Special Issue: Theoretical Consolidation and Innovation in Media Psychology

Theoretical Article

An Explication of Identity Shift

Theory
Getting Our Shift Together
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Caleb T. Carr1 , Yeweon Kim2, Jacob J. Valov3, Judith E. Rosenbaum4 ,

This document is copyrighted by the American Psychological Association or one of its allied publishers.

Benjamin K. Johnson5 , Jeffrey T. Hancock6, and Amy L. Gonzales2

1
School of Communication, Illinois State University, Normal, IL, USA
Department of Communication, University of California – Santa Barbara, CA, USA
2

Department of Communication Arts, University of Wisconsin – Madison, WI, USA

3

4
Department of Communication and Journalism, University of Maine, Orono, ME, USA
5
Department of Advertising, University of Florida, Gainesville, FL, USA
6
Department of Communication, Stanford University, CA, USA

Abstract: First articulated in 2008, the concept of identity shift refers to the process of self-transformation that is the result of intentional
self-presentation in a mediated context. As research into identity shift has become increasingly prevalent, our understanding of the concept’s
mechanisms and constraints has become more detailed and in depth. We now have a greater understanding of the process and limitations of
identity shift: an understanding sufficient to begin articulating a theory of identity shift that explains and predicts the intrapersonal effects of
mediated self-presentations. The present work advances such a theory. We begin by summarizing and synthesizing extant identity shift work –
including published articles, presented manuscripts, and unpublished research – to better understand the identity shift process. We then use
this synthesis to articulate an initial theory (identity shift theory; IST) that specifies the processes, conditions, constraints, and effects of
identity shift based on personal, psychological, and communicative characteristics. Ultimately, the advancement of identity shift theory can
inform future research and practice into the implications of online self-presentation for self-effects.

Keywords: self-presentation, self-concept, computer-mediated communication, self-perception, intrapersonal communication

Identity shift refers to the process of self-transformation emerged in the last decade, and then articulating an initial
resulting from intentional self-presentation in a mediated theory of identity shift based on that expanded perspective.
context. Since its articulation in 2008, the concept of iden-
tity shift has received increasing attention and empirical
assessment, both directly and indirectly. Directly, studies
probing the antecedents, processes, boundaries, and out- Conceptual Foundations
comes of identity shift increasingly appear in journals and
as conference presentations. Indirectly, studies of online Foundational to identity shift scholarship is the idea that
self-presentation, media use, and self-perceptions have individual identity is a product of social interactions. This
called on identity shift as a post hoc explanatory mecha- idea originates from Cooley’s (1902) “looking-glass self,”
nism or have provided information to advance future work a symbolic interactionist perspective that assumes the self
applying identity shift directly. Alongside this surge of pub- is a fluid amalgam of different relationships and environ-
lic scholarship, additional conference discussions, unpub- ments. This paradigm helped lay the groundwork for mul-
lished manuscripts, and unrealized data reveal that the tiple theories in communication and social psychology
contribution captured by the published work is far from that have been central to the evolution of the identity shift
complete. Consequently, there is a need to consider and construct. We thus begin our interrogation by highlighting
explicate just what identity shift is, taking stock of the con- these theoretical foundations and the ways identity shift
cept, its known strengths and limitations, and delineating scholarship has expanded upon them, before outlining the
future research questions. This article advances identity current state of identity shift research and moving into
shift by reviewing and consolidating findings that have our explication of identity shift theory (IST).

Journal of Media Psychology (2021), 33(4), 202–214 Ó 2021 Hogrefe Publishing

https://doi.org/10.1027/1864-1105/a000314
 C. T. Carr et al., Identity Shift Theory 203

The Hyperpersonal Model From these guiding initial frameworks and foundational
study, much research has since been conducted to advance
Building from Walther’s (1996) hyperpersonal framework, our understanding of identity shift processes and effects.
Gonzales and Hancock (2008) proposed identity shift as
the self-transformation that occurs following the online
public self-presentation of the self through observation,
or self-perception (Bem, 1972), of one’s own behavior. In
The Present State of Identity Shift
their foundational study, adapted from offline experiments
(e.g., Kelly & Rodriguez, 2006; Tice, 1992), Gonzales and Shortly after its initial development, research began to
Hancock’s (2008) participants were asked to respond to explore mechanisms related to identity shift beyond
four questions about themselves as if they were either self-presentation, often drawing from the hyperpersonal
introverted or extraverted, providing their deliberate processes that underpin identity shift. These studies
self-statements via either a text document or blog post.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

advanced our understanding of the identity shift concept

This document is copyrighted by the American Psychological Association or one of its allied publishers.

Findings revealed participants in the blog post condition by identifying and extending the boundaries and interven-
self-reported self-perceptions commensurate with their ing mechanisms guiding identity shift, and are discussed
self-presentation – an outcome not demonstrated by par- here.
ticipants randomly assigned to deliberately self-present
in a text document. These findings supported the identity
shift construct, in which individuals internalize their Feedback
strategic online self-presentation.
The mediated nature of self-presentations is critical Of initial interest was the effect of feedback on identity
to identity shift. Unlike offline processes, the hyperper- shift. Although Gonzales and Hancock (2008) experimen-
sonal model suggests online self-presentation requires tally induced private and public self-presentation, self-pre-
greater attention to how one’s self is articulated, involving sentational statements are often met with feedback in situ
more composition and editing, as a function of that self- (Ellison et al., 2014). Thus, the effects of deliberate self-pre-
presentation being visible to the sender on the screen sentation may have limited external validity when studied
before and after it is communicated (Walther, 2007). Per in isolation. As a result, several studies probed the effects
Walther, individuals allocate greater cognitive resources of feedback to one’s self-presentation. Walther et al.
to their mediated self-presentation when they take the time (2011) identified that confirmatory feedback enhanced
to type out (or otherwise self-present) self-statements, identity shift processes, providing an intensifying feedback
and are then concurrently exposed to their own self- loop proposed by the hyperpersonal model. Carr and Fore-
presentation (see also, Bem, 1972). In this way, the most man (2016) found that the source of the feedback mattered,
critical receiver in identity shift is actually the individual so that confirmatory feedback from a close tie (e.g., friend
presenting, who is then subsequently exposed to and expe- or roommate) intensified identity shift compared to identi-
riences self-effects: “[T]he effect of messages on the cogni- cal feedback from a stranger. Further experiments com-
tions (knowledge or beliefs), emotions, attitudes, and pared confirmatory and disconfirmatory feedback, and
behavior of the message creators/senders themselves” found that negative responses to self-presentation could
(Valkenburg et al., 2016, p. 2). ameliorate identity shift (Carr & Hayes, 2019) or even pro-
duce boomerang effects away from the initial self-presenta-
tion (Johnson & Van der Heide, 2015).
Public Commitment
Central to Gonzales and Hancock’s (2008) initial conceptu-
Assumptions About the Audience
alization of identity shift was the distinction between pri-
vate self-presentation (for which they found no effect) Another collection of research has delved into how a self-
and public self-presentation (for which they did find an presenter’s assumptions about the audience moderate iden-
identity shift effect). Building on the principle of public com- tity shift. For example, a sender’s assumption of an audi-
mitment, in which individuals feel more beholden to self- ence’s size and how connected senders felt to this
presentations that are visible to others (Schlenker et al., audience moderated the effects of publicness on self-trans-
1994; Tice, 1992), identity shift was considered an outcome formation effects (Rosenbaum & Johnson, 2019), indicating
of selective self-presentation and public commitment to an perceptions of an audience may be just as critical as the
espoused identity, whereby an individual alters their own actual audience. Moreover, French and Hancock (2016)
self-view to be consonant with the now-public identity. found transformation toward a presented trait is stronger

Ó 2021 Hogrefe Publishing Journal of Media Psychology (2021), 33(4), 202–214

 204 C. T. Carr et al., Identity Shift Theory

when presenters assume the presented trait is valued and self-concept can be influenced by self-statements, and iden-
cared about by the audience, indicating audiences’ per- tity shift may have critical boundary conditions.
ceived values influence identity shift. Further, Carr (2021)
demonstrated messages targeted toward a narrower or
more deliberate audience (e.g., a particular individual Time
rather than a broad group) evoked greater transfor-mation. Finally, duration and development of identity shift over
time are important but understudied. Shifted self-concepts
may fade over time (i.e., return to pre-statement levels),
Self-Concept Outcomes
particularly from single self-presentations. Assessing partic-
Broadly, identity shift has considered transformations in ipants 1 week after their self-presentations, Johnson and
self-concept, or “the collection of beliefs and attitudes of Van der Heide (2015) found feedback effects persisted –
an individual about him or herself” (Valkenburg, 2017, and in some cases were strengthened – over time. However,
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

p. 3). That shift in self-concept refers to a change in the they found less support for the persistence of the publicness
This document is copyrighted by the American Psychological Association or one of its allied publishers.

salience of some beliefs or attitudes about the self. For effects on self-concept. Carr (2019) likewise did not evi-
example, much of the early work on identity shift asked dence a temporal effect of self-transformation, finding no
people to behave in an extraverted or introverted manner, differences in self-reported trait (i.e., altruism) assessed at
in part due to extraversion’s deep roots in psychology as three periods following self-presentation (immediate, 5
a component of self-concept (see Tice, 1992). Identity shift min, or 15 min). More work is needed to determine the per-
assumes enacting these behaviors made extraverted or sistence of effects both from the time of self-presentation
introverted aspects of the self more salient. and due to the frequency of self-presentation.
Research has begun to explore additional personality
traits and characteristics that may also be malleable, and
thus susceptible to identity shift effects. For example,
Johnson and Van der Heide (2015) found that publicly Lessons Learned (But Not Always
sharing attitudes about abstract photos could influence Published)
subsequent attitudes toward art. Carr and Hayes (2019)
extended identity shift to account for changes in individuals’ Although the extant literature seems to point toward a
relationships with a brand identity, feeling more or less stable (albeit expanded) understanding of identity shift, sev-
affinity toward a brand (i.e., Starbucks) commensurate with eral challenges exist within identity shift research. These
their selective self-presentations. And, exploring political challenges are compounded by the publication bias against
communication in social media, Lane and colleagues nonsignificant statistical results (see Levine et al., 2009).
(2019) found individuals who made political self-disclosures Perhaps because identity shift work has typically identified
online were subsequently more politically motivated. small effect sizes for the main effect of deliberate self-pre-
Although research has identified several traits and atti- sentation (Peña and Hill [2020] calculated Cohen’s d =
tudes that can be affected by selective mediated self- 0.36 based on extant articles), several studies remain
presentation, several recent studies have indicated there unpublished. Yet the corpus of identity shift work, pub-
may be some facets of one’s personality not susceptible lished and unpublished alike, can help inform subsequent
to identity shift. Carr (2019) found assigning individuals protocols or boundary conditions through articulating chal-
to deliberately self-present as either generous or miserly lenges and obstacles researchers have faced, and not
did not influence either self-perceived altruism or behav- always overcome.
ioral generosity toward others, revealing an unexpected These challenges also inform the theory of identity shift
boomerang effect whereby those presenting as greedy actu- this article proffers, and can guide future study. The six
ally became slightly more altruistic. Hampton et al. (2020) challenges we identify here have been drawn from pub-
likewise found that individuals making photographic and lished literature, conference presentations, and the personal
textual self-claims about their clothing stylishness did not communication with and experiences of researchers who
experience subsequent changes in their self-perceived fash- have conducted identity shift research but have been
ionability. Meanwhile, Kim and Gonzales (2018) suggested unable to publish. Many of these studies were conducted
self-presentational publicness and subsequent feedback in an attempt to understand the conceptual foundations
could have varying effects on the perception toward feed- and strengthen the external validity of the identity shift
back providers, depending on the desire to preserve one’s construct, incorporating the features and norms of use from
existing view of the self (i.e., self-verification; Swann & a variety of digital platforms, and yield a string of lessons
Read, 1981). These findings suggest not all facets of one’s learned.

Journal of Media Psychology (2021), 33(4), 202–214 Ó 2021 Hogrefe Publishing

 C. T. Carr et al., Identity Shift Theory 205

Individuals Need to Be Identifiable in Their some work has found that self-concept change can occur
Self-Presentations as a function of tying feedback to a history of online self-
presentation (French, 2018), other studies have not been
Individuals must be directly identified in message (e.g., as successful in prompting self-concept change without
name/hometown, profile photo, associated social network- strong self-presentation (Gonzales & Petro, 2020). This is
ing site [SNS] account) for identity shift to occur. In the com- perhaps not surprising given the implicated mechanisms
mon identity shift experimental paradigm, private self- typically involve poignant observation of one’s own behav-
awareness – which focuses on introspection of the internal ior (Bem, 1972; Tice, 1992).
and inward-directed aspects of the self (Froming et al., It is also possible inauthentic self-presentations will com-
1982) – is generated by asking otherwise-anonymous partic- promise identity shift. As one example, Gonzales and Kim
ipants to begin or conclude their self-statements with their (2014) found that, contrary to expectations, posting edited
first and last name and hometown, providing a personal selfies actually reduced posters’ self-concepts (relative to
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

and visible connection between the participant and their non-edited selfies), so that deliberate favorable self-
This document is copyrighted by the American Psychological Association or one of its allied publishers.

statement. When this simple “name and hometown” manip- presentations resulted in diminished self-attitudes. Conver-
ulation is absent, no identity shift effects manifest (Carr & sations with participants after the study indicated that the
Hayes, 2019, Study 1). Following advice from a colleague, lack of naturalism associated with needing to unexpectedly
the study was replicated – but participants included the post selfies on Facebook in the middle of the study was not
name/hometown identifiers – and resulted in the expected viewed as a positive experience for most participants, and
identity shift effects (Carr & Hayes, 2019, Study 2). Even was only made worse for those participants who were
minimal biographic information may create discursive iden- forced to edit those selfies. The tension between internal
tifiability (see Anonymous, 1998) to activate private self- and external validity in these studies could account for
awareness and thus identity shift in a laboratory setting. these mixed or inconclusive results.
Most studies examining identity shift in situ, such as
through participants’ actual social media profiles (i.e., Face-
book in Carr & Foreman, 2016; Facebook and Twitter in “Publicness” Is Complex
Rosenbaum & Johnson, 2019), did not need to have partic- One of the initial premises of identity shift was that for the
ipants take additional steps to identify themselves, as the internalization of external behaviors to occur, people must
system automatically displayed their name and picture next publicly commit to this behavior. Consistent with Schlenker
to their posted self-statements. Demonstrating the challenge et al. (1994), Gonzales and Hancock’s (2008) initial study
of innate identifiability in social media, many of Rosenbaum found that mediated self-presentation via private text docu-
and Johnson’s (2019) participants refused to comply with ment did not yield identity shift effects as did the same
their study’s request to share or like something via their per- mediated self-presentation via a purportedly publicly acces-
sonal Facebook page or Twitter feed, implying concern sible blog post.
about their attachment to their profile and maintenance of In the second identity shift study, Walther and colleagues
a consistent online presence. Ultimately, whether induced (2011) drew on the computers as social actors (CASA;
to do so or occurring innately via system features, individu- Reeves & Nass, 1996) paradigm and found identity shift
als must connect their self-statements to themselves for still occurred when an individual was asked to disclose so
those self-statements to manifest transformative effects. that a nonhuman computer algorithm could assess their
self-presentation, albeit to a lesser degree than when dis-
Self-Presentations Must Be Salient and closing to other humans via an online blog. Most recently,
Carr (2021) demonstrated the mere consideration of an
Authentic
audience was sufficient to activate identity shift processes,
The form, salience, and authenticity of self-presentation are as presenters simply considered specific receiver groups
important, as participants must draw from personal experi- when composing a message they knew would not be trans-
ences to experience self-transformation. Presenters who mitted or accessible. Given these findings, it seems a criti-
purely fabricate or invent experiences do not connect them- cal component of identity shift is the sender’s imagining of
selves to those statements, and thus will not experience a self-presentational message as public, even if the medi-
identity shift. For example, Rosenbaum and Johnson ated message is not actually accessible to others.
(2015) found no impact of public sharing of food diaries Public – and publicly accessible – commitment remains
on Instagram on eating behaviors, and attributed the lack an important factor of identity shift, because it both
of identity shift to respondents’ general apathy toward commits the individual to the presented self (Schlenker
healthy eating. Most identity shift studies elicit relevant et al., 1994; Tice, 1992) and it provides an opportunity for
self-presentations in the context of the study. Although others to access and validate that presentation. Yet in

Ó 2021 Hogrefe Publishing Journal of Media Psychology (2021), 33(4), 202–214

 206 C. T. Carr et al., Identity Shift Theory

recent years, the distinction between private and public has components of our self-concept may not be able to be altered
become increasingly blurred (e.g., Marwick & boyd, 2011), by selective self-presentation. Some research has shown
raising questions about how to define “public.” Walther identity shift to apply to relatively fleeting or fluid char-
et al. (2011) found that identity shift occurs whenever there acteristics of the self, such as people’s brand attitudes
is a sense of co-presence, whether this means in a com- (Rosenbaum & Johnson, 2019) or opinions about art
pletely public setting or for just one other person. An addi- (Johnson & Van der Heide, 2015). More core personal traits,
tional complication comes from most identity shift research such as altruism, have failed to be influenced by deliberate
relying on researchers’ assumptions (and dichotomization) self-presentation (Carr, 2019). Consequently, while some
of public and private, without considering how participants facets of one’s self seem susceptible to influence, others
interpret the publicness of the various platforms. In one seem resistant to identity shift’s transformation. Additional
study that directly explored perceived publicness, Waterloo work is needed to explore these boundaries and their nature.
(2017) found no differences in identity shift between those
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

self-presenting publicly (i.e., using hashtags to increase

The Persistence of Self-Presentation
This document is copyrighted by the American Psychological Association or one of its allied publishers.

searchability, making posts publicly accessible, and using

notification settings to be made aware of public interaction May Matter
with the post) and those self-presenting semi-publicly One question raised by the nature of popular social media
(i.e., without using any of those three features). Likewise, platforms is how persistent self-presentations need to be
although Walther et al. (2011) found that public self- to yield identity shift. Exposure to and reflection on one’s
presentations (i.e., blogs) resulted in greater identity shift own self-presentation is seen as an essential part of the
than did private self-presentations (i.e., text documents), self-persuasion underpinning identity shift (Bem, 1972);
an unreported analysis on those in the public condition yet ephemeral channels (e.g., Snapchat, Twitter fleets)
revealed participants’ perceptions of how many people can make self-presentations appear temporary and inconse-
could see their post (range: 1–1,000,000) did not moderate quential to the sender, not giving senders time to be
the effect of publicness on identity shift. Rosenbaum and exposed and reflect on their self-presentations. Moreover,
Johnson (2019) attempted to account for people’s under- self-presentations via ephemeral media may not afford or
standing of publicness and its impact on identity shift by encourage the degree of editing and deliberation seen in
assessing participants’ perceived publicness of behavior persistent channels, rendering ephemeral presentations less
on the social media platform where they posted; but they persistent and impactful (cf. Choi et al., in press). While
found that high levels of perceived publicness were actually research that involves participants self-presenting on their
negatively related to the impact of behavior on brand atti- own social media posts shows evidence of identity shift,
tude on brand attitude, while lower levels of publicness research has yet to examine whether these participants
led to a positive effect of behavior on brand attitudes. decided to keep these posts on their timeline and if they
These findings complicate the idea that publicness is the did not, how deleting their post impacted the persistence
driver behind identity shift, pointing to the complexities of the identity shift observed earlier. Similarly, repeated
around how social media users understand privacy and self-presentations may have cumulative effects. At the
publicness (see Marwick & boyd, 2011). In short, not only same time, there is some evidence that publicness – not
do social media users have diverse privacy setting prefer- self-perception – led to identity shift, raising questions
ences, people also have different understandings of what whether self-presentation on ephemeral channels will have
constitutes a “public” behavior and who or what their imag- a different impact on one’s self-concept compared with
ined audience is (Litt, 2012). Furthermore, it is possible more persistent presentations (Van der Heide, Dickinson,
publicness does not matter as much as the value a social et al., 2013). As individuals – and particularly youth – contin-
media user attaches to the audience for whom they are – uously construct and present their identities online (Cingel
or think they are – performing (Hayes et al., 2016). Addi- et al., 2014), self-presentations may be incorporated into
tional work is clearly needed to explicate the role of public- individuals’ self-concepts as durable or permanent identity
ness in identity shift as the nature of public and audience shifts.
evolves in social media contexts.

Not All Feedback Is Equal

Some Personal Traits Are Resistant
Although feedback intensifies identity shift, not all feed-
to Change
back – and therefore effects of feedback – is equivocal. Lik-
From the immutable (e.g., you cannot identity shift yourself ing a post on social media takes less effort than reposting
to believe you are an asparagus stalk) to the critical and or leaving a comment, and a like can be perceived as less
deeply held internalized views (e.g., altruism, religion), some deliberate (Carr et al., 2016). Moreover, liking someone

Journal of Media Psychology (2021), 33(4), 202–214 Ó 2021 Hogrefe Publishing

 C. T. Carr et al., Identity Shift Theory 207

Figure 1. Identity shift theory and its component propositions. Proposition 1: Selective self-presentation results in identity shift. Proposition 2:
Identity shift is stronger when selective self-presentation is publicly accessible. Proposition 3: Identity shift is stronger when third-party feedback
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

confirms the selective self-presentation. Proposition 4: To activate identity shift, presenters must be identifiable and connected to their selective
self-presentation. Proposition 5: To activate identity shift, presenters must identify with their selective self-presentation, seeing the presented
self as a potential self.

else’s content may represent different levels of commu- flexibility in constructing messages and filters out nonverbal
nicative and ego involvement than interaction (e.g., com- cues associated with entrained (typically conceptualized as
menting) or rebroadcast (e.g., retweeting). Indeed, in face-to-face) communication (Walther, 1996). The more
their study of political self-expression, Lane et al. (2019) time a medium affords an individual to encode and transmit
found that liking others’ political posts did not have the messages, and the more cues that are filtered out from mes-
same self-transformative effects as composing and sharing sages sent through said medium, the more disentrained the
one’s own political views. Variance in deliberativeness and communication via that medium is. Mediated channels –
involvement in providing feedback to or interacting with an from handwritten letters to emails to videoconferences to
individual’s self-statement may further impact the extent TikTok – afford individuals varying degrees of disentrain-
of identity shift (Rosenbaum & Johnson, 2019). ment, as each filters out various nonverbal cues allowing
for greater opportunities for selective self-presentation. As
users have more time to self-present, they can direct more
cognitive resources toward constructing a strategic and
Advancing Identity Shift Theory intentional message to articulate themselves and appeal to
Consolidating extant scholarship, this final section is their conversation partner, with a particular focus on aspects
devoted to the articulation of an initial theory of identity that can be obfuscated through text-only channels, such as
shift. The theory is informed by the hyperpersonal model, physical characteristics and behaviors (Walther, 1996).
but accommodates the expected boundaries and influences Identity shift occurs when this curated self-presentation
unique to identity shift. Identity shift theory (IST) focuses is presented to an audience and that audience has the
on identity shift as a conceptual mechanism connecting opportunity for giving feedback (Gonzales & Hancock,
self-presentation and identity characteristics as antece- 2008). Because individuals sending a message through a
dents, and the breadth and scope of changes in self-concept disentrained channel know their recipient may not receive
that may be expected as outcomes. Presentation publicness their message as it is composed, they have the opportunity
and feedback loops are articulated as moderating factors in to contemplate their now-visible self-presentation, which
the theory. Conceptually modeled in Figure 1, we explicate then forms the foundation of the identity shift.
several key concepts (e.g., self-presentation, altered percep-
tions and behaviors) and the relations between them, Proposition 2: Identity Shift Is Stronger
advancing several propositions for the relationships among
When Selective Self-Presentation Is
the concepts to constitute an initial theory of identity shift.
Publicly Accessible
Proposition 1: Selective Mediated Although the transformative outcomes of identity shift are
driven by self-statements, identity shift effects are exacer-
Self-Presentation Results in Identity Shift
bated by the degree to which an individual believes their
Identity shift occurs when people engage in selective self- mediated self-presentation is observable by an audience,
presentations through mediated, disentrained communica- or is public, in nature. Early scholars in the study of the self
tion channels. Disentrainment refers to the degree to which suggested the observed presence of an audience prompts
a given communication channel affords users temporal “awareness that one is the object of another’s attention’’

Ó 2021 Hogrefe Publishing Journal of Media Psychology (2021), 33(4), 202–214

 208 C. T. Carr et al., Identity Shift Theory

(i.e., public self-awareness; Orive, 1984, p. 730). That is, self-presentation – seems to ameliorate the internalization
when individuals are publicly self-aware they perceive they of the stated characteristic (Carr & Hayes, 2019), likely
are being observed and evaluated by others, which pro- due to the resultant cognitive dissonance of seeking to
duces apprehension and discomfort, which in turn leads resolve the tension between one’s own self-presentation
them to act in compliance with social norms (Govern & and the other person’s oppositional view. Alternately, confir-
Marsch, 2001). As a result, the social self, an individual’s matory feedback – that which affirms and reinforces the
beliefs about the image they display to others (Leary, individual’s self-presentation – can strengthen the internal-
2001), is established and validated. In many contemporary ization of the stated characteristic, beyond the direct effect
mediated environments, the imagined audience is as of the self-statement. Confirmatory feedback fulfills the
important as an actual audience (Litt, 2012), making appar- intensification loop proffered by the hyperpersonal model
ent the complexity of conceptualizing publicness. as a validation of the individual’s selective self-presentation,
The importance of an audience to identity shift is evident reinforcing the presented traits and encouraging future
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

in its ties to the public commitment framework (Schlenker self-presentations of that desired trait to achieve self-
This document is copyrighted by the American Psychological Association or one of its allied publishers.

et al., 1994). An individual’s public self-awareness aug- presentational goals. In other words, if a facet of the self
ments the effect of self-presentation on identity formation. was seen as favorable once, the individual continues to
In a situation where an individual enacts a particular behav- present that facet in future interactions. The effects of this
ior, they tend to internalize that behavior more strongly feedback loop therefore go beyond mere commitment to a
when in the observed presence (vs. absence) of an audience public presentation. Consistent with both the hyperpersonal
due to the sense of obligation to the public presentation model and Bem (1972), others’ affirming reactions of an
(Kelly & Rodriguez, 2006; Tice, 1992). individual’s self statements provide public validation of that
The extent to which online publicness increases identity public self, which is then more intensely integrated not only
shift has been revealed by comparing different types of into their self but their subsequent self-presentations.
media platforms presenting public versus private settings
such as blogs versus text documents (e.g., Walther et al.,
2011), or varying the size of audience within a single med- Proposition 4: Self-Presentation Needs to
ium (e.g., Carr, 2021). Similarly, publicness has been
Be Connected to the Presenter’s Identity
pointed to as a pivotal component of other forms of online
behavior, including cyberbullying, such that negative for Identity Shift to Occur
impacts are intensified in public compared to semipublic Identity shift occurs when actors are self-identifiable: pri-
or private incidents (Pieschl et al., 2015; Sticca & Perren, vately self-aware the self-presentational messages they
2013). These additional research contexts extend our are creating are linked to their unmediated selves. Self-
understanding of the effects of online publicness, and can identifiability does not innately necessitate messages must
help focus on shifts in intrapersonal assessments resulting be identifiable to others (i.e., third parties can distinguish
from publicly made self-presentations. the source of the message), but rather focuses on the sen-
der being aware that a given message is tied to and reflects
Proposition 3: Identity Shift Is Stronger themselves versus being unattributable to them or attribu-
When Third-Party Feedback Confirms the ted to a disassociated pseudonym (see Anonymous,
1998). Although public self-awareness influences identity
Selective Self-Presentation
shift by moderating identity shift outcomes (see Proposition
Potentially tied to the publicness of self-presentation, 2), private self-awareness is a prerequisite in identity shift,
identity shift effects can be intensified when the individ- whereby self-presenters explicitly understand the self-pre-
ual’s self-presentation is validated by others. Although sentations that they are encoding and articulating are of
self-presentations alone are sufficient for individuals to themselves and linked to their unmediated self. Presenters,
internalize their espoused selves (Tice, 1992), human com- when making mediated self-statements, must allocate cog-
munication is often collaborative, eliciting some form of nitive effort to consider their own personal attitudes and
response from others. As public self-presentation strength- behaviors (Matheson & Zanna, 1988), and know these
ens the internalization of those traits (Proposition 2), it also self-statements are directly linked to themselves.
increases the likelihood of receiving feedback to that self- Experimentally, private self-awareness has been com-
presentation. monly manipulated by letting participants see and focus
Broadly, the feedback an individual receives from a third on themselves while engaging in a communicative task.
party can either disconfirm or confirm the selective self- Early studies achieved this by placing a mirror near the
presentation made by the individual. Disconfirmatory individual or affixing it to a computer monitor while they
feedback – that which rejects or contradicts the individual’s completed an experimental task (e.g., Froming et al.,

Journal of Media Psychology (2021), 33(4), 202–214 Ó 2021 Hogrefe Publishing

 C. T. Carr et al., Identity Shift Theory 209

1982). More recently, private self-awareness was manipu- tion, weakly held social identities (e.g., brand or sports team
lated in videoconferencing by embedding the participant’s identity) may be more mutable, and thus susceptible to
own video feed into the video chat (Joinson, 2001). Both identity shift effects (see Carr & Hayes, 2019). But as indi-
experimental inducements increased participants’ self- viduals likely cannot form an attachment with social cate-
identifiability and led to increased private self-awareness, gories that are disparate from their established and
but also reflect high ecological validity beyond the lab. Many closely held social identities (Brewer, 1999), an individual
contemporary mediated contexts provide multiple cues to cannot likely experience identity shift affecting their citizen-
link a sender’s identity to the message being conveyed, ship, educational attainment, vocation, or other social cate-
which engages the high private self-awareness needed to gory that is salient and central to their sense of self.
activate identity shift. For example, emails are composed Likewise, individuals engaging in outright deception, or pre-
and transmitted with the sender’s identifying email immedi- senting an attribute they know to be unfaithful to their self-
ately evident, social media systems display users’ names and concept to achieve strategic interpersonal goals rather than
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

profile pictures alongside the content they post, and articulate their authentic selves, will likely not experience
This document is copyrighted by the American Psychological Association or one of its allied publishers.

collaboration tools automatically provide metadata identify- self-transformation. Unlike presenting an aspirational or
ing which user contributed which content or comments. ought self (see Ellison et al., 2006), when individuals pre-
This identifiability is necessary to enact identity shift sent an entirely fabricated set of characteristics to persuade
processes. others, their focus is likely on the deceptive act rather than
on the self-presentational effort and their self-concept,
which in turn should lessen the cognitive effort put toward
Proposition 5: Identity Shift Is Activated self-reflection.
Relatedly, an individual’s motivation or desire to alter
When Presenters Perceive the Presented
their self-concept may influence the ability of an aspect of
Self as a Potential Self the self to be transformed, and should not directly affect
Presenters also need to view the self being perceived as a the individual’s physical properties or immutable traits.
viable version of themselves, so that they could see them- Studying fitness-sharing as a form of self-presentation,
selves – and perhaps even be motivated to – imbuing or Zhang (in press) found that sharing of exercise experiences
enacting the characteristics, behaviors, or attributes they on SNS (i.e., self-presentation) mediated the effect of fitness
put forth. Alternatively, individuals who selectively self- motivations on individuals’ use of fitness apps (i.e., a behav-
present a self they cannot identify with will not take on ioral outcome). In other words, deliberate self-presentation
the characteristics they have presented. The self being pre- only influenced self-perceptions when the individual
sented must be palatable and envisionable as a viable form actually desired to be fit, rather than just wanting to be
of themselves for individuals to experience identity shift. perceived as such (Rosenbaum & Johnson, 2015). And
Much extant identity shift research has utilized extraver- although an individual may pretend to be a rhino when
sion/introversion as a manipulated trait and operational- intoxicated (Knapp & Vangelisti, 2005, p. 155), no number
ized outcome. Although it is a superordinate and broadly of self-statements can likely convince the sender they are
stable personality trait within any individual (see Guilford, actually a 2-ton Rhinocerotidae. The congruence of selec-
1977), individuals can move within the extraverted– tive self-presentation and an individual’s own self-view
introverted continuum based on social contexts and needs allows for the perspective-taking and agency required for
(Robinson, 2009). Individuals can likely recall past experi- the cognitive assimilation of that self-presentation (see
ences and envision themselves in future circumstances in Tice, 1992). Contrarily, individuals asked to articulate or
which they clearly demonstrate both extraverted and demonstrate a characteristic they either cannot possess or
introverted characteristics. Individuals can recognize facets find incompatible with their self-concept may not experi-
of their self-concepts at both endpoints of the extraversion/ ence any transformation, finding such presentations incom-
introversion continuum, and thus identify and articulate mensurate with their self-view.
facets of themselves when selectively self-presenting as
either. However, other self-concepts may be too extreme,
polarized, or unrealistic to be realized. Just as Midwestern Moving Identity Shift Forward
individuals may have been stymied presenting themselves
as greedy (a fundamental departure from ingrained cultural These five propositions articulate IST. Embedded in IST is
mores; see Carr, 2019), individuals may not experience the construct of identity shift, which is the actual process of
identity shift when selectively self-presenting qualities they self-transformation. That construct is situated in a process
view as incompatible with their own social and personal whereby deliberate mediated self-presentations result in
identities (e.g., religion; see Ysseldyk et al., 2010). In addi- that transformation, accounting for several theoretically -

Ó 2021 Hogrefe Publishing Journal of Media Psychology (2021), 33(4), 202–214

 210 C. T. Carr et al., Identity Shift Theory

guided antecedents, moderators, and caveats. Consolidat- der Heide, Schumaker, et al., 2013) suggest future research
ing what is known about identity shift processes to date, should do more to account for modality in identity shift.
IST explains the relationships among its constructs and
provides predictive power to inform future research. Build-
ing on the existing constructs and propositions of IST, we Modality of Self-Presentation
recognize the need for further research, particularly in five
Most of the work supporting identity shift has used text-
areas.
based presentations, lending to the presumption the visibil-
ity of composing, editing, and the final self-presentation
Distinguishing Identity Shift activates private self-awareness and draws cognitive focus
to the presented traits (Bem, 1972). Studies using pictorial
Applying and testing IST requires distinguishing identity (Gonzales & Kim, 2014; Hampton et al., 2020) and verbal
shift from other processes, such as public commitment (Peña & Hill, 2020) self-presentations have failed to sup-
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

and the Proteus effect. Identity shift differs from public

This document is copyrighted by the American Psychological Association or one of its allied publishers.

port identity shift hypotheses, potentially due to the lack

commitment (Schlenker et al., 1994) as the latter gives pri- of time or effort individuals spent composing and reflecting
macy to public self-awareness and is not contingent on on self-presentation. And yet, individuals can spend sub-
affordances of various media. Identity shift requires self- stantive time selectively self-presenting online via disen-
presentation enabled by disentrained channels, heightening trained visual or audio channels, taking time to retake
private self-awareness and affording communicators delib- and edit self-portraits, videos, or audio files (e.g., Fox &
erate construction of and reflection on their self-presenta- Vendemia, 2016; Sun, 2009). An important question for
tions. An innately mediated process, identity shift is future work is therefore what aspects of modality are
evident in computer-mediated communication, but also important for identity shift.
manifests in other mediated contexts where individuals
strategically reflect on and present specific facets of them-
selves. For example, identity shift effects could be expected Deliberativeness of Self-Presentational
for workers posting daily affirmation on a pen-and-paper Efforts
bulletin board. Likewise, individuals using a virtual thera-
Perhaps related to the modality of self-presentation is the
pist (i.e., artificial intelligence agent; Xu et al., 2018) may
effort – both temporally and cognitively – involved in crafting
experience identity shift commensurate with their self-
a self-presentation. Some work already suggests deliberate-
statements, as effects can manifest even when individuals
ness may affect identity shift processes: Lane et al. (2019)
self-disclose to nonhuman digital others (Walther et al.,
found that self-statements – but not less-effortful behaviors,
2011).
including liking and sharing others’ political posts – shaped
Another similar process is the Proteus effect: a process
individuals’ later political self-concepts. Low-effort, one-
whereby virtual self-representations (avatars) alter attitudes
click behaviors (i.e., Likes, retweets, shares) have been
and behaviors outside of the virtual space (Yee & Bailenson,
shown to be paradoxical: sometimes used almost thought-
2007). It could be argued the Proteus effect is a virtual real-
lessly to bookmark information or acknowledge viewing
ity-specific instance of IST, yet there are key differences in
social media content, and other times used as thoughtful
specifications of these processes. First, Proteus effects show
and deliberate forms of self-presentation (Hayes et al.,
consistent change in attitudes and behaviors in line with
2016). It may be that the communicative act itself is less crit-
prior self-perception (Ratan et al., 2020). IST can involve
ical to IST than the self-presentational intent and effort.
attitude and behavior change, but is anchored in a changing
self-concept. Second, change in a Proteus effect is attribu-
ted to self-perception of one’s virtual embodiment. Self-
Desire and Ability to Change
observation in an altered environment leads to altered
self-perception. While the customization and modification Research has noted that some traits, such as altruism and
possible in virtual environments allow for an altered self, fashionability, may be resistant to identity shift, but have
the Proteus effect – as theorized and tested – emphasizes only been able to provide post hoc rationalizations as to
the transformative power of virtual embodiment, rather why. Future work may seek to consider two related underly-
than selective self-presentation. Virtual environments often ing mechanisms to identity shift that may account for the
allow for interactions and audiences, yet these are not fac- boundary conditions facing these traits, specifically: one’s
tors implicated in the Proteus effect, underlining again how desire to change and one’s self-efficacy regarding change.
it lacks the self-presentational elements of IST. Whereas An individual’s desire to change may moderate the transfor-
self-representation is reflexive; self-presentation is social. mative effect of deliberate self-presentation. For example,
However, comparisons with the Proteus effect (e.g., Van exploring online dating profiles, Ellison et al. (2006) noted

Journal of Media Psychology (2021), 33(4), 202–214 Ó 2021 Hogrefe Publishing

 C. T. Carr et al., Identity Shift Theory 211

that many individuals presented their ideal self – the self they would automatically delete after 24 hr – resulted in identity
desired to manifest – in their dating profiles, such as under- shifts consistent with the presented introverted/extraverted
stating their weight or presenting a favored activity they in trait, although not self-presentations made via persistent
which they do not actually engage. In both cases, respon- messages. This raises questions about the importance of
dents professed their intention to be able to change and man- message persistence for IST and points to a need for future
ifest that future self (e.g., by exercising, taking up that hobby) research in this area.
by the time they met their suitor face to face. When moti-
vated or desiring to change, individuals may be more readily
able to visualize their transformed self, and thus be more sus-
How New Technologies May Affect
ceptible to the self-effects of their messages. Identity Shift
Beyond desires and intentions, the general malleability Finally, several new AI-powered technologies can alter real-
of one’s self-concept and thus ability to change may like- time video streams to allow speakers to see versions of
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

wise moderate the transformative effect of selective self- themselves acting in new ways, such as excelling at speaking
This document is copyrighted by the American Psychological Association or one of its allied publishers.

presentation. It is possible self-esteem plays a role here. in a confident manner or telling a funny joke. These tech-
Campbell (1990) found individuals with “lower self-esteem nologies, often referred to as deepfakes, employ AI to modify
have less clearly defined concepts of who or what they are behavior in video and raise interesting implications for IST
in terms of their personality attributes” (p. 544), suggesting (Hancock & Bailenson, 2021). What happens to a person’s
the attributes of less-secure individuals may be more mal- sense of self when their behavior is modified without any
leable, and thus experience greater shift. From this perspec- effort or thought on their behalf? Will self-perception
tive, individuals with high self-esteem, either as a general dynamics drive identity shift, or will some level of effort
attribute or with respect to the given trait, may find their be required to observe identity shift? As AI is deployed to
self-concepts more rigid. Individuals with less flexibility in accomplish interpersonal goals, from coming across as more
their self-concept have greater stability in their self-concept. funny or warm or dominant (Hancock et al., 2020), an
As a result, they would likely experience cognitive disso- important question is whether AI can activate identity shift.
nance when presenting unfamiliar traits, causing them to If so, the almost infinite scale at which AI can modify human
resolve the dissonance by disregarding the unfamiliar trait behavior highlights the urgency and importance of under-
that was presented to remain congruous with their prior standing identity shift, its limits, and its potential.
self-concept. Both mechanisms suggest individuals who
want to change or are more susceptible to change should
be more likely to experience identity shift and will experi-
ence greater self-transformation. These ideas are also
Conclusion
evoked in the literature on self-theories and identity nego-
tiation; but it is important to note individuals are often dri- Strategic self-presentation is a common goal online, as indi-
ven by the desire to be consistent (vs. change) with initial viduals seek to present a public persona on web pages,
self-perceptions and seek out interaction opportunities social media, dating services, and organizational websites.
likely to result in confirmatory feedback. Depending on Identity shift theory affords utility to predict, control, and
which of the competing desires (self-verification vs. self- explain the self-effects of those self-presentational efforts.
enhancement) is pursued, different predictions will be Synthesizing what is known about this phenomenon and
made. Future work should thus consider both the mecha- advancing IST, this article clarifies the construct of identity
nisms behind the self-transformation and the aspects of shift and provides a means for scholars to consider the self-
the self-motives that that are susceptible or resistant to transformation of individuals’ self-concepts stemming from
identity shift. mediated self-statements. Identity shift theory offers a
parsimonious tool to theorize the self-effects of mediated
Impact of the Temporality of self-presentation.
Identity Performance
Time matters for identity shift. One aspect of identity perfor- References
mance in contemporary social media environments to con-
sider is the fleeting nature of many of these performances Anonymous. (1998). To reveal or not to reveal: A theoretical model
(e.g., stories on Snapchat disappearing after 24 hr). It is of anonymous communication. Communication Theory, 8(4),
381–407. https://doi.org/10.1111/j.1468-2885.1998.tb00226.x
possible the temporality of the self-statement impacts the
Bem, D. J. (1972). Self-perception theory. In L. Berkowitz (Ed.),
self-transformation. Choi et al. (in press) found that ephem- Advances in Experimental Social Psychology (Vol. 6, pp. 1–62).
eral self-presentations – those made via messages the system Elsevier.

Ó 2021 Hogrefe Publishing Journal of Media Psychology (2021), 33(4), 202–214

 212 C. T. Carr et al., Identity Shift Theory

Brewer, M. B. (1999). The psychology of prejudice: Ingroup love or Gonzales, A.L., & Petro, G. (2020). Adding insult to algorithm: How
outgroup hate? Journal of Social Issues, 55(3), 429–444. unfavorable behavioral advertising impacts self-concept
https://doi.org/10.1111/0022-4537.00126 (Unpublished manuscript).
Campbell, J. D. (1990). Self-esteem and clarity of the self- Govern, J. M., & Marsch, L. A. (2001). Development and validation of
concept. Journal of Personality and Social Psychology, 59(3), the situational self-awareness scale. Consciousness and Cogni-
538–549. https://doi.org/10.1037/0022-3514.59.3.538 tion, 10(3), 366–378. https://doi.org/10.1006/ccog.2001.0506
Carr, C. T. (2021). Identity shift effects of personalization of self- Guilford, J. P. (1977). Will the real factor of extraversion-introver-
presentation on extraversion. Media Psychology, 24(4), 490– sion please stand up? A reply to Eysenck. Psychological Bulletin,
508. https://doi.org/10.1080/15213269.2020.1753540 84(3), 412–416. https://doi.org/10.1037/0033-2909.84.3.412
Carr, C. T. (2019, May). Identity shift effects of selective self- Hampton, N. D., Valov, J. J., Toutia, H., Engstrom, S. K., Mason,
presentation on self-perception and subsequent behaviors. E. J., Le, H., Green, A. M., Crumbaugh, A., & Carr, C. T. (2020,
Paper presented at the annual meeting of the International November). Network diversity and trait desirability have no
Communication Association Conference, Washington, DC. effect on identity shift: We R who we R. Paper presented at the
Carr, C. T., & Foreman, A. C. (2016). Identity shift III: Effects of annual meeting of the National Communication Association,
publicness of feedback and relational closeness in computer- Virtual Conference.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

mediated communication. Media Psychology, 19(2), 334–358. Hancock, J. T., & Bailenson, J. N. (2021). The social impact of
This document is copyrighted by the American Psychological Association or one of its allied publishers.

https://doi.org/10.1080/15213269.2015.1049276 deepfakes. Cyberpsychology, Behavior, and Social Networking,

Carr, C. T., & Hayes, R. A. (2019). Identity shift effects of self- 24(3), 149–152. https://doi.org/10.1089/cyber.2021.29208.jth
presentation and confirmatory & disconfirmatory feedback on Hancock, J. T., Naaman, M., & Levy, K. (2020). AI-mediated
self-perceptions of brand identification. Media Psychology, 22(3), communication: Definition, research agenda, and ethical con-
418–444. https://doi.org/10.1080/15213269.2017.1396228 siderations. Journal of Computer-Mediated Communication,
Carr, C. T., Wohn, D. Y., & Hayes, R. A. (2016). as social support: 25(1), 89–100. https://doi.org/10.1093/jcmc/zmz022
Relational closeness, automaticity, and interpreting social Hayes, R. A., Carr, C. T., & Wohn, D. Y. (2016). One click, many
support from paralinguistic digital affordances in social media. meanings: Interpreting paralinguistic digital affordances in
Computers in Human Behavior, 62, 385–393. https://doi.org/ social media. Journal of Broadcasting & Electronic Media, 60(1),
10.1016/j.chb.2016.03.087 171–187. https://doi.org/10.1080/08838151.2015.1127248
Choi, S., Williams, D., & Kim, H. (in press). A snap of your true self: Joinson, A. N. (2001). Self-disclosure in computer-mediated
How self-presentation and temporal affordance influence self- communication: The role of self-awareness and visual anon-
concept on social media. New Media & Society. https://doi.org/ ymity. European Journal of Social Psychology, 31(2), 177–192.
10.1177/444820977199 https://doi.org/10.1002/ejsp.36
Cingel, D. P., Wartella, E., & Krcmar, M. (2014). The role of Johnson, B. K., & Van der Heide, B. (2015). Can sharing affect
adolescent development in social networking site use: Theory liking? Online taste performances, feedback, and subsequent
and evidence. Journal of Youth Development, 9(1), 29–40. media preferences. Computers in Human Behavior, 46, 181–
https://doi.org/10.5195/jyd.2014.70 190. https://doi.org/10.1016/j.chb.2015.01.018
Cooley, C. H. (1902). Human nature and the social order. Scribner. Kelly, A. E., & Rodriguez, R. R. (2006). Publicly committing oneself
Ellison, N., Heino, R., & Gibbs, J. (2006). Managing impressions to an identity. Basic and Applied Social Psychology, 28(2), 185–
online: Self-presentation processes in the online dating envi- 191. https://doi.org/10.1207/ s15324834basp2802_8
ronment. Journal of Computer-Mediated Communication, 11(2), Kim, Y. W., & Gonzales, A. L. (2018). You don’t know me: Negative
415–441. https://doi.org/10.1111/j.1083-6101.2006.00020.x self-views interact with publicness and feedback to shape
Ellison, N. B., Vitak, J., Gray, R., & Lampe, C. (2014). Cultivating interpersonal impressions online. Computers in Human Behav-
social resources on social network sites: Facebook relationship ior, 88, 195–204. https://doi.org/10.1016/j.chb.2018.06.025
maintenance behaviors and their role in social capital pro- Knapp, M. L., & Vangelisti, A. L. (2005). Relationship stages: A
cesses. Journal of Computer-Mediated Communication, 19(4), communication perspective. In M. L. Knapp & A. L. Vangelisti
855–870. https://doi.org/10.1111/jcc4.12078 (Eds.), Interpersonal communication and human relationships
Fox, J., & Vendemia, M. A. (2016). Selective self-presentation and (pp. 151–158). Allyn & Bacon.
social comparison through photographs on social networking Lane, D. S., Lee, S. S., Liang, F., Kim, D. H., Shen, L., Weeks, B. E.,
sites. Cyberpsychology, Behavior, and Social Networking, 19(10), & Kwak, N. (2019). Social media expression and the political
593–600. https://doi.org/10.1089/cyber.2016.0248 self. Journal of Communication, 69(1), 49–72. https://doi.org/
French, M. R. (2018). Algorithmic mirrors: An examination of how 10.1093/joc/jqy064
personalized recommendations can shape self-perceptions and Leary, M. R. (2001). The self we know the self we show: Self-
reinforce gender stereotypes (Unpublished doctoral disserta- esteem, self-presentation, and the maintenance of interper-
tion). Stanford University. sonal relationships. In G. J. O. Fletcher & M. S. Clark (Eds.),
French, M., & Hancock, J. T. (2016, June). Addressing the audience: Blackwell handbook of social psychology: Interpersonal pro-
Effects of evaluation on identity shift in computer-mediated cesses (pp. 457–477). Blackwell.
environments. Paper presented at the annual meeting of the Levine, T. R., Asada, K. J., & Carpenter, C. (2009). Sample sizes
International Communication Association, Fukuoka, Japan. and effect sizes are negatively correlated in meta-analyses:
Froming, W. J., Walker, G. R., & Lopyan, K. J. (1982). Public and Evidence and implications of a publication bias against non-
private self-awareness: When personal attitudes conflict with significant findings. Communication Monographs, 76(3), 286–
societal expectations. Journal of Experimental Social Psychology, 302. https://doi.org/10.1080/03637750903074685
18(5), 476–487. https://doi.org/10.1016/0022-1031(82)90067-1 Litt, E. (2012). Knock, knock. Who’s there? The imagined audience.
Gonzales, A. L., & Hancock, J. T. (2008). Identity shift in computer- Journal of Broadcasting & Electronic Media, 56(3), 330–345.
mediated environments. Media Psychology, 11(2), 167–185. https://doi.org/10.1080/08838151.2012.705195
https://doi.org/10.1080/15213260802023433 Marwick, A. E., & boyd, d. (2011). I tweet honestly, I tweet
Gonzales, A. L., & Kim, M. (2014). Me, my-selfie, and I: Identity passionately: Twitter users, context collapse, and the imagined
shift through selfie editing and naturalistic feedback (Unpub- audience. New Media & Society, 13(1), 114–133. https://doi.
lished manuscript). org/10.1177/1461444810365313

Journal of Media Psychology (2021), 33(4), 202–214 Ó 2021 Hogrefe Publishing

 C. T. Carr et al., Identity Shift Theory 213

Matheson, K., & Zanna, M. P. (1988). The impact of computer- Van der Heide, B., Schumaker, E. M., Peterson, A. M., & Jones,
mediated communication on self-awareness. Computers in E. B. (2013). The Proteus Effect in dyadic communication:
Human Behavior, 4(3), 221–233. https://doi.org/10.1016/0747- Examining the effect of avatar appearance in computer-
5632(88)90015-5 mediated dyadic interaction. Communication Research, 40(6),
Orive, R. (1984). Group similarity, public self-awareness, and 838–860. https://doi.org/10.1177/0093650212438097
opinion extremity: A social projection explanation of deindivid- Walther, J. B. (1996). Computer-mediated communication: Imper-
uation effects. Journal of Personality and Social Psychology, sonal, interpersonal, and hyperpersonal interaction. Communi-
47(4), 727–737. https://doi.org/10.1037/0022-3514.47.4.727 cation Research, 23(3), 3–43. https://doi.org/10.1177/
Peña, J., & Hill, D. (2020). Examining identity shift effects in virtual 009365096023001001
reality. Cyberpsychology, Behavior, and Social Networking, Walther, J. B. (2007). Selective self-presentation in computer-
23(10), 697–701. https://doi.org/10.1089/cyber.2020.0010 mediated communication: Hyperpersonal dimensions of tech-
Pieschl, S., Kuhlmann, C., & Porsch, T. (2015). Beware of publicity! nology, language, and cognition. Computers in Human Behavior,
Perceived distress of negative cyber incidents and implications 23(5), 2538–2557. https://doi.org/10.1016/j.chb.2006.05.002
for defining cyberbullying. Journal of School Violence, 14(1), Walther, J. B., Liang, Y. J., DeAndrea, D. C., Tong, S. T., Carr, C. T.,
111–132. https://doi.org/10.1080/15388220.2014.971363 Spottswood, E. L., & Amichai-Hamburger, Y. (2011). The effect
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Ratan, R., Beyea, D., Li, B. J., & Graciano, L. (2020). Avatar of feedback on identity shift in computer-mediated communi-
This document is copyrighted by the American Psychological Association or one of its allied publishers.

characteristics induce users’ behavioral conformity with small- cation. Media Psychology, 14(1), 1–26. https://doi.org/10.1080/
to-medium effect sizes: A meta-analysis of the Proteus effect. 15213269.2010.547832
Media Psychology, 23(5), 651–675. https://doi.org/10.1080/ Waterloo, S. F. (2017). Identity expression and self-concept change
15213269.2019.1623698 online: The impact of publicness and customization (Unpub-
Reeves, B., & Nass, C. (1996). The media equation: How people lished doctoral dissertation). University of Amsterdam.
treat computers, television, and new media like real people and Xu, J., De’Aira, G. B., & Howard, A. (2018, August). Would you trust
places. Cambridge University Press. a robot therapist? Validating the equivalency of trust in human-
Robinson, O. C. (2009). On the social malleability of traits: robot healthcare scenarios. Paper presented at the 27th IEEE
Variability and consistency in Big 5 trait expression across International Symposium on Robot and Human Interactive
three interpersonal contexts. Journal of Individual Differences, Communication (RO-MAN), Nanjing, China.
30(4), 201–208. https://doi.org/10.1027/1614-0001.30.4.201 Yee, N., & Bailenson, J. (2007). The Proteus effect: The effect of
Rosenbaum, J. E., & Johnson, B. K. (2015). Picture this: Using transformed self-representation on behavior. Human Commu-
social media to increase health and nutrition awareness nication Research, 33(3), 271–290. https://doi.org/10.1111/
(Unpublished manuscript). j.1468-2958.2007.00299.x
Rosenbaum, J. E., & Johnson, B. K. (2019, May). Self-persuasion Ysseldyk, R., Matheson, K., & Anisman, H. (2010). Religiosity as
through sharing, liking, and retweeting: Identity shift on social identity: Toward an understanding of religion from a social
platforms. Paper presented at the annual meeting of the identity perspective. Personality and Social Psychology Review,
International Communication Association, Washington, DC. 14(1), 60–71. https://doi.org/10.1177/1088868309349693
Schlenker, B. R., Dlugolecki, D. W., & Doherty, K. (1994). The impact Zhang, X. (in press). Keeping up appearances: Testing a moder-
of self-presentations on self-appraisals and behavior: The power ated mediation path of self-presentation motives, self-efficacy
of public commitment. Personality and Social Psychology Bul- beliefs, social sharing of fitness records and fitness app uses.
letin, 20(1), 20–33. https://doi.org/10.1177/0146167294201002 Behaviour & Information Technology. https://doi.org/10.1080/
Sticca, F., & Perren, S. (2013). Is cyberbullying worse than 0144929X.2020.1829709
traditional bullying? Examining the differential roles of medium,
publicity, and anonymity for the perceived severity of bullying.
History
Journal of Youth and Adolescence, 42, 739–750. https://doi.org/
Received April 30, 2020
10.1007/s10964-012-9867-3
Revision received June 15, 2021
Sun, Y.-C. (2009). Voice blog: An exploratory study of language
Accepted June 29, 2021
learning. Language Learning & Technology, 13(2), 88–103.
Published online November 18, 2021
https://doi.org/10125/44182
Swann, W. B. Jr., & Read, S. J. (1981). Self-verification processes:
ORCID
How we sustain our self-conceptions. Journal of Experimental
Caleb T. Carr
Social Psychology, 17(4), 351–372. https://doi.org/10.1016/
https://orcid.org/0000-0002-5700-6636
0022-1031(81)90043-3
Tice, D. M. (1992). Self-concept change and self-presentation: The
Judith E. Rosenbaum
looking glass self is also a magnifying glass. Journal of
https://orcid.org/0000-0003-1295-5325
Personality and Social Psychology, 63(3), 435–451. https://doi.
org/10.1037/0022-3514.63.3.435
Benjamin K. Johnson
Valkenburg, P. M. (2017). Understanding self-effects in social
https://orcid.org/0000-0003-0588-221X
media. Human Communication Research, 43(4), 477–490.
https://doi.org/10.1111/hcre.12113
Caleb T. Carr
Valkenburg, P. M., Peter, J., & Walther, J. B. (2016). Media effects:
School of Communication
Theory and research. Annual Review of Psychology, 67, 315–
Illinois State University
338. https://doi.org/10.1146/annurev-psych-122414-033608
453 Fell Hall, Campus Box 4480
Van der Heide, B., Dickinson, T., Schumaker, E. M., & DeAndrea,
Normal, IL 61790
D. C. (2013, June). Explaining online self-influence: Exploring
USA
the effects of computer-mediation on attitude change. Paper
[email protected]
presented at 63rd Annual Conference of the International
Communication Association, London, UK.

Ó 2021 Hogrefe Publishing Journal of Media Psychology (2021), 33(4), 202–214

 214 C. T. Carr et al., Identity Shift Theory

Caleb T. Carr (PhD, Michigan State Benjamin K. Johnson (PhD, The Ohio
University) researches how com- State University) is an assistant pro-
puter-mediated communication af- fessor of advertising at the University
fects communicative processes, of Florida, USA. His research focuses
including how social media are used on why and how people select and
for organizational uncertainty reduc- share messages in new media set-
tion and to create and maintain tings, especially as it relates to psy-
identity online. He is a Professor of chological processes including
Communication at Illinois State impression management, social
University’s School of Communica- comparison, and self-regulation.
tion, USA.

Jeffrey T. Hancock (PhD, Dalhousie

Yeweon Kim (MA, Seoul National University) researches the psycho-
University) is a PhD candidate in the logical and interpersonal processes
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Department of Communication at the in social media, specializing in using

This document is copyrighted by the American Psychological Association or one of its allied publishers.

University of California, Santa Bar- computational linguistics and exper-

bara, CA, USA. Her research investi- iments to understand how the words
gates the benefits and risks of we use can reveal psychological and
information communication tech- social dynamics. He is a professor in
nologies for people and society and the Department of Communication at
uncovers the mechanisms underlying Stanford University, CA, USA.
these technologies’ ambivalent func-
tions.
Amy L. Gonzales (PhD, Cornell
University) examines the effects of
Jacob J. Valov (MA, Illinois State social interaction via communication
University) studies digital rhetoric technologies on individual identity,
and computer-mediated communi- social support, and well-being and
cation, focusing on the construction the consequences of disrupted ac-
of group identities and individual cess to communication technology.
self-presentations within online She is an Associate Professor in the
spaces, namely, social media. He is a Department of Communication at the
PhD student in the Department of University of California – Santa Bar-
Communication Arts at the University bara.
of Wisconsin – Madison, USA.

Judith E. Rosenbaum (PhD, Radboud

University Nijmegen) conducts re-
search focused on understanding
how people actively use media con-
tent in their daily lives and how this
usage is related to their identities,
relationships, and perceptions of
reality. She is Associate Professor of
Media Studies and Chair of the
Department of Communication and
Journalism at the University of
Maine, USA.

Journal of Media Psychology (2021), 33(4), 202–214 Ó 2021 Hogrefe Publishing