars were experienced as having authentic influence and meaning.

Relationship benefits. PARS were described as providing many different benefits to players — and in some cases to avatars — that can be best understood as being strategic or non-strategic benefits. Strategic benefits were ways that PARs afforded players a competitive, social, or resource advantage in the game. Alex, for instance, said that being dedicated to a single character “lets me continue doing what I like ... not having to re-learn anything,” so that even when class mechanics changed, he still held competitive advantage. Non-strategic benefits were concerned with seeing intrinsic value in the relationship. That is, players and avatars were experienced as being connected simply out of mutual care and respect, and that connection was in itself meaningful. For example, Chas lauded, “I’m proud of my boy.” Heiko noted, “I actually, thought of him as this person ... that I had a fondness for ... pretty early on.” These patterns align with Oliver and Rainey’s (2011) notion of hedonic enjoyment (fun, amusement, and reward) versus eudaimonic appreciation (meaningfulness, goodness, and virtue).

Overall, emotional language, shared experiences, and relationship benefits reflect a continuum of low to high emotional intimacy between players and avatars — findings that align roughly with Sinclair and Dowdy’s (2005) five dimensions of interpersonal emotional intimacy: acceptance, openness to sharing thoughts and feelings, understanding of thoughts and feelings, feelings of deep care, and unconditional reliability.

Perceived agency

Agency in PARs can be best understood as the perception of an entity — player or avatar — being “in charge” of gameplay. In some PARs, avatars were seen as independent agents driving play decisions, and in others as tools completely subjugate to a player’s will. Although no players treated an avatar as completely capable of achieving its own goals, evidence suggests that some players perceived themselves as being subject to an avatar’s motivations, morality, and intention. Overall, perceived agency depended on the governance of both moral and functional dimensions of play.

Moral agency. Moral agency is the ability to carry out moral reasoning — to consider the consequences of one’s actions and to take responsibility for those actions (Kohlberg, 1958). The primary moral agent in PARs was sometimes the avatar and sometimes the player. This sense of being morally in charge of play manifested in two themes: making decisions and taking responsibility for those decisions.

Making decisions. One aspect of being morally in charge of gameplay was having the right or privilege to make decisions. Often, players described themselves as being the decision-maker as they focused on their own motivations and needs. For example, Alex said in relation to choosing combat roles, “I have a lot more control over the success of a group [as a healer] than I did as just a damage dealer and it’s much more engaging.” Sometimes players described their avatars as making decisions about gameplay activities. Berkana, for instance, said that her avatar, a male Troll, was aware of his sameness and difference from other avatars: “He doesn’t flirt with other races because he’s kind of a racist.” This characteristic, explained Berkana, was not her own feelings about WoW races, but rather the Troll’s alone. Considering the avatar as a decision-maker often relied on complex backstories and well-developed avatar personalities, much as fiction-writers create for book characters. Although associated with roleplay, not all PARs with morally agentic avatars were involved with formal roleplaying.

Sometimes players described sharing decision-making with the avatar. This was most often based on player and avatar having common interests, similar preferences, or shared attributes. For example, Mingus explained how he could “be himself” by virtue of his avatar’s race and class lore: “It’s fun to be able to take that part of my personality that’s not really appropriate in real life ... You do it in game as a Warlock, as an Undead, especially ’cause people expect us to be particularly terrible.” As such, decision-making was either a negotiation or a shared experience drawing on motivations of both player and avatar.

Taking responsibility. Being morally in charge was most often a function of who should take responsibility for decisions by realizing benefits of positive actions or paying consequences for negative actions. Most often, players took responsibility when a benefit or reward was presented (e.g., sense of pride, team recognition, community prestige, reward). Other times, players avoided responsibility by blaming their avatars, although some struggled with the habit, as when Heiko noted: “It’s a poor musician who blames his instrument, but he’s a very convenient target if I fail at a raid or I fail at some quest.” Consistent with theories of attribution bias (Heider, 1958), players tended to accept positive responsibility and avoid negative responsibility.

Sometimes avatars were perceived as independently able, even duty-bound, to take responsibility for the in-game actions. As such, avatars were framed as able to know and understand things about the world and about themselves, feel emotions about world events, and sustain a moral code. Dani, for example, described how her avatar became a vegetarian: “The way he liked killing the animals kind of bothered him, so a few years later when he felt a bit better and more in control of himself, he just stopped eating any kind of meat altogether to make up for it.” As with decision-making, avatars’ responsibility for play actions relied on character narratives that included ways that avatars think, feel, and understand meaning in the broader gameworld narrative.

On few occasions, players described joint responsibility for the consequences of jointly or singularly decided and performed actions, as when Heiko described his sense of responsibility to be true to his avatar’s character. In negotiating with his avatar on whether or not to complete violent quests, he noted “ultimately, I would say it serves both of our purposes to have the richest exchange we can. Usually what he wants is what I want from the game too because serving his ambitions, his goals, brings me to a more fulfilling game.” That players sometimes consider the repercussions of actions to the avatars (in addition to themselves) highlights the ways players see their avatars as being influenced by shared experiences.

Functional agency. Independent of moral agency, functional agency is the capability to make manifest the decisions made by the moral agent. It is an “internal instrumentality through which external influences operate mechanistically, without motivation, self-reflection, self-reaction, creative, self-directive properties” [4].

Most often, players were the primary functional agent, focusing on their own physical abilities. Of particular importance is the act of button-pushing: Randy spoke of “mashing a button” to cast a spell, and Dani similarly noted, “You push the button that heals them and then the monsters die faster.” Players also enacted intention in other ways, such as changing servers, switching teams, and communicating with other players.

Sometimes the avatar was the primary functional agent, focusing on mechanics, statistics, and features of the avatar. Sometimes this avatar agency was described explicitly, as when Dani noted, “He joined a guild last November.” For Dani, it was the avatar that actually joined the guild, separate from her own action of clicking buttons to link the avatar with the guild structure. As with decision-making, this enactment was situated in the gameworld narrative — an avatar cannot control interface functions to put its name on the roster, however it can join a guild as a persona within its own character and game narratives. Other times, descriptions of avatars enacting intention were more implicit and functional. Chad, for example, outlined the how his avatar’s class mechanics and defensive abilities offered competitive advantage against a particular boss. In this way, the avatar enacts the player’s intention through its internal mechanics: governing statistical systems narratively framed as spells and abilities.

Some PARs feature both agents jointly enacting intention. Often this was a function of the player and/or the avatar having some deficit, so that an action required combining actual or perceived knowledge, abilities, resources, or materiality of both agents. For example, Chas wanted to high-five other avatars in the game, but was limited by his avatar’s inability to do so: “If you want to interact with somebody else ... you high five ’em, right? But you can’t do that in WoW ... [instead,] when you’re in a dungeon, you’re just kind of like bow and toss a quick heal that’s unnecessary — just brings you closer.” In other words, the player could not physically high-five (as he was not present in the gameworld) nor could the avatar digitally high-five (since the game does not allow avatars to interact thusly). What Chas could do is re-interpret the high-five gesture and type in the commands or press the buttons to enact those gestures. His own cognitive and button-pushing abilities were combined with the avatar’s in-world abilities to jointly perform an intended friendly act.

Considered together, attributions of functional and moral agency were inversely associated. That is, where players saw themselves as the moral agent, the avatar tended to be the functional agent; where players saw the avatar as the moral agent, they tended to see themselves as the functional agent. As such, in PARs sometimes the avatar is a tool to enact the player’s will and other times the player is a tool to enact the avatar’s perceived will. Further, overall emphasis on each type of agency was not equally weighted. Moral agency — the ability to make decisions in particular — was more important to an overall sense of being “in charge” of gameplay. The agent most often seen as driving gameplay according to motivations, preferences, and potential outcomes was seen as the stronger agent.



Player-avatar relationships as variably social

This study examined the potential for characteristics of human social relationships to manifest in PARs. Data suggest that PARs can feature such characteristics — self-differentiation, emotional intimacy, and mutual influence through perceived agency — to different degrees. Further, analysis revealed that where self-differentiation was strongest, emotional intimacy and perceiving the avatar as the primary agent were also strongest. Conversely, where self-differentiation was weakest (that is, where the avatar was merely a tool or an extension of the player), emotional intimacy and perceiving the avatar as the primary agent were also weakest. In this way, PARs can be understood as more or less social according to the ways the avatar is experienced by the player as an independent but beloved agent (see Figure 1).


Sociality of player-avatar relationships as a function of self-differentiation, emotional intimacy, and perceived agency
Figure 1: Sociality of player-avatar relationships as a function of self-differentiation, emotional intimacy, and perceived agency.


PAR sociality and motivations for play

Existing scholarship from a parasocial perspective has linked player-avatar connections with both broad motivations for play (Lewis, et al., 2008) and particular motivations such as prosocial and antisocial behaviors (Bowman, et al., 2012), and hedonic/control versus eudaimonic/introspection entertainment gratifications (Bowman, et al., 2013). This scholarship informs how players may differentially find meaning in, learn from, and be entertained by game content as a function of the one-way connection. As such, it is important to extend understandings of how variations in PAR sociality are related to gameplay motivations. Indeed, in addition to PAR features of self-differentiation, emotional intimacy, and perceived agency, players described different game-driven (e.g., completing a dungeon) and player-driven (such as making friends or escaping stress) play practices. That is, differences in PAR sociality aligned with different motivations for play. While all players described multiple play motivations, analysis revealed that one of four motivation types was central to each PAR.

Combat and competition. In some PARs, combat and competition were the most important aspects of gameplay. Chad, for instance, noted the importance of skill and competitive ranking in playing his Death Knight avatar: “My DK is my pride and joy. I’m really good at DK. I ranked 95 percentile on heroic Hagara last week!” This also manifests in emergent play, as Mow made it a personal challenge to collect all available alchemy recipes. In combat and competition practices, players focused on the game as a rivalry, race, sport, or otherwise contentious endeavor in which success is determined according to how quickly, efficiently, or completely game-defined goals are accomplished. In these PARs, playing WoW is primarily an enterprise to be the best.

Social play and play-as-ritual. Other players emphasized social or ritual motivations for gameplay. Mikey, for instance, highlighted regular social interaction as a reason he continues to play: “Especially if you’re in a guild and you’re in a guild with a bunch of friends. It’s kind of a nice change from maybe always going out to the bars or something.” Often, players described the play as a daily ritual: “It’s just a force of habit to log in ... when I come home from work ... it’s always nice and comfortable just to log into the game.” In PARs driven by social and ritual play, avatars are vehicles facilitating players’ connections with people (especially faraway friends), community membership, and sense of belonging and comfort. For these players, the game is a social arena and a type of “third place” (Steinkuehler and Williams, 2006).

Identity negotiation and sense-making. In other PARs, gameplay was associated with identity negotiation and sense-making activities. That is, some players engaged avatars in attempts to close the gap between current and ideal selves, resolve beliefs about how others see them, or experiment with alternate embodiments and personalities (see Taylor, 2002). “I was not feeling like a fun person,” said Kayne of a particularly difficult time in his life, during which he was closely connected to his avatar. “I think [my avatar] was me without those burdens.” In another case, Colleen suffered from severe social anxiety after childhood bullying, and crafted an outgoing personality for her avatar and practiced being social in the game: “[My avatar] is the person that is trapped inside me. She is my way of being who I want to be as far as personality.” In identity negotiation and sense-making motivations, players and avatars were cooperative agents in the imagining, crafting, performing, and sometimes adopting of possible selves.

Escape and segmentation. Other players said they used the game to escape the boredom or stress of their everyday lives. “It lets me get away from stinky diapers,” said Heiko, a new father. For Dani, WoW was a safe space with friends and activities that were otherwise inaccessible to her. To keep the game a separate and enjoyable place to which they could escape, some players focused on keeping game and non-game worlds as completely separate spaces. In terms of PARs, this practice often included maintaining distinct demarcations of player versus avatar and of everyday life versus game life. For example, Penny said, “I’m ‘Misha the Player’ and she’s ‘Misha the Tauren.’” Berkana opted for character narratives different than her own: “I’ve got a [husband-and-wife story] in real life, why would I roleplay it?” With and through avatars, players created and engaged digital embodiments as real, comfortable alternatives to the rigors of work, family, social and economic strife, and physical limitations.

Broadly, these patterns align with Yee’s (2006) motivations for play: achievement (advancement, mechanics, and competition), social (socializing, relationships, and teamwork), and immersion (discovery, roleplaying, customization, and escapism). That these motivations tended to align with PARs’ positions on the sociality continuum suggests that discrete dimensions of play — such as the unassuming avatar — are closely tied to game engagement factors, as discussed in the next section.

A social typology of player-avatar relationships

By examining alignments in social features and play practices, PARs can be best understood according to a four-point typology. From this framework, PARs are situated along a continuum of socialness, where primary play practices and motivations define demarcations along that continuum (see Figure 2). This typology details PAR socialness and motiviations according to the roles that avatars take up in the relationship — functional objects, player representations, symbiotes, or social others — and serves as a comprehensive framework to amalgamate disparate perspectives on player-avatar relations.


A social typology of player-avatar relationships
Figure 2: A social typology of player-avatar relationships.


Avatar-as-object relationships. Avatar-as-object relationships — in which the avatar acts as a tool — are characterized by low self-differentiation, low emotional intimacy, high player agency, and emphasis on combat and competition practices. They bear little resemblance to interpersonal social relationships. These connections are generally detached and strategic, and players in them effectively and enjoyably play the game without much consideration of the digital body as a body proper. This type of connection aligns with literature characterizing avatars as tools (Linderoth, 2005) and bundles of resources (Castronova, 2005). Playing the avatar as an object is associated with a style of gameplay focused on game-defined goals over goals of social interaction or identity expression. This PAR constitutes an exchange relationship (Mills and Clark, 1982) rather than an intimate, social relationship.

Avatar-as-Me relationships. Avatar-as-Me relationships — in which the avatar acts as a mirror or extension of the player — have low self-differentiation, low to moderate emotional intimacy, high player agency, and emphasize social play and gaming as a ritual or practice. These connections also bear little resemblance to human social relationships because the avatar “is Me,” so there is not an “other” to care for or to share experience with. These relationships tend to feature mild expressions of care, but these convey affection for how the item “is Me,” as one might express care for a favorite t-shirt or childhood toy. In this PAR type, avatars reify players’ senses of Self in the gamespace, both in terms of identity (through roles, appearance, personality) and agency (through validation and functional enactments of intention). Play was approached as a social activity engaged through the avatar, aligning with notions that avatars are social surrogates (Gee, 2006), vehicles for play (Carr, 2002), and identity bricolages (Turkle, 1997).

Avatar-as-symbiote relationships. Avatar-as-symbiote relationships — in which the avatar is intertwined with the player — have moderate or shifting self-differentiation, moderate to high emotional intimacy, mixed player and avatar agencies, and emphasis on negotiating identities and sense-making. These PARs in some ways mirror human social relationships, as players engage avatars in some meaningful transactions. Although this PAR type aligns, in part, with literature characterizing avatars as masks (Galanxhi and Nah, 2007) and costumes (Merola and Peña, 2010), these metaphors are extended by incorporating a form of identity exchange between player and avatar. That is, the player does not merely “wear” the avatar as a mask or costume, rather both player and avatar engage in cooperative processes of becoming more alike, usually toward an ideal self. In other words, players these PARs craft ideal or alternate personas (e.g., sober, brave, strong, happy, social, independent) and use the avatar to practice being that persona. After successful practice, players started bringing dimensions of that persona into non-game spaces. Taking up the avatar as a symbiote, or as a partner in play, is associated with experimentation and exploration within both gameworld and everyday narratives.

Avatar-as-other relationships. Avatar-as-other relationships — in which the avatar is a distinct social agent — feature high self-differentiation, high emotional intimacy, high avatar agency, and an emphasis on escapism and separating game and non-game realities. These are authentic social relationships between player and avatar, where the avatar is experienced as a distinct moral agent with its own governing systems, life history, and trajectory, and as embodying independent existence within the gameworld. Although narrative is key in legitimizing avatars as others (see Webb, 2001), the avatar’s perceived autonomy goes beyond narrative: in these PARs the avatar is a real amalgam of body, personality, behaviors, and beliefs through and about which new narratives emerge over time; emotional intimacy emerges through such narratives (see Wiener and Mehrabian, 1968).



Discussion and conclusion

In summary, analysis revealed that players and avatars relate in different ways that can be understood as patterns in self-differentiation, emotional intimacy, perceived agency that give rise to variations in socialness and align with gameplay motivations. The PAR typology presented in this article is a framework for examining PARs as more or less like human social relationships; existing characterizations of PARs (e.g., totems, masks, possibilities, tools) can be situated in this framework and understood in relation to one another.

Of particular importance in this examination is that some players relate to avatars as real social others, a relation that gives rise to authentic emotional intimacy. Recent scholarship has highlighted the question of whether or not humans can form authentic relationships with technologies made by them and for them (e.g., Turkle, 2007), especially when that creation maps human attributes to those devices (Lee, 2007). For Turkle, this is a question not of whether or not a machine is really capable of emoting in return, but of how humans’ relational stance evokes vulnerability and heightened connection to objects when they thrive in human care. The PAR types presented in this study provide an answer to Lee’s and Turkle’s questions: humans can form authentic relationships with technologies on the basis of real emotional vulnerability, senses of connection, nurturing, confirmation of worth, commitment, information exchanges, reliable companionship, and the experiences of alliance and shared experience. These qualities satisfy the criteria for an authentic, intimate social relationship (see Granovetter, 1973; Sinclair and Dowdy, 2005; Weiss, 1974).

In short, the PAR sometimes goes beyond a user-tool connection to a “we.” These potentially social relationships are particularly significant considering that leading game developers are asking complex questions about what avatars are, what they mean, and how they will evolve. Game designer Jesse Schell (2013) spoke of a vision for “digital companions” that accompany players over their lifetimes and that are inherited by each users’ progeny. In the Webcast revealing the Xbox One gaming console, using the device was described as a “deep companion experience” (Microsoft, 2013). Game narratives and challenges are being written to provoke visceral emotional reactions and moral dilemmas (Posey, 2013; Rouse, 2013), to facilitate the perception of romantic relationships between players and characters, and to guide us in our daily lives (McGonigal, 2013). Not only are avatars and other game objects sometimes taken up in social relationships, but those objects are being designed, specifically, to be social. As such, it is increasingly important to advance understandings of how humans connect with them as social, subjective, human-like agents (Turkle, 2010).

Limitations and future research

The exploratory nature and limited scope of this study carries with it a number of limitations. Although I sought theoretical saturation through careful sampling for a range of player and avatar characteristics, qualitative data and analyses are always subject to the limitations of small sample sizes and self-selection. The author’s current research efforts expand this line of inquiry to a) evaluate generalizability to a larger sample of WoW PARs, b) evaluate generalizability to PARs emerging from other virtual worlds and games, and c) to develop an instrument to measure dimensions of socialness. Finally, as with any interpretive research, findings were subject to my own perspectives and theoretical lens, despite addressing this potential through member checks and triangulations.

This study’s findings suggest future research directions that will advance understandings of social PARs and of human-technology relationships more broadly. Future research should examine the extent to which the PAR features extend other digital bodies and representations (e.g., Facebook profiles, robots, non-humanoid avatars), other gameplay conditions (e.g., assigned avatars or games with extensive customization options), and other degrees of virtuality (e.g., tabletop games and augmented reality scenarios). It is possible that degrees of self-differentiation and socialness in the PAR may moderate how players interact with digital game content and with other players. For example, it is possible that PAR or its constitutive dimensions could contribute to prosocial and antisocial behaviors in-game or outside of gameplay, to particular identity performances, and to self-efficacy and learning. Of further note, the information exchanges relevant to PAR variations in this study were largely graphical and procedural — that is, the avatar’s appearance and behaviors/logics mattered — but the textual and aural avatar elements did not emerge as relevant in players’ experiences; future research should evaluate the role of those modalities in forming and maintaining PARs, since audio cues in particular are understood to diagetically situate the controlled avatar in the immersive gameworld (Jørgensen, 2008).

Additionally, the potential for genuine relationships between humans and media technologies has particular implications for the application of communication theories to human-technology relationships. As suggested by Reeves and Nass (1996), since these relationships are fully social it may be useful to extend theories of interpersonal communication to the study of PARs. For example, the Hyperpersonal Model (Walther, 1996) may be extended to consider the ways that humans deal with “impoverished” cues where the technology does not yet speak back; symbolic interaction perspectives (e.g., Mead, 1934) may be taken up to better understand how meaning emerges through human-technology interactions; social penetration theory (Altman and Taylor, 1973) may frame investigations of human-technology relationships over time.

Finally, an important discovery in this investigation is initial evidence that the nature of PARs may shift over time. For example, some players described that their avatars were mere objects until they were encouraged by friends to craft character narratives, suggesting that narrative may be a key driver in forging feelings of deeper emotional intimacy and perceptions that avatars exist independently. Conversely, some players spoke of how their favorite avatar used to be a “fun toon” and shifted to a “work toon” when they started using it for raiding, suggesting that increased play intensity or optimizing the avatar as a play tool may be associated with a decrease in emotional intimacy and heightened player agency. Although such shifts were beyond the scope of this study, they have not yet been addressed in the literature and future research should address these potentials.End of article


About the author

Jaime Banks is an Assistant Professor in the Department of Communication Studies at West Virginia University.
E-mail: jabanks [at] mail [dot] wvu [dot] edu



1. Meadows, 2008, p. 23.

2. Apter, 2008, para. 1.

3. Little, 1999, p. 3.

4. Bandura, 1989, p. 1,175.



Irwin Altman and Dalmas A. Taylor, 1973. Social penetration: The development of interpersonal relationships. New York: Holt, Rinehart and Winston.

Emily Apter, 2008. “Technics of the subject: The avatar-drive,” Postmodern Culture, volume 18, number 2, at http://www.english.illinois.edu/-people-/faculty/schaffner/w4w/readings/apter.pdf, accessed 19 January 2015.
doi: http://dx.doi.org/10.1353/pmc.0.0021, accessed 19 January 2015.

Albert Bandura, 1989. “Human agency in social cognitive theory,” American Psychologist, volume 44, number 9, pp. 1,175–1,184.
doi: http://dx.doi.org/10.1037/0003-066X.44.9.1175, accessed 19 January 2015.

Marlin C. Bates, 2009. “Persistent rhetoric for persistent worlds: The mutability of the self in massively multiplayer online role-playing games,” Quarterly Review of Film and Video, volume 26, number 2, pp. 102–117.
doi: http://dx.doi.org/10.1080/10509200600737770, accessed 19 January 2015.

Ellen Berscheid and Letitia Anne Peplau, 1983. “The emerging science of relationships,” In: Close relationships. New York: W.H. Freeman, pp. 1–19.

Katherine Bessière, A. Fleming Seay, and Sara Kiesler, 2007. “The ideal elf: Identity exploration in World of Warcraft,” CyberPsychology & Behavior, volume 10, number 4, pp. 530–535.
doi: http://dx.doi.org/10.1089/cpb.2007.9994, accessed 19 January 2015.

Joël Billieux, Martial Van der Linden, Sophia Achab, Yasser Khazaal, Laura Paraskevopoulos, Daniele Zullino, and Gabriel Thorens, 2013. “Why do you play World of Warcraft? An in-depth exploration of self-reported motivations to play online and in-game behaviours in the virtual world of Azeroth,” Computers in Human Behavior, volume 29, number 1, pp. 103–109.
doi: http://dx.doi.org/10.1016/j.chb.2012.07.021, accessed 19 January 2015.

Blizzard Entertainment, 2010. “World of Warcraft subscriber base reaches 12 million worldwide” (7 October), at http://us.blizzard.com/en-us/company/press/pressreleases.html?id=2847881, accessed 19 January 2015.

Ian Bogost, 2012. Alien phenomenology, or, What it’s like to be a thing. Minneapolis: University of Minnesota Press.

Megan Boler, 2007. “Hypes, hopes and actualities: New digital Cartesianism and bodies in cyberspace,” New Media & Society, volume 9, number 1, pp. 139–168.
doi: http://dx.doi.org/10.1177/1461444807067586, accessed 19 January 2015.

Murray Bowen, 1978. Family therapy in clinical practice. New York: J. Aronson.

Nicholas David Bowman, Daniel Schultheiss, and Christina Schumann, 2012. “‘I’m attached, and I’m a good guy/gal!’ How character attachment influences pro- and anti-social motivations to play MMORPGs,” Cyberpsychology, Behavior, and Social Networking, volume 15, number 3, pp. 169–174.

Nicholas David Bowman, Ryan Rogers, Brett Sherrick, and Julia Wooley, 2013. “In control or in their shoes: How character attachment differentially influences video game enjoyment and appreciation,” paper presented at the Broadcast Education Association Research Symposium (7 April).

Virginia Braun and Victoria Clarke, 2006. “Using thematic analysis in psychology,” Qualitative Research in Psychology, volume 3, number 2, pp. 77–101.
doi: http://dx.doi.org/10.1191/1478088706qp063oa, accessed 19 January 2015.

Andrew Burn, 2003. “Poets, skaters and avatars: Performance, identity and new media,” English Teaching: Practice and Critique, volume 2, number 2, pp. 6–21.

Dianne Carr, 2002. “Playing with Lara,” In: Geoff King and Tanya Krzywinska (editors). ScreenPlay: Cinema/videogames/interfaces. London: Wallflower Press, pp. 171–180.

Edward Castronova, 2005. Synthetic worlds: The business and culture of online games. Chicago: University of Chicago Press.

Holtjona Galanxhi and Fiona Fui-Hoon Nah, 2007. “Deception in cyberspace: A comparison of text-only vs. avatar-supported medium,” International Journal of Human-Computer Studies, volume 65, number 9, pp. 770–783.
doi: http://dx.doi.org/10.1016/j.ijhcs.2007.04.005, accessed 19 January 2015.

James Paul Gee, 2006. “Video games as learning machines for a global world,” A Vision Splendid (17 August), at http://ccclearning.edublogs.org/2006/08/17/a-vision-splendid-james-paul-gee, accessed 19 January 2015.

James Paul Gee, 2005. “Pleasure, learning, video games, and life: The projective stance,” E-Learning and Digital Media, volume 2, number 3, pp. 211–223.

Seth Giddings, 2007. “Playing with non-humans: Digital games as techno-cultural form,” In: Suzanne de Castell and Jennifer Jenson (editors). Worlds in play: International perspectives on digital games research. New York: Peter Lang, pp. 115–128.

Barney G. Glaser, 2012. “Stop. Write! Writing grounded theory,” Grounded Theory Review, volume 11, number 1, at http://groundedtheoryreview.com/2012/06/01/stop-write-writing-grounded-theory/, accessed 19 January 2015.

Barney G. Glaser and Anselm L. Strauss, 1967. The discovery of grounded theory: Strategies for qualitative research. Chicago: Aldine.

Erving Goffman, 1959. The presentation of self in everyday life. Garden City, N.Y.: Doubleday.

Mark S. Granovetter, 1973. “The strength of weak ties,” American Journal of Sociology, volume 78, number 6, pp. 1,360–1,380.

Greg Guest, Kathleen M. MacQueen, and Emily E. Namey, 2012. Applied thematic analysis. Los Angeles: Sage.

Charlotte Hagström, 2008. “Playing with names: Gaming and naming in World of Warcraft,” In Hilde G. Corneliussen and Jill Walker Rettberg (editors). Digital culture, play, and identity: A World of Warcraft reader. Cambridge, Mass.: MIT Press, pp. 265–286.

Donna Haraway, 1991. Simians, cyborgs, and women: The reinvention of nature. New York: Routledge.

Graham Harman, 2005. Guerrilla metaphysics: Phenomenology and the carpentry of things. Chicago: Open Court.

John H. Harvey and Brian G. Pauwels, 2009. “Relationship connection: A redux on the role of minding and the quality of feeling special in the enhancement of closeness,” In: Shane J. Lopez and C.R. Snyder (editors). The Oxford handbook of positive psychology. New York: Oxford University Press, pp. 385–392.
doi: http://dx.doi.org/10.1093/oxfordhb/9780195187243.013.0036, accessed 19 January 2015.

Fritz Heider, 1958. The psychology of interpersonal relations. New York: Wiley.

Donald Horton and R. Richard Wohl, 1956. “Mass communication and para-social interaction: Observations on intimacy at a distance,” Psychiatry, volume 19, number 3, pp. 215–229.

Kristine Jørgensen, 2008. “Audio and gameplay: An analysis of PvP battlegrounds in World of Warcraft,” Game Studies, volume 8, number 2, at http://gamestudies.org/0802/articles/jorgensen, accessed 19 January 2015.

Daniel King, Paul Delfabbro, and Mark Griffiths, 2010. “Video game structural characteristics: A new psychological taxonomy,” International Journal of Mental Health and Addiction, volume 8, number 1, pp. 90–106.
doi: http://dx.doi.org/10.1007/s11469-009-9206-4, accessed 19 January 2015.

Lawrence Kohlberg, 1958. “The development of modes of moral thinking and choice in the years 10 to 16,” Ph.D. dissertation, University of Chicago.

Bruno Latour, 2005. Reassembling the social: An introduction to actor-network-theory. New York: Oxford University Press.

Billy Lee, 2007. “Nonverbal intimacy as a benchmark for human-robot interaction,” Interaction Studies, volume 8, number 3, pp. 411–422.
doi: http://dx.doi.org/10.1075/is.8.3.06lee, accessed 19 January 2015.

Lawrence Lessig, 1999. Code and other laws of cyberspace. New York: Basic Books.

George Levinger, 1979. “A social exchange view on the dissolution of pair relationships,” In: Robert L. Burgess and Ted L. Huston (editors). Social exchange in developing relationships. New York: Academic Press, pp. 169–193.

Melissa L. Lewis, René Weber, and Nicholas David Bowman, 2008. “‘They may be pixels, but they’re MY pixels’ Developing a metric of character attachment in role-playing video games,” CyberPsychology & Behavior, volume 11, number 4, pp. 515–518.
doi: http://dx.doi.org/10.1089/cpb.2007.0137, accessed 19 January 2015.

Dong Dong Li, Albert Kien Liau, and Angeline Khoo, 2013. “Player–avatar identification in video gaming: Concept and measurement,” Computers in Human Behavior, volume 29, number 1, pp. 257–263.
doi: http://dx.doi.org/10.1016/j.chb.2012.09.002, accessed 19 January 2015.

Jonas Linderoth, 2005. “Animated game pieces. Avatars as roles, tools and props,” paper presented at Aesthetics of Play conference (14–15 October, University of Bergen, Norway), at http://www.aestheticsofplay.org/papers/linderoth2.htm, accessed 19 January 2015.

Gregory Little, 1999. “An avatar manifesto,” Intertexts, volume 3, number 2, at http://www.intertexts.org/id19.html, accessed 19 January 2015.

Michael Mateas and Andrew Stern, 2006. “Interaction and narrative,” In: Katie Salen and Eric Zimmerman (editors). The game design reader: A rules of play anthology. Cambridge, Mass.: MIT Press, pp. 642–669.

Victoria McArthur, 2008. “World of Warcraft as a ludic cyborg,” Future Play ’08: Proceedings of the 2008 Conference on Future Play: Research, Play, Share, pp. 264–265.
doi: http://dx.doi.org/10.1145/1496984.1497046, accessed 19 January 2015.

Jane McGonigal, 2013. “There is no escape: Designing games for maximum real-life impact,” paper presented at the Game Developers Conference, San Francisco, at http://janemcgonigal.com/2013/03/27/there-is-no-escape-designing-videogames-for-maximum-real-life-impact/, accessed 19 January 2015.

George Herbert Mead, 1934. Mind, self & society from the standpoint of a social behaviorist. Chicago: University of Chicago Press.

Mark Stephen Meadows, 2008. I, avatar: The culture and consequences of having a second life. Berkeley, Calif.: New Riders.

Nicholas Merola and Jorge Peña, 2010. “The effects of avatar appearance in virtual worlds,” Journal of Virtual Worlds Research, volume 2, number 5, at https://journals.tdl.org/jvwr/index.php/jvwr/article/view/843, accessed 19 January 2015.

Microsoft, 2013. “Xbox, a new generation revealed,” at http://www.xbox.com/, Webcast released 30 May 2013.

Judson Mills and Margaret S. Clark, 1982. “Exchange and communal relationships,” In: Ladd Wheeler (editor). Review of personality and social psychology, volume 3. Beverly Hills, Calif.: SAGE, pp. 121–144.

Kristine L. Nowak and Christian Rauh, 2005. “The influence of the avatar on online perceptions of anthropomorphism, androgyny, credibility, homophily, and attraction,” Journal of Computer–Mediated Communication, volume 11, number 1, pp. 153–178.
doi: http://dx.doi.org/10.1111/j.1083-6101.2006.tb00308.x, accessed 19 January 2015.

Mary Beth Oliver and Arthur A. Raney, 2011. “Entertainment as pleasurable and meaningful: Identifying hedonic and eudaimonic motivations for entertainment consumption,” Journal of Communication, volume 61, number 5, pp. 984–1,004.
doi: http://dx.doi.org/10.1111/j.1460-2466.2011.01585.x, accessed 19 January 2015.

W. Barnett Pearce, 2007. Making social worlds: A communication perspective. Malden, Mass.: Blackwell.

Jay Posey, 2013. “Tastes like chicken: Authenticity in a totally fake world,” paper presented at the Game Developers Conference, San Francisco, at at http://schedule2013.gdconf.com/session-id/823740, accessed 19 January 2015.

Byron Reeves and Clifford Nass, 1996. The media equation: How people treat computers, television, and new media like real people and places. Stanford, Calif.: CSLI Publications.

Bob Rehak, 2003. “Playing at being: Psychoanalysis and the avatar,” In: Mark J.P. Wolf and Bernard Perron (editors). The video game theory reader. New York: Routledge, pp. 103–127.

Scott Rettberg, 2008. “Corporate ideology in World of Warcraft,” In: Hilde G. Corneliussen and Jill Walker Rettberg (editors). Digital culture, play, and identity: A World of Warcraft reader. Cambridge, Mass.: MIT Press, pp. 19–38.

Richard Rouse III, 2013. “Seven (or so) techniques for writing a moral game,” paper presented at the Game Developers Conference, San Francisco, at http://www.gdcvault.com/play/1018156/Seven-%28Or-So%29-Techniques-for, accessed 19 January 2015.

Jesse Schell, 2013. “The future of storytelling: How medium shapes story,” paper presented at the Game Developers Conference, San Francisco.

Ulrike Schultze and Matthew Michael Leahy, 2009. “The avatar-self relationship: Enacting presence in Second Life,” ICIS 2009 Proceedings, at http://aisel.aisnet.org/icis2009/12, accessed 19 January 2015.

Vaughn G. Sinclair and Sharon W. Dowdy, 2005. “Development and validation of the emotional intimacy scale,” Journal of Nursing Measurement, volume 13, number 3, pp. 193–206.
doi: http://dx.doi.org/10.1891/jnum.13.3.193, accessed 19 January 2015.

Charles R. Snyder and Shane J. Lopez, 2007. Positive psychology: The scientific and practical explorations of human strengths. Thousand Oaks, Calif.: SAGE.

Constance Steinkuehler and Dmitri Williams, 2006. “Where everybody knows your (screen) name: Online games as ‘third places’,” Journal of Computer-Mediated Communication, volume 11, number 4, pp. 885–909.
doi: http://dx.doi.org/10.1111/j.1083-6101.2006.00300.x, accessed 19 January 2015.

Allucquère Rosanne Stone, 1995. The war of desire and technology at the close of the mechanical age. Cambridge, Mass.: MIT Press.

Anselm L. Strauss and Juliet Corbin, 1990. Basics of qualitative research: Grounded theory procedures and techniques. Newbury Park, Calif.: Sage.

T.L. Taylor, 2009. “The assemblage of play,” Games and Culture, volume 4, number 4, pp. 331–339.
doi: http://dx.doi.org/10.1177/1555412009343576, accessed 19 January 2015.

T.L. Taylor, 2002. “Living digitally: Embodiment in virtual worlds,” In: Ralph Schroeder (editor). The social life of avatars: Presence and interaction in shared virtual environments. London: Springer, pp. 40–62.

John W. Thibaut and Harold H. Kelley, 1959. The social psychology of groups. New York: Wiley.

Sabine Trepte and Leonard Reinecke, 2010. “Avatar creation and video game enjoyment,” Journal of Media Psychology, volume 22, number 4, pp. 171–184.
doi: http://dx.doi.org/10.1027/1864-1105/a000022, accessed 19 January 2015.

Sherry Turkle, 2010. “In good company? On the threshold of robotic companions,” In: Yorick Wilks (editor). Close engagements with artificial companions: Key social, psychological, ethical and design issues. Amsterdam: John Benjamins Publishing, pp. 3–10.
doi: http://dx.doi.org/10.1075/nlp.8.03tur, accessed 19 January 2015.

Sherry Turkle, 2007. “Simulation versus authenticity,” In: John Brockman (editor). What is your dangerous idea? Today’s leading thinkers on the unthinkable. New York: Harper Perennial, pp. 244–247.

Sherry Turkle, 1997. “Multiple subjectivity and virtual community at the end of the Freudian century,” Sociological Inquiry, volume 67, number 1, pp. 72–84.
doi: http://dx.doi.org/10.1111/j.1475-682X.1997.tb00430.x, accessed 19 January 2015.

Sherry Turkle, 1995. Life on the screen: Identity in the age of the Internet. New York: Simon & Schuster.

Jan Van Looy, Cédric Courtois, and Melanie De Vocht, 2010. “Playing a self: An exploration into the effect of avatar identification on gamer ratings of self, ideal self and avatar personality in MMORPGs,” paper presented at the pre-conference to the ECREA 2010 — 3rd European Communication Conference, Avatars and Humans. Representing Users in Digital Games (Hamburg, Germany); version at https://biblio.ugent.be/publication/1059804, accessed 19 January 2015.

Joseph B. Walther, 1996. “Computer-mediated communication: Impersonal, interpersonal, and hyperpersonal interaction,” Communication Research, volume 32, number 1, pp. 3–43.
doi: http://dx.doi.org/10.1177/009365096023001001, accessed 19 January 2015.

Stephen Webb, 2001. “Avatarculture: Narrative, power, and identity in virtual world environments,” Information, Communication & Society, volume 4, number 4, pp. 560–594.
doi: http://dx.doi.org/10.1080/13691180110097012, accessed 19 January 2015.

Robert S. Weiss, 1974. “The provisions of social relationships,” In: Zick Rubin (editor). Doing unto others: Joining, molding, conforming, helping, loving. Englewood Cliffs, N.J.: Prentice-Hall, pp. 17–26.

Norbert Wiener, 1948. Cybernetics. New York: Wiley.

Morton Wiener and Albert Mehrabian, 1968. Language within language: Immediacy, a channel in verbal communication. New York: Appleton-Century-Crofts.

Dmitri Williams, Nicolas Ducheneaut, Li Xion, Nick Yee, and Eric Nickell, 2006. “From tree house to barracks: The social life of guilds in World of Warcraft,” Games and Culture, volume 1, number 4, pp. 338–361.
doi: http://dx.doi.org/10.1177/1555412006292616, accessed 19 January 2015.

Nick Yee, 2006. “Motivations for play in online games,” CyberPsychology & Behavior, volume 9, number 6, pp. 772–775.
doi: http://dx.doi.org/10.1089/cpb.2006.9.772, accessed 19 January 2015.

Nick Yee and Jeremy Bailenson, 2007. “The Proteus effect: The effect of transformed self–representation on behavior,” Human Communication Research, volume 33, number 3, pp. 271–290.
doi: http://dx.doi.org/10.1111/j.1468-2958.2007.00299.x, accessed 19 January 2015.


Editorial history

Received 2 July 2014; revised 23 January 2015; accepted 29 January 2015.

Creative Commons License
“Object, Me, Symbiote, Other: A social typology of player-avatar relationships” by Jaime Banks is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Object, Me, Symbiote, Other: A social typology of player-avatar relationships
by Jaime Banks.
First Monday, Volume 20, Number 2 - 2 February 2015
doi: http://dx.doi.org/10.5210/fm.v20i2.5433

A Great Cities Initiative of the University of Illinois at Chicago University Library.

© First Monday, 1995-2017. ISSN 1396-0466.