Available via license: CC BY-NC 3.0
Content may be subject to copyright.
Avatars and Computer-Mediated Communication:
A Review of the Definitions, Uses, and Effects
of Digital Representations
OPEN ACCESS Top-Quality Science
Peer Reviewed
Open Peer Reviewed
Freely available online
Review of Communication Research
2018, Vol.6
doi: 10.12840/issn.2255-4165.2018.06.01.015
ISSN: 2255-4165
Kristine L. Nowak
University of Connecticut (USA)
Kristine.nowak@uconn.edu
Avatars are growing in popularity and present in many interfaces used for computer-mediated communication (CMC) in-
cluding social media, e-commerce, and education. Communication researchers have been investigating avatars for over
twenty years, and an examination of this literature reveals similarities but also notable discrepancies in conceptual defini-
tions. The goal of this review is to provide a general overview of current debates, methodological approaches, and trends
in findings. Our review synthesizes previous research in four areas. First, we examine how scholars have conceptualized
the term “avatar,” identify similarities and differences across these definitions, and recommend that scholars use the term
consistently. Next, we review theoretical perspectives relevant to avatar perception (e.g., the computers as social actors
framework). Then, we examine avatar characteristics that communicators use to discern the humanity and social potential
of an avatar (anthropomorphism, form realism, behavioral realism, and perceived agency) and discuss implications for at-
tributions and communication outcomes. We also review findings on the social categorization of avatars, such as when
people apply categories like sex, gender, race, and ethnicity to their evaluations of digital representations. Finally, we exam-
ine research on avatar selection and design relevant to communication outcomes. Here, we review both motivations in CMC
contexts (such as self-presentation and identity expression) and potential effects (e.g., persuasion). We conclude with a dis-
cussion of future directions for avatar research and propose that communication researchers consider avatars not just as a
topic of study, but also as a tool for testing theories and understanding critical elements of human communication. Avatar-
mediated environments provide researchers with a number of advantageous technological affordances that can enable
manipulations that may be difficult or inadvisable to execute in natural environments. We conclude by discussing the use
of avatar research to extend communication theory and our understanding of communication processes.
Suggested citation: Nowak, K. L., & Fox, J. (2018). Avatars and computer-mediated communication: A review of the
definitions, uses, and effects of digital representations. Review of Communication Research, 6, 30-53. doi: 10.12840/issn.2255-
4165.2 018.06.01.015
Keywords: avatars, perception, digital environments, video games, computer-mediated communication, digital representations,
social actors, anthropomorphism, human-computer interaction
Received: Oc t. 17th, 2016 Open peer review: Jan. 2nd, 2017 Accepted: Oct. 27th Prepublished online: Oct. 29th
Published: Jan. 2018
Article edited by associate editor: James D. Ivory (Virginia Polytechnic Institute and State University, USA)
Journal editor: Giorgio P. De Marchis (Universidad Complutense de Madrid, Spain).
Reviewers: T. Franklin Waddell (University of Florida, USA); Malte Elson (Ruhr-University Bochum, Germany)
Abstract
Jesse Fox
The Ohio State University
fox.775@osu.edu
31
Avatars and Computer-Mediated Communication
2018, 6, 30-53
Content
• The term “avatar” is not consistently defined either operationally or conceptually by researchers in communication.
•
In computer-mediated communication, scholars agree that avatars are digital representations of the user in a digital
environment.
• The avatar influences perception and attribution of sources and messages in a digital environment.
• The computers as social actors framework and the model of social influence in digital environments lend insight into
avatars.
• Perceptions of avatar agency and social potential (e.g., anthropomorphism and behavioral realism) have implications for
digital interactions.
•
Avatar characteristics afforded by an interface can augment or limit people’s ability to self-present and engage in the
digital environment.
• Scholars can use avatars in research to facilitate experimental control and the investigation of complex communication
processes and theories.
Highlights
INTRODUCTION ........................................................................................................32
THE EMERGENCE AND DEFINITION OF AVATARS IN COMMUNICATION
RESEARCH ..................................................................................................................33
Similarities in Conceptualization ................................................................................33
Differences in Conceptualization ................................................................................33
Conceptualizing Avatar for Future Research ...............................................................34
AVATAR PERCEPTION: PROCESSING DIGITAL BODIES AND PEOPLE ..............35
Perceptions of Avatars: Humanity, Agency, and Social Potential ..................................35
Agency. ................................................................................................................. 36
Anthropomorphism. ..............................................................................................36
Realism. ................................................................................................................37
The uncanny valley. ............................................................................................... 37
Social Categorization of Avatars .................................................................................38
Sex and gender. . .................................................................................................... 38
Race and ethnicity. ................................................................................................ 39
Similarity and homophily. ......................................................................................39
AVATAR SELECTION AS SELF-REPRESENTATION ............................................... 40
Avatars as identity expression. ...................................................................................40
Identity exploration and deception. ............................................................................40
EFFECTS OF AVATAR EMBODIMENT .....................................................................41
FUTURE DIRECTIONS: USING AVATARS TO STUDY HUMAN
COMMUNICATION .................................................................................................... 42
Refining Research on Avatars .....................................................................................42
Topics for Future Directions for Avatar Research.........................................................43
Ethical Considerations ...............................................................................................43
CONCLUSION .............................................................................................................44
REFERENCES .............................................................................................................45
Kristine L. Nowak & Jesse Fox
32
www.rcommunicationr.org
Introduction
When communicating in computer-mediated environ-
ments, individuals often rely on some form of avatar, or a
digital representation that symbolizes the self in the interac-
tion. Depending on the context and definition of the term,
these representations may range from a simple screen name
or a graphical icon to a lifelike, animated three-dimension-
al character. Researchers across several disciplines and
scholarly traditions have posed questions about the influ-
ence of these avatars, both within and outside of the com-
puter-mediated environment.
Users are able to manipulate, control, embody, and in-
teract via avatars in a variety of contexts, which is of inter-
est to communication scholars because these representations
can shape computer-mediated communication (CMC) ex-
periences. For example, when users interact in digital en-
vironments, they make judgments and attributions based
on the names, appearance, and behaviors of others’ avatars;
further, users’ avatars may influence their behaviors in
interactions (E. J. Lee, 2007; Nakamura, 2002; Nowak &
Rauh, 2005). Whether intended by the sender or not, ava-
tars may also be perceived as messages in and of themselves,
such as when individuals use a political symbol as a graph-
ical avatar on a social networking site.
An avatar’s characteristics may be determined by sev-
eral factors, including user preferences, social norms, ex-
periences within the environment, and technological
affordances or constraints of the system (Blascovich &
Bailenson, 2011; Nowak, 2015; Stromer-Galley & Martey,
2009; Yee, 2014). These characteristics may include appear-
ance, traits, abilities, or behaviors that are ref lective of
human capacities and norms or complete fantasy. Differ-
ences in these representations are notable given that avatars
have been shown to influence beliefs, attitudes, and behav-
iors in a variety of contexts, including interpersonal com-
munication (Kotlyar & Ariely, 2013; Van Der Heide,
Schumaker, Peterson, & Jones, 2013; Waddell & Ivory,
2015), health communication (Ahn, 2015; Fox, 2012), group
communication (Van Der Land, Schouten, Feldberg, Huys-
man, & Van Den Hooff, 2015), environmental communica-
tion (Ahn, Bostick, Ogle, Nowak, McGillicuddy, &
Bailenson, 2016), nonverbal communication (Bente &
Krämer, 2011; Hasler & Friedman, 2012), organizational
communication (Park & Lee, 2013), and advertising (Ahn
& Bailenson, 2011).
Our aim for this article is to provide a broad overview of
avatar research in the field of communication. In our litera-
ture search, we examined every article including the term
“avatar” in its text according to the EBSCO Communication
& Mass Media database. Further, we reviewed titles and
abstracts in major communication journals based on search-
es for terms such as “virtual,” “computer,” and “digital” (e.g.,
Journal of Communication, Communication Research, Human
Communication Research, Journal of Computer-Mediated
Communication, Communication Monographs) to identif y po-
tentially relevant articles that may not have used the word
“avatar.” We also conducted a keyword search for “avatar”
in WorldCat and identif ied several relevant articles written,
or often cited by communication scholars. Based on these
searches and sifting through references cited within avatar-
related papers, we also identified communication scholars
who publish research on avatars and identified some of their
relevant work published outside of communication, though
we maintained our focus their work specific to avatars and
communication processes. Although these methods are not
perfectly systematic, this use of multiple approaches helped
identify a broad range of work of interest to communication
scholars. Obviously, we cannot discuss every article uncov-
ered in this process; thus, we identified common themes and
focused predominantly on articles relevant to communica-
tion processes rather than user experience, human-comput-
er interaction, or psychological effects of avatars independent
of a social context (e.g., individual experiences of identifica-
tion or embodiment).
In this article, we begin with a review of select definitions
of the term “avatar,” variations in the use of the term in the
field of communication, and how those variations influence
the ability to replicate results, apply theory, and make mean-
ingful, generalizable conclusions. Next, we provide an over-
view of the literature on how users select avatars and what
is known or hypothesized about the effects of these choices.
Then, we review research examining how people perceive
and make attributions about avatars in communication con-
texts. Finally, we conclude with a discussion of how avatars
can be used for theory development in communication and
pose critical questions for future research.
Avatars and Computer-Mediated Communication
33 2018, 6, 30-53
The Emergence and Definition of Avatars in
Communication Research
Before discussing the uses and effects of avatars, we will
explore the various conceptualizations of the term avatar in
the communication literature. Our examination of published
articles demonstrates that many researchers use the term
avatar but do not explicitly define it; others examine what
many would consider avatars without using the term; and
some scholars employ inconsistent definitions across their
body of work. Although it is not uncommon within the social
sciences to see conceptual definitions differ across studies,
these variations can influence not only the individual study’s
conclusions, but also the interpretation of results more
broadly within the context of communication theories and
processes (Chaffee, 1991). The absence of agreement on
conceptual definitions, inconsistent operationalization, and
appropriate manipulation checks also contribute to the rep-
lication crisis discussed by Kahneman (2012) and others.
Scientists have had difficulty replicating results even when
they are trying to test the same concepts and theories with
the same protocols (Open Science Collaboration, 2015).
Thus, variations in the use of avatar make it difficult for re-
searchers to ascertain the scope of relevant research and
complicates the process of replicating findings, leading to
further difficulty for those seeking to understand the influ-
ence of avatars. We begin our review by identifying several
recurring similarities and differences in the conceptualiza-
tion of the term.
Similarities in Conceptualization
Among articles that discussed the origins of the term,
there was agreement that avatar originated in Hinduism and
is adapted from the Sanskrit word for “descent.” In this
context, an avatar is human embodiment of a deity or a
spirit which allows them to experience earth from the per-
spective of, or to interact with, humans (Blascovich &
Bailenson, 2011; Castronova, 2005; Nowak, 2015). Although
there were earlier uses of avatar referring to computer-based
contexts, Neal Stephenson’s (1992) science fiction novel Snow
Crash is generally credited with popularizing the use of the
word to refer to representations in online or digital environ-
ments (Blascovich & Bailenson, 2011; Nowak, 2004).
Descriptions of avatars by communication scholars near-
ly universally acknowledge the avatar as a digital representa-
tion (e.g., Biocca, 1997; Blascovich & Bailenson, 2011;
Nakamura, 2002; Nowak & Biocca, 2003). The representa-
tion serves as a symbol or marker of its associated entity
within the digital space. Another commonality across most
definitions is that an avatar represents the user in a digital
environment (e.g., Kapidzic & Herring, 2011; Klang, 2004;
Nakamura, 2002), although there are variations in how
scholars label other controllers’ representations. Further,
most definitions state or imply that the purpose of an avatar
is to enable the user to experience and interact within the
spaces of digitally mediated worlds (Biocca, 1997; Yee &
Bailenson, 2009), and with other users (Meadows, 2007).
Although fully interactive digital embodiments provide a
more immersive interaction in the digital world, even a
simple screen name or static image in an online chat room
can facilitate this ability to experience digital worlds and
interact with others. Everything from a screen name to a 3D
embodiment can give receivers information about others in
the environment, help identify who is speaking, and provide
cues for conversation regulation and turn-taking (Schroeder,
2002). In virtual worlds and video games, avatars provide a
form of embodiment that enables navigation through vir-
tual spaces. In some platforms, avatars can also facilitate
more complex actions including nonverbal communication
via gestures, body posture, proxemics, and even haptics
(Biocca, 1997; Blascovich & Bailenson, 2011). These com-
monalities provide a foundation outlining the definition of
an avatar, though scholars demonstrate disagreement and
inconsistency on other aspects of the construct.
Differences in Conceptualization
Across the literature, communication scholars have em-
ployed several different conceptualizations of avatars. Here,
we focus on three distinctions that seemed most common
and consequential. First, many scholars specified a more
conservative definition that either explicitly or implicitly
required that an avatar have a graphical embodiment or vi-
sual presentation. One example is Nakamura (2002), who
described an avatar as “a visual digital representation of a
self in cyberspace” (p. 153). Several scholars have included
this qualification (e.g., Bailenson, Yee, Merget, & Schroeder,
2006; Kim & Sundar, 2012; Martey & Consalvo, 2011;
Nowak & Biocca, 2003; Peña, Hancock, & Merola,2009;
Schroeder, 2002; Van Der Heide et al., 2013; Webb, 2001).
Other researchers adopted a broader definition that
Kristine L. Nowak & Jesse Fox
34
www.rcommunicationr.org
but not others. Optimally, both conceptualization and theo-
rizing about avatars would be more generalizable and not
constrained to specif ic platforms.
Conceptualizing Avatar for Future Research
In summary, the most liberal definition of the term avatar
would entail any representation of any controller. Even in
the context of CMC, the types of representations that could
be considered an avatar are quite broad: a photograph on a
social networking or online dating site profile; a non-play-
able, computer-controlled agent in a video game; a graphical
icon in an online forum or chat; a floating hand in an aug-
mented-reality environment; one’s image in a videoconfer-
ence; a caller’s personalized ringtone on a phone; or one’s
virtual body in an immersive virtual environment. Other
definitions would limit avatars to only visual representa-
tions, or even more conservatively, to three-dimensional,
animated, human-like representations controlled by humans
in real time. Each of these different conceptualizations is
likely to influence the conclusions researchers make about
the effects of avatars.
Based on our review of the research, we believe it is im-
portant to set some boundaries for appropriate use of the
term avatar. To do so, we must acknowledge both historical
and ongoing research in this area. We must also consider
existing capabilities, modalities, and affordances without
being short-sighted and constraining the definition in a way
that would exclude potential future computer-mediated in-
terfaces. Thus, we endorse a more open definition and argue
that an avatar is a digital representation of a human user that
facilitates interaction with other users, entities, or the envi-
ronment. For communication scholars, this definition high-
lights the communicative potential of avatars.
Perhaps the most common element we noted across exist-
ing definitions that we opted not to include in ours was the
more restrictive requirement that avatars be “visual” or
“graphical.” Although current CMC environments may be
largely visual—and existing research ref lects the prominence
of this mode—text-based and auditory representations are
not uncommon. We did not want to restrict our definition
based on the type of sensory input given the growing richness
of many digital environments. A final consideration is that
humans vary in their sensory abilities, and some users or
platforms may employ non-visual representations to increase
accessibility. For example, developers have worked to replace
encompassed any form of digital representation and includ-
ed non-graphical representations such as usernames, sounds,
or text-based descriptions (e.g., Chan & Vorderer, 2006; Fox
& Ahn, 2013).
Another distinction across conceptual definitions is
whether the agency or control of the representation is a hu-
man in real time or an automated computer program. Some
definitions of avatar do not make a distinction; for example,
Nowak and Rauh (2005) describe avatars as “computer gen-
erated visual representations of people or bots” (p. 153).
Other scholarship uses the term agent to describe a represen-
tation whose actions are controlled by a computer, whereas
avatar is used to describe a representation whose actions are
controlled by a human (e.g., Biocca, 1997; Fox, Ahn, Jans-
sen, Yeykelis, Segovia, & Bailenson, 2015; Lim & Reeves,
2010; Nowak & Biocca, 2003). Researchers have theorized
that human-controlled representations are more likely to
inf luence users than computer-controlled ones (Blascovich,
Loomis, Beall, Swinth, Hoyt, & Bailenson, 2002; Nowak &
Biocca, 2003). There is no clear line to be drawn in terms of
human and computer control, however. In practice, human-
controlled representations rely on computers to control at
least some functionality, whether making the human user’s
typed text appear on a screen or having a video game char-
acter execute a sequence of animations based on the human
user’s keypresses. Although rare, some avatar definitions
acknowledge aspects of both human and computer control
(e.g., Bente, Rüggenberg, Krämer, & Eschenburg, 2008).
A final notable variation in definitions is tied to the fidel-
ity of avatars in terms of detail, richness, appearance, or
ability. Some authors define the avatar in terms of physical
realism (how lifelike it appears), behavioral realism (how
authentic its actions are), or anthropomorphism (how simi-
lar it is to human morphology or behavior; e.g., Nowak,
Hamilton, & Hammond, 2009). Early definitions were like-
ly inf luenced by the particular environments researchers
were studying at the time; hence, some definitions include
terms like “cartoon” or “two-dimensional” (e.g., Haythorn-
thwaite & Wellman, 2002; Webb, 2001). As technologies
evolved, so did definitions. Later definitions refer to avatars
as “three-dimensional” (Bailenson, Yee, Blascovich, & Gua-
dagno, 2008) and “animated” (Bente et al., 2008). These
more specific definitions are quite limiting, as the nature of
an avatar is determined by what a particular computer-me-
diated platform affords; in essence, these definitions mean
that avatars can only exist in certain digital environments
Avatars and Computer-Mediated Communication
35 2018, 6, 30-53
corporeal body. First, it is the digital or corporeal body that
allows people to experience the environment. Second, both
avatars and corporeal bodies help identify a person and
distinguish them from others. Further, both avatars and
corporeal bodies are a form of self-presentation and identity
expression that provide information to other interactants;
this information is used in making social judgments and
attributions of others (Ash, 2015; Blascovich & Bailenson,
2011; Guadagno, Blascovich, Bailenson, & McCall, 2007;
Hamilton & Nowak, 2010; Lee & Nass, 2002; Nowak et al.,
2009).
Likely because of these similarities, people have carried
the categories and processing strategies they have developed
from a lifetime of off-line experiences with corporeal bodies
into their experiences in digital worlds with avatars. This
practice is consistent with theoretical perspectives and mod-
els such as computers as social actors (CASA, Nass & Moon,
2000; derived from the media equation, Reeves & Nass,
1996), which suggests that humans treat computer generated
entities and digital representations as social others. Thus,
the off-line person perception process has been shown to help
predict online avatar perception with some exceptions, as
discussed below.
Perceptions of Avatars: Humanity, Agency, and
Social Potential
Among the first judgments made of a representation in
a digital environment is determining agency or humanity
(Bailenson, Swinth, Hoyt, Persky, Dimov, & Blascovich,
2005; Nowak, 2004; Nowak & Biocca, 2003). People are
likely to feel a stronger connection with an avatar with hu-
man features at a basic biological level (Sheehan & Sosna,
1991), which leads to the assessment of the social potential
of the entity being represented by the avatar (Nass & Moon,
2000; Nowak, 2015).
Information processing theory (McGuire, 1968) provides
additional insights into why these assessments of humanity
are crucial to understanding avatar perception. According
to this theory, people pay more attention to sources that have
more dynamism, which makes it more likely they will be
perceived to be human and have more social potential (Mc-
Guire, 1985). Those with more social potential also activate
schema and category assignment that lead to attributions
that are traditionally reserved for humans such as intention-
ality, emotions, or categorizations such as gender or age.
visual stimuli in digital games with auditory or haptic rep-
resentations for blind and visually impaired players (Yuan,
Folmer, & Harris, 2011). Thus, we opted for a more inclusive
scope in our definition.
We acknowledge that our definition is broad, but it is
intended to serve as an umbrella term independent of spe-
cific platforms or affordances. Researchers are encouraged
to employ more specific and precise terms to describe subsets
of avatars. For example, the term embodied avatar has been
used to describe representations that have a bodily form to
control via naturally mapped movements (e.g., Groom,
Bailenson, & Nass, 2009), and the term virtual human is often
used to describe highly realistic representations of people in
immersive virtual environments (e.g., Blascovich & Bailens-
on, 2011). Further, what is excluded from our definition
indicates that other terms are necessary to describe other
types of representations. Our definition is limited to digital
representations, which excludes physical entities such as
game pieces, sock puppets, or robots. Because our definition
necessitates a human user, it implies that other terms should
be used to describe computer-controlled entities, bots, and
algorithms (e.g., computer agent).
Though we are making an argument for this definition,
we recognize that not all researchers will agree. Whether
researchers adopt our def inition or not, it is critical for re-
searchers to clearly and carefully explicate their use of the
term. They should also consider how avatars are being con-
ceptualized and operationalized in others’ research when
framing hypotheses, theorizing, or drawing conclusions from
their f indings. Such clarity and precision are necessary be-
cause existing research demonstrates that these differences
are not merely semantic. For example, a meta-analysis re-
vealed that people perceive representations differently when
they believe representations are controlled by humans rath-
er than computers (Fox et al., 2015). Researchers must con-
sider the scope and potential boundary conditions of how
they conceptualize avatars to promote appropriate general-
izing, enable suitable replications, and facilitate theory build-
ing.
Avatar Perception: Processing Digital Bodies
and People
An avatar allows a person to experience and interact in
a digital world. In many ways, the avatar is analogous to the
Kristine L. Nowak & Jesse Fox
36
www.rcommunicationr.org
clarified, some researchers have found that individuals ex-
perience higher levels of physiological arousal when they
believe they are interacting with another human compared
to when they believe they are interacting with a bot (e.g., Lim
& Reeves, 2010; Ravaja, 2009). This higher arousal indicates
that at a physiological level, the body is making a distinction
between interacting with what it perceives as a human en-
tity compared to an object. Future research using technolo-
gies such as EEG or fMRI may provide further insights into
variations in physiological response when interacting with
human- as opposed to computer-controlled representations.
Anthropomorphism. Anthropomorphism is the perception
or assignment of human traits or qualities such as mental
abilities (Kennedy, 1992), cognitions (Tamir & Zohar, 1991),
intentions and emotions (Breazeal, 2003), or behavior (Nass,
Lombard, Henriksen, & Steuer, 1995) to entities that may or
may not be human. This concept is commonly used in avatar
research (e.g., Banks & Bowman, 2016; E. J. Lee, 2010;
Nowak & Biocca, 2003; Nowak & Rauh, 2008; Verhagen,
Van Nes, Feldberg, & Van Dolen, 2014). Others have incor-
porated terms such as humanoid or human-like when discuss-
ing anthropomorphic digital representations (e.g., Gong &
Nass, 2007; Martey & Consalvo, 2011).
Understanding factors that inf luence perceived anthro-
pomorphism is critical to the role of avatars because cues to
humanity are believed to provide clues to an avatar’s social
potential (Nass & Moon, 2000; Nowak et al., 2009). One
factor that increases perceived anthropomorphism is the
extent to which an image has a human-like appearance
(Gong & Nass, 2007; Hamilton & Nowak, 2010), which can
be called form anthropomorphism. Another factor is behavioral
anthropomorphism, or avatars speaking, moving, or acting in
ways that may be expected of humans. These behaviors may
include responding appropriately to stimuli, interacting au-
tonomously, displaying intelligence or emotion, or satisfying
interaction goals (Breazeal, 2003; Nowak et al., 2009; Reeves
& Nass, 1996).
Researchers have investigated how anthropomorphic
representations inf luence communicative outcomes and
found that more human-like representations are judged more
favorably; people consider them more attractive, credible,
and competent (Gong, 2008; Nowak & Rauh, 2005; Nowak
et al., 2009; Westerman, Tamborini, & Bowman, 2015).
Higher levels of anthropomorphism also lead to higher in-
volvement, social presence, and communication satisfaction
(Bailenson et al., 2006; Breazeal, 2003; Kang & Watt, 2013).
Research has replicated this finding with digital representa-
tions: more human-like images were perceived to have great-
er social potential even with simple, static, two-dimensional
avatars (Hamilton & Nowak, 2010; Nowak et al., 2009). The
more human-like people perceive avatars to be, the more
likely it is that theories and findings from human communi-
cation will apply to avatar-based interactions. Thus, under-
standing the how people perceive the social potential of
avatars will help predict communication processes and out-
comes in computer-mediated environments.
Across the literature, scholars have focused on three
overlapping aspects that influence users’ perceptions of the
social potential of avatars: agency, anthropomorphism, and
realism. Researchers must differentiate perceived agency
(whether or not an entity is perceived to be human), anthro-
pomorphism (having human form or behavior), and realism
(having accurate form or behavior). Here, we define these
concepts and review research relevant to understanding
avatars.
Agency. In the modern physical world, there is a clear
boundary between human and not human, as there are cer-
tain visual characteristics, traits, behaviors, and abilities that
are unique to humans (Sheehan & Sosna, 1991). There are
not necessarily any visible differences between computer-
controlled agents and human-controlled avatars in online
interactions. A computer agent may be represented by a
human-like image that moves and speaks f luidly, whereas a
person’s avatar could look like a bison or a stapler with halt-
ing speech and unnatural movements. Given there are no
clear indicators, users may not always be able to distinguish
agency, or whether a representation is controlled by a human
or a bot (Kim & Sundar, 2012; Nowak, 2004; Nowak &
Biocca, 2003).
According to the model of social influence in virtual
environments (SIVE; Blascovich & Bailenson, 2011; Blasco-
vich et al., 2002), people try to determine whether the digital
representation they are interacting with is a person or a bot.
This perceived agency influences people’s responses in the
interaction regardless of who or what is actually controlling
the representation. A meta-analysis of studies comparing
agents and avatars found that both agency and perceived
agency mattered: representations controlled by humans were
more persuasive than those controlled by bots, and represen-
tations believed to be controlled by humans were more per-
suasive than those believed to be controlled by bots (Fox et
al., 2015). Although the exact mechanism has not yet been
Avatars and Computer-Mediated Communication
37 2018, 6, 30-53
equip people with additional limbs (Won, Bailenson, Lee,
& Lanier, 2015), allow every receiver to see consistent eye
contact from the same source (Bailenson, Beall, Loomis,
Blascovich, & Turk, 2005), or make a message source look
exactly like the receiver (Bailenson & Segovia, 2010; Fox &
Bailenson, 2009a). Such manipulations are not possible out-
side of digital environments, but all have notable implica-
tions for communication. For example, additional body parts
provide new opportunities for studying nonverbal commu-
nication; augmented gaze and appearance mimicry have
clear implications for persuasion. Thus, the ability to ma-
nipulate the anthropomorphism and realism of avatars in
near-inf inite ways may offer new insights into communica-
tion.
The uncanny valley. People rate avatars with abnormal
or exaggerated features as unpleasant (Seyama & Negayama,
2007) and expect anthropomorphic bodies to be animated
authentically (Dalibard, Magnenat-Thalmann, & Thalmann,
2012). The uncanny valley hypothesis (Mori, 1970) suggests
that there is a general trend for humans to like things that
demonstrate human features. There is a point at which high
levels of anthropomorphism, however, evoke negative reac-
tions. When representations are perceived as too human-like,
but not yet human, they are perceived as creepy and unset-
tling. Consistent with this hypothesis, research on virtual
human representations has shown that too much anthropo-
morphism can have negative outcomes in social interactions;
if a representation appears too human-like, participants like
it less, trust it less, and experience discomfort (Groom, Nass,
Chen, Nielsen, Scarborough, & Robles, 2009). Stein and
Ohler (2017) also argue that an “uncanny valley of the mind”
exists, such that people also have negative reactions when a
computer-controlled agent demonstrates a certain level of
behavioral anthropomorphism, such as autonomous deci-
sion-making.
Some argue that the uncanny valley is triggered not only
when digital stimuli have high anthropomorphism alone,
but also when there is a mismatch between the level of form
and behavioral anthropomorphism, or when levels of realism
and anthropomorphism do not match (Bailenson et al., 2005;
Hamilton & Nowak, 2010). People anticipate that anthropo-
morphic avatars have more social potential and expect them
to demonstrate intelligence, responsiveness, appropriateness,
and sociability (Bailenson, Swinth et al., 2005; Nowak et al.,
2009). Participants evaluate avatars who look human (i.e.,
have high form anthropomorphism) but do not act human
People also communicate more naturally with more anthro-
pomorphic avatars (Heyselaar, Hagoort, & Segaert, 2017).
Anthropomorphism is also tied to social influence, as more
human-like representations can be more persuasive (Gong,
2008; Guadagno et al., 2007). Perceived anthropomorphism
is a key determinant of the way information and people are
judged, which likely inf luences the extent to which theories
about human-human interaction can apply to avatar-based
communication. Thus, researchers should continue to inves-
tigate the various ways that avatars can resemble humans,
and the avatar features and individual differences that influ-
ence perceived anthropomorphism.
Realism. Realism is the perception that something could
realistically or possibly exist in a non-mediated context (Bus-
selle, 2001; Busselle & Greenberg, 2000). Avatar realism
could be assessed on many levels (Bailenson et al., 2006;
Nowak et al., 2009). An avatar could be judged on its level
of fidelity to what an object would look or move like in the
off-line world. This fidelity may include details in appear-
ance; rendering such as shading and depth; fluidness of
motion; or the naturalness of auditory cues. For example, an
avatar could appear cartoon-like or be photorealistic. Real-
ism could also entail an assessment of whether that repre-
sentation could exist in the physical world or is complete
fantasy. In this way, an avatar of a dog may be seen as more
realistic than an avatar of a flying purple dragon.
Some scholars have described avatars’ similarity to hu-
mans, or having human form, using terms such as form
realism, behavioral realism, communicative realism, or simply
avatar realism (e.g., Bente et al., 2008; Guadagno et al., 2007;
Guadagno, Swinth, & Blascovich, 2011; James, Potter, Lee,
Kim, Stevenson, & Lang, 2015). We argue that this concep-
tualization and operationalization would more accurately
be considered anthropomorphism because these studies spe-
cifically explored determinations and representations of
humanity. Researchers should distinguish anthropomor-
phism from realism because they are distinct judgments with
different implications for understanding communication.
For example, a person may be represented by a highly ac-
curate and lifelike avatar of a fir tree. Although this avatar
is realistic, other users may be less likely to attribute social
potential to it—and less likely to attempt to communicate
with it—because it is not anthropomorphic.
Digital environments provide researchers with novel
ways to manipulate and study the roles of anthropomor-
phism and realism. For example, virtual environments can
Kristine L. Nowak & Jesse Fox
38
www.rcommunicationr.org
Sex and gender. Determinations of sex, assessed through
typically dimorphic biological features, is one of the most
common categorizations humans make of others, perhaps
due to the evolutionary drive of sexual reproduction. Al-
though some categorizations may be made based on physi-
cality and biological attributes, the sociocultural spectrum
of gender is often equally salient. Indeed, people make at-
tributions of sex even when physical or biological informa-
tion is not available, likely because people believe that sex
categorization provides information that is useful in under-
standing others, predicting behaviors, and identifying ap-
propriate interaction scripts (Lakoff, 1987). Thus, it is
unsurprising that attributions of sex and gender have re-
mained salient in computer-mediated contexts (Biocca &
Nowak, 2002; Fox, Ralston, Cooper, & Jones, 2015; Nass &
Brave, 2005; Reeves & Nass, 1996; Turkle, 1995).
Findings from several studies have supported the hypoth-
esis that the sex and gender stereotypes used to evaluate
humans are also applied to computer-mediated representa-
tions. For example, people expect gendered avatars to have
gendered knowledge. Children have been shown to trust
female voices more than male voices on topics such as prin-
cesses and makeup, but trust male voices more than female
voices on topics such as football and dinosaurs (K. M. Lee,
Liao, & Ryu, 2007). Similarly, adults trusted simulated male
voices more than female voices on a math tutorial but trust-
ed the simulated female voices more than the male voices in
a tutorial on relationships (Reeves & Nass, 1996). Stereo-
typical attributions also hold when people embody gendered
representations. Regardless of their biological sex, partici-
pants who were given a male avatar in a virtual environment
and competed against two female avatars had a higher per-
formance on a math task compared to those who embodied
a female avatar in the presence of male avatars (J. E. R. Lee,
Nass, & Bailenson, 2014).
These effects are also observed in naturalistic online set-
tings, such as virtual worlds and video games. Similar to
face-to-face contexts, male-male avatar dyads maintain
greater interpersonal distance than in male-female or female-
female dyads (Yee, Bailenson, Urbanek, Chang, & Merget,
2007), and female avatars are subject to more sexual harass-
ment than male avatars (Behm-Morawitz & Schipper, 2016).
Other studies have found that exposure to stereotypical or
sexualized representations in digital environments is similar
to findings with other media. Stereotypical virtual represen
-
tations of women evoked more sexism than nonstereotypical
(i.e., have low behavioral anthropomorphism) negatively and
consider their communication lower in quality compared to
other avatars (Bailenson, Swinth et al., 2005; Hamilton &
Nowak, 2010). This failure to meet expected levels of social
potential leads to disappointment and negative evaluations,
such as lower likeability and credibility (Nowak, 2004; Slat-
er & Steed, 2002).
In summary, further research is needed to identify the
extent to which avatars must demonstrate anthropomorphic
qualities to accomplish the same communicative outcomes
as one would expect in off-line environments, and which
qualities allow enhanced or augmented communication out-
comes. More research should investigate the individual and
interactive roles of perceived agency, anthropomorphism,
and realism in avatar-based communication. To complicate
this process further, evaluations of anthropomorphism, real-
ism, and social potential are subjective and vary widely
across individuals and may be influenced by previous expe-
riences, contextual factors, or other cues provided by the
source (Bailenson et al., 2006; Busselle, 2001; Nowak et al.,
2009). This complexity makes it difficult to specify what
human-like or realistic characteristics have the greatest im-
pact when communicating. In the next section, we will
discuss the ways in which avatars are perceived in ways
similar to human communicators and the implications of
such perceptions for CMC research.
Social Categorization of Avatars
Several theories of social identity and intergroup com-
munication acknowledge that individuals evaluate others as
members of various groups (see Gaertner, Dovidio, & Hou-
lette, 2010, for a review). Because avatars are perceived as
social entities, humans often engage in similar categoriza-
tion processes, applying the same heuristics and stereotypes
they associate with human members of that category (Nowak
et al., 2009). Information processing theory would predict
that this categorization is more likely to occur when the
avatars are perceived to have more social potential. Continu-
ing to make the same category assignments to avatars that
are made of humans, including sex and race, is consistent
with predictions of the media equation and CASA (Nass &
Moon, 2000; Reeves & Nass, 1996). Here, we review some
of the research findings on avatars, social categorization,
and intergroup communication in digital environments.
Avatars and Computer-Mediated Communication
39 2018, 6, 30-53
Black avatar requesting assistance than a White avatar (East-
wick & Gardner, 2009). Similarly, in an emergency simula-
tion, White participants were less likely to help Black avatars
than White avatars (Gamberini, Chittaro, Spagnolli, & Car-
lesso, 2015). Some studies have shown that embodying or
interacting with Black avatars triggers racial stereotypes
(Groom et al., 2009), particularly if the portrayals are ste-
reotypical (Burgess, Dill, Stermer, Burgess, & Brown, 2011;
Cicchirillo, 2015).
Alternatively, some studies have shown that the experi-
ence of embodying a non-White avatar can reduce racial bias
(Behm-Morawitz, Pennell, & Speno, 2016; Maister, Sebanz,
Knoblich, & Tsakiris, 2013; Peck, Seinfeld, Aglioti, & Slater,
2013). More research is needed to understand the disparities
in these f indings, though it seems prosocial outcomes may
be more likely when users identify more with their avatar
and engage in less structured tasks (Behm-Morawitz et al.,
2016; Peck et al., 2013).
Similarity and homophily. Homophily, or perceived
visual or psychological similarity to the self, inf luences per-
ception and attribution in communicative contexts. People
respond more positively to and prefer others who are similar
to themselves. Similar to outcomes in face-to-face contexts,
people prefer more homophilous avatars, which are seen as
more credible and likeable (Nowak et al., 2009; Nowak,
2013) as well as more persuasive (Guadagno et al., 2007).
Avatars do not have to resemble the actual self to be
persuasive, however; they can also persuade when the user’s
avatar matches others’ avatars. As predicted by the social
identity model of deindividuation effects, several studies
have shown that when interactants’ avatars have common
features or feel they belong to the same social categories or
groups, this similarity enhances social identity and bolsters
positive impressions of partners or teammates (E. J. Lee,
2007). Visually similar representations also promote persua-
sion and conformity effects (Ahn & Bailenson, 2011; Fox &
Bailenson, 2009a; E. J. Lee, 2004, 2007). Combining both
similarity to the self and similarity to the other in one’s ava-
tar may optimize outcomes. One experiment examined
team-similar and self-similar avatars and found that avatars
that both resembled the user and matched other teammates’
avatars yielded the highest levels of social attraction as well
as task performance (Van Der Land et al., 2015).
In summary, the characteristics of avatars influence how
receivers interpret sources and their messages. Moreover,
representations (Fox & Bailenson, 2009b). Interacting with
sexualized representations has been shown to encourage
men and women to perceive women as less intelligent (Behm-
Morawitz & Mastro, 2009), make men more tolerant of
sexual harassment (Dill, Brown, & Collins, 2008), and in-
crease men’s likelihood to sexually harass (Yao, Mahood, &
Linz, 2010). Embodying or playing a video game as a sexual-
ized avatar has been associated with self-objectification (Fox,
Bailenson, & Tricase, 2013; Fox, Ralston, et al., 2015; Van-
denbosch, Driesmans, Trekels, & Eggermont, 2017). Find-
ings on helping behavior are mixed, however. One study
found that female avatars are more likely to receive help than
male avatars (Lehdonvirta, Nagashima, Lehdonvirta, &
Baba, 2012) whereas a second study found differences were
based on interactions with the user’s sex or the avatar’s at-
tractiveness (Waddell & Ivory, 2015).
In some cases, interactants may not be able to make a sex
or gender categorization. Users feel more uncertain with
androgynous avatars that lack clear indications of gender
(Nowak & Rauh, 2005, 2008). It is possible that an inability
to make this categorization leads to an undesirable state of
uncertainty, particularly in a simulated environment with
the absence of a corporeal body. At this time, additional
research is needed to understand how people process and
interpret androgynous avatars.
Race and ethnicity. As with other categories, people
often rely on visual cues to determine race or ethnicity, as
they believe this information may help them predict behav-
ior. Making this attribution may indicate perceived social
potential given that race would only be relevant for humans.
In certain conditions, the perceived race of others’ avatars
in a digital environment influences perceptions of a message
source (Spence, Lachlan, Westerman, & Spates, 2013), per-
ceptions of an interaction partner (Vang & Fox, 2014), or a
willingness to disclose one’s own race (J. E. R. Lee & Park,
2011; J. E. R. Lee, 2014).
Consistent with CASA, several studies have indicated
that users assign avatars to racial categories and apply the
same associated stereotypes as they would humans. Dotsch
and Wigboldus (2008), for example, found that White par-
ticipants approaching Black avatars in a digital environment
experienced higher physiological arousal (measured through
skin conductance) and maintained greater interpersonal
distance compared to White participants approaching White
avatars. In a virtual world, users were less likely to help a
Kristine L. Nowak & Jesse Fox
40
www.rcommunicationr.org
fairly accurate portrayal of the user. Several studies have also
indicated that people choose to convey elements of their
social identities in the avatars they select, such as sex, gen-
der, race, or age (e.g., Cheong & Gray, 2011; Gerbaudo, 2015;
J. E. R. Lee, 2014; Martey & Consalvo, 2011; Nowak &
Rauh, 2008). Alternatively, they may select avatars that
depict a more idealized or aspirational version of the self
(e.g., Bessière, Seay, & Kiesler, 2007; Lee-Won, Tang, &
Kibbe, 2017, Sah, Ratan, Tsai, Peng, & Sarinopoulos, 2016).
Some digital environments make it easy for users to ex-
press elements of their identities through their avatars (e.g.,
users can select an avatar that shows identification in a group
or that matches their race or gender) or select an option in
their profile (e.g., users can upload pictures of themselves,
select their age, race/ethnicity, or other characteristics on
online dating sites). People in these systems create avatars
that represent them, and some even use these systems to
reveal aspects of their true or desired selves that they are
uncomfortable or unwilling to present face-to-face (Bargh,
McKenna, & Fitzsimons, 2002; Turkle, 1995). Even so, it
may not always be possible or desirable to be represented by
an avatar that accurately represents the off-line self.
Identity Exploration and Deception
Some users will select avatars that accurately represent
something about them, but that inaccurately present other
aspects of the self. While this may sometimes be a choice,
other times it is not. Some digital environments make it dif-
ficult to convey one’s identity authentically due to techno-
logical constraints or social norms. For example, it can be
difficult to portray race or ethnicity due to limitations on
avatar options such as skin tones, facial features, or hair-
styles (Kafai, Cook, & Fields, 2010; Martey & Consalvo,
2011), or even an absence of avatars with diverse gender or
race options (Brock, 2011; Nakamura, 2002). Users may have
to choose between accurately presenting their sex and ac-
curately presenting their personality, favorite sport, or other
aspects of identity (Nowak, 2013).
Alternatively, users may select avatars to “try on” or
explore different identities out of curiosity or to see how it
feels to be an “other” (Bessière et al., 2007; Turkle, 1995).
Identity exploration via avatars may have some positive
benefits, such as individuals self-disclosing to others, build-
ing relationships, and gaining self-acceptance of their iden-
tity. This experimentation can inf luence people while in the
avatar characteristics affect outcomes such as communica-
tion satisfaction, social influence, and task performance.
Whether examining or manipulating avatars in CMC envi-
ronments, it is crucial that researchers recognize how users
are evaluating these representations. These same factors also
inf luence the avatars people choose to represent themselves.
Avatar Selection as Self-Representation
According to Goffman (1959), people carefully manage
how they present themselves to optimize their ability to
fulfill social goals. In digital environments, avatars are used
for self-presentation, and inf luence how people evaluate
digital bodies as they are considering how and when they
may select them as avatars (Nowak & Rauh, 2005; Nowak,
2013). Choosing an avatar in the virtual world is in some
ways analogous to the process of choosing an outfit to wear
in the physical world, though the selections are contingent
on technological limitations rather than the clothes in a
person’s closet.
Compared to face-to-face settings, digital environments
typically present users with greater flexibility and control in
modifying their self-presentation. Because CMC environ-
ments are often asynchronous and lack cues available in
face-to-face settings, users can capitalize on the affordance
of editability. This process of selective self-presentation allows
users to tailor their presence to a particular context or inter-
actant (Walther, 1996). Users make judgments of avatars
they encounter as described above, which leads them to se-
lect avatars they believe will help them meet interaction
goals, which could include revealing or concealing elements
of their identity to other users. Here, we review several stud-
ies regarding the types of avatars people choose and their
motivations for doing so.
Avatars as Identity Expression
People typically have multiple goals when selecting an
avatar to represent themselves. One common goal inf luenc
-
ing avatar selection is the desire to identify and express the
self to others. Many users prefer avatars that accurately
represent something about them either physically or psycho-
logically (Kang & Yang, 2006; Nowak & Rauh, 2008;
Nowak, 2013). On social networking sites and online dating
sites, for example, it is expected that a profile picture is a
Avatars and Computer-Mediated Communication
41 2018, 6, 30-53
curate depiction of the self can also influence self-perception,
identification, and attitudes, as discussed in the next section.
Effects of Avatar Embodiment
As noted, the f lexibility of avatars in many contexts
means that a user’s self-representation can be modified in
meaningful ways that may be dissimilar to the physical self.
According to the Proteus effect, the user’s behavior conforms
to the modified self-representation regardless of the true
physical self (Yee & Bailenson, 2007, 2009; Yee, Bailenson,
& Ducheneaut, 2009). Taking on the characteristics of an
avatar may influence how a user communicates both online
and off-line. When participants embody attractive avatars
in a virtual environment, they disclose more personal infor-
mation and approach their partner’s avatar more closely.
When participants embody taller avatars, they are more
confident and aggressive when negotiating with another
person (Yee & Bailenson, 2007). Research from the Proteus
effect paradigm has demonstrated that the attractiveness
(Van Der Heide et al., 2013), gender (J. E. R. Lee, Nass, &
Bailenson, 2014), race (Ash, 2015), or sexualization (Fox et
al., 2013) of one’s avatar inf luences self-perception, attitudes
towards others, and behavior. Other research has also pro-
vided support for the Proteus effect without explicitly adopt-
ing the paradigm. For example, Palomares and Lee (2010)
found that participants experienced linguistic assimilation
with their avatars. Men and women in avatars that matched
their gender used more gender-typical language; when they
embodied an avatar of a different gender, they adopted lan-
guage suited to the avatar’s gender. These findings suggest
that avatar selection may inf luence communicative outcomes
not only based on the receiver’s impression, but also based
on the sender’s experience.
Priming has been put forth as an alternative explanation
for the effects of avatars (e.g., Peña et al., 2009; Peña, 2011).
Direct comparisons of the two perspectives, however, favor
the Proteus effect (e.g., Ash, 2015; Yee & Bailenson, 2009).
In addition, many priming findings in other areas have not
held up to replications, casting some doubt on the validity
of priming effects (Open Science Collaboration, 2015). As
with many effects, it is likely that there are uncovered mech-
anisms and conditional effects at work. Given small sample
sizes, changing technologies, and inconsistent findings, re-
searchers should seek to replicate priming studies, Proteus
environment as well as after the experience has ended (Bargh
et al., 2002; Turkle, 1995). Some argue, however, that this
practice of identity tourism may have negative outcomes. For
example, if users embody the avatar of a person of color and
enact stereotypically consistent behaviors, they may rein-
force negative stereotypes for themselves and the people with
whom they interact (Nakamura, 2002).
Other users may select avatars with the intention to con-
ceal elements of their identity. In some cases, this conceal-
ment has a protective function. For example, users may
adopt avatars that do not resemble them to maintain ano-
nymity in a health support group (Green-Hamann, Eich-
horn, & Sherblom, 2011). Choosing an avatar that masks
one’s gender is not uncommon for women in video games
and online virtual worlds (e.g., Huh & Williams, 2010; Mar-
tey, Stromer-Galley, Banks, Wu, & Consalvo, 2014); some
women report engaging in this behavior to prevent harass-
ment (Fox & Tang, in press). Given nonwhite avatars are
often subject to prejudice, people of color may also choose
not to disclose their race or ethnicity to avoid being treated
in a stereotypical fashion, discriminated against, or harassed
(Nakamura, 2009; Yee, 2014). Although targeted individuals
may benefit from masking their identities through the ava-
tars they select, there are downsides. Limiting the visibility
of women, people of color, and other groups may feed into
the illusion that they are not present in these environments
and reinforce the default assumption that the vast majority
of users are White males (Brock, 2011; J. E. R. Lee, 2014).
Of course, these affordances can be used maliciously:
people may select an avatar that does not accurately repre-
sent them with the intent to mislead or deceive. For example,
someone may choose a deceptive avatar on an online dating
site with the intention to “catfish” a target and coerce them
into sending money. More commonly, these choices are more
self-enhancing than malicious. For example, people often
use idealized photographs of themselves on online dating
sites that make them appear younger, thinner, and more at-
tractive (Hancock & Toma, 2009). When the person does
not resemble their avatar in a face-to-face meeting, however,
the other person may be disappointed or angry. Because
interactants are often aware of the capacity for misrepresen-
tation online, they often capitalize on other affordances to
evaluate senders and their messages (DeAndrea, 2014; Wal-
ther & Parks, 2002). In this way, motivated receivers mind-
fully evaluate provided cues to avoid deception and other
negative outcomes. Embodying an avatar that is not an ac-
Kristine L. Nowak & Jesse Fox
42
www.rcommunicationr.org
across various types or characteristics of avatars and indicate
anticipated boundary conditions.
We also advise researchers to provide considerably more
detail in the reporting of their manipulations and methods.
In our literature search, far too many studies only provided
brief, text-based descriptions of the avatars or their digital
environments. Given the nearly infinite possibilities in creat-
ing avatars across platforms, it is difficult to accurately rep-
licate a study that only offers vague descriptions such as “a
male avatar.” Within publications, richer descriptions, fig-
ures, or links to online content can help resolve this issue.
On a broader scale, more open scientific procedures will ease
the burden on other scholars looking to build upon, extend,
or replicate existing avatar research. If journals cannot ac-
commodate additional information, supplementary materi-
als (such as avatar stimuli or scripts of interactions) can be
shared on researchers’ websites or in online repositories.
As the research continues to show the effects of experi-
ences with and as different bodies online, it becomes critical
to consider the potential of long-term effects caused by ava-
tar choices. Some studies have included post-experimental
measures ranging from 24 hours to a few weeks (e.g., Ahn
& Bailenson, 2011; Ahn et al., 2016; Fox & Bailenson,
2009a). Although a handful of studies have collected data
on avatar use over time (e.g., Bailenson & Yee, 2006; Yee,
Ducheneaut, Yao, & Nelson, 2011), longitudinal experimen-
tal research is rare, which limits the ability to understand the
long-term effects of interacting with, and as, avatars. In ad-
dition to long-term studies, researchers should consider po-
tential effects both within and outside of the digital
environment (Yee & Bailenson, 2007). For example, there
are considerably different implications for someone who
imitates an aggressive avatar within a virtual context (e.g.,
attacking another player’s character in a video game) as op-
posed to imitating that behavior outside that environment
(e.g., becoming physically violent with a sibling). Finally, as
illustrated by studies suggesting that embodying Black ava-
tars entrenched racial stereotypes for some White partici-
pants (Burgess et al., 2011; Groom et al., 2009), researchers
should consider and attempt to illuminate both the intended
and unintended effects of avatars (Cho & Salmon, 2007). For
example, embodying an attractive avatar may be intended
to increase self-esteem, but it may also evoke self-objectifi-
cation, narcissism, or a beauty bias in judging others’ ava-
tars.
effect studies, and any number of other findings regarding
avatars.
In summary, the flexibility of avatars in many contexts
means that a user’s self-representation can be modified in
meaningful ways that may be dissimilar to the physical self.
Given mixed findings across multiple contexts, researchers
should continue to investigate the frequency and rationale
people have for selecting avatars that do not accurately rep-
resent them. Further, researchers should examine how em-
bodying avatars with dissimilar characteristics inf luence
behavior and attitude change. For communication research-
ers, it is particularly important to examine interpersonal and
contextual elements that may influence the outcomes of
various types of embodiment.
Future Directions: Using Avatars to Study
Human Communication
Although researchers have varied in their ways of con-
ceptualizing avatars, the lack of a cohesive definition has not
prevented research on avatars from having a significant
impact on our understanding of certain communication
processes both online and off-line. Here, we sifted through
communication scholarship to clarify definitional approach-
es and identify common threads in avatar-based research.
Based on our review, we have distilled some recommenda-
tions for avatar researchers going forward.
Refining Research on Avatars
Our literature review indicated several directions in
which avatar scholars can improve the construction, meth-
ods, and reporting of research. First, given the diversity of
approaches we identified in our literature search, we recom-
mend that future researchers adopt a universal and consis-
tent definition of the term. Consistent conceptualization and
appropriate operationalization are essential for building
theory about communicating via avatars and increase the
potential for replication of findings across studies. Even if
disagreement about a universal definition remains, research-
ers must provide clear definitions within their publications
so that other scholars can assess the applicability of a par-
ticular finding and understand the nature of the avatars used
in the study. Further, when discussing their results, research-
ers should evaluate the generalizability of their findings
Avatars and Computer-Mediated Communication
43 2018, 6, 30-53
interactions (i.e., transformed social interaction; Bailenson,
Beall, et al., 2005). For example, avatars could portray real-
time physiological indicators such as heartbeat or pupil dila
-
tion so that users could track others’ attention or involvement
in a conversation. Alternatively, avatars could be used to
illustrate past selves, aspirational future selves, or other it-
erations of the self to provide more information than may be
visible in real time. Although these portrayals may have
benefits, there are potential downsides: too much discrep-
ancy between the avatar and the current self could backfire,
or too much information could create cognitive overload.
Future research is needed to parse out what additional in-
formation avatars can effectively communicate as well as the
positive and negative effects of this information.
Avatars can provide experiences beyond just entertain-
ment, and these interactions may influence off-line interac-
tions and processes. Experience with anthropomorphic
others who are not actually human, or non-anthropomorphic
others who are human, may inf luence the meaning of “hu-
manness” as people develop new scripts and categorization
schemes for digital others. These experiences may lead peo-
ple to alter their reliance on avatars in perceiving the entities
they represent. It may also influence how people interact.
For example, avatars with additional body parts, tails, or
color-changing halos provide greater bandwidth for com-
municating nonverbally and may lead to the development of
novel forms of interaction. In this way, avatars are both a
novel topic of research and a method for understanding exist-
ing communication processes and their evolution.
Ethical Considerations
Avatars also present some methodological advantages for
scholars. Virtual worlds represent constantly evolving com-
municative contexts (Castronova, 2006; Williams, 2010).
Virtual environments can also enable social scientists to
examine questions that are impractical, unethical, or even
impossible to study in natural environments (Blascovich et
al., 2002; Nowak, 2015; Schönbrodt & Asendorph, 2011),
though some uses of avatars may yield antisocial outcomes.
Given that the effect of embodying different avatars one time
in a lab can influence off-line behaviors and attitudes days
or weeks later (Ahn et al., 2016; Klimmt, Hefner, Vorderer,
Roth, & Blake, 2010), researchers and designers must care-
fully consider how these experiences may influence others
long term.
Topics for Future Directions for Avatar
Research
Topically, our review indicates that avatars present two
opportunities for communication scholars: they can serve as
a novel context to study how people use virtual worlds, and
they can be used as tools to understand existing communica-
tive processes (Fox, Arena, & Bailenson, 2009). Avatars can
facilitate new and infinitely flexible ways for individuals
interact with others and to self-present, ideally allowing a
diversity of presentations that can provide insight into com-
munication processes (Nowak, 2015). How does communi-
cation behavior vary when one’s self-representation is
highly dissimilar to one’s physical or psychological self? Is
information presented by an avatar more or less useful than
the corporeal body in developing an accurate mental model
of a person?
As objects of study, it is crucial to investigate many of the
distinguishing characteristics of avatars we identified in our
review. Within the avatar literature, many studies have
shown variation in outcomes based on avatar features such
as the degree of human agency, levels of realism, and various
types of anthropomorphism (e.g., Bailenson et al., 2006;
Kang & Watt, 2013; Nowak & Rauh, 2005; Rosenthal-von
der Pütten, Krämer, Gratch, & Kang, 2010; Seyama &
Negayama, 2007; Stein & Ohler, 2017). As indicators of
humanity and social potential, these features may play a
critical role in determining how existing theories of human
communication apply when interacting with or via avatars.
Future researchers should also consider how attribution
and perceptual processes occur in CMC as compared to
face-to-face environments. It is possible that information
gleaned from the avatar a person selected for an interaction
could provide a more accurate mental model of the person
than using their corporeal body. Users consciously choose
the avatar they are embodying in many interactions to fa-
cilitate their interaction goals (Nowak, 2013). It is likely that
this presents a part of themselves they want others to under-
stand and may serve as a type of disclosure, whether inten-
tional or not. Future research can examine how much
consideration people give to their avatar, and whether their
interaction partners can accurately perceive either the person
behind the avatar or what the user intended to portray.
Although many theories characterize computer-mediated
channels as having depleted cues, avatars can also benefit
users by providing more information to interactants in a
digital environment than is available in face to face
Kristine L. Nowak & Jesse Fox
44
www.rcommunicationr.org
engaged in interactions mediated by phones and computers,
which is affecting how they present themselves, how they
perceive others, what they learn, and how they relate to the
world around them. The research on avatars is in its relative
infancy, but has demonstrated rather consistently that people
follow similar processes in communicating via computer
media that they follow face-to-face (Nass & Moon, 2000),
and that the type of avatar used in an interaction inf luences
perceptions, attributions, and behaviors (Nowak & Rauh,
2008; Yee & Bailenson, 2007, 2009).
One contribution of this review is identifying the varied
definitions of the term avatar employed by Communication
scholars, which underscores the importance of researchers
clearly explicating and defining the way they are using the
term. Many terms never have a single accepted definition;
thus, a consensus on the meaning of the term may not be
needed as long as researchers provide clear conceptual and
operational definitions. The varied use of the term, however,
can make it difficult to generalize or fully understand the
effects of avatars on communication processes. Finally, with
-
out consistent meaning of a construct, it is difficult to syn-
thesize existing research, replicate findings, or advance
theory (Chaffee, 1991).
In synthesizing the work on avatars in person perception,
this review clarified some existing theoretical approaches
relevant to evaluating and making attributions regarding
avatars. This review has also shown that there are specific
constraints in avatar-mediated environments that determine
the range of possible selves one can convey. These limitations
can have unintended effects on both the person creating the
avatar as well as other users, who will naturally make infer-
ences about the person based on their avatar. Researchers
and users in these environments should consider these con-
straints and their implications as they interact and conduct
research. Erroneous inferences and attributions may be made
about the person based on the avatar that represents them,
or people may feel a person intentionally deceived them even
though the system’s constraints may have forced them to
make those choices. For example, a woman may not want to
represent herself as hypersexualized and scantily clad, but
in many video games, these are her only options for female
avatars. Other players may judge her negatively or target her
for harassment because she “chose” to represent herself as a
hypersexualized female without acknowledging that it was
her only choice if she wanted a female avatar. A woman in
this situation could choose a male avatar to avoid being
Although there may be prosocial effects of embodying
avatars depicting races, or identities different from our cor-
poreal bodies, researchers, designers, and users must be
careful to avoid antisocial effects such as the reinforcement
of stereotypes and the entrenchment of bias. Designers must
also ensure that people have diverse and appropriate options
for self-presentation so as not to marginalize underrepre-
sented groups (Brock, 2011; J. E. R. Lee, 2014), and be cau-
tious about when they allow avatars to display
stereotypically consistent behaviors (Fox & Bailenson,
2009b; Ratan & Sah, 2015).
Additionally, although we have discussed the malleabil-
ity of one’s own avatar, we should also consider the potential
for users to control others’ avatars in digital environments
and how that might affect interactions. There may be ben-
efits to allowing people to select the avatar that represents
their interaction partner or presents them with information.
For example, participants who were allowed to design a
salesperson’s avatar rated the source and the brand more
highly than participants who were not allowed to design the
person’s avatar (Hanus & Fox, 2015). At the same time, there
may be antisocial effects of this affordance, as malicious
users could take the opportunity to create negative represen-
tations such as offensive stereotypes or use this information
to manipulate people (e.g., to create political propaganda).
Regardless, this level of control over the message source is
unprecedented in face-to-face environments and presents
interesting challenges to our understanding of the commu-
nication process.
In summary, existing research on avatars has not only
allowed us to understand more about how people use, per-
ceive, and are affected by avatars, but it has also illuminated
communication processes more broadly. Going forward,
there is no shortage of ways in which avatar research can
lend further insight into human communication, but re-
searchers must carefully consider the ethical implications of
research designs and the potential for long-term effects of
avatar interactions.
Conclusion
The use of avatars in computer-mediated communication
and as stimuli to extend our understanding of processes in
communication research is both theoretically and practi-
cally important. People are spending considerable time
Avatars and Computer-Mediated Communication
45 2018, 6, 30-53
ining the role avatars on communication processes, high-
lighting major advancements and limitations while also
making recommendations for future communication re-
searchers. Although this work relatively nascent, the preva-
lence of CMC in ever yday social interaction underscores its
importance and the necessity for further introspection. As
is generally true, the more one understands about elements
of a communication process, the more complicated it seems.
Thus, there is an ongoing need for more nuanced and fo-
cused research to explicate these processes. We look forward
to watching this area of research continue to mature and
expand as researchers outside of the domain of digital envi-
ronments begin to appreciate how avatar research can inform
their inquiry as well.
hypersexualized, though she may be accused of deception.
Distinctions in processing may become more influential
as advances in graphics capabilities, speech agents, and ar-
tificial intelligence make virtual entities more and more re-
alistic and human-like. Over time, as humans gain
experience interacting with avatars, robots, or other com-
puter agents, they may make rational decisions to determine
what social categories are useful for evaluation. We may see
the development of new categories that will become relevant
in perceiving others represented by avatars. Further research
on this question can enlighten our understanding of person
perception more broadly (Blascovich & Bailenson, 2011;
Blascovich et al., 2002; Nowak, 2015; Schroeder, 2002).
The goal of this review was to synthesize research exam-
References
Ahn, S. J. (2015). Incorporating immersive virtual environments in health promotion campaigns: A construal level theory
approach. Health Communication, 30, 545-556. doi: 10.1080/10410236.2013.869650
Ahn, S. J., & Bailenson, J. N. (2011). Self-endorsing versus other-endorsing in virtual environments. Journal of Advertising,
40, 93-106. doi: 10.2753/JOA0091-3367400207
Ahn, S. J., Bostick, J., Ogle, E., Nowak, K. L., McGillicuddy, K., & Bailenson, J. N. (2016). Experiencing nature: Embodying
animals in immersive virtual environments increases inclusion of nature in self and involvement with nature. Journal of
Computer-Mediated Communication. d oi: 10.1111/jcc4.12173
Ash, E. (2015). Priming or Proteus effect? Examining the effects of avatar race on in-game behavior and post-play aggressive
cognition and affect in video games. Games and Culture, 11, 422-440. doi: 10.1177/1555412014568870
Bailenson, J. N., Beall, A. C., Loomis, J., Blascovich, J., & Turk, M. (2005). Transformed social interaction, augmented
gaze, and social inf luence in immersive virtual environments. Human Communication Research, 31, 511-537. do i: 10.1111/
j.1468-2958.2005.tb00881.x
Bailenson, J. N., & Segovia, K.Y. (2010). Virtual doppelgangers: Psychological effects of avatars who ignore their owners.
In W.S. Bainbridge (Ed.), Online worlds: Convergence of the real and the virtual (pp. 175-186). London, UK: Springer-Verlag.
doi: 10.1007/978-1-84882-825-4_14
Bailenson, J. N., Swinth, K. R., Hoyt, C. L., Persky, S., Dimov, A., & Blascovich, J. (2005). The independent and interactive
effects of embodied agent appearance and behavior on self-report, cognitive, and behavioral markers of copresence in
immersive virtual environments. Presence: Teleoperators and Virtual Environments, 14 , 379-393. doi:10.1162/105474605774785235
Bailenson, J. N., & Yee, N. (2006). A longitudinal study of task performance, head movements, subjective report, simulator
sickness, and transformed social interaction in collaborative virtual environments. Presence: Teleoperators & Virtual
Environments, 15, 699-716. doi: 10.1162/pres.15.6.699
Bailenson, J. N., Yee, N., Blascovich, J., & Guadagno, R. E. (2008). Transformed social interaction in mediated interpersonal
communication. In E. A. Konijn, S. Utz, M. Tanis, & S. B. Barnes (Eds.), Mediated interpersonal communication (p p. 77-9 9).
New York: Routledge.
Kristine L. Nowak & Jesse Fox
46
www.rcommunicationr.org
Bailenson, J. N., Yee, N., Merget, D., & Schroeder, R. (2006). The effect of behavioral realism and form realism of real-time
avatar faces on verbal disclosure, nonverbal disclosure, emotion recognition, and copresence in dyadic interaction.
Presence: Teleoperators and Virtual Environments, 15, 359-372. doi:10.1162/pres.15.4.359
Banks, J., & Bowman, N. D. (2016). Emotion, anthropomorphism, realism, control: Validation of a merged metric for
player–avatar interaction (PAX). Computers in Human Behavior, 54, 215-223. doi: 10.1016/j.chb.2015.07.030
Bargh, J. A., McKenna, K. Y. A., & Fitzsimons, G. M. (2002). Can you see the real me? Activation and expression of the
“true self” on the Internet. Journal of Social Issues, 58(1), 33-48. doi: 10.1111/1540-4560.00247
Behm-Morawitz, E., & Mastro, D. (2009). The effects of the sexualization of female video game characters on gender
stereotyping and female self-concept. Sex Roles, 61, 808-823. doi:10.1007/s11199-009-9683-8
Behm-Morawitz, E., Pennell, H., & Speno, A. G. (2016). The effects of virtual racial embodiment in a gaming app on reducing
prejudice. Communication Monographs, 83, 396 -418. doi: 10.1080/03637751.2015.1128556.
Behm-Morawitz, E., & Schipper, S. (2016). Sexing the avatar: Gender, sexualization, and cyber-harassment in a virtual
world. Journal of Media Psychology, 28, 161-174. doi: 10.1027/1864-1105/a000152
Bente, G., & Krämer, N. C. (2011). Virtual gestures: embodiment and nonverbal behavior in computer-mediated
communication. In A. Kappas & N. C. Krämer (Eds.), Face-to-face communication over the Internet (pp. 176-209). New York:
Cambridge.
Bente, G., Rüggenberg, S., Krämer, N. C., & Eschenburg, F. (2008). Avatar-mediated networking: Increasing social presence
and interpersonal trust in net-based collaborations. Human Communication Research, 34, 287-318. doi:
10.1111/j.1468-2958. 20 08.0 0322 .x
Bessière, K., Seay, A. F., & Kiesler, S. (2007). The ideal elf: Identity exploration in World of Warcraft. CyberPsychology &
Behavior, 10, 530-535. doi: 10.1089/cpb.2007.9994
Biocca, F. (1997). The cyborg’s dilemma: Progressive embodiment in virtual environments. Journal of Computer-Mediated
Communication, 3. doi:10.1111/j.1083-6101.1997.tb00070.x
Biocca, F., & Nowak, K. L. (2002). Plugging your body into the telecommunication system: Mediated embodiment, media
interfaces, and social virtual environments. In D. Atkin & C. Lin (Eds.), Communication technology and society: Audience
adoption and uses (pp. 407-447). Cresskill, NJ: Hampton Press.
Blascovich, J., & Bailenson, J. (2011). Infinite reality: Avatars, eternal life, new worlds, and the dawn of the virtual revolution. New
York: Harper Collins.
Blascovich, J., Loomis, J., Beall, A. C., Swinth, K. R., Hoyt, C. L., & Bailenson, J. N. (2002). Immersive virtual environment
technology as a methodological tool for social psychology. Psychological Inquiry, 13, 103-124. doi: 10.1207/
S15327965PLI1302_ 01
Breazeal, C. (2003). Emotion and sociable humanoid robots. International Journal of Human-Computer Studies, 59, 119-155. doi:
10.1016/S1071-5819(03)00018-1.
Brock, A. (2011). ‘‘When keeping it real goes wrong’’: Resident Evil 5, racial representation, and gamers. Games &Culture,
6, 429-452. doi: 10.1177/1555412011402676.
Burgess, M. C., Dill, K. E., Stermer, S. P., Burgess, S. R., & Brown, B. P. (2011). Playing with prejudice: The prevalence and
consequences of racial stereotypes in video games. Media Psychology, 14, 289-311. doi: 10.1080/15213269.2011.596467
Busselle, R. (2001). Television exposure, perceived realism, and exemplar accessibility in the social judgment process. Media
Psychology, 3, 43-67. doi: 10.1207/S1532785XMEP0301_03
Busselle, R., & Greenberg, B. (2000). The nature of television realism judgments: A reevaluation of their conceptualization
and measurement. Mass Communication & Society, 3, 249-268. doi:10.1207/S15327825MCS0323_05
Castronova, E. (2005). Synthetic worlds: The business and culture of online games. Chicago, IL: University of Chicago Press.
Castronova, E. (2006). On the research value of large games: Natural experiments in Norrath and Camelot. Games and
Culture, 1, 163–186.
Chaffee, S. H. (1991) Communication concepts 1: Explication. Newbury Park, CA: Sage.
Avatars and Computer-Mediated Communication
47 2018, 6, 30-53
Chan, E., & Vorderer, P. (2006). Massively multiplayer online games. In P. Vorderer & J. Bryant (Eds.), Playing video games:
Motives, responses, and consequences (pp. 77-88). Mahwah, NJ: Erlbaum.
Cheong, P. H., & Gray, K. (2011). Mediated intercultural dialectics: Identity perceptions and performances in virtual worlds.
Journal of International & Intercultural Communication, 4, 265-271. doi: 10.1080/17513057.2011.5980 47
Cho, H., & Salmon, C. T. (2007). Unintended effects of health communication campaigns. Journal of Communication, 57,
29 3-317. do i: 10.1111/j .14 60 -24 6 6. 20 07.0 03 4 4. x
Cicchirillo, V. (2015). Priming stereotypical associations: Violent video games and African American depictions.
Communication Research Reports, 32, 122-131. doi: 10.1080/08 824 096.2015.1016148
Dalibard, S., Magnenat-Thalmann, N., & Thalmann, D. (2012). Anthropomorphism of artificial agents: A comparative
survey of expressive design and motion of virtual characters and social robots. Paper presented at the Workshop on
Autonomous Social Robots and Virtual Humans at the 25th Annual Conference on Computer Animation and Social
Agents (CASA 2012), Singapore.
DeAndrea, D. C. (2014). Advancing warranting theory. Communication Theory, 24, 186-204. doi: 10.1111/comt.12033
Dill, K. E., Brown, B. P., & Collins, M. A. (2008). Effects of exposure to sex-stereotyped video game characters on tolerance
of sexual harassment. Journal of Experimental Social Psychology, 44, 1402-1408. doi:10.1016/j.jesp.2008.06.002
Dotsch, R., & Wigboldus, D. H. (2008). Virtual prejudice. Journal of Experimental Social Psychology, 44, 1194-1198. doi:
10.1016/j.jesp.2008.03.003
Eastwick, P. W., & Gardner, W. L. (2009). Is it a game? Evidence for social inf luence in the virtual world. Social Influence,
4, 18-32. doi: 10.1080/15534510802254087
Fox, J. (2012). Avatars in health communication contexts. In S. M. Noar & N. G. Harrington (Eds.), eHealth applications:
Promising strategies for behavior change (pp. 96-109). New York: Routledge.
Fox, J., & Ahn, S. J. (2013). Avatars: Portraying, exploring, and changing online and off line identities. In R. Luppicini (Ed.),
Handbook of research on technoself: Identity in a technological society (pp. 255-271). Hershey, PA: IGI Global. doi: 10.4018/978-
1-466 6-2211-1.ch014
Fox, J., Ahn, S. J., Janssen, J. H., Yeykelis, L., Segovia, K. Y., & Bailenson, J. N. (2015). Avatars versus agents: A meta-
analysis quantifying the effects of agency on social inf luence. Human-Computer Interaction, 30, 401-432. doi:
10.1080/07370024.2 014.921494
Fox, J., Arena, D., & Bailenson, J. N. (2009). Virtual reality: A survival guide for the social scientist. Journal of Media
Psychology, 21, 95-113. doi: 10.1027/186 4-1105. 21.3.95
Fox, J., & Bailenson, J. N. (2009a). Virtual self-modeling: The effects of vicarious reinforcement and identification on exercise
behaviors. Media Psychology, 12, 1-25. doi: 10.1080/15213260802669474
Fox, J., & Bailenson, J. N. (2009b). Virtual virgins and vamps: The effects of exposure to female characters’ sexualized
appearance and gaze in an immersive virtual environment. Sex Roles, 61, 147-157. doi: 10.1007/s11199-009-9599-3
Fox, J., Bailenson, J. N., & Tricase, L. (2013). The embodiment of sexualized virtual selves: The Proteus effect and experiences
of self-objectification via avatars. Computers in Human Behavior, 29, 930-938. doi: 10.1016/j.chb.2012.12.027
Fox, J., Ralston, R. A., Cooper, C. K., & Jones, K. A. (2015). Sexualized avatars lead to women’s self-objectification and
acceptance of rape myths. Psychology of Women Quarterly, 39, 349-362. doi: 10.1177/0361684314553578
Fox, J., & Tang, W. Y. (in press). Women’s experiences with harassment in online video games: Rumination, organizational
responsiveness, withdrawal, and coping strategies. New Media & Society.
Gaertner, S. L., Dovidio, J. F., & Houlette, M. A. (2010). Social categorization. In J. F. Dovidio, M. Hewstone, & V. M. Esses
(Eds.), Handbook of prejudice, stereotyping and discrimination (pp. 526-543). Thousand Oaks, CA: Sage.
Gamberini, L., Chittaro, L., Spagnolli, A., & Carlesso, C. (2015). Psychological response to an emergency in virtual reality:
Effects of victim ethnicity and emergency type on helping behavior and navigation. Computers in Human Behavior, 48,
104 -113. doi: 10.1016/j.chb.2015.01.0 40
Gerbaudo, P. (2015). Protest avatars as memetic signifiers: political profile pictures and the construction of collective identity
on social media in the 2011 protest wave. Information, Communication & Society, 18, 916 -929. doi: 10.1080/1369118X.2015.1043316
Kristine L. Nowak & Jesse Fox
48
www.rcommunicationr.org
Goffman, E. (1959). Presentation of self in everyday life. Garden City: NY: Anchor Books.
Gong, L. (2008). How social is social responses to computers? The function of the degree of anthropomorphism in computer
representations. Computers in Human Behavior, 24, 1494-1509. doi: 10.1016/j.chb.2007.05.007
Gong, L., & Nass, C. (2007). When a talking-face computer agent is half-human and half-humanoid: Human identity and
consistency preference. Human Communication Research, 33, 163-193. doi: 10.1080/15213260801906489
Green-Hamann, S., Campbell Eichhorn, K., & Sherblom, J. C. (2011). An exploration of why people participate in Second
Life social support groups. Journal of Computer-Mediated Communication, 16, 4 65-491. doi: 10.1111/j.1083- 6101.2011.01543.x
Groom, V., Bailenson, J. N., & Nass, C. (2009). The influence of racial embodiment on racial bias in immersive virtual
environments. Social Influence, 4, 1-18. doi: 10.1080/15534510802643750
Groom, V., Nass, C., Chen, T., Nielsen, A., Scarborough, J. K., & Robles, E. (2009). Evaluating the effects of behavioral
realism in embodied agents. International Journal of Human-Computer Studies, 67, 842-849. doi: 10.1016/j.ijhcs.2009.07.001
Guadagno, R. E., Blascovich, J., Bailenson, J. N., & McCall, C. (2007). Virtual humans and persuasion: The effects of
agency and behavioral realism. Media Psychology, 10, 1-22. doi: 10.1109/MIS.2002.1024753
Guadagno, R. E., Swinth, K. R., & Blascovich, J. (2011). Social evaluations of embodied agents and avatars. Computers in
Human Behavior, 27, 2380-2385. doi: 10.1016/j.chb.2011.07.017
Hamilton, M., A., & Nowak, K. L. (2010). Advancing a model of avatar evaluation and selection. PsychNology Journal, 8(1),
33- 65.
Hancock, J. T., & Toma, C. L. (2009). Putting your best face forward: The accuracy of online dating photographs. Journal
of Communication, 59, 367-386. doi: 10.1111/j.1460-2466.2009.01420.x
Hanus, M. D., & Fox, J. (2015). Persuasive avatars: The effects of customizing a virtual salesperson’s appearance on brand
liking and purchase intentions. International Journal of Human-Computer Studies, 84, 33-40. doi: 10.1016/j.ijhcs.2015.07.004
Hasler, B. S., & Friedman, D. A. (2012). Sociocultural conventions in avatar-mediated nonverbal communication: A
cross-cultural analysis of virtual proxemics. Journal of Intercultural Communication Research, 41, 238-259. doi:
10.1080/17475759.2012.728764
Haythornthwaite, C., & Wellman, B. (2002). The internet in everyday life: An introduction. In B. Wellman &
C. Haythornthwaite (Eds.), The internet in everyday life (pp. 1-41). Malden, MA: Blackwell.
Heyselaar, E., Hagoort, P., & Segaert, K. (2017). In dialogue with an avatar, language behavior is identical to dialogue with
a human partner. Behavior Research Methods, 49, 46-60. doi: 10.3758/s13428-015-0688-7
Huh, S., & Williams, D. (2010). Dude looks like a lady: Gender swapping in an online game. In W. S. Bainbridge, Online
worlds: Convergence of the real and the virtual (pp. 161-174). London, UK: Springer London.
James, T. W., Potter, R. F., Lee, S., Kim, S., Stevenson, R. A., & Lang, A. (2015). How realistic should avatars be? An initial
fMRI investigation of activation of the face perception network by real and animated faces. Journal of Media Psychology,
27, 109 -117. doi: 10.1027/1864-1105/a0 00156
Kafai, Y. B., Cook, M. S., & Fields, D. A. (2010). ‘‘Blacks deserve bodies too!’’ Discussion and design about diversity and
race in a tween virtual world. Games & Culture, 5, 43-63. doi: 10.1177/1555412009351261
Kahneman, D. (2012). A proposal to deal with questions about priming effects. Retrieved from https://www.nature.com/
polopoly_fs/7.6716.1349271308!/suppinfoFile/Kahneman Let ter.pdf
Kang, S. H., & Watt, J. H. (2013). The impact of avatar realism and anonymity on effective communication via mobile
devices. Computers in Human Behavior, 29, 1169-1181. doi: 10.1016/j.chb.2012.10.010
Kang, H. S., & Yang, H. D. (2006). The visual characteristics of avatars in computer-mediated communication: Comparison
of Internet Relay Chat and Instant Messenger as of 2003. International Journal of Human-Computer Studies, 64, 1173-118 3.
doi: 10.1016/j.ijhcs.2006.07.003
Kapidzic, S., & Herring, S. C. (2011). Gender, communication, and self-presentation in teen chatrooms revisited: Have
patterns changed? Journal of Computer-Mediated Communication, 17, 39 -59. doi: 10.1111/j.1083 -6101.2011.01561.x
Kennedy, J. S. (1992). The new anthropomorphism. Cambridge, UK: Cambridge University Press.
Avatars and Computer-Mediated Communication
49 2018, 6, 30-53
Kim, Y., & Sundar, S. S. (2012). Anthropomorphism of computers: Is it mindful or mindless? Computers in Human Behavior,
28, 241-250. doi: 10.1016/j.chb.2011.09.006
Klang, M. (2004). Avatar, from deity to corporate property: A philosophical inquiry into digital property in online games.
Information, Communication & Society, 7, 389-402. doi: 10.1080/1369118042000284614
Klimmt, C., Hefner, D., Vorderer, P., Roth, C., & Blake, C. (2010). Identification with video game characters as automatic
shift of self-perceptions. Media Psychology, 13, 323-338. doi: 10.1080/15213269.2010.524911
Kotlyar, I., & Ariely, D. (2013). The effect of nonverbal cues on relationship formation. Computers in Human Behavior, 29,
54 4-551. doi: 10.1016/j.chb.2012.11.02 0
Lakoff, G. (1987). Women, fire, and dangerous things: What categories reveal about the mind. Chicago, IL: The University of Chicago
Press.
Lee, E. J. (2004). Effects of visual representation on social influence in computer-mediated communication. Human
Communication Research, 30, 234-259. doi: 10.1111/j.1468-2958.2004.tb00732.x
Lee, E. J. (2007). Character-based team identification and referent informational influence in computer-mediated
communication. Media Psychology, 9, 135-155. doi: 10.1080/15213260701279622
Lee, E. J. (2010). What triggers social responses to flattering computers? Experimental tests of anthropomorphism and
mindlessness explanations. Communication Research, 37, 191-214. doi: 10.1111/j.1460-2466.2008.00386.x
Lee, E. J., & Nass, C. (2002). Experimental tests of normative group influence and representation effects in computer-mediated
communication. Human Communication Research, 28, 349-381. doi: 10.1111/j.1468-2958.2002.tb00812.x
Lee, J. E. R. (2014). Does virtual diversity matter? Effects of avatar-based diversity representation on willingness to express
offline racial identity and avatar customization. Computers in Human Behavior, 36, 190-197. doi: 10.1016/j.chb.2014.03.040
Lee, J. E. R., & Park, S. G. (2011). “Whose Second Life is this?” How avatar-based racial cues shape ethno-racial minorities’
perception of virtual worlds. CyberPsychology, Behavior, & Social Networking, 14, 637-642. doi: 10.1089/cyber.2010.0501
Lee, J. E. R., Nass, C. I., & Bailenson, J. N. (2014). Does the mask govern the mind? Effects of arbitrary gender representation
on quantitative task performance in avatar-represented virtual groups. Cyberpsychology, Behavior, & Social Networking, 17,
248-254. doi: 10.1089/cyber.2013.0358
Lee, K. M., Liao, K., & Ryu, S. (2007). Children’s responses to computer-synthesized speech in educational media: Gender
consistency and gender similarity effects. Human Communication Research, 33, 310-329. doi: 10.1111/j.1468-2958.2007.00301.x
Lee-Won, R., Tang, W. Y., & Kibbe, M. R. (2017). When virtual muscularity enhances physical endurance: Masculinity
threat and compensatory avatar customization among young male adults. Cyberpsychology, Behavior, & Social Networking,
20, 10-16. doi: 10.1089/cyber.2016.0418
Lehdonvirta, M., Nagashima, Y., Lehdonvirta, V., & Baba, A. (2012). The stoic male: How avatar gender affects help-seeking
behavior in an online game. Games & Culture, 7, 29-47. doi: 10.1177/1555412012440307
Lim, S., & Reeves, B. (2010). Computer agents versus avatars: Responses to interactive game characters controlled by a
computer or other player. International Journal of Human-Computer Studies, 68, 57-68. doi: 10.1080/03637750903074685
Maister, L., Sebanz, N., Knoblich, G., & Tsakiris, M. (2013). Experiencing ownership over a dark-skinned body reduces
implicit racial bias. Cognition, 128, 170-178. doi: 10.1016/j.cognition.2013.04.002
Martey, R. M., & Consalvo, M. (2011). Performing the looking-glass self: Avatar appearance and group identity in Second
Life. Popular Communication, 9, 165-180. doi: 10.1080/15405702.2011.583830
Martey, R. M., Stromer-Galley, J., Banks, J., Wu, J., & Consalvo, M. (2014). The strategic female: Gender-switching and
player behavior in online games. Information, Communication & Society, 17, 286-300. doi: 10.1080/1369118X.2013.874493
McGuire, W. J. (1968). Personality and attitude change: An information-processing theory. In A. Greenwald, T. Brock, &
T. Ostrom (Eds.), Psychological foundations of attitudes (pp. 171–196). New York: Academic Press.
McGuire, W. J. (1985). Attitudes and attitude change. In G. Lindzey & E. Aronson (Eds.) Handbook of social psychology
(pp. 233-346). New York, NY: Random House.
Meadows, M. S. (2007). I, avatar: The culture and consequences of having a second life. Berkeley, CA: New Riders.
Mori, M. (1970). The uncanny valley. Energy, 7(4), 33-35.
Kristine L. Nowak & Jesse Fox
50
www.rcommunicationr.org
Nakamura, L. (2002). Cybertypes: Race, ethnicity, and identity on the Internet. New York, NY: Routledge.
Nakamura, L. (2009). Don’t hate the player, hate the game: The racialization of labor in World of Warcraft. Critical Studies
in Media Communication, 26, 128-144. doi: 10.1080/15295030902860252
Nass, C. I., & Brave, S. (2005). Wired for speech: How voice activates and advances the human-computer relationship. Cambridge,
MA: MIT Press.
Nass, C., Lombard, M., Henriksen, L., & Steuer, J. (1995). Anthropocentrism and computers. Behavior & Information
Technology, 14, 229-238. doi:10.1080/01449299508914636.
Nass, C., & Moon, Y. (2000). Machines and mindlessness: Social responses to computers. Journal of Social Issues, 56, 81-103.
do i: 10.1111/00 22 -4537.0 0153.
Nowak, K. L. (2004). The influence of anthropomorphism and agency on social judgment in virtual environments. Journal
of Computer Mediated Communication, 9, doi: 10.1111/j.1083- 6101.2004.tb 00284.x.
Nowak, K. L. (2013). Choosing buddy icons that look like me or represent my personality: Using buddy icons for social
presence. Computers in Human Behavior, 29, 1456-1464. doi: 10.1016/j.chb.2013.01.027
Nowak, K. L. (2015). Examining perception and identification in avatar mediated interaction. In S.S. Sundar (Ed.), The
handbook of the psychology of communication technology (pp. 89-114). Hoboken, NJ: Wiley-Blackwell.
Nowak, K. L., & Biocca, F. (2003). The effect of agency and anthropomorphism on users’ sense of telepresence, social
presence and copresence in virtual environments. Presence: Teleoperators & Virtual Environments, 12, 481-494. MIT Press.
doi: 10.1162/105474603322761289.
Nowak, K. L., Fox, J., & Ranjit, Y. (2015). Inferences about avatars: Sexism, appropriateness, anthropomorphism, and the
objectification of female virtual representations. Journal of Computer-Mediated Communication, 20, 554-569. doi: 10.1111/
jcc4.12130
Nowak, K. L., Hamilton, M. A., & Hammond, C. C. (2009). The effect of image features on judgments of homophily,
credibility, and intention to use as avatars in future interactions. Media Psychology, 12, 50-76. doi: 10.1080/15213260802669433.
Nowak, K. L., & Rauh, C. (2005). The influence of the avatar on online perceptions of anthropomorphism, androgyny,
credibility, homophily, and attraction. Journal of Computer-Mediated Communication, 11(1), 153-178. doi:
10.1111/j.1083-6101.20 06.tb003 08.x.
Nowak, K. L., & Rauh, C. (2008). Choose your “buddy icon” carefully: The influence of avatar androgyny, anthropomorphism
and credibility in online interactions. Computers in Human Behavior, 24, 1473-1493. doi: 10.1016/j.chb.2007.05.005.
Open Science Collaboration. (2015). Estimating the reproducibility of psychological science. Science, 349 (6251), aac4716.
doi: 10.1126/science.aac4716
Palomares, N. A., & Lee, E. J. (2010). Virtual gender identity: The linguistic assimilation to gendered avatars in
computer-mediated communication. Journal of Language & Social Psychology, 29, 5-23. doi: 10.1177/0261927X09351675
Park, H., & Lee, H. (2013). Show us you are real: The effect of human-versus-organizational presence on online relationship
building through social networking sites. Cyberpsychology, Behavior, & Social Networking, 16, 265-271. doi:
10.1089/cyber.2012.0 051
Peck, T. C., Seinfeld, S., Aglioti, S. M., & Slater, M. (2013). Putting yourself in the skin of a black avatar reduces implicit
racial bias. Consciousness & Cognition, 22, 779-787. doi: 10.1016/j.concog.2013.04.016
Peña, J. F. (2011). Integrating the influence of perceiving and operating avatars under the automaticity model of priming
effects. Communication Theory, 21, 150-168. doi: 10.1111/j.1468-2885.2011.01380.x
Peña, J., Hancock, J. T., & Merola, N. A. (2009). The priming effects of avatars in virtual settings. Communication Research,
36, 838-856. doi: 10.1177/0093650209346802
Ratan, R., & Dawson, M. (2015). When Mii is me: A psychophysiological examination of avatar self-relevance. Communication
Research, 43, 1065-1093. doi: 10.1177/0093650215570652
Ratan, R., & Sah, Y. J. (2015). Leveling up on stereotype threat: The role of avatar customization and avatar embodiment.
Computers in Human Behavior, 50, 367-374. doi: 10.1016/j.chb.2015.04.010
Avatars and Computer-Mediated Communication
51 2018, 6, 30-53
Ravaja, N. (2009). The psychophysiology of digital gaming: The effect of a non co-located opponent. Media Psychology, 12,
268-294. doi: 10.1177/016327879501800306
Reeves, B., & Nass, C. (1996). The media equation: How people treat computers, television, and new media like real people and places.
Stanford, CA: CSLI Publications.
Rosenthal-von der Pütten, A. M., Krämer, N. C., Gratch, J., & Kang, S.-H. (2010). “It doesn’t matter what you are!”
Explaining social effects of agents and avatars. Computers in Human Behavior, 26, 1641-1650. doi: 10.1016/j.chb.2010.06.012
Sah, Y. J., Ratan, R., Tsai, H. S., Peng, W., & Sarinopoulos, I. (2016). Are you what your avatar eats? Health-behavior effects
of avatar-manifested self-concept. Media Psychology, DOI : 10.10 80/15213269.2016.1234397
Schönbrodt, F. D., & Asendorpf, J. B. (2011). Virtual social environments as a tool for psychological assessment: Dynamics
of interaction with a virtual spouse. Psychological Assessment, 23, 7–17. doi:10.1037/a00210 49
Schroeder, R. (2002). Social interaction in virtual environments: Key issues, common themes, and a framework for research.
In R. Schroeder (Ed.), The social life of avatars: Presence and interaction in shared virtual environments. London: Springer-Verlag.
doi: 10.10 07/978 -1-4471- 0277-9_1
Seyama, J., & Nagayama, R. S. (2007). The uncanny valley: Effect of realism on the impression of artificial human faces.
Presence: Teleoperators & Virtual Environments, 16, 337-351. doi:10.1162/pres.16.4.337
Sheehan, J., & Sosna, M. (Eds.) (1991). The boundaries of humanity: Humans, animals, machines. Oxford, England: University
of California Press.
Slater, M., & Steed, A. (2002). Meeting people virtually: experiments in shared virtual environments. In R. Schroeder (Ed.),
The social life of avatars: Presence and interaction in shared virtual environments (pp. 146-171). London, UK: Springer-Verlag.
doi: 10.10 07/978 -1-4471- 0277-9_9
Spence, P. R., Lachlan, K. A., Westerman, D., & Spates, S. A. (2013). Where the gates matter less: Ethnicity and perceived
source credibility in social media health messages. Howard Journal of Communications, 24, 1-16. doi:
10.1080/106 46175.2013.748593
Stein, J. P., & Ohler, P. (2017). Venturing into the uncanny valley of mind: The influence of mind attribution on the acceptance
of human-like characters in a virtual reality setting. Cognition, 160, 43-50. doi.org/10.1016/j.cognition.2016.12.010.
Stephenson, N. (1992). Snow crash. New York, NY: Bantam Books.
Stromer-Galley, J., & Martey, R. M. (2009). Visual spaces, norm governed places: The influence of spatial context online.
New Media & Society, 11, 1041-1060. doi: 10.1177/1461444809336555
Tamir, P., & Zohar, A. (1991). Anthropomorphism and teleology in reasoning about biological phenomena. Science Education,
75, 57– 67.
Turkle, S. (1995). Life on the screen: Identity in the age of the Internet. New York, NY: Simon & Schuster.
Vandenbosch, L., Dreismans, K., Trekels, J., & Eggermont, S. (2017). Sexualized video game avatars and self-objectification
in adolescents: The role of gender congruency and activation frequency. Media Psychology, 20, 221-239. doi.org/10.1080/
15213269.2016.1142380.
Van Der Heide, B., Schumaker, E. M., Peterson, A. M., & Jones, E. B. (2013). The Proteus effect in dyadic communication:
Examining the effect of avatar appearance in computer-mediated dyadic interaction. Communication Research, 40, 838-860.
doi: 10.1177/0093650212438097
Van Der Land, S. F., Schouten, A. P., Feldberg, F., Huysman, M., & Van Den Hooff, B. (2015). Does avatar appearance
matter? How team visual similarity and member-avatar similarity inf luence virtual team performance. Human
Communication Research, 41, 128-153. doi: 10.1111/hcre.12044
Vang, M. H., & Fox, J. (2014). Race in virtual environments: Competitive versus cooperative games with black or white
avatars. CyberPsychology, Behavior, & Social Networking, 17, 235-240. doi: 10.1089/cyber.2013.0289
Verhagen, T., Van Nes, J., Feldberg, F., & Van Dolen, W. (2014). Virtual customer service agents: Using social presence and
personalization to shape online service encounters. Journal of Computer-Mediated Communication, 19, 529-545. doi: 10.1111/
jcc4.12066
Kristine L. Nowak & Jesse Fox
52
www.rcommunicationr.org
Waddell, T. F., & Ivory, J. D. (2015). It’s not easy trying to be one of the guys: The effect of avatar attractiveness, avatar sex,
and user sex on the success of help-seeking requests in an online game. Journal of Broadcasting & Electronic Media, 59,
112-129. doi: 10.1080/08838151.2014.998221
Walther, J. B. (1996). Computer-mediated communication: Impersonal, interpersonal, and hyperpersonal interaction.
Communication Research, 23, 3-43. doi: 10.1177/009365096023001001
Walther, J. B., & Parks, M. R. (2002). Cues filtered out, cues filtered in: Computer-mediated communication and relationships.
In M. L. Knapp & J. A. Daly (Eds.), Handbook of interpersonal communication (3rd ed., pp. 529-563). Thousand Oaks, CA:
Sage.
Webb, S. (2001). Avatarculture: Narrative, power and identity in virtual world environments. Information, Communication &
Society, 4, 560-594. doi: 10.1080/13691180110097012
Westerman, D., Tamborini, R., & Bowman, N. D. (2015). The effects of static avatars on impression formation across
different contexts on social networking sites. Computers in Human Behavior, 53, 111-117. doi: 10.1016/j.chb.2015.06.026
Williams, D. (2010). The mapping principle, and a research framework for virtual worlds. Communication Theory, 20, 451-470.
do i: 10.1111/j.14 68-2 885.2010.01371.x
Won, A. S., Bailenson, J., Lee, J., & Lanier, J. (2015). Homuncular flexibility in virtual reality. Journal of Computer-Mediated
Communication, 20, 241-259. doi: 10.1111/jcc4.12107
Yao, M. Z., Mahood, C., & Linz, D. (2010). Sexual priming, gender stereotyping, and likelihood to sexually harass:
Examining the cognitive effects of playing a sexually-explicit video game. Sex Roles, 62, 77-88. doi: 10.1007/s11199-009-9695-4
Yee, N. (2014). The Proteus paradox: How online games and virtual worlds change us--and how they don’t. New Haven, CT: Yale
University Press.
Yee, N., & Bailenson, J. (2007). The Proteus effect: The effect of transformed self-representation on behavior. Human
Communication Research, 33, 271-290. doi: 10.1111/j.1468-2958.2007.00299.x
Yee, N., & Bailenson, J. N. (2009). The difference between being and seeing: The relative contribution of self-perception
and priming to behavioral changes via digital self-representation. Media Psychology, 12, 195-209. doi:
10.1080/15213260902849943
Yee, N., Bailenson, J. N., & Ducheneaut, N. (2009). The Proteus effect: Implications of transformed digital self-representation
on online and offline Behavior. Communication Research, 36, 285-312. doi: 10.1177/0093650208330254
Yee, N., Bailenson, J. N., Urbanek, M., Chang, F., & Merget, D. (2007). The unbearable likeness of being digital: The
persistence of nonverbal social norms in online virtual environments. CyberPsychology & Behavior, 10, 115-121. doi: 10.1089/
cpb.20 06.9 984
Yee, N., Ducheneaut, N., Yao, M., & Nelson, L. (2011, May). Do men heal more when in drag? Conflicting identity cues
between user and avatar. In Proceedings of the SIGCHI conference on human factors in computing systems (pp. 773-776). ACM.
Yuan, B., Folmer, E., & Harris, F. C. (2011). Game accessibility: A survey. Universal Access in the Information Society, 10, 81-100.
doi: 10.10 07/s10209-010- 0189-5
53
Copyrights and Repositories
This work is licensed under the Creative Commons Attribution-NonCommercial-3.0 Unported License.
This license allows you to download this work and share it with others as long as you credit the authors and the journal.
You cannot use it commercially without the written permission of the authors and the journal.
Attribution
You must attribute the work to the author and mention the journal with a full citation, whenever a fragment or the full text
of this paper is being copied, distributed or made accessible publicly by any means.
Commercial use
The licensor permits others to copy, distribute, display, and perform the work for non-commercial purposes only, unless you
get the written permission of the authors and the journal.
The above rules are crucial and bound to the general license agreement that you can read at:
http://creativecommons.org/licenses/by-nc/3.0/
Corresponding authors
Kristine L. Nowak, Ph.D.
University of Connecticut
337 Mansfield Rd, Unit 1259, Storrs, CT 06269, USA
Email: Kristine.nowak@uconn.edu
Jesse Fox, Ph.D.
The Ohio State University
3016 Derby Hall, 154 North Oval Mall, Columbus, OH 43210-1339, USA
Email: fox.775@osu.edu
Attached is a list of permanent repositories where you can find the articles published by RCR:
Academia.edu @ http://independent.academia.edu/ReviewofCommunicationResearch
Internet Archive @ http://archive.org (collection “community texts”)
Social Science Open Access Repository, SSOAR @ http://www.ssoar.info/en/home.html
Kristine L. Nowak & Jesse Fox Avatars and Computer-Mediated Communication
Review of Communication Research 2018, 6, 30-53
