ChapterPDF Available
Chapter 16
Self and Identity
From the beginning, psychology s relationship with the
self has been a tempestuous one. When, for example,
William James (1890/1950) marched the self to psychology s
center stage in his classic text, the field promptly ushered
it to the wings. There it languished for more than half a
century, ignored by a psychological mainstream whose
embrace of positivism made it squeamish about constructs
that seemed to lack clear empirical referents (e.g., Allport,
1943). And when the self finally did gain admission into
the social psychological mainstream in the 1960s, it had
been stripped of some crucial features of the construct
that James introduced. Whereas James saw the self as a
source of continuity that gave the individual a sense of
“ connectedness ” and “ unbrokenness ” (p. 335), the 1960s
were dominated by an ephemeral, shape - shifting self that
routinely reinvented itself in the service of winning social
approval (e.g., Scheibe, 1985).
Happily, over the last few decades, conceptualiza-
tions of the self have reclaimed much of the richness and
integrity with which James (1890/1950) first imbued the
construct. Moreover, contemporary social - personality psy-
chologists have warmly embraced these emerging, neo -
Jamesian visions of the self: Between 1972 and 2002,
the percentage of self - related studies published in the
Journal of Personality and Social Psychology increased
fivefold (Swann & Seyle, 2005). The rejuvenated image
of the self is multidimensional. Most researchers now
assume that the self has a rich history, some of which is
conscious and accessible through self - reports and some
of which is presumably nonconscious and accessible pri-
marily through indirect measures. Although a strong belief
still exists in the prepotency of a desire to win approval
from others, most theorists acknowledge the significance
of rival motivational forces, particularly in non - Western
cultural settings (e.g., Banaji & Prentice, 1994). And
modern researchers have complemented their long - stand-
ing interest in personal self - views or identities (we use
these terms interchangeably) with investigations of social
identities. It was this growing interest in social identity
that prompted us to cover this work and title the chapter
Self and Identity instead of simply The Self, the title
of Baumeister s (1998) earlier contribution to this volume.
Before turning to the specific substantive issues that
we cover here, we place our analysis in historical context.
In particular, we briefly describe the chain of events that led
to the legitimization of a multifaceted, enduring conception
of the self.
Psychology s failure to follow up on James s (1890/1950)
initial investigation of the self left a void that scholars
from other fields quickly stepped in to fill. Two of the most
prominent such scholars, the sociologists Charles Horton
Cooley and George Herbert Mead, rallied behind the banner
of a theoretical perspective known as symbolic interaction-
ism (Cooley, 1902; Mead, 1934). This theory was designed
to illuminate the nature and origins of self - knowledge,
especially the reactions of others and the roles people play.
We know ourselves, the theory assumed, by observing
how we fit into the fabric of social relationships and how
others react to us. In its emphasis on the social construc-
tion of the self, symbolic interactionism zeroed in on the
aspect of self that James dubbed the social self and about
which he famously noted that a man has as many social
selves as there are individuals who recognize him and
carry an image of him in their mind (1890/1950, p. 294).
We are grateful to Jennifer Beer, Matt Brooks, Serena Chen, Susan Fiske, Dan Gilbert, Josh Hart, Steven Heine, Stan Klein, Tracy
Kwang, Andrea Lindzey, Thomas Pyszczynski, and Morgan Ward, for their comments on earlier versions of this chapter.
CH16.indd 589CH16.indd 589 10/22/09 2:50:02 PM10/22/09 2:50:02 PM
590 Self and Identity
Conspicuously absent from these accounts were the other,
more enduring aspects of the self that figured prominently in
James ’ s account, notably the “ empirical self, ” which includes
the physical self, and the spiritual self, which consists of
beliefs about one s qualities. As symbolic interaction-
ism assumed center stage in the scientific community s
emerging understanding of the nature of the self, James s
relatively enduring forms of self - knowledge faded into
Several decades later, the dominance of the social self
was augmented by one of symbolic interactionism s most
prominent intellectual progeny, the dramaturgical move-
ment. Spearheaded by Goffman (1959), this movement
assumed that people are like actors in a play who perform
for different audiences. As people take on various identi-
ties, the self is merely a consequence, rather than a cause,
of the performance, a product of the scene that comes off
(p. 252). Once people lay claim to an identity, they are obli-
gated to remain in character until they move to the next
scene, at which point the former self is discarded in favor
of a self that fits the new context. For Goffman, there was
no enduring sense of self; instead, Goffman envisioned the
self as an ahistorical construction that emerged and van-
ished at the whim of the situational cues that regulated its
form and structure.
When mainstream social psychologists developed an
interest in the systematic study of the self in the 1960s,
they looked to sociology for a promising paradigm. They
were smitten with Goffman s (1959) newly minted vision
of self and identity. Goffman s influence is most obvious
in accounts of impression management (Jones, 1964),
accounts that were later embellished by Edward Jones s
students (e.g., Roy Baumeister and Fred Rhodewalt), as
well as others (Schlenker, 1980; Tedeschi, 1981). These
theorists proved to be extremely influential in shaping
early social psychological views of the self. But Goffman s
vision of the self had broader impacts as well. First, if any-
one could assume any identity that the situation demanded,
then people were essentially interchangeable. This senti-
ment helped legitimize a situationist approach to the self
and identity. Second, the theatre metaphor that Goffman
used to exemplify social interaction led researchers to focus
narrowly on a single goal: gaining the approval of the audi-
ence (i.e., other people). From this vantage point, people
were presumably in the business of constructing whichever
identities they believed would help them win the favor of
their interaction partners, with the only proviso being that
they should strive to prevent observers from viewing them
as inconsistent or dishonest (e.g., Schlenker, 1980, 1985).
Nowhere in this scheme was there an intrinsic need to rec-
oncile the presented self with an enduring, underlying, or
authentic sense of self. For social psychologists of the day,
the world was, as Daniel Webster put it, governed more
by appearances than by reality (D. Webster, F. Webster,
Sanborn, 1857, p. 146).
Even when researchers became interested in motives that
seemed superficially incompatible with approval seeking
or “ self - enhancement, ” these motives were not informed
by an enduring sense of self. For example, when research-
ers began to examine self - consistency, they typically left
the enduring self out of the equation. Dissonance research-
ers, for instance, would subtly persuade participants to
behave in ways that made them look more or less deficient
and then observe their subsequent efforts to save face (e.g.,
Aronson, 1968). Again, social actors were presumed to be
interchangeable. Consequently, researchers had no need
to consider how an enduring sense of self might influence
people s reactions to the situations in which they found
It was not until the 1970s that the paradigm began to
shift and the enduring sense of self began to gain currency
within mainstream social psychology. Snyder (1974)
developed a personality measure (the self - monitoring
scale) that distinguished people who were thought to be
perpetually engaged in Goffman - esque impression man-
agement activities from those whose actions were guided
by a deep - seated, enduring sense of self that valued cross -
situational consistency. In a somewhat parallel effort that
drew on developments in cognitive psychology, Markus
(1977) introduced the idea that some people possessed
enduring “ self - schemas ” that systematically guided infor-
mation processing about the self. Shortly afterwards,
Kuiper and Rogers (1979) provided evidence that people
store representations of the self in memory and that these
mental representations facilitate the retrieval of self -
relevant information.
By 1980, the stage had been set for a wide - ranging
examination of the nature and consequences of a multi-
faceted self that featured enduring, as well as relatively
fleeting, components (Markus & Wurf, 1987; Swann,
1983). No longer were social psychologists conceptualiza-
tions of the self hitched to the wagon of pretense stubbornly
intent on self - enhancement. Increasingly, researchers were
abandoning the stage - acting metaphor of the self and the
superficial relationships it illuminated and instead turning
their attention to the relatively stable identities that people
negotiated in their ongoing social relationships. This is not
to say that all prominent social psychologists followed this
trend. But even the few who continued to emphasize the
ephemeral self over the enduring self updated and refined
their analyses considerably (e.g., Gergen, 1991). And when
more mainstream self theorists began to acknowledge
people s stable identities, they quickly came to embrace the
richness and complexity of the multifaceted, neo - Jamesian
CH16.indd 590CH16.indd 590 10/22/09 2:50:03 PM10/22/09 2:50:03 PM
Self as a Mental Representation 591
conception of self. In the following section, we begin to
examine the fruits of these efforts by turning to work that
conceptualizes the self as a mental representation.
Although we believe that psychology s love affair with
logical positivism explains most of its historical ambiva-
lence toward the self, an additional problem has been
that the term self has been used in multiple, sometimes
contradictory ways (Leary, 2004). In its most common
usage, the self refers to a representation or set of represen-
tations about oneself, parallel to the representations people
have of other individuals. This is the most straightforward
and common usage of the term and the one on which we
focus most of our attention in this chapter. It is the me, or
self - as - object, about which James (1890/1950) wrote — the
entire set of beliefs, evaluations, perceptions, and thoughts
that people have about themselves.
Nevertheless, the term self has also been substituted
for “ behavior, ” as in “ self - regulation. ” Our review does not
focus on work exemplifying the latter usage, partly because
this work was covered comprehensively in Baumeister s
(1998) chapter. In addition, however, we are concerned that
if the boundary conditions of the subarea self are relaxed
to encompass all research that involves behavior, then vir-
tually any activity can be incorporated within the domain
of self - psychology simply by prefixing it with “ self - . ” For
these reasons, our review focuses on work that directly or
indirectly involves the represented self.
Types of Self - Representations
In what follows, we identify and define several important
distinctions that underlie people s mental representations
of self. Although not exhaustive, this list is intended to
capture most major forms that self - views (self - concepts
and self - esteem) assume.
Active Versus Stored Self - Knowledge
The amount of self - knowledge — beliefs, thoughts, mem-
ories, and feelings about the self that people possess is
theoretically unlimited in quantity and scope. As such, it
cannot all be brought to attention at once. Beginning in the
late 1960s, researchers began to acknowledge this fact by
differentiating between active and stored self - knowledge.
Active self - knowledge includes information about oneself
that is held in consciousness. It has been referred to as the
phenomenal self (Jones & Gerard, 1967), the spontaneous
self - concept (McGuire, McGuire, Child, & Fujioka, 1978),
and the working self - concept (Markus & Kunda, 1986).
In contrast, stored self - knowledge includes information about
the self that is held in memory but is not being attended to.
Thus, whereas most self - knowledge is at least theoretically
accessible to conscious awareness, only information in the
working self - concept is available for immediate reflection.
The working self - concept is highly responsive to con-
text, such that people are particularly likely to bring to mind
aspects of the self that stand out or differentiate them from
others (e.g., McGuire et al., 1978). One consequence of
this malleability is that self - knowledge can shift somewhat
easily to fit the demands of the current situation, without
eliciting troubling feelings of inconsistency or inauthentic-
ity (e.g., Swann, Bosson, & Pelham, 2002). We have more
to say about this later.
Semantic Versus Episodic Self - Knowledge
Based on Tulving s (1983) distinction between two types of
declarative memory, Klein and Loftus (1993) distinguished
semantic and episodic representations of the self. Semantic
memory is relatively abstract, context - free knowledge such
as Elephants are heavy and George H. W. Bush was
considered unpopular until his son brought new meaning
to the word. Although semantic memory is not necessarily
linked to the self, it can consist of propositions about the
self (e.g., I have brown hair ). More relevant here,
the semantic memory system may contain a subsystem
in which information about one s qualities, traits, and
social roles is stored (e.g., I can be assertive if pushed ).
Such a system would be useful to those who are asked to
describe themselves quickly and succinctly. For example,
first dates, job interviews, and other first - time encounters
often compel people to generate global self - characterizations
with little time to consult the evidence on which such char-
acterizations are based.
As the name implies, episodic memories encapsulate
specific episodes or events that occurred in a person s life.
When accessed, the retrieved events are experienced in
conjunction with a conscious awareness that they actually
occurred in the person s life (e.g., Suddendorf & Corballis,
1997). Most people can recall hundreds if not thousands
of episodic memories, including events in the distant
past (e.g., their first kiss) or only moments ago (e.g., the
sentence they just finished reading).
Although it is obvious that episodic self - knowledge is
based on specific events in people s lives, it is less obvious
where semantic self - knowledge comes from. At first blush,
it might seem that self - knowledge is organized inductively,
with specific episodes of episodic self - knowledge giving
rise to and supporting semantic knowledge about the self.
Although this surely occurs in some instances, the research
literature shows clearly that this is not always so. Instead,
at least some semantic beliefs about the self seem to be
CH16.indd 591CH16.indd 591 10/22/09 2:50:03 PM10/22/09 2:50:03 PM
592 Self and Identity
formed and stored quite independently of specific episodic
memories. Early support for this generalization came from
studies of normal college students who completed priming
tasks. The results showed that priming a trait stored in
semantic memory (e.g., “ Does ’ stubborn ’ describe you? ” )
does not facilitate the recall of corresponding episodic
memories, namely, behavioral incidents that exemplify the
trait (e.g., Klein & Loftus, 1993; Klein, Loftus, Trafton, &
Fuhrman, 1992). Moreover, semantic and episodic self -
representations seem to be stored in different regions of
the brain. For example, participants who were asked to
judge trait adjectives for self - descriptiveness showed acti-
vation of cortical areas associated with semantic memory
retrieval (left frontal regions) but not of areas associated
with episodic memory (right frontal regions; e.g., Kelley
et al., 2002).
Converging evidence for the independence of seman-
tic and episodic representations of self comes from case
studies of people with various cognitive impairments (e.g.,
amnesia, autism, and Alzheimer s dementia). For example,
patients with brain injuries that make them unable to access
and recall episodic memories are nevertheless able to make
accurate judgments about their own traits. In addition,
people with impaired episodic memories are capable of
updating their semantic memories to accommodate newly
acquired self - knowledge (e.g., Klein, Loftus, & Kihlstrom,
1996; for a review, see Klein, 2004).
Implicit Versus Explicit Self - Knowledge
Like other types of knowledge stored in memory, self -
knowledge varies in how explicit it is. Whereas explicit
self - knowledge is relatively controllable and deliberate,
implicit self - knowledge is relatively uncontrollable and
automatic (Fazio, Sanbonmatsu, Powell, & Kardes, 1986).
Moreover, explicit self - knowledge is readily reported but
implicit self - knowledge is often gleaned indirectly by
observing its effects on people s feelings and automatic
behaviors. Indeed, Greenwald and Banaji (1995, p. 11)
defined implicit self - esteem as the introspectively uniden-
tified (or inaccurately identified) effect of the self - attitude
on evaluation of self - associated and self - dissociated
objects. ”
Although indirect measures might ordinarily seem less
desirable than measures that assess the target construct
directly, some suggest that implicit self - esteem measures
circumvent self - presentational processes to lay bare the
unvarnished self (Farnham, Greenwald, & Banaji, 1999).
Others propose that implicit self - esteem measures circum-
vent deliberative thought processes and thus reveal the
“ intuitive ” self (Jordan, Whitfield, & Zeigler - Hill, 2007).
So enticing is the prospect of bypassing respondents
deliberative self - views that research on the nature, origins,
and consequences of the implicit self has grown at a
remarkable rate. At the time of this writing, a PsycINFO
search for publications with keywords including implicit
and either “ self or “ identity ” yielded 292 publications
between 2000 and 2009, as compared to only 50 such pub-
lications during the preceding decade.
Research on the implicit self explores several themes.
Some work focuses on documenting a positivity bias on
implicit measures of self - knowledge that parallels the
positivity bias observed with explicit measures of self -
knowledge (Taylor & Brown, 1988). At the trait level, people
generally display highly favorable self - views and high self -
esteem when these variables are measured implicitly (e.g.,
Kitayama & Karasawa, 1997; Yamaguchi et al., 2007). At
the group level, however, members of minority and low -
status groups display relatively weak implicit liking for
their own social group relative to comparison majority or
high - status groups (Jost, Pelham, & Carvallo, 2002; Nosek,
Banaji, & Greenwald, 2002). These findings are consis-
tent with work suggesting that members of disadvantaged
groups can preserve their personal feelings of self - worth
while still recognizing that their social groups are devalued
by the wider culture (Crocker & Major, 1989).
Another research theme explores the manner in which
implicitly measured self - knowledge shapes people ’ s thoughts,
feelings, or behaviors. For example, in their work on implicit
egotism an automatic preference for things that resemble
the self Pelham, Carvallo, and Jones (2005) argue that
people s implicit feelings about the self guide many of their
most important life decisions, including choice of occupa-
tion, romantic partner, and residence. Because most people
feel quite favorably toward the self, they tend to seek out
people, places, and things that remind them of the self.
A third theme in research on the implicit self focuses on
the emotional and behavioral implications of discrepancies
between people s implicitly and explicitly measured self -
knowledge. For instance, some work reveals that people
who display favorable self - views on explicit measures, but
relatively unfavorable self - views on implicit measures, are
characterized by heightened levels of self - aggrandizement
(e.g., Bosson, Brown, Zeigler - Hill, & Swann, 2003; Jordan,
Spencer, Zanna, Hoshino - Browne, & Correll, 2003), ver-
bal defensiveness (Kernis, Lakey, & Heppner, 2008), and
belief conviction (McGregor & Marigold, 2003). Thus, it
appears that discrepancies between implicitly and explic-
itly measured self - knowledge may predict a defensive
tendency to present the self in an overly zealous manner.
Despite the attention that implicit self - knowledge and
implicit self - esteem in particular — has commanded in
recent years, troubling questions have been raised regarding
several fundamental issues, including what the construct
is. There are currently (at least) two competing schools of
CH16.indd 592CH16.indd 592 10/22/09 2:50:03 PM10/22/09 2:50:03 PM
Self as a Mental Representation 593
thought on the distinction between explicit and implicit
self - knowledge. One perspective, exemplified in Epstein s
(1994) cognitive experiential self theory, assumes that
explicit self - knowledge and implicit self - knowledge rep-
resent fundamentally distinct constructs that derive from
different types of learning experiences; have independent
effects on thoughts, feelings, and behaviors; and may even
be processed via separate systems in the brain. Some advo-
cates of this perspective assume further that implicit self -
knowledge is nonconscious. That is, not only do people
lack conscious awareness of the effects of implicit self -
knowledge on their behavior as Greenwald and Banaji
(1995) initially asserted about implicit self - esteem but
people may also lack conscious awareness of the contents
of their implicit self - knowledge (e.g., Devos & Banaji,
2003). As such, the same self - views measured via explicit
and implicit methods should not necessarily be expected to
correlate; in fact, they often do not (e.g., Bosson, Swann, &
Pennebaker, 2000).
The other perspective, exemplified in Fazio s motivation
and opportunity as determinants model (Fazio & Towles -
Schwen, 1999), holds that explicit and implicit measures
of the same self - view often do access the same underlying
attitude. According to this perspective, factors such as
people s opportunity and motivation to control their
behavioral responses determine the degree of correspon-
dence between a self - view that is measured by self - report
(an explicit method) and the same self - view measured by an
implicit method such as response latency. Advocates of
this perspective assume that both types of measures tap the
same self - view but that explicit measures afford respon-
dents more opportunities to influence the manner in which
they present the self than do implicit measures (Olson &
Fazio, 2008; Olson, Fazio, & Hermann, 2007). Thus, explicit
and implicit measures of the same self - view predict dif-
ferent outcomes, but this need not imply that these mea-
sures access different underlying constructs. At present,
the debate between these two perspectives continues.
Concerns have also been raised about whether implicit
measures can, in principle, deliver on their promise. Initial
enthusiasm for measures of implicit self - esteem was based
on the hope that they would tap an unvarnished or true
form of self - esteem and would therefore outpredict measures
of explicit self - esteem in at least some domains. This theory
has received some support (e.g., Spalding & Hardin, 1999)
but likely not as much as hoped. One reason for this may be
that, like explicit self - esteem, implicit self - esteem is a
broad - based construct that has a wide bandwidth (Marsh &
Craven, 2006; Swann, Chang - Schneider, & McClarty, 2007).
If so, it may not be feasible to assess implicit self - esteem
by way of a simple association between one or more spe-
cific characteristics and the self. Instead, it seems likely
that people have many nonconscious associations with
various aspects of themselves. This might explain why dif-
ferent measures of implicit self - esteem are often uncorre-
lated with one another (Bosson et al., 2000).
A final concern is specific to measures of implicit
self - esteem that are based on minimizing the ability of
participants to reflect before responding (Farnham et al.,
1999). This approach, which is used in some of the most
popular measures such as the Implicit Association Test,
presumably reduces the capacity of respondents to engage
in self - presentation. This is not necessarily true, as self -
presentational activity can be automatized (Paulhus, 1993).
In addition, depriving respondents of the opportunity to
reflect may have the additional effect of preventing them
from accessing autobiographical knowledge, an activity that
requires cognitive work. Therefore, when they are deprived
of cognitive resources, people with negative and positive
self - views tend to respond similarly to self - relevant feedback
(Hixon & Swann, 1993; Swann, Hixon, Stein - Seroussi, &
Gilbert, 1990). Such findings raise the possibility that
measures that diminish the capacity to reflect may unin-
tentionally throw out the self - knowledge baby with the
self - presentational bathwater. This possibility is supported
by evidence that responses to the Implicit Association
Test predict theoretically relevant outcomes more strongly
when they are contaminated by recently activated explicit
beliefs about the self (e.g., Bosson et al., 2000).
Together, these considerations raise vexing questions
about whether implicit measures of self - esteem do in
fact provide clearer insight into people s self - views than
do explicit measures. An alternative view is that implicit and
explicit measures both reveal valid information about people s
self - knowledge but that the image of self that emerges from
such measures may differ as a result of various underlying
processes and situational features. Recent research and the-
ory offers insight into the processes that underlie responses
to implicit and explicit measures of attitudes in general (e.g.,
Gawronski & Bodenhausen, 2006; Nosek, 2005). Time will
tell whether these insights from the general attitudes litera-
ture will generalize to attitudes toward the self.
Actual Versus Possible Self - Views
Whereas we have restricted our discussion thus far to self -
knowledge that people hold about themselves in the pres-
ent, several influential theories focus instead on potential
or possible self - knowledge. For example, E. Tory Higgins ’ s
(1987; Higgins, Klein, & Strauman, 1985) self - discrepancy
theory proposes that people store self - knowledge not only
in the form of actual beliefs about the self but also in the
form of ideal and ought beliefs about the self. The ideal self
contains people s beliefs about their personal aspirations,
as well as their beliefs about important others hopes for
CH16.indd 593CH16.indd 593 10/22/09 2:50:04 PM10/22/09 2:50:04 PM
594 Self and Identity
them; the ought self contains people s beliefs about their
personal obligations and duties, as well as their beliefs
about important others expectations for them. According to
self - discrepancy theory, discrepancies between actual and
ideal selves are associated with heightened levels of sad-
ness and dejection, while actual ought discrepancies are
associated with fear and anxiety. Thus, the ideal and ought
selves serve as guides that motivate behaviors aimed at
minimizing existing discrepancies. Initial support for these
predictions (e.g., Higgins, Bond, Klein, & Strauman, 1986;
Strauman & Higgins, 1988) was followed by the publica-
tion of some inconsistent findings (Tangney, Niedenthal,
Covert, & Barlow, 1998). In an effort to reconcile these
inconsistencies, researchers subsequently identified modera-
tors of the effects such as the magnitude and importance of
the self - discrepancy, the accessibility of the self - discrepancy,
and the applicability and relevance of the self - discrepancy in
a current context. Eventually, Higgins (1998) developed his
ideas into a new theory of regulatory focus.
Similar to Higgins s (1987) self - guides, Markus and
Nurius (1986) proposed the construct of possible selves,
which are people s projections about what they might
become, would like to become, and are afraid to become
in the future. Possible selves motivate behaviors intended
to achieve desired possible selves and to avoid feared ones
(e.g., Oyserman, Bybee, Terry, & Hart - Johnson, 2004).
Nevertheless, possible selves alone may not be sufficient
to motivate effective behaviors unless they are accompa-
nied by plausible strategies for achieving desired goals
(Oyserman, Bybee, & Terry, 2006).
Global Versus Specific Self - Knowledge
Self - views vary in their breadth or specificity, which corre-
sponds directly to the amount of information they convey
(Hampson, John, & Goldberg, 1987). At the broadest level,
global self - views are generalized beliefs that encompass a
range of personal qualities (e.g., I am worthwhile and I
like myself ). At the narrowest level, specific self - views or
self - concepts pertain to relatively specific qualities (e.g.,
“ I am a world - class guitarist ” ). Between these extremes
lie midlevel self - views that convey a moderate amount
of information about the self (e.g., I am cooperative and
“ I lack common sense ” ).
The distinction between global and specific self - views
offers an alternative means of conceptualizing self - esteem.
Instead of conceptualizing self - esteem as primarily affec-
tive (i.e., how people feel about the self) and self - concepts
as primarily cognitive (i.e., what people believe about
the self), as have some theorists (Baumeister, Campbell,
Krueger, & Vohs, 2003), it is possible to think of self -
esteem as a global belief about the self and self - concepts as
relatively specific beliefs about the self (Marsh & Craven,
2006; Shavelson, Hubner, & Stanton, 1976; Swann et al.,
2007). The latter approach assumes that self - knowledge is
structured hierarchically in memory, with global self - esteem
at the top of the hierarchy. Beneath global self - esteem lie
more specific self - concepts nested within domains such as
academic, physical, and social. Empirical investigations
support this hierarchical model. For example, evidence
indicates that individual self - concepts, measured sepa-
rately, combine statistically to form a superordinate global
self - esteem factor (Marsh & Hattie, 1996).
Conceptualizing self - esteem as a global representa-
tion of the self can shed light on an ongoing debate in
the self - esteem literature. Whereas some suggest that
global self - esteem lacks predictive ability when it comes
to important life outcomes (e.g., Baumeister et al., 2003),
others find that global self - esteem does predict important
outcomes, as long as those outcomes are measured at a
global level, such as several outcomes bundled together
(e.g., Donnellan, Trzesniewski, Robins, Moffitt, & Caspi,
2005; Trzesniewski et al., 2006). Thus, recognizing that
self - knowledge assumes both global and specific forms may
bear practical fruit by increasing researchers ability to pre-
dict criterion variables of interest (e.g., Swann, et al., 2007).
Some theorists seek a middle ground between concep-
tualizing self - esteem as a single global entity and seeing it
as numerous specific self - views. Based on the assumption
that agency and communion represent universal dimen-
sions that underlie much of human behavior and thought
(e.g., Fiske, Cuddy, Glick, & Xu, 2002; Wiggins, 1979),
this middle ground approach identifies two components
of global self - esteem that correspond to agency and com-
munion (e.g., Franks & Marolla, 1976; Gecas, 1971).
Tafarodi and Swann (2001) labeled these components self -
competence , an evaluation of one s ability to bring about
desired outcomes, and self - liking , an evaluation of one s
goodness, worth, and lovability. Supporting this distinc-
tion, research indicates that self - competence and self - liking
predict unique outcomes (e.g., Bosson & Swann, 1999;
Tafarodi & Vu, 1997).
Personal Versus Social Self - Knowledge
Within social psychology, social identity theorists were among
the first to distinguish personal from social self - knowledge
(Tajfel & Turner, 1979, 1986). Whereas personal self - views
refer to individual, trait - like attributes (e.g., submissive, intel-
ligent), social self - views consist of people s knowledge of the
social groups to which they belong, along with their feelings
about those groups. One important consequence of this dis-
tinction is the recognition that people can derive feelings of
value and worth not only from their personal qualities but also
from their associations with valued groups (e.g., Luhtanen &
Crocker, 1992; Tajfel & Turner, 1986).
CH16.indd 594CH16.indd 594 10/22/09 2:50:04 PM10/22/09 2:50:04 PM
Self as a Mental Representation 595
Further refining the personal social distinction, some
theorists propose the existence of several levels at which
self - knowledge is represented (e.g., Brewer & Gardner,
1996). According to these perspectives, self - knowledge
pertaining to people s distinct traits and qualities, or per-
sonal self - views, is stored at the individual level. At the
interpersonal level reside relational self - views, which
describe qualities that are relevant to people s social roles
and relationships (e.g., protective older sister). Finally,
two types of self - views associated with group member-
ships can be distinguished, collective self - views and group
identities. Collective self - views refer to personal qualities
that are associated with people s group memberships (e.g.,
open - minded Democrat; Chen, Chen, & Shaw, 2004), and
group identities refer to characteristics of a group that may
or may not describe an individual member of that group
(Lemay & Ashmore, 2004). For example, people may
hold convictions about the groups to which they belong
( Spaniards are impulsive ) that conflict with their per-
sonal self - views ( “ I am cautious ” ).
Although all people presumably store self - knowledge at
all three levels (personal, relational, and group), there exist
stable individual differences in the extent to which people
focus on, value, and derive self - esteem from each form
of self - view. For example, people from collectivistic cul-
tures tend to focus more on their relational and collective
self - knowledge, whereas those from individualistic cul-
tures tend to focus more on their personal self - knowledge
(Cousins, 1989; Markus & Kitayama, 1991). Moreover,
whereas women emphasize their relational self - views
more than men, men emphasize their collective self -
views more than women (Gabriel & Gardner, 1999). These
differences in chronic focus suggest that people s cultural
background and gender play important roles in the types
of self - relevant information they are most likely to notice,
recall, and be influenced by.
Metacognitive Aspects of Self - Knowledge
Metacognitive aspects of self - knowledge refer to charac-
teristics such as importance, certainty, and stability that
differentiate some self - views from others. Here, we cover
several metacognitive aspects of self - knowledge that have
attracted substantial empirical scrutiny.
Valence of Self - Knowledge
Not surprisingly, robust associations exist between the
valence of people s specific self - views and their global feel-
ings of self - esteem, such that people higher in self - esteem
tend to have more positive self - views and fewer negative
ones (Brown, 1998; Pelham & Swann, 1989). Theorists
have explained this relation in two ways. According to the
bottom - up perspective, global self - esteem derives from
the overall valence of individual self - views in the self -
concept (e.g., Marsh, 1990). As such, a woman who thinks
of herself as intelligent, sociable, and attractive has higher
global self - esteem than a woman who thinks of herself
as unintelligent, socially awkward, and unattractive. The
competing, top - down perspective, holds that feelings of
global self - esteem are the driving force behind the valence
of people s relatively specific self - views (e.g., Brown,
Dutton, & Cook, 2001). According to this perspective,
people develop global feelings of self - esteem early in life,
and their global regard for the self determines whether they
subsequently develop positive or negative beliefs about the
self within specific domains. Thus, a man who has high
global self - esteem thinks of himself as more intelligent,
sociable, and attractive than a man who has low global
self - esteem. At present, evidence exists for both of these
perspectives, pointing to an interactive effect wherein bidi-
rectional, direct and indirect links are found between the
valence of self - knowledge and the valence of global self -
esteem (Showers & Zeigler - Hill, 2006).
Importance of Self - Knowledge
James (1890/1950) first observed that self - views can vary in
importance and that such variation can have important impli-
cations for the self. In fact, this observation led to his classic
formula in which self - esteem equals success (actual achieve-
ments) divided by pretensions (desired achievements). Thus,
competent performance in important domains fosters self -
esteem, but incompetence does not threaten self - esteem if it
occurs in devalued domains. As James put it, I, who for the
time have staked my all on being a psychologist, am mor-
tified if others know much more psychology than I. But I
am contented to wallow in the grossest ignorance of Greek
(1890/1950, p. 310). Initial attempts to test James s formula
were unsupportive (Hoge & McCarthy, 1984; Marsh, 1986).
Later work, however, revealed that the importance of self -
views is related to self - esteem primarily among people who
have relatively negative self - views overall but are highly cer-
tain of their positive self - views (Pelham, 1995; Pelham &
Swann, 1989).
The importance that people place on their specific self -
views predicts other self - relevant phenomena. For example,
when people deem a self - view high in importance they
are more likely to behave in accordance with it (Pelham,
1991), and they demonstrate higher levels of cross -
situational consistency in their self - descriptions of it
(English & Chen, 2007). Indeed, people behave so as to
protect and maintain their highly important self - views. For
instance, people exhibit stronger resistance to challenges
to highly important self - views than to self - views that are
less important (Markus, 1977), and they work especially
CH16.indd 595CH16.indd 595 10/22/09 2:50:04 PM10/22/09 2:50:04 PM
596 Self and Identity
hard to surround themselves with people and feedback
that verify the self - views they deem most important (Chen
et al., 2004; Swann & Pelham, 2002). Moreover, people
avoid painful social comparisons when the domain of com-
parison is linked to their most important self - views (Wood,
1989), and they may even distance themselves from
close friends who outperform them in such domains
(Tesser, 1988).
Just as specific self - views can vary in importance, so can
collective self - views and group identities. Whereas some
people place great importance on their memberships in
various social groups, others attribute little significance to
“ being male ” or “ being Native American ” (e.g., Luhtanen &
Crocker, 1992; Turner & Brown, 2007). Placing a lot of
stock in collective self - views is linked to both positive and
negative outcomes. On the one hand, for members of nega-
tively stereotyped social groups, placing importance on the
collective self can serve as a buffer against the hurtful effects
of discrimination on self - esteem and well - being (Crocker &
Major, 1989; Wong, Eccles, & Sameroff, 2003). On the other
hand, those who value strongly their group memberships
are particularly vulnerable to the harmful effects of stereo-
types about their group, and they display heightened levels
of conformity to maladaptive group norms. We have more
to say about these effects of group identification in our
discussions of social identity (Tajfel & Turner, 1979) and
stereotype threat (Steele, 1997) theories.
Certainty and Clarity of Self - Knowledge
The certainty with which people hold self - views has impor-
tant implications for the self. Increases in the certainty
of people s self - views, for example, are associated with
increases in global self - esteem (Baumgardner, 1990; Story,
2004). Conversely, low levels of self - view certainty are
associated with increased tendencies toward maladaptive
psychological conditions, such as social phobia (Wilson &
Rapee, 2006).
As with important self - views, people work espe-
cially hard to maintain their highly certain self - views.
For instance, people who are more certain of their self -
views tend to behave more consistently across situations
(Baumgardner, 1990). Similarly, people are more likely to
seek (Pelham, 1991) and receive (Pelham & Swann, 1994)
interpersonal feedback that is consistent with self - views of
which they are highly certain. When confronted with feed-
back that challenges highly certain self - views, people dis-
play resistance (Swann & Ely, 1984), and such resistance
efforts may further buttress the certainty of their self - views
(Swann, Pelham, & Chidester, 1988).
Closely related to self - view certainty is self - concept
clarity, which is defined as the extent to which self - views
are clear, confident, consistent, and stable across time
(Campbell et al., 1996). Like certainty, the clarity of people s
self - views is associated with higher global self - esteem
(Campbell, 1990). Moreover, heightened self - concept clarity
is associated with decreased neuroticism (Campbell et al.,
1996), more adaptive coping skills (Smith, Wethington, &
Zhan, 1996), and increased psychological adjustment
(Campbell, Assanand, & Di Paula, 2003).
Stability of Self - Knowledge
Despite an overall tendency toward stability across long
periods (see the section on Identity Negotiation and
Change), some self - views fluctuate a great deal across
shorter time frames. Much of the research on short - term
fluctuations in self - knowledge focuses on individual dif-
ferences in self - esteem stability. Whereas some people
provide similar ratings of their global self - esteem from
one measurement to the next, others experience relatively
frequent, transient fluctuations in state self - esteem (e.g.,
Kernis, Cornell, Sun, Berry, & Harlow, 1993). Moreover,
although there is a modest, positive association between
the level and stability of self - esteem (e.g., Kernis, Paradise,
Whitaker, Wheatman, & Goldman, 2000), high levels of
instability may occur at any level of global self - esteem.
In general, higher levels of self - esteem stability are asso-
ciated with superior psychological well - being. For example,
independent of their self - esteem level, people with more
stable self - esteem are more likely to pursue everyday
goals for intrinsic reasons (e.g., interest and enjoyment)
rather than extrinsic reasons (e.g., feeling forced), and they
feel less anxiety associated with the pursuit of such goals
(Kernis et al., 2000). People higher in self - esteem stability
also report fewer depressive symptoms in the face of daily
stressors (Kernis, Grannemann, & Barclay, 1989).
To date, much of the research on self - esteem stability
focuses on the experiences of people with stable versus
unstable high self - esteem. Compared with their stable high
self - esteem peers, individuals with unstable high self - esteem
appear hypervigilant for social feedback, and they react to
negative performance feedback with heightened anger, hos-
tility, and defensiveness (e.g., Kernis et al., 1989). Because
of its high reactivity to events that challenge the self, Kernis
(2003) calls unstable high self - esteem a form of fragile
high self - esteem.
Note that actual stability of self - knowledge and perceived
stability of self - knowledge are independent. For instance,
those who assume that their belief structures tend to
remain stable across time may perceive greater consis-
tency between their past and their present attitudes than
is actually the case (Ross, 1989). Conversely, those whose
implicit theories lead them to expect that training programs
will improve their skills (e.g., Conway & Ross, 1984), or
that personal and social adjustment generally increase with
CH16.indd 596CH16.indd 596 10/22/09 2:50:05 PM10/22/09 2:50:05 PM
Self as a Mental Representation 597
age (Woodruff & Birren, 1972), may perceive less stability
across time in these self aspects than they actually display.
For these reasons and related ones, people s beliefs about
the stability of their self - knowledge may fail to track the
actual stability of such knowledge.
Organization of Self - Knowledge
Research suggests that people differ in terms of how they
organize self - knowledge in memory. Much of this research
considers four features of the structure of self - knowledge.
First is the number of different self - aspects — superordinate
traits or roles (e.g., wife and social self) that house all
lower - order pieces of self - knowledge in the self - concept
(e.g., Linville, 1987). Next is the valence of self - knowledge,
often measured as a function of the ratio of positive to neg-
ative self - views in the self - concept (Showers, 1992). Third
is the level of compartmentalization versus integration that
characterizes the self - aspects. Compartmentalization refers
to the tendency to store positive and negative self - views
within separate self - aspects, whereas integration refers to
the tendency to store both positive and negative self - views
within the same self - aspects. Finally, some researchers
consider the importance that people place on their different
self - aspects, with the assumption that more important self -
aspects — and their accompanying contents — are likely to
be activated most frequently (Showers, 1992).
Consideration of these features of the self - concept has
led to several important insights into the links between
self - concept and mental health. For example, work done
by Showers and colleagues (Showers, 1992; Showers &
Kling, 1996) reveals that compartmentalization is gener-
ally associated with higher self - esteem and reduced depres-
sion among people who place importance on their positive
self - aspects. For such individuals, compartmentalization
limits their cognitive access to painful or threatening self - rel-
evant information. Conversely, integrative self - structures are
associated with higher self - esteem and lower depression for
people who place importance on their negative self - aspects,
because experiences that activate negative self - aspects call
to mind both negative and positive pieces of self - knowledge.
Integration also promotes resilience in the face of extreme
stress or adversity (Showers & Zeigler - Hill, 2007) or intense
negative mood states (Showers & Kling, 1996).
Contingency of Self - Esteem
The contingency of self - esteem refers to the extent to which
people base their feelings of self - worth on their ability to
achieve specific outcomes or match specific standards. As
noted by Kernis (2003), at least two different approaches
exist to the study of contingent self - esteem. One approach
assumes that most people have contingent self - esteem but
that they differ in the particular domains on which they
base their self - esteem (Crocker & Wolfe, 2001). According
to this perspective, college students differ reliably in
the extent to which they base their self - esteem on their
accomplishments within seven broad domains: academics,
appearance, approval from others, competition, family sup-
port, God s love, and virtue (Crocker, Luhtanen, Cooper, &
Bouvrette, 2003). Moreover, consistent with James s
(1890/1950) self - esteem formula, people s feelings of
global self - esteem tend to fluctuate as a function of their
successes and failures primarily within domains in which
they are psychologically invested (Crocker, Karpinski,
Quinn, & Chase, 2003; Crocker, Sommers, & Luhtanen,
2002). Although some work suggests that contingencies
of self - worth can interfere with adaptive functioning (e.g.,
Crocker & Luhtanen, 2003), some domains of contingency
may be healthier than others. For instance, basing self -
esteem on internal contingencies, such as virtue or God s
love, is associated with fewer signs of psychological dis-
tress than basing self - esteem on external contingencies,
such as appearance or others approval.
The other approach to contingent self - esteem assumes
individual differences exist in the overall extent to which
people possess contingent versus true (noncontingent)
self - esteem (Deci & Ryan, 1995). According to this perspec-
tive, contingent and true self - esteem represent two types of
self - esteem that derive from different learning experiences.
Contingent self - esteem develops when individuals learn
that their worth and lovability depend on their attainment
of specific outcomes. Having internalized this belief, indi-
viduals with contingent self - esteem tend to pursue goals for
extrinsic reasons (e.g., others approval) rather than intrinsic
reasons (e.g., interest), and they show heightened levels of
conformity to external forces (Gagn é , Ryan, & Bargmann,
2003; Ryan, Sheldon, Kasser, & Deci, 1996). Conversely,
true high self - esteem develops when individuals learn that
they are valued for who they are and receive high levels of
care and autonomy in the pursuit of their goals. Not surpris-
ingly, true self - esteem is associated with a reduced focus
on extrinsic reinforcers and higher levels of psychological
adjustment (Kasser, Ryan, Zax, & Sameroff, 1995).
Although these approaches focus on different aspects
of contingent self - worth, they need not be viewed as
antagonistic. Indeed, while people do differ in the specific
domains on which they base their self - esteem (Crocker,
Luhtanen, et al., 2003), meaningful overall differences
also occur in the degree to which people exhibit contingent
versus true self - esteem (Kernis et al., 2008; Neighbors,
Larimer, Markman Geisner, & Knee, 2004).
Narcissism: A Special Case of Fragile Self - Esteem
Whereas most metacognitive features of self - knowledge
discussed thus far have been unidimensional, narcissism
CH16.indd 597CH16.indd 597 10/22/09 2:50:05 PM10/22/09 2:50:05 PM
598 Self and Identity
is a multidimensional construct. Some qualities associated
with narcissism were discussed earlier, including highly
positive self - views (in agentic domains; Campbell, Bosson,
Goheen, Lakey, & Kernis, 2007; Campbell, Rudich, &
Sedikides, 2002), low levels of certainty and clarity, and
contingent self - esteem within competitive, but not affili-
ative, domains (Crocker, Luhtanen,et al., 2003; Zeigler -
Hill, Clark, & Pickard, 2008). We include narcissism in
our list of metacognitive features because narcissism has
broad effects on how people value, select, organize, store,
and activate self - knowledge (Morf & Rhodewalt, 2001;
Rhodewalt, 2005).
Freud (1914/1957) first introduced the idea of narcis-
sism to the psychoanalytical literature, viewing it as a dis-
order that arises when individuals attach too much libido ,
or psychic energy, to the self and not enough to their inter-
nalized representations of relationship partners. As a result,
the individual develops excessive levels of self - regard but
does not feel sufficient love for others. Later psychoana-
lytical theorists (e.g., Kernberg, 1986; Kohut, 1966, 1971)
differed from Freud in their understanding of the origins
of narcissism but still conceptualized it as an outgrowth of
troubled relationships and unmet needs early in life (see
Bosson et al., 2008).
Although narcissism is typically viewed as a person-
ality disorder among clinical psychologists (American
Psychiatric Association, 2000; Westen, 1990), social and
personality psychologists often treat it as an individual dif-
ference variable that can be assessed meaningfully within
any population. This approach gave rise to the Narcissistic
Personality Inventory (Raskin & Hall, 1981), a scale
designed to measure narcissistic tendencies within normal,
nonpathological populations. When treated as a unidimen-
sional scale, the Narcissistic Personality Inventory assesses
grandiose narcissism , which is characterized by high self -
esteem, vanity, entitlement, a willingness to manipulate
and exploit others for personal gain, and high levels of
defensiveness in response to self - threats (e.g., Paulhus,
Robins, Trzesniewski, & Tracy, 2004; Raskin, Novacek, &
Hogan, 1991). Like individuals with unstable high self -
esteem and contingent self - esteem, those high in grandi-
ose narcissism appear to have fragile high self - esteem
because their self - esteem is easily threatened and requires
constant validation (Kernis, 2003).
Recently, theorists have given increasing attention to a
second narcissistic subtype referred to as vulnerable nar-
cissism (see Dickinson & Pincus, 2003). Like grandiose
narcissists, vulnerable narcissists entertain self - aggran-
dizing fantasies about themselves, and they demonstrate
a heightened sense of entitlement and a willingness to
exploit others. In contrast to grandiose narcissists, how-
ever, vulnerable narcissists report feelings of inferiority,
shame - proneness, and low self - esteem (Cooper &
Ronningstam, 1992; Gramzow & Tangney, 1992). Moreover,
vulnerable narcissists tend to hide their feelings of grandi-
osity behind a fa ç ade of modesty. Thus, whereas grandiose
narcissists demand admiration and respect from others,
vulnerable narcissists crave approval but are too inhibited
to demand it.
The forms and features of self - knowledge described in the
preceding sections do not arise in a vacuum. Self - knowledge
is shaped by numerous interacting forces, both biological
and social. Here, we outline some of the raw materials of
self - knowledge, as well as the mechanisms through which
people develop mental representations of the self. We also
consider questions and findings concerning the accuracy of
people s representations of the self.
Biological Origins of the Self and Identity
Where, in the brain, is the self represented? Although this
question defies a simple answer, researchers have begun
to specify the neurological correlates of various aspects of
self - knowledge. In general, much of this work converges on
the conclusion that self - referential tasks such as thinking
about one s traits or feelings or evaluating the self trigger
heightened activity in the medial prefrontal cortex (MPFC;
e.g., Johnson et al., 2002; Kelley et al., 2002; Mitchell,
Banaji, & Macrae, 2005; Ochsner et al., 2004; Saxe,
Moran, Scholz, & Gabrieli, 2006). Notably, heightened
MPFC activity is also observed when people think about
the traits and mental states of close others (Ochsner et al.,
2005), suggesting that the MPFC may be part of a network
of brain systems that mediates social knowledge in general.
Moreover, different areas of the brain become active when
people report on the self - descriptiveness of trait terms
associated with domains with which they have high versus
low levels of personal experience (Lieberman, Jarcho, &
Satpute, 2004). This latter work is interpreted as evidence
that different brain systems process evidence - based (high
personal experience) and intuition - based (low personal
experience) self - knowledge. The larger point here is that
no single brain system or area of the brain appears to
be, of itself, responsible for our sense of self. Instead, mul-
tiple systems work together to create the sense of a unitary
self, and some of the same systems that mediate self -
knowledge are involved in mediating knowledge about
others traits and states.
CH16.indd 598CH16.indd 598 10/22/09 2:50:05 PM10/22/09 2:50:05 PM
Origins of Self-Representations 599
Genes and Heredity
Much of what is known about the genetic bases of the
self comes from the personality literature, which typically
assesses personality by having people report their self -
views (e.g., Vazire, 2006). Based largely on twin studies,
this research indicates a substantial genetic basis to people s
self - views within the broad, Big Five, personality
factors of extraversion, neuroticism, agreeableness, con-
scientiousness, and openness. Specifically, approximately
40% to 60% of the population variance in self - reports of
the Big Five factors is accounted for by genes (for a review,
see Bouchard & Loehlin, 2001). The remaining variance
is typically attributed to environmental influences, gene
environment interactions, and chance factors.
Recently, there has been a surge of interest in under-
standing the genetic bases of global self - esteem. Thus far,
researchers in this area have demonstrated that self - esteem
is heritable (McGuire et al., 1999) and that genes explain
approximately 30% to 50% of the population variance in self -
esteem (Kamakura, Ando, & Ono, 2007; Kendler, Gardner, &
Prescott, 1998). Heredity also appears to explain a substantial
amount of the variance in self - esteem stability and change
across time (Neiss, Sedikides, & Stevenson, 2006).
Despite the vigor with which some theorists have inte-
grated genetic influences into broad, biosocial models of
the self (e.g., Tesser, 2002), the biology of the self remains
an area of inquiry in which research lags behind theory.
One interesting challenge for future researchers will be to
specify the biological bases of distinctions that are basic
to the area, such as enduring versus weakly held identities.
One possibility is suggested by a general theory of learning
known as Hebb s (1949) law. The law states that if one neu-
ron (A) is repeatedly involved in causing another neuron
(B) to fire, metabolic changes occur in one or both cells
that enhance the ability of A to cause B to fire. Simply put,
cells that fire together, wire together. Insofar as this prin-
ciple applies to the clusters of neurons or pathways that are
associated with self - knowledge, then it may be that endur-
ing self - representations are simply ones that have been
activated repeatedly in the past (which makes them more
readily activated in the future). Although this particular
account is speculative, it is clear that further elaboration of
the neural bases of self - knowledge could be extremely ben-
eficial to the development of theory within the subarea.
Interpersonal Origins of the Self and Identity
Attachment Relationships
According to attachment theory (Bowlby, 1969; Hazan &
Shaver, 1994), infants begin to formulate rudimentary
schemas — or working models — about their lovability
and worth early in life, before they have self - awareness.
Working models reflect the consistency and responsiveness
of treatment that infants receive from primary caregiv-
ers. Specifically, caregiving that is both consistent and
adequately responsive to infants needs should convince
them that they are worthy of love and capable of effica-
cious action. This, in turn, should instill in children the
foundations of favorable self - concepts and high global
self - esteem (Bretherton, 1988; Verschueren, Marcoen, &
Schoefs, 1996). Conversely, caregiving that is inconsis-
tent, unresponsive, neglectful, or abusive teaches children
that they are not valuable, that others are not trustworthy
and dependable, or both. In such cases, relatively negative
self - concepts and low esteem will likely result.
This is not to suggest, however, that young children with
negative models of self will describe themselves in unfa-
vorable terms. In fact, young children appear to display
what Swann and Schroeder (1995, p. 1310) refer to as a
“ positive tropism ” — a cognitively simplistic, automatic,
and adaptive propensity to seek positive evaluations and
avoid negative ones. Indeed, research reveals that young
children generally describe their qualities and skills in an
extremely positive manner (Harter, 1999), and the ten-
dency to endorse positive self - descriptions emerges before
the tendency to endorse negative ones (e.g., Benenson &
Dweck, 1986; Stipek & Tannatt, 1984). This may occur
because, before middle childhood, children lack the cogni-
tive capacity to differentiate between their actual and ideal
selves, and they answer questions about the self primar-
ily in terms of their ideals (Harter, 2006; Turner, 1968).
Alternatively, it may be that children in the West are social-
ized to embrace positive evaluations spontaneously and
without reflection (e.g., Heine & Hamamura, 2007). In any
event, around middle childhood, children begin to display
a more nuanced understanding of the self, and stable indi-
vidual differences in self - concepts and self - esteem emerge.
Specifically, children at this age begin to internalize
the appraisals of others (see the next section). Thus, it
may be that the working models developed during infancy
provide a lens through which children interpret others
reactions to them. Indeed, some research suggests that
the working models that are set in place during infancy
continue to influence people s interpretations of social
feedback into adulthood (e.g., Hazan & Shaver, 1987). It
is important to note, however, that a substantial minority
(30 45%) of people change their attachment style their
characteristic pattern of relating to others across time
(e.g., Cozzarelli, Karafa, Collins, & Tagler, 2003).
Whereas working models presumably filter people s
interpretations of self - relevant experiences and feedback,
CH16.indd 599CH16.indd 599 10/22/09 2:50:05 PM10/22/09 2:50:05 PM
600 Self and Identity
appraisals are part of the raw materials from which people
derive specific beliefs about the self. As noted earlier,
Cooley (1902) and Mead (1934) relied on the theory of
symbolic interactionism to explain how people translate
others reactions into self - knowledge. According to this
theory, people come to know their own abilities, traits, and
qualities by seeing themselves through other people s eyes.
More specifically, the symbolic interactionists described a
sequence in which we (a) observe others reactions to our
behaviors; (b) use others reactions to construct reflected
appraisals , or inferences about how others perceive us;
(c) internalize these reflected appraisals as elements of the
self - concept; and (d) use the self - concept to guide subse-
quent behaviors. Thus, the self is created socially and is
subsequently sustained in a cyclical, self - perpetuating
Because children younger than 8 years lack the
perspective - taking skills to evaluate themselves through
the eyes of others (e.g., Selman, 1980), they do not typi-
cally demonstrate an awareness of others appraisals until
middle childhood. It is most likely for this reason that clear
individual differences in self - reported self - esteem do not
emerge until middle childhood (Harter, 1999). At this point,
people who perceive that they are respected, admired, and
loved accordingly internalize these appraisals as positive
self - views, whereas those who perceive that they are eval-
uated unfavorably develop negative self - views. In turn,
people s self - views shape their subsequent interpretations
of others reactions to them: Whereas people high in self -
esteem believe that others perceive them quite favorably,
those low in self - esteem tend to underestimate how favor-
ably they are appraised by others (Bohrnstedt & Felson,
1983; Murray, Holmes, & Griffin, 2000).
Despite the intuitive appeal of symbolic interactionism
in its original form, research has necessitated significant
refinements of some of its tenets (see Tice & Wallace,
2003, for a review). In particular, while it is clear that
people s reflected appraisals correlate strongly with their
self - views that is, people see themselves the way they
believe that others see them it is not clear that people s
reflected appraisals correspond to others actual evalua-
tions of them (Shrauger & Schoeneman, 1979). Moreover,
little evidence indicates that people observe specific oth-
ers reactions to them and then base their self - views on that
feedback. Instead, people s own beliefs about the self seem
to shape their assumptions about how others view them
(Kenny & DePaulo, 1993). Finally, people have a more
accurate understanding of how they are perceived by others
in general than of how they are perceived uniquely by
specific others (Kenny & Albright, 1987), a finding that
further challenges the notion that people internalize the
feedback they receive from specific others. Thus, although
people undoubtedly base their self - knowledge to some
degree on the feedback they receive from others, they have
other influential sources of self - knowledge.
Social Comparisons
According to Festinger s (1954) social comparison theory,
people develop self - knowledge by comparing their own
traits, abilities, opinions, and emotions with those of similar
others (for reviews, see Suls & Wills, 1991; Taylor & Lobel,
1989). Moreover, the direction of comparison that people
make — upward versus downward — is assumed to influence
their resulting self - views and feelings of self - esteem. For
example, while comparing oneself with someone who is
better than the self on a particular dimension of evaluation
(an upward comparison) can diminish a person s feeling
of self - esteem, comparing oneself with someone who is
worse off than the self (a downward comparison) tends
to boost self - esteem (e.g., Helgeson & Mickelson, 1995;
Marsh & Parker, 1984).
Besides increasing people ’ s self - knowledge, social
comparisons serve an important motivational purpose.
For instance, sufferers of stressful events and painful life
experiences can facilitate their own coping and improve
their affective state by comparing themselves with others
who are worse off than them (Wood, Taylor, & Lichtman,
1985). Indeed, a large body of research suggests that people
tend to make downward social comparisons when under
conditions of threat (Taylor & Lobel, 1989). When moti-
vated to improve the self, however, people may make
upward comparisons with those who embody excellence
along particular dimensions (Blanton, Buunk, Gibbons, &
Kuyper, 1999).
As with reflected appraisals, research suggests that
people do not develop the cognitive ability to compare the
self explicitly with others until middle childhood (Harter,
1999; Ruble, Boggiano, Feldman, & Loebl, 1980). Once this
ability emerges, social comparisons tend to occur spontane-
ously, effortlessly, and even unintentionally. For example,
some work demonstrates that people change their self -
views automatically on comparison with both appropriate
and inappropriate comparison partners. Given adequate
mental resources, however, people mentally undo modi-
fications to the self - concept that are based on inappropriate
comparisons (Gilbert, Giesler, & Morris, 1995).
Incorporating Others ’ Qualities
As we saw in the previous three sections, people can
acquire novel self - knowledge through their interactions
with significant relationship partners. Self - expansion the-
ory (Aron & Aron, 1996), however, highlights yet another
route through which people s interactions with close others
can lead to changes in the self. The theory predicts that
CH16.indd 600CH16.indd 600 10/22/09 2:50:06 PM10/22/09 2:50:06 PM
Origins of Self-Representations 601
as closeness between two people grows, they gradually
come to experience a cognitive overlapping of their self -
concepts. As a consequence, relationship partners begin
to act as if the resources, perspectives, and characteristics
of the close other are at least partially their own (Aron,
Aron, Tudor, & Nelson, 1991). Thus, in close relation-
ships, people acquire novel self - knowledge in the form of
features of the close other that have been subsumed into
the self - concept. Support for these ideas can be found in
research demonstrating that people cognitively confuse
the self more with close others than they do with non-
close, but equally familiar, others (e.g., Mashek, Aron, &
Boncimino, 2003). Similarly, people s self - concepts
contain more self - descriptive information in the weeks
immediately after, versus before, they fall in love (Aron,
Paris, & Aron, 1995), suggesting that features of the new
loved one are incorporated into the self.
Influences of Culture and Gender
To a large degree, people s self - knowledge reflects the cul-
ture in which they are socialized. Researchers interested
in the effects of culture on the self have long assumed that
the broad dimensions of collectivism and individualism
differentiate not only the normative rules and structures of
societies but the self - structures of individuals as well (for
a review, see Oyserman, Coon, & Kemmelmeier, 2002).
Individualism refers to a set of beliefs and values that has, at
its core, the assumption that individuals are ascendant over
the groups to which they belong. Conversely, collectiv-
ism holds that individuals are mutually interdependent and
that groups take priority over individuals (Hofstede, 1980;
Triandis, 1995). Given these different emphases on the
individual versus the group, it is not surprising that people
who are socialized in different cultures often display
self - concepts with remarkably different structures, proper-
ties, and contents.
In their review of cross - cultural differences in the self,
Markus and Kitayama (1991) proposed that collectivistic
cultures generally give rise to people with interdependent
selves, whereas individualistic cultures engender people
with independent selves. In the interdependent self, the
individual is connected to significant others, relatively
undifferentiated, and fluid across contexts and time; in
the independent self, the individual is distinct from oth-
ers, autonomous, and stable across contexts and time.
Consistent with the different values that underlie collec-
tivism versus individualism, people with interdependent
versus independent selves tend to exhibit divergent motiva-
tions regarding the self. For instance, whereas people with
interdependent selves appear to value modesty and self -
criticism (Heine, Lehman, Markus, & Kitayama, 1999), those
with independent selves prefer being better than others
(Alicke & Govorun, 2005; Taylor & Brown, 1988). As a
result, people who are raised in individualistic cultures
report substantially higher global self - esteem than do peo-
ple raised in collectivistic cultures (Heine & Hamamura,
2007). In analyses that treat culture as the unit of analysis,
there is a strong positive correlation between a culture s
individualism and the global self - esteem of its inhabit-
ants (Oyserman et al., 2002).
Using a two - component definition of self - esteem,
however, may lead to a more nuanced understanding of
the effects of culture on the positivity of the self - concept.
Consider research showing that people raised in collectivistic
cultures demonstrate relatively high levels of self - liking,
whereas those raised in individualistic cultures demonstrate
relatively high levels of self - competence (Tafarodi, Lang, &
Smith, 1999; Tafarodi & Swann, 1996). As noted earlier,
self - liking reflects people s evaluations of themselves in
terms of qualities that link them to others; conversely, self -
competence reflects people s evaluations of themselves in
terms of qualities that distinguish the self from others. If
collectivistic cultures value the group over the individual,
then it makes sense that people with interdependent selves
derive self - esteem primarily from those qualities that make
them good group members, such as loyalty (Sedikides,
Gaertner, & Vevea, 2005; but see Heine, Kitayama, &
Hamamura, 2007, who raise serious questions regarding the
evidentiary basis of this assertion). It is also no surprise
that people with independent selves tend to derive self -
esteem primarily from the qualities that make them stand
out from others. Within their cultural contexts, each way of
constructing self - esteem makes adaptive sense.
Mirroring these cultural differences are gender differ-
ences in the extent to which people exhibit interdependent
versus independent selves. Whereas girls are often social-
ized to prioritize the qualities that align them to others,
boys are taught to prioritize the qualities that distinguish
and differentiate them from others (e.g., Spence, Deaux, &
Helmreich, 1985). Accordingly, women tend to develop
more interdependent selves, and men tend to develop more
independent selves (Cross, Bacon, & Morris, 2000; Cross &
Madson, 1997). Moreover, women tend to link their self -
esteem to their relational qualities, whereas men link
their self - esteem to their independent qualities (Josephs,
Markus, & Tafarodi, 1992).
Intrapsychic Origins of the Self and Identity
Self - Perception
Like outside observers, people sometimes learn about them-
selves by observing their own behavior, and the situation in
which it occurs, and then inferring their underlying attitudes
and dispositions (Bem, 1972). According to self - perception
CH16.indd 601CH16.indd 601 10/22/09 2:50:06 PM10/22/09 2:50:06 PM
602 Self and Identity
theory, people are most likely to acquire self - knowledge
through observation of their own behavior when their self -
knowledge is weak, ambiguous, or difficult to interpret.
Moreover, many attributional principles that guide people s
perceptions of others also operate when they infer their own
dispositions via self - perception. Thus, for example, the self -
knowledge that people acquire through self - perception of
their behavior is less certain to the extent that multiple pos-
sible causes exist for that behavior (Kelley, 1971). As such,
an overjustification effect occurs when people lose intrinsic
motivation to perform a certain activity because extrinsic
rewards convince them that they are performing the behav-
ior merely to obtain the rewards (for reviews, see Deci,
Koestner, & Ryan, 1999; Lepper, Henderlong, & Gingras,
As noted earlier, research on the inaccuracy of reflected
appraisals casts doubt on the notion that others truly serve
as looking glasses in which people see the self reflected
(Cooley, 1902; Mead, 1934). On this point, self - perception
theory may provide an answer. Instead of accurately per-
ceiving themselves through the eyes of others, people may
engage in self - perception of their own behaviors and then
attribute these perceptions of the self to others (Kenny &
DePaulo, 1993). If so, then self - perceptions of behavior
may play an important role in two related, but distinct,
self - knowledge processes: First, self - perceptions are used
as a basis for inferring one s own internal qualities, traits,
attitudes, and the like, and second, self - perceptions are
used as a basis for inferring how others view the self.
In an interesting twist on self - perception theory, Goldstein
and Cialdini (2007) proposed that people can learn about
their own internal states at least their temporary ones by
observing the behavior of others with whom their identities
are merged. The logic is that, when viewing a close other
perform a behavior, people vicariously infer novel informa-
tion about the self, resulting in a change in the self - concept.
Although the notion of vicarious self - perception is rela-
tively new, it fits well with the general idea that people mod-
ify the self to achieve greater congruence with the presumed
beliefs, self - views, and attitudes of close others (Baldwin,
Carrell, & Lopez, 1990; Hinkley & Andersen, 1996).
Unlike self - perception, which involves observing one s
overt behaviors and using them to infer one s internal
qualities, introspection involves deliberate attempts to
achieve self - knowledge by directing attention “ inward. ”
While introspecting about the self seems like a fairly
obvious route to self - knowledge, research suggests that
people spend surprisingly little time (about 8% of total
thoughts) reflecting on themselves (Csikszentmihalyi &
Figurski, 1982).
When people do engage in introspection, the fruits
of their efforts are sometimes rather bitter. For example,
introspection about the reasons behind one ’ s attitudes,
behaviors, and feelings is likely to produce inaccurate self -
knowledge (Wilson, Laser, & Stone, 1982). As a conse-
quence, when people introspect about the reasons behind
their feelings, they sometimes change their feelings to
match the reasons they generate (Wilson & Kraft, 1993),
which can lead them to make decisions that they later
regret (Wilson et al., 1993).
Introspection can also lead people to compare their cur-
rent achievements and behaviors with their beliefs about
how they should or ought to be, which can create dis-
comfort if there is a disparity (Duval & Wicklund, 1972;
Higgins, 1987). Falling short of internal standards can pro-
duce painful feelings of shame for those who are prone to
attribute their shortcomings to their whole self (Tangney &
Dearing, 2002). Some propose that this state of self - aware-
ness can be so troubling that people go to great lengths to
escape the self through activities such as drinking, drug
use, binge eating, and even suicide (Baumeister, 1991).
Introspection can also have desirable effects, however.
Introspecting about who one is, for example, can produce
accurate self - knowledge if people have sufficient cogni-
tive resources (Hixon & Swann, 1993). Similarly, when
self - reflection reveals that one meets or exceeds one ’ s
standards, positive feelings result (Greenberg & Musham,
1981; Silvia & Abele, 2002). Moreover, people are more
likely to behave in line with their personal values when
in a state of self - awareness, suggesting that introspection
can promote adaptive (or at least self - consistent) self -
regulation (e.g., Beaman, Klentz, Diener, & Svanum, 1979;
Gibbons, 1978).
Experiencing the Subjective Self
Yet another source of self - knowledge is the continual,
ever - changing “ stream of consciousness ” about which
James (1890/1950) wrote the spontaneous thoughts,
feelings, and reactions that constitute the self - as - subject
(or I ). Experiencing the subjective self differs from both
self - perception and introspection in fundamental ways. For
instance, whereas both self - perception and introspection
involve reflection on the self, experiencing the subjective
self can involve an outward focus, a full engagement in
the moment that draws attention away from the self (e.g.,
Csikszentmihalyi, 1990). Moreover, whereas behaviors
provide the raw material for self - perception processes,
private thoughts and feelings provide the raw material for
subjective self processes.
Research suggests that people rely more on their sub-
jective experiences than on their overt behaviors when con-
structing self - knowledge (Andersen, 1984; Andersen & Ross,
CH16.indd 602CH16.indd 602 10/22/09 2:50:06 PM10/22/09 2:50:06 PM
Motivational Properties of the Self 603
1984), and when people encounter others who share their
subjective reactions to the world, they feel especially
attracted to them (Pinel, Long, Landau, Alexander, &
Pyszczynski, 2006). Apparently, when people sense that
they and others perceive the world through the same psy-
chological lens, their confidence in the validity of their
own visions of reality is reinforced. Such I - sharing may
constitute a powerful antidote to the problem of existential
Accuracy of Self - Knowledge
Humans routinely assert that they know themselves. Most
people assert that they know themselves better than oth-
ers do (Pronin, Kruger, Savitsky, & Ross, 2001). And it
is not just laypeople who harbor this belief in the fidelity
of self - knowledge: Behavioral scientists harbor this same
conviction. For example, in 2003, 70% of the studies pub-
lished in a leading personality psychology journal (Journal
of Research in Personality) relied on self - reports as the
index of personality (Vazire, 2006). Such confidence in
the veracity of self - knowledge is challenged by a spate of
demonstrations, mostly conducted in the laboratory, indicat-
ing that some aspects of self - knowledge are simply wrong
(e.g., Epley & Dunning, 2006; Wilson & Gilbert, 2003;
for non - laboratory studies, see Gosling, John, Craik, &
Robins, 1998; Vazire & Mehl, 2008). At least two potential
sources of such errors exist. In the tradition of Freud, many
have argued that people unconsciously or consciously sup-
press unwanted thoughts and feelings. Although intrigu-
ing, these processes have been notoriously difficult to
demonstrate empirically (for reviews, see Erdelyi, 1974,
1993). The other source of errors in self - knowledge is that
people simply lack access to many processes that give rise
to self - relevant behaviors, which throws a wrench into
the process of introspection. As a result, people err when
asked about the causes of their actions (Nisbett & Wilson,
1979), their attitudes (e.g., Galdi, Arcuri, & Gawronski,
2008), or their future emotional reactions (e.g., Wilson &
Gilbert, 2003; for a review, see Wilson & Dunn, 2004). In
the end, such errors could undermine the veracity of peo-
ple s representations of themselves.
Such sharp clashes between people s intuitions and the
results of systematic research invariably led theorists and
researchers to ask which source — individuals or research
findings really is more accurate. Such questions regard-
ing the validity of self - knowledge are particularly vexing
due to the criterion problem: It is easy to say whether or
not someone has brown eyes or even a pleasant smile, but
the problem of assessing the validity of self - knowledge is
knotty indeed (e.g., Kruglanski, 1989; Swann, 1984). For
example, when it comes to high - level, global self - views
such as worthwhile, questions of accuracy are impossible
to answer definitively because choosing one or more crite-
ria is inherently subjective. More specific self - views such
as “ extroverted ” or “ fastidious, ” however, have relatively
clear empirical referents. Empirical assessments of accu-
racy have therefore focused on lower - level self - views.
Some of the most telling studies of accuracy of self -
knowledge involve comparing the capacity of people s
self - ratings and the ratings of peers to predict some objec-
tive outcome, such as the ratings of observers, or some
naturally occurring outcome, such as success in military
training. Different methodologies lead to different conclu-
sions, but the studies using the strongest methodologies
generally conclude that well - acquainted observers are
at least as accurate as are people themselves. The results
of one recent study (Vazire & Mehl, 2008) suggest that
the specific content of criterion behaviors may be criti-
cal. These researchers compared the ability of individuals
and acquainted others to predict naturally occurring behav-
iors over a 4 - day period. Findings revealed that individuals
were more accurate in predicting some of their behaviors
(e.g., deliberate behaviors such as arguing) but acquain-
tances were more accurate in predicting other behaviors
(e.g., spontaneous behaviors such as talking one on one).
Although it is too early to draw definitive conclusions
from this work, it is probably safe to say that people
are accurate about the self within some arenas but that
others particularly close others may be better able to
predict people s reactions within other arenas, especially
when people s own wishes and desires compromise their
objectivity. Therefore, some analysts have concluded that
accurate self - knowledge is best obtained not from intro-
specting but instead from consulting with friends and
acquaintances or observing one s own behavior (Wilson &
Dunn, 2004).
My thinking is first and last and always for the sake of my
William James (1890/1950, p. 333)
Although James believed that the major function of self -
knowledge was to guide action, this proposition has proven
surprisingly controversial in certain quarters. Indeed, some
prominent thinkers have gone as far as to take the oppo-
site position, suggesting instead that self - knowledge is
an epiphenomenal product of social relations that has no
causal status. In self - perception theory, for example, Daryl
Bem (1972) proposed that the flow of influence between
behavior and the self is unidirectional: We infer who we
by observing our own behavior and the conditions under
CH16.indd 603CH16.indd 603 10/22/09 2:50:06 PM10/22/09 2:50:06 PM
604 Self and Identity
which it unfolds, but self - knowledge has no impact on
subsequent action. Theorists from different theoretical tra-
ditions have echoed Bem s assertions. Group researcher
John Turner, for example, contended that personal self -
views are there to be explained, not in themselves expla-
nations (Turner, Reynolds, Haslam, & Veenstra, 2006,
p. 25). Such pockets of skepticism notwithstanding, it
is fair to say that today the self s motivational proper-
ties are widely accepted among most students of the self
(e.g., Higgins & Pittman, 2008; Pittman & Zeigler, 2007).
Testimony to contemporary enthusiasm for motivational
processes is offered by the burgeoning literature on self -
enhancement processes.
Self - Enhancement Motive
The self - enhancement motive has been defined in many
ways, but it is most commonly conceptualized as a desire
to maximize the positivity of one s self - views (e.g.,
Leary, 2007). The notion that people prefer and seek self -
enhancement is enormously popular, with one landmark
statement of the viewpoint Taylor and Brown s (1988)
literature review garnering more than 2,200 citations.
The popularity of the self - enhancement motive is easy to
understand. After all, the notion that people want positive
evaluations seems like a relatively simple and readily test-
able argument. In addition, and perhaps more importantly,
the notion that self - enhancement is a basic human motive
lends scientific credibility to the cultural assumptions and
normative behaviors of the researchers conducting the
research, most of whom happen to be Westerners.
Little wonder, then, that self - enhancement assumptions
have found their way into an extraordinarily wide range
of contemporary theories (e.g., Hoyle, Kernis, Leary, &
Baldwin, 1999; Sedikides & Gregg, 2003). Purported
evidence for the theory abounds. One popular phenom-
enon is the “ better - than - average effect, ” wherein most
people assert that they are above average, a mathematical
impossibility. For example, college students overwhelm-
ingly report that they are above - average drivers (Svenson,
1981). Ironically, when told of the existence of such posi-
tive illusions, people claim that they are less susceptible
to them than most others are (Pronin, Gilovich, & Ross,
2004). At this juncture, dozens of such self - enhancing
illusions have been reported in the research literature (for
reviews, see Helweg - Larsen & Shepperd, 2001; Sedikides &
Gregg, 2008).
Yet, in recent years, researchers have begun to voice
reservations about social psychology s motivational cen-
terpiece. Some findings suggest that some specific effects
reflect the failure of participants to comprehend fully what
they are being asked when they are encouraged to estimate
their standing relative to others. Kruger and Dunning
(1999), for example, had participants estimate their perfor-
mance on dimensions such as humor, grammar ability, and
logical reasoning. Regardless of their actual performance,
participants estimated that they scored in the 60th to 70th
percentile. The result was that low scorers overestimated
their performance but high scorers under estimated their
performance. Similarly, other findings reinforce the notion
that people s performance estimates should not be taken at
face value. Indeed, in estimating performance relative to
average performance, people seem to rely on a heuris-
tic that leads them to rate everyone including unknown
strangers slightly above average (Klar & Giladi, 1997).
To be sure, some researchers (Alicke, Klotz,
Breitenbecher, Yurak, & Vredenburg, 1995) have shown
that people display self - enhancing performance estimates
even when asked to compare themselves against specific
others. Nevertheless, evidence that such biases are stronger
when people have positive self - views (Brown, 1986) raises
a further issue regarding the mechanism that gives rise to
these effects. That is, almost all studies that are taken as evi-
dence of self - enhancement suffer from a serious potential
confound. Researchers have been aware of this confound
for some time but have failed to appreciate its full implica-
tions. For example, in their review of the positive illusions
literature, Taylor and Brown (1988) noted the following:
One caveat, however, deserves mention. A considerable amount
of the research cited demonstrates that people solicit and receive
self - confirming feedback, not necessarily positive feedback. For
example, a woman who thinks of herself as shy may seek and
receive feedback that she is (see Swann, 1983). At first, these
results may seem contradictory with the position that social feed-
back fosters positive self - conceptions, but in fact, they are not.
Because most people think well of themselves on most attributes,
confirming feedback is typically positive feedback. (p. 202)
Taylor and Brown (1988) were alluding to an ambiguity
imposed by almost all research on self - enhancement having
been conducted on unselected participants, roughly 70% of
whom have positive self - views (Diener & Diener, 1995).
Therefore, evidence of self - enhancement may reflect, to an
unknown degree, a desire for confirmation of chronic self -
views (e.g., Kwan, John, Kenny, Bond, & Robins, 2004;
Kwan, John, Robins, & Kuang, 2008).
To illustrate the import of Taylor and Brown s (1988)
caveat, consider one of the most robust findings in the self -
enhancement literature: the tendency for people to make
self - serving attributions (e.g., Mezulis, Abramson, Hyde, &
Hankin, 2004; Miller & Ross, 1975). When researchers
conducted a parallel study in which they measured the self -
views of participants, they discovered that those with posi-
tive self - views displayed the self - serving pattern but those
CH16.indd 604CH16.indd 604 10/22/09 2:50:07 PM10/22/09 2:50:07 PM
Motivational Properties of the Self 605
with negative self - views displayed precisely the opposite
pattern, perceiving negative evaluators to be more accurate
than positive evaluators (e.g., Swann, Predmore, Griffin, &
Gaines, 1987). This pattern of data clearly indicates that a
desire to confirm, rather than enhance, self - views under-
lies participants ’ responses.
Another finding that has widely been attributed to
self - enhancement strivings is the tendency for people
to selectively recall positive feedback about themselves
(Sanitioso & Wlodarski, 2004). Again, when researchers
measured the self - views of participants in such studies,
they discovered that their responses seemed to be driven by
a desire for self - confirmation. That is, only those with posi-
tive self - views preferentially recalled positive feedback;
people with negative self - views displayed the opposite ten-
dency, recalling more negative than positive feedback (e.g.,
Story, 1998; Swann & Read, 1981). Similarly, people s ten-
dency to define virtues as qualities they possess and vices as
qualities they lack (Dunning, Perie, & Story, 1991) appears
to be due primarily to people who have relatively positive
views of themselves (Beauregard & Dunning, 2001).
The upshot of such findings is simple: Although self -
enhancement strivings seem to be pervasive, the motive that is
driving such strivings may be self - confirmation rather than
self - enhancement. And even if it turns out that such puta-
tive self - enhancement strivings among people with positive
self - views are indeed compelled by a self - enhancement
motive, self - enhancement theory still cannot account for
the responses of those with negative self - views. This is
problematic for a propensity that is alleged to represent a
basic human motivation. If the self - enhancement motive
is so basic, why has this news escaped the attention of the
roughly 30% of the population who possess negative self -
views? The nonenhancing responses of people with negative
self - views are particularly perplexing when we consider that
when most human needs are frustrated people redouble their
efforts to gratify those needs. Instead, people with nega-
tive self - views actually embrace negative evaluations (for a
review, see Swann, Chang - Schneider, & Angulo, 2007).
Even if evidence of self - enhancement from participants
in the West could be assumed to reflect a desire for self -
enhancement, growing evidence indicates that such striv-
ings do not generalize across cultures. Japanese people,
for example, posses a relatively strong desire to be viewed
as modest, and this desire causes them to eschew positive
evaluations at times. In addition, a relatively strong interest
in self - improvement among the Japanese seems to make
them more receptive to negative feedback than Americans
are (e.g., Heine, Kitayama, & Lehman, 2001). Japanese
participants are also less unrealistically optimistic about
their futures when compared with American participants
(Chang, Asakawa, & Sanna, 2001).
Some counter such contentions by arguing that the East
West difference in behaviors related to self - enhancement
reflects a difference not in the strength of the self - enhance-
ment motive itself but only in how people pursue this
motive (e.g., Kurman, 2003; Yik, Bond, & Paulhus, 1998).
Japanese are modest, the argument goes, as a means of
attaining social acceptance, which is considered self -
enhancing in Japanese culture (e.g., Sedikides, Gaertner, &
Toguchi, 2003). Although this tactic may appear to rescue
self - enhancement theory from disconfirmation, it does so at
the cost of redefining self - enhancement from a theory about
a preference for positive evaluations to a theory about a
desire for social acceptance or communion. As we argue later,
the two motives are quite different in form, structure, and
consequence. Moreover, to the best of our knowledge,
no one ever contended that the Japanese eschewed social
acceptance or communion. To the contrary, it would seem
that, if anything, Japanese individuals are especially inter-
ested in social acceptance and communion (Kitayama,
Markus, Matsumoto, & Norasakkunkit, 1997).
A final concern grows out of mounting evidence that
self - enhancement strivings can degrade the quality of
people s relationships and even their well - being (Colvin,
Block, & Funder, 1995; Paulhus, 1998; Robins & Beer,
2001; for a review, see Crocker & Park, 2004). Such evi-
dence leads one to wonder why the pursuit of a suppos-
edly basic human motivation should be associated with
dysfunctional outcomes.
There are, then, reasons to ask whether the superficial
charms of self - enhancement theory and research are out-
weighed by some fundamental difficulties with the theory
and the data that ostensibly support it. We think that
the answer to this question is yes, and we accordingly pro-
pose an alternative approach to self - related motivation in
the next section. Our goal is not to banish self - enhancement
theory. Rather, we seek to partition it into two motives that
we perceive as more viable, both logically and empiri-
cally. We then add an additional motive to the mix. The
result is three broad self - motives that serve to guide most
identity - relevant functioning.
Before turning to our three - motive scheme, let us add
three caveats. Our goal is to identify three broad self motives.
One can surely make fine - grained distinctions among vari-
ants of each of our motives, and it is no doubt useful to do so
in certain contexts. In addition, our scheme is limited to iden-
tity - related motives and hence excludes biological motives
(e.g., sex), as well as other motives that have appeared in
formulations that are broader in scope (e.g., Fiske, 2004).
Finally, although we believe that this scheme offers a useful
lens for viewing the self literature, we see its role as lim-
ited to just that; we make no claim to having discovered the
motives that underlie all human social behavior.
CH16.indd 605CH16.indd 605 10/22/09 2:50:07 PM10/22/09 2:50:07 PM
606 Self and Identity
Tripartite Motivational Approach
There is no doubt that people prefer and enjoy positive over
negative evaluations when they perceive that such evalua-
tions are appropriate and deserved (e.g., Swann, Krull, &
Pelham, 1989). Nevertheless, we suggest that obtain-
ing positive evaluations cannot be an end in itself (e.g.,
Leary, 2007). Like paper currency, positive evaluations
are valuable more for what they symbolize than for their
intrinsic properties. In particular, positive evaluations are
valued because they are markers of one s social worth (and
thus satisfy a desire for communion) and competence
(and thus satisfy a desire for agency). From this vantage
point, motives for communion (belonging and interpersonal
connectedness) and agency (autonomy and competence)
may be responsible for the phenomena that researchers
have attributed to self - enhancement strivings.
The desires for communion and agency have a promi-
nent history in the psychological literature. The needs for
communion and agency are assumed to underlie many
aspects of personality and social behavior (Baumeister &
Leary, 1995; Wiggins & Broughton, 1991), and theories of
optimal functioning emphasize the importance of meeting
both needs (e.g., Ryff, 1989). In the domain of attitudes,
researchers suggest that constructs similar to communion
and agency (i.e., warmth and competence) represent the
two basic dimensions of attitudes (e.g., Abele & Wojciszke,
2007; Fiske, Cuddy, & Glick, 2007; Fiske et al., 2002;
Judd, James - Hawkins, Yzerbyt, & Kashima, 2005). Of
even more relevance here, communion and agency corre-
spond with the dual forms of self - esteem discussed ear-
lier in this chapter (e.g., Franks & Marolla, 1976; Gecas,
1971). Using Tafarodi and Swann s (2001) terms, the need
for communion motivates a desire for self - liking and the
need for agency motivates a desire for self - competence .
Replacing the self - enhancement motive with the com-
munion and agency motives evades the reservations raised
earlier regarding self - enhancement. For example, whereas
the self - enhancement formulation requires that people
with both positive and negative self - views prefer positive
evaluations over negative ones, our formulation does not.
Therefore, evidence that people with negative perceptions
of their social worth and competence fail to embrace posi-
tive evaluations of themselves on these dimensions (e.g.,
Bosson & Swann, 1999) does not challenge the assumption
that such individuals want to enjoy feelings of communion
and agency. Rather, communion and agency motives theo-
retically encourage people to achieve actual communion
and agency rather than merely seeking positive evaluations
for their own sake. Finally, although evidence indicates
that people from Southeast Asia are less inclined to self -
enhance than Westerners (e.g., Heine et al., 1999), the needs
for communion and agency appear to be pancultural.
In addition to the desires for communion and agency,
we propose a third motive: the desire for coherence. In
our usage, coherence encompasses feelings of regularity,
predictability, meaning, and control. Coherence is distinct
from consistency, which emerges whenever any two psy-
chological elements follow logically from each other (e.g.,
Festinger, 1957). Thus, to maintain consistency between
two elements, such as a behavior and a related identity,
one can change either element. In contrast, coherence is
a special case of consistency that refers specifically to the
degree of correspondence between one s enduring self -
concept and the other elements in one s psychological uni-
verse (English, Chen, & Swann, 2008).
In some respects, the coherence motive may be even more
fundamental than the desires for communion and agency
(Guidano & Liotti, 1983; Popper, 1963). After all, those who
lack the conviction that their knowledge system offers coher-
ent and trustworthy insights into the world around them are
unable to evaluate evidence of social worth or competence
because they are not confident that they know such evi-
dence when they see it. If the self - views serve as the lenses
through which people perceive reality, incoherence degrades
the vision of reality that these lenses offer. Deprived of a
clear vision of reality, people have little means of knowing
whether what they see faithfully reflects reality. Indeed, if
people completely lose faith in the veracity of their knowl-
edge system, their sense of self begins to unravel and they fall
into a state of disintegration anxiety (Kohut, 1971). Deprived
of stable self - knowledge, people feel that they have no basis
for knowing how to act, and guiding action is the primary
objective of thinking in the first place (James, 1890/1950).
Consider evidence that people who experienced events
that bolstered their feelings of communion or agency, or
both, also suffered deficits in mental and physical health
if those events challenged their need for coherence (e.g.,
Swann et al., 2007). This research was based on the
assumption that experiences that challenge one s enduring
self - views are stressful enough that, over time, they may
actually be physically debilitating. The first two studies
(Brown & McGill, 1989) examined the impact of posi-
tive life events on the health outcomes of people with low
and high self - esteem. Positive life events (e.g., improve-
ment in living conditions or getting a high grade) pre-
dicted increases in health among high self - esteem people
but decreases in health among people low in self - esteem.
A more recent study (Shimizu & Pelham, 2004) extended
these results by demonstrating that the effects replicated
even while controlling for negative affectivity (thus under-
mining the rival hypothesis that negative affect influenced
both self - reported health and reports of symptoms).
But if the desire for coherence may sometimes override
the desires for communion and agency, we do not mean
to imply this is always the case. Whereas some degree
CH16.indd 606CH16.indd 606 10/22/09 2:50:07 PM10/22/09 2:50:07 PM
Motivational Properties of the Self 607
of coherence may be necessary for people to effectively
pursue their communion and agency needs, the opposite
may also be true.
At first blush, it might seem that the three - motive con-
ceptualization overlooks several motives that other theo-
rists have deemed important. For example, whereas some
propose motives that are somewhat overlapping with ours
(acceptance, status, and meaning; Hogan & Shelton, 1998),
others differentiate between motives that we instead classify
together (autonomy and competence; Deci & Ryan, 1995)
or introduce other motives into the mix (self - assessment;
Sedikides & Strube, 1997). Still others propose a six - motive
scheme (i.e., self - esteem, continuity, distinctiveness, belong-
ing, efficacy, and meaning; Vignoles, Regalia, Manzi,
Golledge, & Scabini, 2006) that appears broader than our
approach. Nevertheless, our approach incorporates most of
these other motives if one allows that several of the motives
distinguished by previous researchers in fact reflect the same
core motive with different criteria used to gauge its gratifica-
tion. In what follows, we consider several examples of these
The communion motive is designed to maximize feelings
of acceptance, belongingness, and social worth. Humans
evolved in the context of small, close - knit groups, and the
need for communion remains a constant theme in the con-
struction and maintenance of the self and identity (Bowlby,
1969; McAdams, 1989). On a biological level, evidence sug-
gests that people require a minimum number of close, posi-
tive, interpersonal connections to thrive. Those who lack such
connections exhibit relatively poor physical health, weak-
ened immune functioning, and even higher mortality rates
(House, Landis, & Umberson, 1988; Uchino, Cacioppo, &
Kiecolt - Glaser, 1996). On a psychological level, people
who lack positive affiliations with others experience trou-
bling feelings of loneliness (Archibald, Bartholomew, &
Marx, 1995; Newcomb & Bentler, 1986), while those with
rich social networks report higher levels of happiness and
life satisfaction (Diener, Suh, Lucas, & Smith, 1999). So
central is communion to humans existence that Baumeister
and Leary (1995) deemed the need to belong a fundamen-
tal human motive. Although researchers generally agree
that the criterion for this motive should be the appraisals
of others, the precise nature of these appraisals has varied.
Whereas advocates of the self - enhancement motive argue
for the importance of positive evaluations (e.g., Murray,
Holmes, & Griffin, 1996; Sedikides & Strube, 1997), others
emphasize feelings of connectedness or belonging (e.g.,
Baumeister & Leary, 1995).
In recent years, the desire for communion assumed
center stage in one prominent approach: Leary s socio-
meter theory (Leary & Downs, 1995). Leary and colleagues
assumed that people are profoundly invested in estimating
the extent to which they are valued by interaction partners,
group members, and relationship partners. Signs of rejec-
tion trigger an alarm reaction that is punctuated by a loss
of self - esteem. Thus, self - esteem is a psychological fuel
gauge that is sensitive to variations in perceived inclusion.
Support for sociometer theory comes from evidence that
manipulations that convey rejection, disapproval, or disin-
terest on the part of others tend to lower participants state
self - esteem (e.g., Leary, Tambor, Terdal, & Downs, 1995;
Nezlek, Kowalski, Leary, & Holgate, 1997). Similarly,
field studies demonstrate that self - esteem dips when
people experience rejection (Murray, Griffin, Rose, &
Bellavia, 2003) and ostracism (e.g., Williams, 2001).
Moreover, longitudinal research shows that perceived rela-
tional value is linked to changes in self - esteem over time
(Srivastava & Beer, 2005).
Sociometer theory has performed the useful service of
focusing attention on the utility of people s efforts to forge
connections with others. From an evolutionary perspec-
tive, positive evaluations and the feelings of self - esteem
that they foster are useful not because of their intrinsic
value but because they are markers of acceptance within a
larger social group whose protection and shared resources
were vital to humans survival. Conceivably, the argu-
ment could be taken even further, such that all self - views
serve as indices of the manner in which we are perceived
by others. Accepting this broader interpretation, however,
could lead one to question the novel contribution of soci-
ometer theory, because the notion that self - views provide
a window into the perceptions of others has been around
for more than a century (e.g., Cooley, 1902). These and
other considerations led some theorists to raise troubling
indictments of the theory (e.g., Pyszczynski, Greenberg,
Solomon, Arndt, & Schimel, 2004). A further limitation
of the model, however, is that although people are aware of
how others perceive them in general, they seem fairly inept
at discriminating the appraisals of specific other individu-
als (e.g., Kenny & DePaulo, 1993). Therefore, the self -
esteem fuel gauge sometimes offers faulty information.
In addition, it is clear that self - esteem tracks more than
simply social acceptance. For example, self - esteem seems
acutely sensitive to indicators of agency.
The agency motive is theoretically designed to maximize
feelings of autonomy (e.g., self - determination) and compe-
tence. The need for agency begins to guide behavior early
in life, such as when infants strain to escape their caregivers
so that they can explore and manipulate the world around
them (Bowlby, 1969). Later in life, people s sense of effi-
cacy forms a core component of personality (Bandura, 1991)
and contributes to psychological well - being (Ryff, 1989).
CH16.indd 607CH16.indd 607 10/22/09 2:50:07 PM10/22/09 2:50:07 PM
608 Self and Identity
In the most general case, the agency motive encourages
people to strive for successful performance of valued activ-
ities. A special case of agency strivings emerges when peo-
ple seek to improve themselves (e.g., Heine, Kitayama, &
Lehman, 1999; Taylor, Neter, & Wayment, 1995). Self -
improvement strivings are presumably initiated when peo-
ple focus on negative aspects of the self that they need to
change to meet personal or social standards.
Some theorists argue that the prevalence of the self -
improvement motive is strongly influenced by cultural
factors. In particular, researchers suggest that East Asian
cultures place an emphasis on self - improvement at the
expense of self - enhancement (e.g., Heine et al., 1999;
Kitayama & Markus, 1999). The rationale for this conten-
tion rests largely on the relative levels of individualism ver-
sus interdependence in East Asian versus Western cultures.
Individualistic cultures (e.g., those in United States and
Australia) place a premium on independence and therefore
emphasize individual needs, goals, and rights. In contrast,
collectivistic cultures (e.g., those in East Asia and Latin
America) emphasize ingroup goals, needs, and obligations
and thus strongly value interdependence. In such cultures,
it is particularly crucial to attend to others perspectives
so as to meet the expectations of ingroup members and
maintain interpersonal harmony (Heine et al., 1999). This
greater sensitivity to social standards presumably explains
the relative eagerness of East Asians to improve them-
selves to meet others expectations.
Widespread support exists for the notion that people have a
deep - seated need for psychological coherence (Guidano &
Liotti, 1983; Heine, Proulx, & Vohs, 2006). As Popper
(1963) contended, infants are born with a predisposition to
identify patterns and regularities. Without this predisposi-
tion, they would be incapable of learning:
The expectation of finding a regularity . . . connected with an
inborn propensity to look for regularities, or with a
need to
find regularities. . . . This “ instinctive ” expectation of find-
ing regularities . . . is logically a priori to all observational
experience, for it is prior to any recognition of similarities . . .
and all observation involves the recognition of similarities (or
dissimilarities). (pp. 47 – 48)
The coherence construct has gone by several labels,
including security in Maslow s (1954) motivational hierar-
chy, need for closure in Kruglanski s (1990) formulation,
need for structure in Neuberg and Newsom s (1993) model,
and meaning according to Hogan and Shelton (1998).
Evidence of the coherence need has surfaced in many cul-
tures in addition to North America, including countries in
Eastern and Western Europe, as well as in Australia, Korea,
and China, suggesting that it is not limited to a specific
cultural milieu (e.g., Heine et al., 2006).
Moreover, two self theories feature a desire for coherence
as the primary motivational mechanism: self - assessment
(e.g., Trope, 1983) and self - verification (Swann, 1983).
Each theory focuses on different criteria for assessing the
coherence of self - related information. Self - assessment
theorists have suggested that when people are uncertain
of their self - views they seek relatively objective, diagnos-
tic information about themselves. A series of laboratory
studies provides support for this general approach (e.g.,
Brown, 1990; Strube, 1990). While the issue of the relative
potency of self - assessment strivings remains to be deter-
mined, its range of application is limited to self - views of
which people are uncertain. This is an important limita-
tion, for people tend to be highly certain of the qualities
that they care about. Such highly certain self - views have
been the province of various self - confirmation theories
(e.g., Lecky, 1945; Secord & Backman, 1965), the most
recent of which is self - verification theory (Swann, 1983).
Self - verification theory assumes that, out of a desire for
social worlds that are coherent and predictable, people
want others to see them as they see themselves. This desire
can be understood on both epistemic and pragmatic levels.
Epistemically, receiving self - verifying evaluations reas-
sures people that their self - views accurately reflect social
reality and that they can count on their self - views to guide
their behavior. Pragmatically, self - verifying appraisals
signal to people that others hold appropriate expectations
of them and that their interactions will therefore proceed
smoothly. Among people with positive self - views, the
desire for self - verification works with the desires for com-
munion or agency, as all of these motives encourage people
who view themselves positively to embrace positive evalu-
ations. Among people with negative self - views, however,
self - verification theory predicts that they will seek nega-
tive evaluations (e.g., Swann et al., 1989). Self - verification
theory thus makes divergent predictions for people with
enduring positive versus negative self - views.
One focus of research has been on the variables that
determine when people will prioritize self - verification
over the competing desire for positive feedback. The
desire for self - verification prevails (e.g., people with nega-
tive self - views prefer and seek negative evaluations) when
the self - view is firmly held (i.e., certain and important;
Pelham & Swann, 1994; Swann & Pelham, 2002) or extreme
(Giesler, Josephs, & Swann, 1996), when the relationship
is relatively enduring (Campbell, Lackenbauer, & Muise,
2006; Swann, De La Ronde, & Hixon, 1994), and when
people have the cognitive resources needed to compare
the feedback against a relevant mental self - representation
(Hixon & Swann, 1993; Swann et al., 1990). In addition,
CH16.indd 608CH16.indd 608 10/22/09 2:50:08 PM10/22/09 2:50:08 PM
Motivational Properties of the Self 609
challenges to a self - view will trigger compensatory activity
that shores up that self - view or some other component of
the self - system (Swann & Hill, 1982; Swann, Wenzlaff, &
Tafarodi, 1992). Interestingly, such compensatory reac-
tions are symmetrical with respect to self - esteem; just as
high self - esteem people strive to reaffirm their positive
self - views in the wake of negative feedback, those with
low self - esteem strive to reaffirm their negative self - views
in the wake of positive feedback (Swann & Read, 1981;
Swann, Wenzlaff, & Tafarodi, 1992). Finally, whereas the
foregoing research focused on personal self - views, other
research extended the findings to collective self - views
(Chen et al., 2004; Lemay & Ashmore, 2004) and group
identities (G ó mez, Seyle, Huici, & Swann, in press).
Researchers have identified several distinct strategies
of self - verification. For example, people gravitate toward
self - verifying environments, such as interaction partners
who see them congruently and who are apt to provide them
with self - verification (e.g., Robinson & Smith - Lovin,
1992; Swann, Wenzlaff, Krull, & Pelham, 1992). Once
in a given setting, people display identity cues (i.e., overt
signs of who they are, such as clothing or office d é cor)
that communicate their identities to others (Gosling, Ko,
Mannarelli, & Morris, 2002). In addition, people can elicit
self - verifying reactions by behaving in ways that evoke
such responses (Swann & Hill, 1982; Swann & Read,
1981). And if these strategies fail to evoke self - verifying
evaluations, people can distort nonverifying feedback
through preferential attention and recall (Swann & Read,
1981), construe the feedback in ways that make it fit with
their enduring self - views (Swann et al., 1987), or even
leave the relationship (Swann & Pelham, 2002).
Hybrid Theories
Elements of the three motives described above can be
found in two major social psychological theories of
the self, self - affirmation theory and terror management
theory. Self - affirmation theory (Steele, 1988) combines
elements of the agency and coherence motives (see also
self - esteem maintenance theory, Tesser, 1988). This
theory focuses on how people react when they encoun-
ter challenges to their positive self - views specifically,
challenges to the sense of being a moral, adaptive, and
capable person. The theory assumes that the self - system
is composed of many interrelated parts that interact with
one another. As a result, shoring up one component of
the system can buttress other components against threats.
Thus, whereas people ordinarily respond defensively
when they receive information that challenges a positive
self - view, these defensive reactions can be attenuated by
inoculating them with positive feedback.
Interestingly, people strive to affirm their positive self -
views in ways that may have little to do with the nature of
the self - threat (Aronson, Cohen, & Nail, 1999; Tesser &
Cornell, 1991). In some early research, the researchers
used a cognitive dissonance paradigm to show that the
self - threat that arises from counterattitudinal behavior can
be alleviated by having participants first affirm an impor-
tant, self - relevant value in a domain unrelated to that of the
dissonant behavior (Steele & Liu, 1983). In a similar vein,
later studies indicated that people were more willing to
examine useful but potentially threatening feedback about
themselves if they first enjoyed success on an unrelated
task (Trope & Pomerantz, 1998).
One especially fruitful line of research was designed to
explore the health implications of self - affirmation theory.
The results of one study indicated that people were more
willing to examine potentially threatening information
related to AIDS prevention after an important but unre-
lated value was affirmed (e.g., Sherman, Nelson, & Steele,
2000). In this and related studies, researchers appear to
have uncovered an effective strategy for neutralizing the
defensive reactions that have long impeded efforts to enlist
the compliance of people who engage in risky behav-
iors (e.g., Jemmott, Ditto, & Croyle, 1986; Liberman &
Chaiken, 1992).
Within our scheme, another hybrid approach is terror
management theory (Solomon, Greenberg, & Pyszczynski,
2004). At its core, this theory is concerned with people s
attempts to evade the existential anxiety that arises from
awareness of their own mortality. To quell the fear of
death, people work to convince themselves that they are
worthwhile actors who are playing an important role in
a world that has meaning and purpose. More specifically,
people rely on their self - concepts (beliefs about the self
relative to culturally valued standards), their cultural
worldviews (sets of socially shared beliefs and values),
and their close relationships to help them manage the fear
of death. When people encounter challenges to any compo-
nents of this belief system, death awareness increases and
existential anxiety ensues (Schimel, Hayes, Williams, &
Jahrig, 2007). Note that self - views, worldviews, and rela-
tionships provide much (if not all) of the raw material
through which people meet their needs for agency, coher-
ence, and communion. Moreover, clear parallels can be
drawn between the needs for self - esteem, meaning, and
relationships, as discussed in terror management theory,
and the three motives that we emphasize here. Thus,
while sharing our recognition of the importance of the tri-
partite self - motives, terror management theory uniquely
proposes that people pursue these three self motives as
a means of buffering themselves against a primitive and
basic fear of death.
CH16.indd 609CH16.indd 609 10/22/09 2:50:08 PM10/22/09 2:50:08 PM
610 Self and Identity
Not surprisingly, people react strongly when their mortal-
ity is made salient (for a review, see Solomon et al., 2004). In
dozens of inventive and provocative studies, researchers have
shown that those who are reminded of their own mortality are
more concerned with having high self - esteem (Greenberg,
Solomon, et al., 1992) and are more inclined to behave in
ways that defend and maximize self - esteem (Taubman Ben -
Ari, Florian, & Mikulincer, 1999). Mortality salience manip-
ulations also have a bearing on group relations. For example,
those high in mortality salience are especially motivated to
defend their cultural worldviews by derogating people who
challenge these beliefs (e.g., Florian & Mikulincer, 1997;
Rosenblatt, Greenberg, Solomon, Pyszczynski, & Lyon,
1989). Notably, the tendency for mortality salience to trig-
ger such activities is diminished among people with elevated
levels of self - esteem (e.g., Greenberg, Solomon, et al., 1992)
and among those in whom hope for an afterlife has been
primed (Dechesne et al., 2003).
Terror management theory has also helped illuminate
recent political trends in the United States. For many
Americans, the attack on the World Trade Center on
September 11, 2001, was a naturally occurring manipulation
of mortality salience. Terror management theory suggests
that under such conditions people reach out for a strong -
willed and decisive leader who promises to defend and
protect them. U.S. President George W. Bush represented
just such a leader to many, and as the theory would predict,
his popularity soared after the attacks on the twin towers.
More impressive evidence for the theory was provided by
a series of experiments indicating links among the attacks,
mortality salience, and endorsement of Bush. For exam-
ple, subliminal exposure to stimuli related to September
11 increased participants death - related thoughts, as well
as their support for Bush. Furthermore, mortality salience
made participants more inclined to vote for Bush in the
upcoming presidential election and less inclined to vote for
his opponent, Senator John Kerry (Landau et al., 2004).
More generally, death anxiety appears to be a robust pre-
dictor of rightwing, conservative thinking (Jost, Glaser,
Kruglanski, & Sulloway, 2003), and salient encounters
with mortality - threatening events can compel people to
adopt more politically conservative values and beliefs
(Bonanno & Jost, 2006). It is important to note, however,
that the link between mortality salience and political con-
servatism is not always so direct. Among both strong adher-
ents of political liberalism and those in whom the value
of tolerance has been primed, mortality salience actually
increases acceptance of differing worldviews (Greenberg,
Simon, Pyszczynski, Solomon, & Chatel, 1992).
In short, terror management theory raises the interesting
possibility that many everyday behaviors that have tradition-
ally been chalked up to motives such as agency, coherence,
and communion are really performed in the service of
fending off anxiety associated with death. Yet some have
raised serious challenges to this assumption. For example,
Hart and colleagues independently threatened each of these
three self - motives and found that, at least among some indi-
viduals (those who had insecure attachment styles), each
type of threat produced defensive reactions that were similar
in character to those produced by mortality salience manip-
ulations (Hart, Shaver, & Goldenberg, 2005). From this
vantage point, reminders of death may gain their potency
because they represent a triple whammy: they simultane-
ously undermine the assumptions that we have a future
self to which we can aspire (challenging coherence needs),
we have enduring relationships (challenging communion
needs), and we will accomplish things in the future (chal-
lenging agency needs; for a similar view, see McGregor,
Gailliot, Vasquez, & Nash, 2007). This reasoning raises a
critically important question: Is fear of death the ultimate
motivator of behavior that terror management theory would
have us believe it is, or are other high - level motives (such
as our tripartite motives) of themselves responsible, with
the potency of mortality salience manipulations residing in
their capacity to activate all three motives simultaneously?
At this juncture, this question remains unanswered.
The idea that the self is socially constructed was first elabo-
rated by the symbolic interactionists (Cooley, 1902; Mead,
1934). Since then, numerous theorists have emphasized the
fundamentally interpersonal nature of the self (for a review,
see Markus & Cross, 1990). So crucial are social interac-
tions to the construction and maintenance of the self - concept
that people surely would not possess self - views were it not
for their interactions with others. Consider Gallup s (1977)
seminal work on self - awareness in chimpanzees, which
compared the self - recognition abilities of chimps raised
in isolation with those of chimps raised with conspecifics.
Whereas the chimps with prior social experience readily
recognized their own reflections in a mirror, those raised
in isolation showed no signs of self - recognition. Although
they undoubtedly saw themselves reflected in the surface
before them, the isolated chimps possessed no basis for
understanding exactly who or what they were looking at.
Our genetic similarity to chimps suggests that a similar
fate might befall humans raised in isolation.
To make sense of the vast theoretical and empirical
literatures on the interpersonal self, Markus and Cross
(1990) identified three ways in which others shape the self.
First, individuals come to know who they are, within a
larger social structure, through their interactions with others.
CH16.indd 610CH16.indd 610 10/22/09 2:50:08 PM10/22/09 2:50:08 PM
Self in Relationship to Others 611
This type of interpersonal influence occurs when individuals
internalize the values, attitudes, thoughts, feelings, and
social roles to which they are exposed. It refers to those
aspects of the self that are acquired via ongoing interac-
tions with significant others and that become internalized
so thoroughly by the individual that they seem the natural
and inevitable consequences of his or her own thoughts
(Markus & Cross, 1990, p. 582). Next, people rely on feed-
back and information from others to form the basis of their
self - knowledge, as well as to evaluate, maintain, and regu-
late the self. This type of interpersonal influence is exem-
plified by work on symbolic interactionism (Cooley, 1902;
Mead, 1934), social comparisons (Festinger, 1954), self -
presentation (Tedeschi, 1981), and self - verification (Swann,
1983), among other theories. What these approaches share
is an emphasis on the ways in which the self is influenced
by others real, perceived, and imagined reactions. Finally,
people s interpersonal relationships themselves become
part of the self, as when individuals store mental representa-
tions of close others alongside information about the self in
memory. Work on self - expansion (Aron & Aron, 1996) and
individual differences in individualism versus collectivism
(Markus & Kitayama, 1991; Triandis, 1995) illustrates this
type of interpersonal influence on the self. An assumption
that underlies this work is that close relationship partners
do not just exist outside of us in a real sense, close part-
ners get inside our heads.
In what follows, we consider theories of self and iden-
tity that illustrate each of Markus and Cross s (1990) three
types of interpersonal influence. Note that much of the
material discussed elsewhere in this chapter (e.g., symbolic
interactionism, attachment dynamics, and self - motives)
could fit just as easily in this section. To avoid redundancy,
however, we confine this section to material that we have
not discussed at length elsewhere. Moreover, the placement
of theoretical approaches into one of the three categories of
influence is, admittedly, rough at times. Certainly, the
processes by which people construct, maintain, and store
self - knowledge will, at times, reflect multiple forms of
interpersonal influence. Nonetheless, we impose order by
discussing social identity, self - categorization, and stereo-
type approaches under the Constructing the Self heading;
desires for self - consistent versus overly positive partner
appraisals under the Evaluating and Maintaining the Self
heading; and broad cognitive models of the interpersonal
self under the Including Others in the Self heading.
Constructing the Self
Social Identity Approach
Social identities refer to people s knowledge of their mem-
berships in social groups and the emotional significance
that they attach to these memberships (Tajfel, 1981;
Tajfel & Turner, 1979). These identities presumably
emerge throughout the life span, beginning when children
learn, through interactions and communications with care-
givers and others, the normative behaviors, feelings, and
values associated with the various social groups to which
they belong. Once formed, social identities seem to exert a
powerful influence on social thought and behavior. Indeed,
some argue that because social identities are the building
blocks of personal identities, social identities are more apt
to influence behavior than are personal identities (Turner
et al., 2006).
One version of social identity theory assumes that people
enter groups that they perceive as both positive and distinc-
tive as a means of self - enhancement (e.g., Abrams & Hogg,
1988). Consistent with this idea, evidence indicates that
people display a strong ingroup bias, or tendency to favor
their own group relative to outgroups (e.g., Brewer & Kramer,
1985; Tajfel, Billig, Bundy, & Flament, 1971). Combined
with the outgroup homogeneity effect the perception of
greater similarity among the members of outgroups as com-
pared with ingroups (Linville & Jones, 1980) this bias
facilitates people s ability to dehumanize members of out-
groups by perceiving them as lacking in human qualities.
Dehumanization, in turn, plays a role in the justification and
maintenance of intergroup prejudice and conflict (Cortes,
Demoulin, Rodriguez, Rodriguez, & Leyens, 2005; Vaes,
Paladino, Castelli, Leyens, & Giovanazzi, 2003).
In recent years, social identity approaches have shifted
away from an emphasis on self - enhancement as the opera-
tive motive. Self - categorization theory avoids the issue
of motivation altogether, stressing instead that the per-
ceptual processes that prompt humans to parse the world
into “ us ” and “ them ” are hardwired and basic to human
existence (Turner, 1985; Turner, Hogg, Oakes, Reicher, &
Wetherell, 1987). Other approaches argue that social
identities reduce uncertainty (e.g., Hogg, 2000; Hogg &
Mullin, 1999), make the world more sensible and coher-
ent (e.g., Ellemers & Van Knippenberg, 1997), or pro-
tect people from the existential terror of death (Castano,
Yzerbyt, Paladino, & Sacchi, 2002). Whatever the nature
of the motive that causes people to identify with groups, it
is ironic that although group memberships are essential for
survival, they also place people in grave danger, such as
when social identities motivate people to confront or even
kill one another.
Of course, considerable individual differences exist in
how central of a role social identities play in people s lives.
Most people perceive gender and ethnicity to be impor-
tant social identities, but variation occurs in the strength
of people s identification with these groups (Luhtanen &
Crocker, 1992). Moreover, the centrality of people s social
CH16.indd 611CH16.indd 611 10/22/09 2:50:08 PM10/22/09 2:50:08 PM
612 Self and Identity
identities varies not only as a function of the desirability of
the group (Tajfel & Turner, 1986) but also as a function
of the group s structure, such as its size and distinctiveness
relative to outgroups (Brewer, 2003).
Placing importance on one s social identities can yield
both positive and negative consequences. On the positive
side, for members of low - status groups, higher levels of
group identification can provide a psychological buffer
against the negative effects of discrimination on self -
esteem, well - being, and achievement (e.g., Wong et al.,
2003). Some propose that strong identification with stig-
matized ingroups at least partially explains the relatively
high self - esteem and favorable self - views of members of
many low - status groups (Branscombe, Schmitt, & Harvey,
1999; see also Crocker & Major, 1989). On the negative
side, those who identify more strongly with their social
groups are more likely to display the perceptual and moti-
vational processes that contribute to intergroup conflict
and prejudice (e.g., Branscombe & Wann, 1994; Feather,
1994; Jetten, Spears, & Manstead, 1996). Strong identifi-
cation with social groups can also encourage rigid com-
pliance with the group s behavioral norms, even when
noncompliance would be beneficial. For example, some
research reveals that members of ethnic minority groups
avoid beneficial health behaviors such as exercise and
eating healthy foods to the extent that they perceive those
behaviors as violating their ingroup s norms (Oyserman,
Fryberg, & Yoder, 2007). Thus, stronger identification
with ingroups increases people s motivation to defend and
uphold group norms even when these norms are harmful.
Since its inception, social identity theory has offered a
powerful and generative framework for understanding how
individuals connect themselves to the larger social structure
and rely on groups to provide them with self - knowledge,
meaning, and purpose. Although the seeds of the theory
were sewn during post World War II Europe (e.g., Tajfel,
Jaspars, & Fraser, 1984), social identity theory s impact
now extends far beyond the continent of its birth and it
is considered a major social psychological theory on an
international level. As evidence of the theory s genera-
tivity, a PsycINFO search of articles and chapters with
keywords of “ social identity, ” “ ingroup, ” or “ outgroup ”
produced more than 3,000 publications at the time of this
writing. Moreover, the theory has been revitalized by
new approaches (e.g., Oyserman et al., 2007; Vaes et al.,
2003), as well as applications to diverse subareas within
the behavioral sciences.
Interplay of Personal and Social Selves
Self - categorization theory (Turner et al., 1987) proposes
that the relationship between personal and social self -
views is hydraulic. For example, the principle of functional
antagonism posits that as the salience of group identities
increases, the salience of personal identities decreases.
Similarly, the theory argues that when people enter groups,
they undergo a depersonalization process wherein they
come to see themselves as categorically interchangeable
with other group members. Recently, some theorists have
suggested that these principles may not always apply
(e.g., Postmes & Jetten, 2006; Simon, 2004). A case in
point is offered by a recent study of compensatory self -
verification among fused people, that is, people whose
personal and social self - views have fused. Compensatory
self - verification refers to the tendency for people to react
to self - discrepant (i.e., overly positive or negative) evalu-
ations by intensifying their efforts to elicit self - verifying
evaluations (e.g., Swann & Read, 1981). Because the per-
sonal and social identities of fused people are functionally
equivalent, challenging either type of identity should fuel
behavioral efforts to reaffirm the other type of identity.
Consistent with this reasoning, when researchers presented
participants with overly positive feedback that challenged
the validity of their personal self - views, fused participants
(but not nonfused participants) compensated by affirming
their social self - views. Specifically, they expressed greater
willingness to fight and die for their country (Swann,
G ó mez, Seyle, Morales, & Huici, 2009). Such findings
indicate that personal and social self - views may sometimes
combine synergistically rather than at cross - purposes (see
also Jetten, Branscombe, & Spears, 2002).
In a related vein, optimal distinctiveness theory sug-
gests that just as people have an inherent drive to identify
with groups, they also have an opposing drive for individu-
ation (Brewer, 1991). To cope with these conflicting agen-
das, people strike a balance by finding a point of optimal
distinctiveness, an identity that simultaneously addresses
their needs for affiliation and individuation. This approach
shares with self - verification theory the assumption that
group members remain interested in being individuated
and attaining verification of their personal identities when
they enter groups. One could go even further to suggest that
people may affiliate (at least in part) as a means of obtain-
ing verification for their personal identities. An example of
this would be a woman who joins a chess club to verify her
personal identity as highly intelligent.
Researchers operating outside the social identity tradi-
tion have independently investigated the interplay between
personal and social self - views. One line of research
focused on what happens when the social stereotypes of
some individuals ( perceivers ) channeled their behavior
toward other individuals ( targets ). This work revealed
that perceivers elicited behaviors from targets that con-
firmed their stereotypes (Snyder, Tanke, & Berscheid,
1977). Subsequent investigations examined the conditions
CH16.indd 612CH16.indd 612 10/22/09 2:50:09 PM10/22/09 2:50:09 PM
Self in Relationship to Others 613
under which targets who confirmed the expectations of
perceivers internalized these expectations into correspond-
ing self - views (for a review, see Snyder & Klein, 2005).
In more recent years, researchers have suggested that,
even if stereotypes do not cause perceivers to behaviorally
constrain the response options of targets, the mere exis-
tence of a stereotype may shape the behavior of targets in
undesirable ways. In part, this research was a reaction to
indictments of Black American culture that can be traced
to the anthropologist John Ogbu. On the basis of anecdotal
evidence, Ogbu suggested that in the United States, the
Black minority culture gradually developed an opposi-
tional orientation that encouraged them to disengage from
the educational system, which was perceived as a White
domain (e.g., Fordham & Ogbu, 1986). This argument
gained widespread acceptance in the popular media, which
used it to explain the Black White achievement gap.
Although the validity of Ogbu s assertions was never
established, the psychological literature offers some evi-
dence that people who identify themselves as having low
status sometimes embrace these negative identities (e.g.,
Spears, Jetten, & Scheepers, 2002) or even the political
systems that perpetuate their low status (Jost, Banaji, &
Nosek, 2004). Moreover, the more people face discrimi-
nation, the more they emphasize the devalued identity
(Branscombe, Ellemers, Spears, & Doosje, 1999; Jetten,
Branscombe, Schmitt, & Spears, 2001).
More direct evidence that cultural groups differ in the
extent to which they promote academic achievement has
come from researchers who asked why Asian Americans
(specifically Chinese and Korean immigrants) outperform
both Blacks and Whites on most indices of academic per-
formance. Adopting an interactionist framework, some
argue that the relatively high academic performance of
Asian Americans is multiply determined (Portes & Zhou,
1993). For example, factors such as immigration selectivity,
above - average levels of pre - and postmigration socioeco-
nomic status, and ethnic social structures are thought to
interact with immigrant optimism and the belief in edu-
cation to override blocked mobility (Zhou & Kim, 2006).
From this perspective, it is overly simplistic to blame the
underperformance of some groups on constructs such as
oppositional culture, since social structural variables
must surely play a role as well. At a minimum, a culture of
achievement requires economic resources to support it.
Others have developed formulations that attempt to
explain the underachievement of minorities and other
negatively stereotyped groups without referring to cultural
variables. For example, Steele (1997) contended that for
the marginalized, stereotypes represent a threat in the
air that can trigger anxiety even when the stereotypes
are recognized as fallacious. Research supports the notion
that, in performance settings, anxiety due to stereotypes
can distract the individual and cause poor performance
and failure. Such failure may, in turn, cause the marginal-
ized group member to disengage from the activity. If the
activity happens to involve education, such disengagement
may undercut the future socioeconomic viability of the
marginalized group (see Aronson & Steele, 2005). Such
effects appear to be strongest when the targets of stereo-
types value strongly their group identities. That is, targets
who value their group memberships are more likely to
perceive discrimination against their group (Eccleston &
Major, 2006; Major, Quinton, & Schmader, 2003), and
their performance is more likely to suffer when they
are reminded of negative stereotypes about their group
(Kiefer & Sekaquaptewa, 2007; Schmader, 2002).
Evaluating and Maintaining the Self
In Orson Welles ’ s (1941) Citizen Kane , Charles Foster
Kane holds grandiose illusions about his wife s talents
as an opera singer. So smitten is Kane with Susan s sing-
ing voice that he uses his wealth and power to secure her
headlining roles in world - class venues. Sadly, however,
Susan knows that she is not the musical virtuoso that Kane
believes her to be. Painfully aware of her vocal limitations,
Susan grows increasingly traumatized by the humiliation
of having to parade her mediocrity in front of an audience.
Eventually, the pressure of Kane s misguided illusions
grows too much for Susan to bear, and she tries to escape
by taking her own life. Although Susan survives, she never
forgives Kane for refusing to see her for who she really is.
The relationship experiences of Kane one of the most
unique and memorable movie characters of all time are
by no means typical. Nonetheless, we believe that his con-
flict with Susan illustrates a fairly common relationship
problem. Specifically, when couples disagree about who
is who within the relationship, unhappiness ensues.
As noted in our discussion of the coherence motive,
people desire appraisals from their relationship partners that
verify their highly certain and important self - views, even if
these self - views are negative. Moreover, the need for self -
confirming appraisals runs particularly strong in the context
of relationships characterized by high levels of interdepen-
dence. For example, among both college roommates and
married couples, people with positive self - concepts prefer
partners who view them favorably, whereas those with
negative self - concepts prefer partners who view them nega-
tively (Swann & Pelham, 2002; Swann et al., 1994). And
when people s spouses view them in a manner that is dis-
crepant with their stable self - views, their relationships are
characterized by high levels of marital distress (Schafer,
Wickrama, & Keith, 1996). Indeed, as illustrated by the
CH16.indd 613CH16.indd 613 10/22/09 2:50:09 PM10/22/09 2:50:09 PM
614 Self and Identity
case of Susan Kane, psychological health and well - being
may suffer when people s close relationship partners dis-
agree with them about who they are (Swann et al., 2007).
And yet, despite the results of research demonstrating
that people desire consistent appraisals from their close
relationship partners, some research suggests instead that
people prefer overly positive evaluations from their part-
ners. For example, Murray and colleagues find that even
people with negative self - views feel most satisfied in their
relationships, and closest to their partners, when partners
view them more favorably than they view themselves (e.g.,
Murray et al., 1996). Furthermore, Murray suggests that the
tendency to idealize romantic partners facilitates relation-
ship success by assuaging people s doubts and giving them
the confidence to trust each other (Murray et al., 2000).
From this perspective, it is overly positive appraisals
not self - confirming ones — that members of intimate rela-
tionships crave.
What might account for the apparent discrepancy
between these two bodies of work? One possibility con-
cerns the level of abstraction (e.g., global versus specific)
at which these two sets of researchers typically mea-
sure partners self - views and perceptions of each other.
Whereas most desires for self - verifying appraisals occur at
the level of specific self - views ( My partner think I am
at the 55th percentile in cooking ability ), most desires
for illusory appraisals occur at the level of global self -
views ( My partner thinks I am loving and kind ). Thus,
it may be that members of happy couples maintain ador-
ing appraisals of their partners at a global level while also
appraising their partners accurately at a more specific level
(e.g., Neff & Karney, 2002). Indeed, the results of longitu-
dinal investigations suggest that marriages are most likely
to endure over the long haul when partners global love for
each other is based on an accurate understanding of each
other s specific traits and qualities (Neff & Karney, 2005).
As such, overly positive appraisals of partners may promote
relationship satisfaction, provided that they are grounded
in reality.
One shortcoming of the global - enhancement, specific -
verification argument is that there is no theoretical reason
people should suspend their self - verification attempts once
self - views exceed some threshold of globality. After all,
if the primary function of self - views is to enable people
to understand the world and guide behavior, then people
should be motivated to verify their global self - views in
non - relational contexts. In fact, research on depression
(e.g., Giesler et al., 1996; Swann, Wenzlaff, Krull, &
Pelham, 1992; Swann, Wenzlaff, & Tafarodi, 1992 ) and
self - esteem (Wiesenfeld, Swann, Brockner & Bartel, 2007)
has shown that people are indeed motivated to verify their
global negative self - views .
An approach that confronts this issue directly assumes
that it is a covariate of self - view globality relationship
relevance that is critical. Whereas Swann and colleagues
(1994) tend to measure people s self - views within domains
that vary in their relevance to relationship satisfaction and
functioning (e.g., intelligence, social skills, artistic abili-
ties, and physical attractiveness), Murray and colleagues
limit their focus to domains that are high in relationship
relevance, such as warmth, kindness, and dependability
(e.g., Murray et al., 1996). Stated differently, whereas
Swann and colleagues consider self - views that fall along
both agentic and communal dimensions, Murray and col-
leagues focus primarily on communal self - views. Given
the critical importance of communal qualities in the context
of relationships, it is perhaps not surprising that Murray s
research participants were particularly happy with part-
ners who held idealized images of their communal traits.
Consistent with this possibility, Swann et al. (2002) found
that romantic partners preferred appraisals from their part-
ners that matched their self - views in most domains but
desired overly positive appraisals in domains that they
considered highly important for relationship satisfaction,
such as physical attractiveness.
Including Others in the Self
Several theorists have examined the ways in which close
relationships alter and influence the cognitive contents of
the self. For instance, as noted earlier, research on self -
expansion reveals that people incorporate the perspectives,
resources, and characteristics of close others into their own
self - concepts (Aron et al., 1991). Thus, closeness with
others inevitably leads to an expansion of the self, as the
self - concept grows to incorporate new features (e.g., Aron
et al., 1995). One consequence of this cognitive overlap
between self and other is that people sometimes display
“ self – other confusions ” (Mashek et al., 2003), suggest-
ing that information about close others and the self is pro-
cessed similarly.
Whereas self - expansion research focuses on the ten-
dency to assume the features of close relationship part-
ners, other approaches look at how cognitive processes
are shaped more broadly by relationships. For example,
Baldwin (1992) proposes that people store information
about relationship partners in the form of relational sche-
mas , or mental models consisting of scripts that describe
typical interaction patterns, as well as representations of
the self and other that capture how the self typically feels
and behaves when interacting with the other. As such, cues
that bring a particular relationship partner to mind also
activate people s mental representations of self with other
(Ogilvie & Ashmore, 1991) and call to mind those aspects
CH16.indd 614CH16.indd 614 10/22/09 2:50:09 PM10/22/09 2:50:09 PM
Identity Negotiation and Change 615
of self - knowledge that are contained within that relational
schema (Hinkley & Andersen, 1996).
Similar assumptions form the foundation of Andersen
and Chen s (2002) relational self theory. In their theory,
Andersen and Chen suggest that people s stored represen-
tations of self and others play important roles in shaping
personality and the self by guiding the cognitive, affec-
tive, motivational, and behavioral patterns that become
activated in particular contexts. Because representations
of the self are linked in memory with representations of
significant others, any chronic or transient reminders
of significant others activate particular relational selves
and their accompanying styles of thinking, feeling, and
acting. Thus, personality and the self are interpersonal
patterns that reflect the various selves an individual has
constructed in the context of relationships with signifi-
cant others. Although they acknowledge that the self most
likely contains some aspects that are not directly related to
representations of significant others, Andersen and Chen
propose that the bulk of self - knowledge is acquired in the
context of relationships and that significant others are thus
“ basic to self - experience ” (p. 638).
People can take on numerous identities. The same man, for
example, may be warm with his children, guarded with his
co - workers, and a blend of both with his neighbors. This
fact of social life can prove challenging for those who are
trying to predict what their partners are going to do next.
We propose that people meet this challenge through the
process of identity negotiation , which allows relationship
partners to establish who is who via ongoing, mutual,
give - and - take interactions with each other. Once people
establish a working consensus that is agreeable to both
parties (e.g., Goffman, 1959; Swann & Bosson, 2008),
their mutually agreed on expectations transform discon-
nected individuals into collaborators who have common
obligations, goals, and often, a modicum of commitment
to each other. In this way, identity negotiation processes
provide the interpersonal glue that allies people with
one another. More generally, just as identities define
people and make them viable as humans, identity negotia-
tion processes define relationships and make them viable
as a foundation for organized social activity.
Identity negotiation theory (Swann & Bosson, 2008;
Swann, Johnson, & Bosson, in press) elaborates on the
interpersonal principles that guide identity negotiations.
People follow these processes, albeit largely unintention-
ally, during each of several successive stages of social
interaction. Typically, the principles of identity negotiation
encourage people to negotiate identities that are compatible
with their chronic self - views. At times, however, target
individuals may encounter partners who are unable or
unwilling to honor their chronic identities. To the extent
that targets are invested in the identity (e.g., it is high in
certainty and importance), they may intensify their efforts
to elicit self - verifying evaluations. If investment is low and
resistance from the perceiver is high, however, targets may
behaviorally confirm the expectations of perceivers (e.g.,
Snyder & Klein, 2005). Eventually, they may internalize
the new behaviors into their self - concept, resulting in iden-
tity change.
Identity Negotiation in the East and West
Like most psychological structures, identities survive only
insofar as they receive periodic nourishment from the
social environment. Therefore, characteristics of the social
milieu are a key determinant of how much identity change
any given individual experiences. Consider Western versus
Eastern cultures. Western cultures encourage identity sta-
bility by placing a premium on consistency in the identities
people negotiate both over time and across settings. Given
this, it is not surprising that identity stability seems to be
relatively high in samples of Western participants. For
instance, people s self - descriptions on the Big Five factors
of extraversion, neuroticism, conscientiousness, agreeable-
ness, and openness remain stable (correlations between
.30 and .50) over periods of up to 20 years (e.g., Conley,
1985). Stability estimates are even higher (correlations in
the .55 .85 range) if the measurement period is reduced to
several years (e.g., Costa & McCrae, 1988) or if the self -
views are high in importance or certainty (English & Chen,
2007; Pelham, 1991).
Relative to Western culture, the culture of East Asians
places more emphasis on relationships, connectedness, and
belonging. This makes East Asians particularly inclined to
conform their identities to the expectations and preferences
of their current interaction partner. As a result, relative to
North Americans, East Asians show lower cross - situational
stability in their self - descriptions (Kanagawa, Cross, &
Markus, 2001; Suh, 2002). Furthermore, when describing
themselves, East Asians are especially inclined to endorse
semantically opposite self - views (Choi & Choi, 2002) and
contradictory statements about themselves (Cousins, 1989;
Spencer - Rodgers, Peng, Wang, & Hou, 2004).
On the surface, evidence that the identities of East
Asians appear to be relatively situation specific may seem
to undermine the notion that there exists a universal desire
for coherence. But perhaps not. For those who define
themselves in relational terms, coherence may hinge on
the propensity to honor identities that are negotiated with
CH16.indd 615CH16.indd 615 10/22/09 2:50:10 PM10/22/09 2:50:10 PM
616 Self and Identity
specific others. Consistent with this reasoning, among the
Japanese, cross - situational consistency is less valued and
more weakly related to the sense of having a true self
than it is among Westerners (Kashima et al., 2004).
From this vantage point, there appear to be some intriguing
differences in identity negotiation processes in the East
versus the West. It is tempting, for example, to conclude
that the tendency for behavioral confirmation to trump self -
verification is more prevalent in East Asian than in Western
cultures. Although this characterization seems technically
accurate, it is probably somewhat misleading because self -
verification may simply take a different form in Eastern
cultures. More specifically, highly relational cultures may
prioritize the tendency for relationship partners to remain
true to the identities that they have negotiated within the
relationship while being relatively unconcerned with
the partner s behavior outside the confines of the relation-
ship (see the discussion of circumscribed versus global accu-
racy in Gill & Swann, 2004). More generally, evidence that
East Asians display less cross - situational consistency
than Westerners does not necessarily imply that Asians
routinely experience true changes in their identities. In
the section that follows, we suggest that for true identity
change to occur a relatively dramatic shift in the social
environment must occur that supports the change.
Antecedents of Identity Change
Our emphasis on the importance of coherence thus far
would suggest that changes in identity are typically unwel-
come phenomena that can confuse or even derail the process
of identity negotiation (as well as cause stress and under-
mine health at the individual level). Yet as wrenching
as identity changes may sometimes be, they are a natu-
ral and critically important part of life (Robins, Noftle,
Trzesniewski, & Roberts, 2005). Although many identity
changes are triggered by events over which the person has
no control, at times people disregard their coherence striv-
ings and deliberately attempt to change their identities.
Note, for example, that communion or agency motives
may sometimes override the desire for coherence, and the
former motives may sometimes require identity change if
they are to be gratified. In what follows, we consider four
sets of conditions that foster identity change.
Sociocultural and Environmental Changes
Over the past half century, the United States saw sweeping
changes in cultural expectations regarding groups that were
historically saddled with minority status. For example, the
Civil Rights and Women s Liberation movements altered cul-
tural expectations for Blacks and women, respectively. These
shifting expectations gradually influenced the identities of
members of these groups. The Women s Liberation movement,
for instance, led to the erosion of cultural stereotypes that char-
acterized women as weak and dependent (e.g., Spence et al.,
1985). As these stereotypes lost force and more egalitarian
attitudes took hold, girls and women adopted corresponding
changes in their identities and associated behaviors.
Changes in people s immediate social environment can
also foster identity change. When, for example, people
enter college or move they may encounter relatively unique
expectations and behavioral norms among the locals (e.g.,
Iyer, Jetten, & Tsivrikos, 2008). By altering the way people
relate to others, new settings may encourage people to alter
their self - views (Hormuth, 1990). In addition, new envi-
ronments may foster identity change because they lack
the opportunity structures (McCall & Simmons, 1966)
that once nurtured and sustained the original identity. For
people to sustain their identities, on entering a new set-
ting they must remoor their identities within the new social
structure (Ethier & Deaux, 1994). Failure to do so results
in identity change.
Developmental Growth and Role Changes
When the community recognizes a significant change in an
individual, it may set in motion a sequence of events that
produces identity change. Examples of such community -
initiated changes include changes in age (e.g., when
adolescents become adults), status (e.g., when graduate
students become professors), or social role (e.g., when sin-
gles get married). When such transformations occur, com-
munities may abruptly alter the way they treat the person.
Even if targets of such differential treatment resist change
at first, eventually they recognize the inevitable, become
less invested in maintaining the initial identity, and bring
their identities into agreement with the treatment they
receive. Studies of adolescence support this reasoning.
Theory and research alike suggest that late adolescence
marks a developmental period during which changing
treatment and expectations trigger dramatic identity change
(Arnett, 2000; Erikson, 1959; Pals, 1999).
Acquisition and Loss of Abilities
The process of maturation is marked by the acquisition
of new competencies and the loss of established ones.
Whether one gains or loses an ability, the associated iden-
tity needs to be updated. This may explain why people s
identities are especially turbulent early and late in life.
Although both gains and losses are sprinkled throughout
the life span, gains tend to occur often during the early
years (e.g., acquiring the ability to scale mountains and
drive automobiles) and losses tend to occur during the later
years (e.g., losing the ability to scale mountains and drive
automobiles). Indeed, one of the most troubling aspects of
CH16.indd 616CH16.indd 616 10/22/09 2:50:10 PM10/22/09 2:50:10 PM
References 617
the aging process is not age itself but the loss of capacities
that have become essential to the person s feelings of
agency. From this vantage point, it is easy to understand
how the physical changes that accompany aging can take a
toll on people s identities (Whitbourne, 1996).
Self - Initiated Changes
When people recognize that an identity is undermining
their capacity to achieve a valued goal, they may negotiate
a different identity within a circumscribed set of circum-
stances. If the fruits of such negotiations remain in effect
for an extended period, the changes may generalize to other
settings and eventually lead to permanent identity change.
Consider, for example, a woman who suspects that her low
self - perceived attractiveness will block her efforts to win
the heart of a would - be lover. Recognizing the dilemma, she
may strive to be exceptionally attractive in the presence of
her love interest (Swann et al., 2002). If she succeeds and
wins her beloved s affections, she may internalize his
appreciation of her beauty and upgrade her self - perceived
attractiveness (Jones, Gergen, & Davis, 1962).
The foregoing scenario is just one example of a larger
class of instances in which people initiate an identity change
either because they want to repair an unsatisfying life situ-
ation or because they aspire to self - improvement. Some
evidence indicates that such intentional identity change
requires a self - focused state of mental preparedness or sub-
jective readiness to change (Anthis & LaVoie, 2006). Even
for those who feel prepared for change, the tendency to
ensconce oneself in self - verifying social environments (e.g.,
Swann et al., 2007) may complicate the business of identity
change. For the effects of self - initiated identity change to
be permanent, people must change not only their own self -
views and narratives but also the social environments that
typically support those self - views and narratives.
What does it mean to have a sense of self ? Is there a
single self, or does the self have multiple, independent
components? Are there aspects of the self that cannot be
accessed consciously, and if so, can they be measured?
How do people derive and maintain a sense of self, and
once they do, can it be altered? Do people from different
cultures experience the self in the same way? And what
are the personal and social consequences of our represen-
tations of self?
This chapter was designed to address these and related
questions. We began with a brief history of the self, noting
that in recent decades social psychology has at long last
embraced some seminal conceptions of the self offered by
James (1890/1950) more than a century ago. The result of
this neo - Jamesian approach to the self has been an unprec-
edented explosion of conceptual and methodological inno-
vations that have breathed new life into the subarea.
But if this sudden burst of creativity has had clear bene-
fits, it has had costs as well. To us, the most worrisome risk
is that the subarea will become so broad that it will begin
to lose focus. Eventually, newcomers may begin to wonder
whether there is any there, there. To avert this unhappy
outcome, we suggest setting a boundary condition for the
subarea: Work on the self should involve some consider-
ation of the self as a mental representation.
A secondary concern is the lack of integration both
within and across topic areas. This is understandable
given the complexity of the subject matter and the result-
ing challenge of constructing meaningful integrations.
Nevertheless, if the field is ever to develop a unified theory
of the self, it is critical that theorists continue to forge con-
nections among different themes in the literature.
As future researchers rise to the challenge of forging such
connections, they will build on the fundamental truths that
are gradually emerging in the literature. One truth that has
already been distilled is that the self is, as the symbolic inter-
actionists have long emphasized (e.g., Stryker, 2000), a social
phenomenon. From our first inklings of self - awareness to our
final reflections on the meaning of life, our social interactions
define, nurture, and alter our sense of self. This basic truth
has given rise to a range of investigations that are continuing
to provide crucial insights into the nature of the self.
Many more such truths are within the grasp of contem-
porary self psychologists. Indeed, if our review of the lit-
erature has left us with a single impression, it is that as
the topic area has increasingly attracted attention, the rate
of scientific advances has accelerated proportionally. We
are left with a strong feeling of optimism and a conviction
that, in the future, the answers will come even faster.
Abele, A. E., & Wojciszke, B. (2007). Agency and communion from the
perspective of self versus others.
Journal of Personality and Social
Psychology, 93
, 751 – 763.
Abrams, D., & Hogg, M. A. (1988). Comments on the motivational
status of self - esteem in social identity and intergroup discrimination.
European Journal of Social Psychology, 18
, 317 – 334.
Alicke, M. D., & Govorun, O. (2005). The better - than - average effect. In
M. D. Alicke, D. A. Dunning, & J. I. Krueger (Eds.),
The self in social
(pp. 85 106). New York: Psychology Press.
Alicke, M. D., Klotz, M. L., Breitenbecher, D. L., Yurak, T. J., &
Vredenburg, D. S. (1995). Personal contact, individuation, and the
better - than - average effect.
Journal of Personality and Social
Psychology, 68
, 804 – 825.
CH16.indd 617CH16.indd 617 10/22/09 2:50:10 PM10/22/09 2:50:10 PM
618 Self and Identity
Allport, G. W. (1943). The ego in contemporary psychology.
Review, 50
, 451 – 478.
American Psychiatric Association. (2000).
Diagnostic and statistical man-
ual of mental disorders
(4th ed., text rev.). Washington, DC: Author.
Andersen, S. M. (1984). Self - knowledge and social inference: II. The
diagnosticity of cognitive/affective and behavioral data.
Journal of
Personality and Social Psychology, 46
, 294 – 307.
Andersen, S. M., & Chen, S. (2002). The relational self: An interpersonal
social - cognitive theory.
Psychological Review, 109
, 619 – 645.
Andersen, S. M., & Ross, L. (1984). Self - knowledge and social inference:
I. The impact of cognitive/affective and behavioral data.
Journal of
Personality and Social Psychology, 46
, 280 – 293.
Anthis, K., & LaVoie, J. C. (2006). Readiness to change: A longitudinal
study of changes in adult identity.
Journal of Research in Personality,
, 209 – 219.
Archibald, F. S., Bartholomew, K., & Marx, R. (1995). Loneliness in early
adolescence: A test of the cognitive discrepancy model of loneliness.
Personality and Social Psychology Bulletin, 21
, 296 – 301.
Arnett, J. J. (2000). Emerging adulthood: A theory of development from
the late teens through the twenties.
American Psychologist, 55
469 – 480.
Aron, E. N., & Aron, A. (1996). Love and expansion of the self: The state
of the model.
Personal Relationships, 3
, 45 – 58.
Aron, A., Aron, E. N., Tudor, M., & Nelson, G. (1991). Close relation-
ships as including other in the self.
Journal of Personality and Social
Psychology, 60
, 241 – 253.
Aron, A., Paris, M., & Aron, E. N. (1995). Falling in love: Prospective
studies of self - concept change.
Journal of Personality and Social
Psychology, 69
, 1102 – 1112.
Aronson, E. (1968). A theory of cognitive dissonance: A current perspective.
In L. Berkowitz (Ed.),
Advances in experimental social psychology
(Vol. 4, pp. 1 – 34). New York: Academic Press.
Aronson, J., Cohen, G., & Nail, P. R. (1999). Self - affirmation theory: An
update and appraisal. In E. Harmon - Jones & J. Mills (Eds.),
dissonance: Progress on a pivotal theory in social psychology
(pp. 127 147). Washington, DC: American Psychological Association.
Aronson, J., & Steele, C. M. (2005). Stereotypes and the fragility of
academic competence, motivation, and self - concept. In A. J. Elliot &
C. S. Dweck (Eds.),
Handbook of competence and motivation
(pp. 436 – 456). New York: Guilford.
Baldwin, M. W. (1992). Relational schemas and the processing of informa-
Psychological Bulletin, 112
, 461 – 484.
Baldwin, M. W., Carrell, S. E., & Lopez, D. F. (1990). Priming relationship -
schemas: My advisor and the pope are watching me from the back of
my mind.
Journal of Experimental Social Psychology, 26
, 435 – 454.
Banaji, M. R., & Prentice, D. A. (1994). The self in social contexts.
Review of Psychology, 45
, 297 – 332.
Bandura, A. (1991). Social cognitive theory of self - regulation.
Behavior and Human Decision Processes, 50
, 248 – 287.
Baumeister, R. F. (1991).
Escaping the self: Alcoholism, spirituality,
masochism, and other flights from the burden of selfhood
. New York:
Basic Books.
Baumeister, R. F. (1998). The self. In D. T. Gilbert, S. T. Fiske, &
G. Lindzey (Eds.),
The handbook of social psychology
(pp. 680 – 740).
New York: McGraw - Hill.
Baumeister, R. F., Campbell, J. D., Krueger, J. I., & Vohs, K. D. (2003).
Does high self - esteem cause better performance, interpersonal success,
happiness, or healthier lifestyles?
Psychological Science in the Public
Interest, 4
, 1 – 44.
Baumeister, R. F., & Leary, M. R. (1995). The need to belong: Desire
for interpersonal attachments as a fundamental human motivation.
Psychological Bulletin, 117
, 497 – 529.
Baumgardner, A. H. (1990). To know oneself is to like oneself: Self -
certainty and self - affect.
Journal of Personality and Social Psychology,
, 1062 – 1072.
Beaman, A. L., Klentz, B., Diener, E., & Svanum, S. (1979). Self - awareness
and transgression in children: Two field studies.
Journal of Personality and
Social Psychology, 37
, 1835 – 1846.
Beauregard, K. S., & Dunning, D. (2001). Defining self - worth: Trait self -
esteem moderates the use of self - serving trait definitions in social judg-
Motivation and Emotion, 25
, 135 – 161.
Bem, D. J. (1972). Self - perception theory. In L. Berkowitz (Ed.),
in experimental social psychology
(Vol. 6, pp. 1 62). New York:
Academic Press.
Benenson, J., & Dweck, C. (1986). The development of trait explana-
tions and self - evaluations in the academic and social domains.
Development, 57
, 1179 – 1187.
Blanton, H., Buunk, B. P., Gibbons, F. X., & Kuyper, H. (1999). When
better - than - others compare upward: Choice of comparison and
comparative evaluation as independent predictors of academic
Journal of Personality and Social Psychology, 76
420 – 430.
Bohrnstedt, G. W., & Felson, R. B. (1983). Explaining the relations among
children s actual and perceived performances and self - esteem: A com-
parison of several causal models.
Journal of Personality and Social
Psychology, 45
, 43 – 56.
Bonanno, G. A., & Jost, J. T. (2006). Conservative shift among high -
exposure survivors of the September 11th terrorist attacks.
Basic and
Applied Social Psychology, 28
, 311 – 323.
Bosson, J. K., Brown, R. P., Zeigler - Hill, V., & Swann, W. B., Jr. (2003).
Self - enhancement tendencies among people with high explicit self -
esteem: The moderating role of implicit self - esteem.
Self and Identity,
, 169 – 187.
Bosson, J. K., Lakey, C. E., Campbell, W. K., Zeigler - Hill, V., Jordan, C. H., &
Kernis, M. H. (April, 2008). Untangling the links between narcissism
and self - esteem: A theoretical and empirical review. Personality and
Social Psychology Compass,
(3), 1415 – 1439.
Bosson, J., & Swann, W. B., Jr. (1999). Self - liking, self - competence, and
the quest for self - verification.
Personality and Social Psychology
Bulletin, 25
, 1230 – 1241.
Bosson, J. K., Swann, W. B., Jr., & Pennebaker, J. W. (2000). Stalking
the perfect measure of implicit self - esteem: The blind men and the
elephant revisited?
Journal of Personality and Social Psychology, 79
631 – 643.
Bouchard, T. J., Jr., & Loehlin, J. C. (2001). Genes, evolution, and person-
Behavior Genetics, 31
, 243 – 273.
Bowlby, J. (1969).
Attachment and loss: Attachment
(Vol. 1). New York:
Basic Books.
Branscombe, N. R., Ellemers, N., Spears, R., & Doosje, B. (1999). The con-
text and content of social identity threats. In N. Ellemers, R. Spears, &
B. Doosje (Eds.),
Social identity: Context, commitment, content
35 – 58). Oxford: Blackwell.
Branscombe, N. R., Schmitt, M. T., & Harvey, R. D. (1999). Perceiving
pervasive discrimination among African - Americans: Implications for
group identification and wellbeing.
Journal of Personality and Social
Psychology, 77
, 135 – 149.
Branscombe, N. R., & Wann, D. L. (1994). Collective self - esteem con-
sequences of outgroup derogation when a valued social identity is on
European Journal of Social Psychology, 6
, 641 – 657.
Bretherton, I. (1988). Open communication and internal working models:
Their role in the development of attachment relationships.
Symposium on Motivation, 36
, 57 – 113.
Brewer, M. B. (1991). The social self: On being the same and different
at the same time.
Personality and Social Psychology Bulletin, 17
475 – 482.
CH16.indd 618CH16.indd 618 10/22/09 2:50:10 PM10/22/09 2:50:10 PM
References 619
Brewer, M. B. (2003). Optimal distinctiveness, social identity, and the self.
In M. R. Leary & J. P. Tangney (Eds.),
Handbook of self and identity
(pp. 480 – 491). New York: Guilford.
Brewer, M. B., & Gardner, W. (1996). Who is this “ we ” ? Levels of col-
lective identity and self representations.
Journal of Personality and
Social Psychology, 71
, 83 – 93.
Brewer, M. B., & Kramer, R. M. (1985). The psychology of intergroup
attitudes and behavior.
Annual Review of Psychology, 36
, 219 – 243.
Brown, J. D. (1986). Evaluations of self and others: Self - enhancement
biases in social judgments.
Social Cognition, 4
, 353 – 376.
Brown, J. D. (1990). Evaluating one s abilities: Shortcuts and stumbling
blocks on the road to self - knowledge.
Journal of Experimental Social
Psychology, 26
, 149 – 167.
Brown, J. D. (1998).
The self
. New York: McGraw - Hill.
Brown, J. D., Dutton, K. A., & Cook, K. E. (2001). From the top down: Self -
esteem and self - evaluation.
Cognition and Emotion, 15
, 615 – 631.
Brown, J. D., & McGill, K. J. (1989). The cost of good fortune: When
positive life events produce negative health consequences.
Journal of
Personality and Social Psychology, 55
, 1103 – 1110.
Campbell, J. D. (1990). Self - esteem and clarity of the self - concept.
of Personality and Social Psychology, 59
, 538 – 549.
Campbell, J. D., Assanand, S., & Di Paula, A. (2003). The structure of the
self - concept and its relation to psychological adjustment.
Journal of
Personality, 71
, 115 – 140.
Campbell, J. D., Trapnell, P. D., Heine, S. J., Katz, I. M., Lavallee, L. F., &
Lehman, D. R. (1996). Self - concept clarity: Measurement, personality
correlates, and cultural boundaries.
Journal of Personality and Social
Psychology, 70
, 1114.
Campbell, L., Lackenbauer, S. D., & Muise, A. (2006). When is being
known or adored by romantic partners most beneficial? Self -
perceptions, relationship length, and responses to partner s verifying
and enhancing appraisals.
Personality and Social Psychology Bulletin,
, 1283 – 1294.
Campbell, W. K., Bosson, J. K., Goheen, T. W., Lakey, C. E., & Kernis, M. H.
(2007). Do narcissists dislike themselves deep down inside ” ?
Psychological Science, 18
, 227 – 229.
Campbell, W. K., Rudich, E. A., & Sedikides, C. (2002). Narcissism, self -
esteem, and the positivity of self - views: Two portraits of self - love.
Personality and Social Psychology Bulletin, 28
, 358 – 368.
Castano, E., Yzerbyt, V., Paladino, M. - P., & Sacchi, S. (2002). I belong,
therefore, I exist: Ingroup identification ingroup entitativity, and ingroup
Personality and Social Psychology Bulletin, 28
, 135 – 143.
Chang, E. C., Asakawa, K., & Sanna, L. J. (2001). Cultural variations in
optimistic and pessimistic bias: Do Easterners really expect the worst
and Westerners really expect the best when predicting future life
Journal of Personality and Social Psychology, 81
, 476 – 491.
Chen, S., Chen, K., & Shaw, L. (2004). Self - verification motives at the
collective level of self - definition.
Journal of Personality and Social
Psychology, 86
, 77 – 94.
Choi, I., & Choi, Y. (2002). Culture and self - concept flexibility.
and Social Psychology Bulletin, 28
, 1508 – 1517.
Colvin, C. R., Block, J., & Funder, D. C. (1995). Overly positive self -
evaluations and personality: Negative implications for mental health.
Journal of Personality and Social Psychology, 68
, 1152 – 1162.
Conley, J. J. (1985). Longitudinal stability of personality traits: A
multitrait – multimethod – multioccasion analysis.
Journal of Personality
and Social Psychology, 49
, 1266 – 1282.
Conway, M., & Ross, M. (1984). Getting what you want by revising
what you had.
Journal of Personality and Social Psychology, 47
738 – 748.
Cooley, C. H. (1902).
Human nature and the social order
. New York:
Scribner ’ s.
Cooper, A. M., & Ronningstam, E. (1992). Narcissistic personality disorder.
Review of Psychiatry, 11
, 80 – 97.
Cortes, B. P., Demoulin, S., Rodriguez, R. T., Rodriguez, A. P., & Leyens, J. - P.
(2005). Infrahumanization or familiarity? Attribution of uniquely
human emotions to the self, the ingroup, and the outgroup.
and Social Psychology Bulletin, 31
, 243 – 253.
Costa, P. T., Jr., & McCrae, R. R. (1988). Personality in adulthood: A six -
year longitudinal study of self - reports and spouse ratings on the NEO
Personality Inventory.
Journal of Personality and Social Psychology,
, 853 – 863.
Cousins, S. D. (1989). Culture and self - perception in Japan and the United
Journal of Personality and Social Psychology, 56
, 124 – 131.
Cozzarelli, C., Karafa, J. A., Collins, N. L., & Tagler, M. J. (2003). Stability
and change in adult attachment styles: Associations with personal
vulnerabilities, life events, and global constructs of self and others.
Journal of Social and Clinical Psychology, 22
, 315 – 346.
Crocker, J., Karpinski, A., Quinn, D. M., & Chase, S. K. (2003). When
grades determine self - worth: Consequences of contingent self - worth
for male and female engineering and psychology majors.
Journal of
Personality and Social Psychology, 85
, 507 – 516.
Crocker, J., & Luhtanen, R. K. (2003). Level of self - esteem and contin-
gencies of self - worth: Unique effects on academic, social, and finan-
cial problems in college students.
Personality and Social Psychology
Bulletin, 29
, 701 – 712.
Crocker, J., Luhtanen, R. K., Cooper, M. L., & Bouvrette, A. (2003).
Contingencies of self - worth in college students: Theory and measure-
Journal of Personality and Social Psychology, 85
, 894 – 908.
Crocker, J., & Major, B. (1989). Social stigma and self - esteem: The self -
protective properties of stigma.
Psychological Review, 96
, 608 – 630.
Crocker, J., & Park, L. E. (2004). The costly pursuit of self - esteem.
Psychological Bulletin, 130
, 392 – 414.
Crocker, J., Sommers, S. R., & Luhtanen, R. K. (2002). Hopes dashed
and dreams fulfilled: Contingencies of self - worth and admissions to
graduate school.
Personality and Social Psychology Bulletin, 28
1275 – 1286.
Crocker, J., & Wolfe, C. T. (2001). Contingencies of self - worth.
Psychological Review, 108
, 593 – 623.
Cross, S. E., Bacon, P. L., & Morris, M. L. (2000). The relational interde-
pendent self - construal and relationships.
Journal of Personality and
Social Psychology, 78
, 791 – 808.
Cross, S. E., & Madson, L. (1997). Models of the self: Self - construals and
Psychological Bulletin, 122
, 5 – 37.
Csikszentmihalyi, M. (1990).
Flow: The psychology of optimal experience
New York: HarperCollins.
Csikszentmihalyi, M., & Figurski, T. J. (1982). Self - awareness and aver-
sive experience in everyday life.
Journal of Personality, 50
, 15 – 28.
Dechesne, M., Pyszczynski, T., Arndt, J., Ransom, S., Sheldon, K., van
Knippenberg, A., & Janssen, J. (2003). Literal and symbolic immortal-
ity: The effect of evidence of literal immortality on self - esteem striving
in response to mortality salience.
Journal of Personality and Social
Psychology, 84
, 722 – 737.
Deci, E. L., Koestner, R., & Ryan, R. M. (1999). A meta - analytic review
of experiments examining the effects of extrinsic rewards on intrinsic
Psychological Bulletin, 125
, 627 – 668.
Deci, E. L., & Ryan, R. M. (1995). Human autonomy: The basis for true
self - esteem. In M. H. Kernis (Ed.),
Efficacy, agency, and self - esteem
(pp. 31 49). New York: Plenum Press.
Devos, T., & Banaji, M. R. (2003). Implicit self and identity. In M. R. Leary &
J. P. Tangney (Eds.),
Handbook of self and identity
(pp. 153 – 175).
New York: Guilford.
Dickinson, K. A., & Pincus, A. L. (2003). Interpersonal analysis of grandiose
and vulnerable narcissism.
Journal of Personality Disorders, 17
, 188 – 207.
CH16.indd 619CH16.indd 619 10/22/09 2:50:11 PM10/22/09 2:50:11 PM
620 Self and Identity
Diener, E., & Diener, M. (1995). Cross - cultural correlates of life satisfaction
and self - esteem.
Journal of Personality and Social Psychology, 68
653 – 663.
Diener, E., Suh, E. M., Lucas, R. E., & Smith, H. L. (1999). Subjective
well - being: Three decades of progress.
Psychological Bulletin, 125
276 – 302.
Donnellan, B., Trzesniewski, K., Robins, R., Moffitt, T., & Caspi, A.
(2005). Low self - esteem is related to aggression, antisocial behavior,
and delinquency.
Psychological Science, 16
, 328 – 335.
Dunning, D., Perie, M., & Story, A. L. (1991). Self - serving prototypes as
social categories.
Journal of Personality and Social Psychology, 61
957 – 968.
Duval, S., & Wicklund, R. A. (1972).
A theory of objective self - awareness
Oxford: Academic Press.
Eccleston, C. P., & Major, B. N. (2006). Attributions to discrimination and
self - esteem: The role of group identification and appraisals.
Processes and Intergroup Relations, 9
, 147 – 162.
Ellemers, N., & Van Knippenberg, A. (1997). Stereotyping in social con-
text. In R. Spears, P. J. Oakes, N. Ellemers, & S. A. Haslam (Eds.),
The social psychology of stereotyping and group life
(pp. 208 – 235).
Malden, MA: Blackwell.
English, T., & Chen, S. (2007). Culture and self - concept stability:
Consistency across and within contexts among Asian Americans and
European Americans.
Journal of Personality and Social Psychology,
, 478 – 490.
English, T., Chen, S., & Swann, W. B., Jr. (2008). A cross - cultural analysis
of self - verification motives. In R. Sorrentino & S. Yamaguchi (Eds.),
Handbook of motivation and cognition across cultures
(pp. 119 – 142).
San Diego: Elsevier.
Epley, N., & Dunning, D. (2006). The mixed blessings of self - knowledge
in behavioral prediction: Enhanced discrimination but exacerbated
Personality and Social Psychology Bulletin, 32
, 641 – 655.
Epstein, S. (1994). Integration of the cognitive and the psychodynamic
American Psychologist, 49
, 709 – 724.
Erdelyi, M. H. (1974). A new look at the new look: Perceptual defense and
Psychological Review, 81
, 1 – 25.
Erdelyi, M. H. (1993). Repression: The mechanism and the defense. In
D. M. Wegner & J. W. Pennebaker (Eds.),
Handbook of mental control
(pp. 126 148). Englewood Cliffs, NJ: Prentice - Hall.
Erikson, E. H. (1959).
Identity and the life cycle
. New York: International
University Press.
Ethier, K. A., & Deaux, K. (1994). Negotiating social identity when con-
texts change: Maintaining identification and responding to threat.
Journal of Personality and Social Psychology, 67
, 243 – 251.
Farnham, S. D., Greenwald, A. G., & Banaji, M. R. (1999). Implicit self -
esteem. In D. Abrams & M. Hogg (Eds.),
Social identity and social
(pp. 230 248). Cambridge, MA: Blackwell.
Fazio, R. H., Sanbonmatsu, D. M., Powell, M. C., & Kardes, F. R. (1986).
On the automatic activation of attitudes.
Journal of Personality and
Social Psychology, 50
, 229 – 238.
Fazio, R. H., & Towles - Schwen, T. (1999). The MODE model of attitude
processes. In S. Chaiken & Y. Trope (Eds.),
Dual - process theories in
social psychology
(pp. 97 116). New York: Guilford.
Feather, N. T. (1994). Values, national identification and favouritism
towards the in - group.
British Journal of Social Psychology, 33
467 – 476.
Festinger, L. (1954). A theory of social comparison processes.
Relations, 7
, 117 – 140.
Festinger, L. (1957).
A theory of cognitive dissonance
. Oxford: Row,
Fiske, S. T. (2004).
Social beings: A core motives approach to social psy-
. New York: Wiley.
Fiske, S. T., Cuddy, A. J. C., & Glick, P. (2007). Universal dimensions
of social cognition: Warmth and competence.
Trends in Cognitive
Science, 11
, 77 – 83.
Fiske, S. T., Cuddy, A. J. C., Glick, P., & Xu, J. (2002). A model of (often
mixed) stereotype content: Competence and warmth respectively follow
from perceived status and competition.
Journal of Personality and
Social Psychology, 82
, 878 – 902.
Florian, V., & Mikulincer, M. (1997). Fear of death and the judgment of
social transgressions: A multidimensional test of terror management
Journal of Personality and Social Psychology, 73
, 369 – 380.
Fordham, S., & Ogbu, J. U. (1986). Black students school success: Coping
with the burden of acting white.
Urban Review, 18
, 176 – 206.
Franks, D. D., & Marolla, J. (1976). Efficacious action and social approval
as interacting dimensions of self - esteem: A tentative formulation
through construct validation.
Sociometry, 39
, 324 – 341.
Freud, S. (1957). On narcissism: An introduction. In J. Strachey (Ed. and
The standard edition of the complete psychological works
of Sigmund Freud
(Vol. 14, pp. 67 104). London: Hogarth Press.
(Original work published 1914).
Gabriel, S., & Gardner, W. L. (1999). Are there “ his ” and “ her ” types of
interdependence? The implications of gender differences in collective
versus relational interdependence for affect, behavior, and cognition.
Journal of Personality and Social Psychology, 77
, 642 – 655.
Gagn é , M., Ryan, R. M., & Bargmann, K. (2003). Autonomy support
and need satisfaction in the motivation and well - being of gymnasts.
Journal of Applied Sport Psychology, 15
, 372 – 390.
Galdi, S., Arcuri, L., & Gawronski, B. (2008). Automatic mental associa-
tions predict future choices of undecided decision - makers,
, 1100.
Gallup, G. G. (1977). Self - recognition in primates: A comparative
approach to the bidirectional properties of consciousness.
Psychologist, 32
, 329 – 338.
Gawronski, B., & Bodenhausen, G. V. (2006). Associative and proposi-
tional processes in evaluation: An integrative review of implicit and
explicit attitude change.
Psychological Bulletin, 132
, 692 – 731.
Gecas, V. (1971). Parental behavior and dimensions of adolescent self -
Sociometry, 34
, 466 – 482.
Gergen, K. (1991).
The saturated self: Dilemmas of identity in contempo-
rary life
. New York: Basic Books.
Gibbons, F. X. (1978). Sexual standards and reactions to pornography:
Enhancing behavioral consistency through self - focused attention.
Journal of Personality and Social Psychology, 36
, 976 – 987.
Giesler, R. B., Josephs, R. A., & Swann, W. B., Jr. (1996). Self - verification
in clinical depression.
Journal of Abnormal Psychology, 105
, 358 – 368.
Gilbert, D. T., Giesler, R. B., & Morris, K. A. (1995). When comparisons
Journal of Personality and Social Psychology, 69
, 227 – 236.
Gill, M. J., & Swann, W. B., Jr. (2004). On what it means to know someone:
A matter of pragmatics.
Journal of Personality and Social Psychology,
, 405 – 418.
Goffman, E. (1959).
The presentation of self in everyday life
. Garden City,
NY: Doubleday – Anchor.
Goldstein, N. J., & Cialdini, R. B. (2007). The spyglass self: A model
of vicarious self - perception.
Journal of Personality and Social
Psychology, 92
, 402 – 417.
G ó mez, A., Seyle, D. C., Huici, C., & Swann, W. B., Jr. (in press). Can
self - verification strivings fully transcend the self - other barrier? Seeking
verification of ingroup identities. Manuscript submitted for publica-
Journal of Personality and Psychology
Gosling, S. D., John, O. P., Craik, K. H., & Robins, R. W. (1998). Do people
know how they behave? Self - reported act frequencies compared with
on - line codings by observers.
Journal of Personality and Social
Psychology, 74
, 1337 – 1349.
CH16.indd 620CH16.indd 620 10/22/09 2:50:11 PM10/22/09 2:50:11 PM
References 621
Gosling, S. D., Ko, S. J., Mannarelli, T., & Morris, M. E. (2002). A room
with a cue: Personality judgments based on offices and bedrooms.
Journal of Personality and Social Psychology, 82
, 379 – 398.
Gramzow, R., & Tangney, J. P. (1992). Proneness to shame and the
narcissistic personality.
Personality and Social Psychology Bulletin,
, 369 – 376.
Greenberg, J., & Musham, C. (1981). Avoiding and seeking self - focused
Journal of Research in Personality, 15
, 191 – 200.
Greenberg, J., Simon, L., Pyszczynski, T., Solomon, S., & Chatel, D.
(1992). Terror management and tolerance: Does mortality salience
always intensify negative reactions to others who threaten one ’ s world-
Journal of Personality and Social Psychology, 63
, 212 – 220.
Greenberg, J., Solomon, S., Pyszczynski, T., Rosenblatt, A., Burling, J.,
Lyon, D., Simon, L., & Pinel, E. (1992). Why do people need self -
esteem? Converging evidence that self - esteem serves an anxiety -
buffering function.
Journal of Personality and Social Psychology,
, 913 – 922.
Greenwald, A. G., & Banaji, M. R. (1995). Implicit social cognition:
Attitudes, self - esteem, and stereotypes.
Psychological Review, 102
4 – 27.
Guidano, V. F., & Liotti, G. (1983).
Cognitive processes and emotional
disorders: A structural approach to psychotherapy
. New York:
Guilford Press.
Hampson, S. E., John, O. P., & Goldberg, L. R. (1987). Category - breadth
and social - desirability values for 573 personality terms.
Journal of Personality, 1
, 241 – 258.
Hart, J., Shaver, P. R., & Goldenberg, J. L. (2005). Attachment, self - esteem,
worldviews, and terror management: Evidence for a tripartite security
Journal of Personality and Social Psychology, 88
, 999 – 1013.
Harter, S. (1999).
The construction of the self
. New York: Guilford Press.
Harter, S. (2006). The development of self - esteem. In M. Kernis (Ed.),
Self - esteem issues and answers: A sourcebook of current perspectives
(pp. 144 150). New York: Psychology Press.
Hazan, C., & Shaver, P. R. (1987). Romantic love conceptualized as an
attachment process.
Journal of Personality and Social Psychology, 52
511 – 524.
Hazan, C., & Shaver, P. R. (1994). Attachment as an organizational frame-
work for research on close relationships.
Psychological Inquiry, 5
1 – 22.
Hebb, D. O. (1949).
The organization of behavior
. New York: Wiley.
Heine, S. J., & Hamamura, T. (2007). In search of East Asian self -
Personality and Social Psychology Review, 11
, 1 – 24.
Heine, S. J., Kitayama, S., & Hamamura, T. (2007). The inclusion of
additional studies yields different conclusions: A reply to Sedikides,
Gaertner, & Vevea (2005), JPSP.
Asian Journal of Social Psychology,
, 49 – 58.
Heine, S. J., Kitayama, S., & Lehman, D. R. (2001). Cultural differences in
self - evaluation: Japanese readily accept negative self - relevant informa-
Journal of Cross - Cultural Psychology, 32
, 434 – 443.
Heine, S. J., Lehman, D. R., Markus, H. R., & Kitayama, S. (1999). Is there
a universal need for positive self - regard?
Psychological Review, 106
766 – 794.
Heine, S. J., Proulx, T., & Vohs, K. D. (2006). The meaning maintenance
model: On the coherence of social motivations.
Personality and Social
Psychology Review, 10
, 88 – 110.
Helgeson, V. S., & Mickelson, K. D. (1995). Motives for social compari-
Personality and Social Psychology Bulletin, 21
, 1200 – 1209.
Helweg - Larsen, M., & Shepperd, J. A. (2001). Do moderators of the opti-
mistic bias affect personal or target risk estimates? A review of the
Personality and Social Psychology Review, 5
, 74 – 95.
Higgins, E. T. (1987). Self - discrepancy: A theory relating self and affect.
Psychological Review, 94
, 319 – 340.
Higgins, E. T. (1998). Promotion and prevention: Regulatory focus as a
motivational principle. In M. P. Zanna (Ed.),
Advances in experimental
social psychology
(Vol. 30, pp. 1 – 46). New York: Academic Press.
Higgins, E. T., Bond, R. N., Klein, R., & Strauman, T. (1986). Self -
discrepancies and emotional vulnerability: How magnitude, accessibility,
and type of discrepancy influence affect.
Journal of Personality and
Social Psychology, 51
, 5 – 15.