Social Cognition, Vol. 33, No. 3, 2015, pp. 169–185
© 2015 Guilford Publications, Inc.
Address correspondence to Rebecca J. Schlegel, Department of Psychology, Texas A&M University,
4235 TAMU, College Station, TX 77843-4235; E-mail: firstname.lastname@example.org.
BENCH ET AL.
Discovery in the Self and Others
Thinking abouT Change in The Self and oTherS:
The role of Self-diSCovery MeTaphorS
and The True Self
Shane W. Bench
Washington State University
Rebecca J. Schlegel and William E. Davis
Texas A&M University
Montana State University
People change over the course of their lives, yet little is known about
how people think about these changes. We expected that evaluative
judgments of changes would relate to the type of metaphors people
use to describe those changes. Specically, we predicted that the more
positively a change is evaluated, the more likely it is to be perceived as
a self “discovery” (i.e., a change driven by discovering something with-
in the self). Study 1 established a correlational relationship between
perceived positivity and self-discovery in changes in both the self and a
close other. Study 2 manipulated the valance of the change and found
that positive changes were more likely to be endorsed as self-discover-
ies than negative changes. These ndings highlight the importance of
self-discovery metaphors in understanding how people make sense of
changes in the self and close others. Implications for meaning making,
well-being, and narrative research are discussed.
There is little doubt that people change over the course of their lives (e.g.,
Hopwood et al., 2011; Robins, Noftle, Trzesniewski, & Roberts, 2005; Rob-
erts, Walton, & Viechtbauer, 2006). Yet, little is known about how people
170 benCh eT al.
evaluate and make sense of changes in both the self and other people. We
suggest a strong link between the perceived valence of a change and peo-
ple’s beliefs about how the change came about. Speciﬁcally, we hypoth-
esized that the more positively a change is evaluated, the more likely it is to
be perceived as a case of self “discovery.” The logic of this hypothesis rests
on widespread lay beliefs that true selves are discovered (Schlegel, Vess,
& Arndt, 2012) and that true selves are “fundamentally good” (Newman,
Bloom, & Knobe, 2014). Accordingly, a positive change should be seen as a
movement toward the true self and thus be perceived as a self-discovery. In
contrast, a negative change should be seen as a movement away from the
true self and consequently less driven by self-discovery.
Self-diSCovery MeTaphorS and The True Self
Waterman (1984) proposed that self-discovery metaphors help people
make sense of the process of identity formation. Discovery metaphors are
likened to the activities of scientists and explorers and refer to the act of
ﬁnding something that already exists. The use of a self-discovery metaphor
suggests that one has “found” a characteristic that deﬁnes who he or she
is and that this part of the self was always, to some degree, within the self.
Consistent with this idea, Heatherton and Nichols (1994) report that over
70% of people asked to write a story describing a life change mentioned
that they gained increased self-knowledge. Discovered aspects of the self
are typically thought of as the constituents of a person’s true self (Schlegel
& Hicks, 2011; Schlegel et al., 2012), something that most people believe
to be a relatively immutable set of characteristics (Johnson, Robinson, &
Mitchell, 2004) that are important to deﬁning who someone really is (Harter
& Monsour, 1992).
The link between self-discovery and the true self is important to the cur-
rent research because people tend to think of true selves as “fundamentally
good” (Newman et al., 2014). For example, people like their true selves
better than their actual selves (i.e., their outward behavior in their daily
lives; Schlegel, Hicks, Arndt, & King, 2009) and their “false selves” (Harter,
2002; Harter & Monsour, 1992). They also experience a host of positive psy-
chological beneﬁts when they feel like they are “in touch” with their true
selves, such as increased self-esteem (Andersen & Williams, 1985), reduced
defensiveness (e.g., Arndt, Schimel, Greenberg & Pyszcynski, 2002), less
ego-involved emotions (Vess, Schlegel, Hicks, & Arndt, 2014), increased
meaning in life (Schlegel et al., 2009; Schlegel Hicks, King, & Arndt, 2011),
and increased decision satisfaction (Schlegel, Hicks, Davis, Hirsch, & Smith,
2013). Further, most people hold a lay theory that the true self should be
used to guide one’s decision making (Schlegel et al., 2013), suggesting that
they explicitly believe the true self is fundamentally good and important.
diSCovery in The Self and oTherS 171
These tendencies to see one’s own true self as positive and psychological-
ly beneﬁcial could be dismissed as simple examples of self-serving biases
(e.g., Baumeister, 1998; Sedikides, Gaertner, & Toguchi, 2003). However,
Newman and colleagues (2014) intriguingly suggest that people may think
of all true selves as fundamentally good. Across three studies, Newman
and colleagues presented participants with ﬁctitious targets who changed
in ways that could be viewed as morally good (e.g., changing from a “dead-
beat” dad to a loving and caring father) or morally bad (e.g., changing from
a teetotaler to an alcoholic). Participants in all three studies believed that
the changes toward the morally good behaviors reﬂected an emergence
of the target’s true self. The authors concluded that people possess a lay
theory that suggests “deep inside every individual, there is something mo-
tivating him or her to behave in ways that are virtuous” (p. 203). This work
by Newman and colleagues suggests that people not only believe they are
good inside (as might be predicted from a self-serving bias perspective),
but also hold the broader conviction that all people are good inside.
Building from the research described above, we examined whether evalua-
tions of changes in the self and others are linked to people’s perceptions of
how those changes came about. We predicted that the perceived positivity
of changes in the self and others would positively predict the belief that
those changes were examples of self-discoveries. We made our predictions
based on the previously described research linking the true self to self-
discovery metaphors (Schlegel et al., 2012) and research demonstrating
that positive changes in others are viewed as more representative of an
authentic self (Newman et al., 2014). Thus, a change for the better should
be viewed as more representative of the true self and therefore more con-
sistent with the notion of true self-discovery, and vice versa.
One notable aspect of our approach is that we examined the perception
of changes in both the self and close others in both studies. This allowed us
to examine the potential speciﬁcity and generality of any observed rela-
tionships. Newman and colleagues (2014) suggested that people may think
all true selves are good; however their work only examined hypothetical
targets. It is possible that changes in people that are actually known are
judged differently than such hypothetical targets. After all, changes in a
close other may lead the observer to question his or her understanding
of the close other and wonder if the inferences he/she has previously
made about the actor were ever accurate. In turn, this could threaten the
desired belief that one’s understanding of the world is correct (Festinger,
1957; Fromm, 1941; Hardin & Higgins, 1996). In this way, changes in “real”
people may be more likely to be degraded than changes in hypothetical
172 benCh eT al.
Of course, changes in the self are also at odds with consistency motives
(e.g., Greenwald, 1980; Kernis & Goldman 2006; Swann, 1990) and research
suggests that both negative personality changes (e.g., increasing neuroti-
cism) and positive personality changes (e.g., decreasing neuroticism, in-
creasing conscientiousness) are associated with poorer subjective well-be-
ing and health (Eizenman, Nesselroade, Featherman, & Rowe, 1997; Hu-
man, Biesanz, Miller, Chen, Lachman, & Seeman, 2013; Mroczek & Spiro,
2007). Such evidence might imply that changes in the self would be seen
as negative. However, a self-enhancement perspective suggests (e.g., Bau-
meister, 1998; Sedikides, Gaertner, & Toguchi, 2003) that it less likely that
changes in the self will be degraded. Thus, in light of these divergent per-
spectives on perceiving change in the self and others, it seems important to
examine perceptions of change in both the self and “real” others to test the
full contours of our guiding analysis.
Study 1 was designed as an initial investigation into the relationship be-
tween discovery endorsement and positivity judgments. Participants were
ﬁrst asked to describe a change in themselves or a close other. This ap-
proach was thus broader than the approach of Newman et al. (2014) be-
cause it examined non-hypothetical changes that did not necessarily have
speciﬁc moral implications. An examination of the essays in Study 1 indeed
revealed a wide variety of changes including changes in traits, religious be-
liefs, behaviors, attitudes, and motivation. After writing about a change, all
participants evaluated the valence of the change and indicated the extent
to which the change reﬂected a process of self-discovery. We predicted that
the more positively a change was evaluated, the more likely it would be
considered a form of self-discovery.
Study 1 also explored the idea that self-discovery metaphors may be
associated with measures of personal well-being. Previous work on self-
discovery metaphors has found that endorsing the idea that true selves
are discovered (in general) positively predicts meaning in life judgments
(Schlegel et al., 2012). Thus, we conducted exploratory correlation analyses
to examine the relationship between discovery beliefs, change evaluation,
and three well-being measures (meaning in life, self-esteem, and satisfac-
tion with life).
Participants. Ninety-ﬁve (76 females) introductory psychology students
at Texas A&M University participated for course credit. Participants were
predominantly White (67%). In accordance with recent data collection
practices (Simmons, Nelson, & Simonsohn, 2011), we established upfront
diSCovery in The Self and oTherS 173
to collect as many cases as possible in one week, with a maximum of 100
(i.e., 50 cases in each condition). This resulted in 95 participants complet-
ing the study. All collected observations were included in the reported data
Materials and Procedure. Participants completed the study on individual
computers in private laboratory cubicles in small groups.
They were randomly assigned to write an essay about a change in either
themselves or a close other. Those assigned to the close other condition
were told to “think of a speciﬁc person who is a close friend of yours,” and
were asked to type the name of the person into a text box. The software
was programmed to place the name that was entered in this box into the
subsequent manipulation and dependent measures. This was done to keep
participants focused on the same individual throughout the study. Partici-
pants then read the following prompt (modiﬁcations for close other condi-
tion presented in brackets):
People often change over time. We’d like you to take a few minutes and think
about one way that you have [(friend’s name) has] changed over time. Below,
please describe the change that occurred and reﬂect on why/how this change
may have occurred. Please try to really get into the writing task, and provide
as much detail as possible about this topic.
After the writing task, participants completed a series of items designed to
assess evaluations of the change and perceptions that the change was due
For the overall evaluative judgment of the change, participants respond-
ed to ﬁve items: “The change I wrote about made me [(friend’s name)] a
better person”; “The change I wrote about made me [(friend’s name)] more
likable”; “I am happy about the change in who I am [(friend’s name) is)]”;
“I am glad I [(friend’s name)] changed”; and “If I could go back in time, I
would still want [(friend’s name)] to make the same change.” Responses
were made on a 1 (strongly disagree) to 7 (strongly agree) scale and averaged
into a single composite (M = 5.55, SD = 1.29, α = .92).
For the endorsement of discovery beliefs, participants responded to three
items: “The change I wrote about is a product of me [(friend’s name)] dis-
covering more about who I am [he/she is] inside”; “The change I wrote
about is something I [(friend’s name)] discovered about my [his/her] self”;
and “The change I wrote about reﬂects me [(friend’s name)] becoming
more like my [his/her] TRUE self.” Responses were made on a 1 (strongly
disagree) to 7 (strongly agree) scale and averaged into a single composite (M
= 5.50, SD = 1.16, α = .82).
Finally, participants completed measures of well-being. Meaning in life
was measured by the ﬁve-item presence subscale of the Meaning in Life
Questionnaire (MLQ; Steger, Frazier, Oishi, & Kaler, 2006), (M = 5.00, SD =
1.39, α = .92; sample item “I understand my life’s meaning”), self-esteem
174 benCh eT al.
was assessed by the ten-item Rosenberg Self-Esteem Scale (RSES; Rosen-
berg, 1965; M = 5.24, SD = 1.04, α = .86; sample item “I feel that I have
a number of good qualities”), and life satisfaction was measured by the
ﬁve-item Satisfaction with Life Scale (SWLS; Diener, Emmons, Larsen, &
Grifﬁn, 1985; M = 4.92, SD = 1.24, α = .85; sample item “I am satisﬁed with
RESULTS AND DISCUSSION
The primary analysis was to examine the relationship between evaluative
judgments and discovery endorsement. These two variables were signiﬁ-
cantly correlated at the bivariate level (r = .60, p < .001). Further, a hierar-
chical regression predicting discovery endorsement from evaluative judg-
ments, target (i.e., self vs. other), and their interaction, revealed a main ef-
fect of evaluative judgments, b = .59, t = 7.52, p < .001, such that participants
who felt more positive about the change more strongly endorsed discovery
beliefs. By comparison, the main effect of target, b = -.39, t = 1.91, p = .06,
and the interaction between the two variables, b = .10, t = .05, p = .62, were
not signiﬁcant (though the effect of target was marginal, suggesting a ten-
dency to endorse changes in the self as discoveries more than changes in
An ancillary independent samples t-test revealed that participants had
more positive evaluations of a change in the self (M = 6.02, SD = 0.80),
compared to a change in a close friend (M = 5.15, SD = 1.49), t(80.82) =
3.62, p < .001, Cohen’s d = .81. This is consistent with research that shows
people tend to judge themselves more favorably than others (Brown, 1986;
Chambers & Windschitl, 2004). A second t-test also indicated that discov-
ery endorsement did not differ between the self (M = 5.57, SD =1.05) and
close other conditions (M = 5.44, SD = 1.25), t(93) = .51, p = .61, Cohen’s d =
.11, suggesting that participants were equally likely to judge changes in the
self and others as reﬂective of discovery.
Supplementary Data Coding and Analyses. As a supplementary analysis,
we asked ﬁve undergraduate coders (naive to study hypotheses) to read
the essays generated by the participants and answer a few questions about
each. The purpose of this analysis was to determine whether simply read-
ing essays describing real personality changes would naturally elicit the
same relationship between (perceived) valence and discovery. Observing
a similar relationship among naive coders would provide evidence for
the robustness of the link between these two constructs in people’s minds
and further help to address concerns that our results are driven by self-en-
hancement motives. Coders speciﬁcally answered two questions regarding
the positivity of the change: “In my opinion, the person in the essay seems
to have made a change for the better”; and “In my opinion, the change
written about makes the person more likeable”; and one question regard-
diSCovery in The Self and oTherS 175
ing the extent to which the essay conveyed a sense of self-discovery: “In
my opinion, the change in the essay is a product of the person discovering
more about who they are inside.” All ratings were made on a 1 (not at all)
to 7 (very much) scale. The coders were instructed not to “worry about what
words they literally use, we’re just interested in your general impressions based on
the information provided in each essay.” Despite instructions that allowed for
a great deal of subjectivity when answering the questions, the ﬁve raters
scored the essays very similarly (assessed via two-way mixed intraclass
correlation coefﬁcients for each of the three questions; average measure
ICCs = .90, .85, .80) and we averaged their ratings. The two positivity items
were highly correlated (r = .90, p < .001) and were thus combined into a
composite. Interestingly, the naive coder’s ratings were very similar to the
participant’s ratings (r = .74, p < .001, for positivity ratings; r = .65 for dis-
covery ratings), suggesting convergence not only among the naive coders
but across all raters.
We then examined the correlations between the positivity ratings and the
discovery item. The results revealed that the more positive the essays were
judged to be, the more likely they were to also be judged as more indicative
of self-discoveries (r = .77, p < .001). This suggests that when people read es-
says about changes for the better, they also see those changes as likely to be
self-discoveries, even if they do not know the target. As self-enhancement
biases can extend, albeit to a lessened degree, to close others (e.g., Lench,
Smallman, Darbor, & Bench, 2015; Pahl, Eiser, & White, 2009), the ratings
of coders that did not know the targets provides further evidence that the
relationship between these constructs is not driven by self-enhancement
biases. Rather, it seems that the connection between these constructs forms
such a strong lay theory for people that simply reading about a change that
is for the better naturally conjures a sense of self-discovery (a proposition
that is consistent with the ﬁndings of Newman et al., 2014). As an illustra-
tive example, below is the text from one essay rated by the coders as one of
the two most positive essays in the sample (M = 6.5):
I’ve known X since our ﬁrst day of elementary school. X was always very shy
and very humble. She began cheerleading classes when she was in ﬁrst grade,
and being that it was her ﬁrst year, she wasn’t very good. As time progressed,
though, she gained more skill and more knowledge. By the time she entered
high school, she stuck out amongst her peers when it came to cheerleading.
Because of this, she was named captain of the freshman cheer squad. Later in
her high school career, she was junior varsity captain and then varsity cap-
tain. Naturally, along with skill comes conﬁdence. She was ﬁnally getting to
the point where she could cheer in front of hundreds of people without get-
ting nervous. I’ve seen such a change in conﬁdence over the years in X. The
most important thing, though, is that the humility stayed. X now cheers for X
University not in front of hundreds, but in front of thousands of people each
and every weekend. This is not something anyone would ever assume if they
knew X in the ﬁrst grade.
176 benCh eT al.
Compare this to the text of an essay rated by the coders as the least positive
(M = 1.8):
Junior year of high school X started sleeping with a lot of boys and partying a
lot. Before, she was the most innocent girl, and even a little awkward around
boys. She started talking about me behind my back and pretending that she
was better than everyone else. I think all of this was happening because she
had just become one of the cheer captains, and was just beginning to be no-
ticed by boys.
Neither makes explicit mention of the words “discovery,” “true self,” or
“genuine,” yet the text itself seems to naturally conjure up the idea that the
person written about is either becoming more or less like who they “truly
are” inside. In this way, “discovery sequences” may be a common theme
in people’s life stories, much like redemptive sequences (McAdams, 2005).
Exploratory Well-Being Analyses. Discovery beliefs (speciﬁc to the change
participants wrote about) were positively correlated with presence in
meaning in life (r = .24, p = .02), self-esteem (r = .33, p = .001), and satisfac-
tion with life (r = .32, p = .003). These correlations appear to be indepen-
dent from the positivity of those changes, as evaluation scores were not
signiﬁcantly correlated with any of these three well-being measures (ps >
.17). Further, these relationships were not moderated by target (i.e., self
vs. other) when the interaction between valance and target was tested in a
multiple regression. This suggests that self-discovery metaphors relate to
well-being, even when they are applied to close others. This provides some
indication (consistent with Schlegel et al., 2012) that self-discovery beliefs
are a marker of (or can help foster) positive psychological functioning.
Study 2 was conducted to extend the ﬁndings of Study 1 by ruling out al-
ternative explanations for the relationship between the perceived positiv-
ity of changes in the self and others and self-discovery beliefs. Due to the
very open-ended nature of the prompt used in Study 1, a host of individual
differences could account for the observed relationship. For instance, par-
ticipants with high level of trait optimism may be more likely to think that
the change they wrote about is both positive and a self-discovery. Study 2
better accounts for these types of alternative explanations by manipulating
the valence of the change participants were asked to write about. That is,
the study was a 2 (target: self, other) × 2 (valence: positive, negative) design
that crossed the manipulation from Study 1 with a manipulation of change
valence. Speciﬁcally, participants were randomly assigned to reﬂect upon
a change they considered to have been for the better or for the worse in
either the self or a close friend, and then reported their endorsement of dis-
diSCovery in The Self and oTherS 177
covery beliefs using the scale from Study 1. We predicted that positive (vs.
negative) changes would signiﬁcantly increase discovery endorsement, re-
gardless of whether or not the change occurred in the self or a close other.
Participants. One hundred eighty-six (111 females, 30 did not report gen-
der) introductory psychology students at Texas A&M University partici-
pated for course credit. Participants were predominantly White (62%). Our
goal was to collect 200 participants (i.e., 50 cases in each condition), how-
ever, the semester ended during data collection, closing our participant
pool, resulting in 186 participants completing the study. All observations
are reported and included in all analyses.
Materials and Procedure. The study procedure was largely the same as
Study 1. Participants in the positive, self condition read the following
prompt (changes for the negative condition in brackets):
People often change over time. Some of these changes are positive, meaning
the person changed for the better, and some of these changes are negative,
meaning the person changed for the worse. We’d like you to take a few minute
and think about one way that you have changed over time that you believe
was a positive [negative] change (for the better [worse]). Below, please de-
scribe the positive [negative] change that occurred and reﬂect on why/how
this change may have occurred. Please try to really get into the writing task,
and provide as much detail as possible about this topic.
Participants in the close other condition read the same prompt except that
their friend’s name was inserted into the text where appropriate (as in
Study 1). After completing the writing task, participants were asked the
same self-discovery questions from Study 1 (M = 4.66, SD = 1.39, α = .76).
Participants also completed other measures not relevant to our primary
hypotheses (e.g., lay theories of human nature; all measures available upon
RESULTS AND DISCUSSION
A 2 (target: self, other) × 2 (valence: positive, negative) ANOVA was used to
evaluate the effect of the two manipulations on the endorsement of discov-
ery beliefs. There was a main effect of valence, F(1, 186) = 130.66, p < .001;
η2 = .41, with participants reporting stronger discovery endorsement when
considering a positive change (M = 5.60, SD = 0.92) than a negative change
(M = 3.80, SD = 1.88, d = 1.22). There was not a signiﬁcant interaction be-
tween valence and target, F(1, 186) = 0.001, p = .97, η2 = .00, nor was there
a signiﬁcant main effect of target, F(1, 186) = 1.96, p = .16, η2 = .01. Study
178 benCh eT al.
2 provides converging evidence that when considering a positive change,
people are more likely to think of that change as a case of self-discovery.
Supplementary Data Coding and Analyses. We asked four undergraduate
coders (naive to study hypotheses) to code the essays generated by the
participants on the same three questions used in Study 1. The raters again
scored the essays very similarly (average measure ICCs = .95, .94, .83) and
we averaged their ratings. The two valence items were again highly cor-
related (r = .99, p < .001) and combined into a composite.
The correlation between participant and coder-rated discovery was
smaller than in Study 1, but still signiﬁcant (r = .270, p < .001). Participants
didn’t report on the perceived valence of the change, thus we could not
compare coder and participant ratings on valence. We did compare the
coder’s valence ratings across conditions as a manipulation check, how-
ever. The results revealed the expected signiﬁcant effect of the valence ma-
nipulation, F(1, 186) = 82.06, p < .001, η2 = .31 (MNegCond = 2.42, SDNegCond =
3.38; MPosCond = 5.83, SD PosCond = .71; d = 1.40). There was not a signiﬁcant
interaction between valence and target, F(1, 186) = 1.71, p = .19, η2 = .01, nor
was there a signiﬁcant main effect of target, F(1, 186) = .54, p = .46, η2 = .003.
Of primary interest, coder’s ratings of perceived valence and discovery
were highly correlated (r = .94, p < .001), replicating Study 1. Further, when
the coder’s ratings were compared across conditions, the same pattern of
results that emerged for the participant ratings was observed. Speciﬁcally,
there was a strong main effect of valence, F(1, 186) = 23.39, p < .001, η2 = .11,
with coders detecting stronger discovery endorsement when considering a
positive change (M = 5.46, SD = .60) than a negative change (M = 3.69, SD
= 3.24, d = .76). There was not a signiﬁcant interaction between valence and
target, F(1, 186) = 2.02, p = .16, η2 = .01, nor was there a signiﬁcant main ef-
fect of target, F(1, 186) = .39, p = .53, η2 = .002. This provides further support
to the idea that positive changes naturally seem to conjure ideas related to
Despite the fact that people often change over the course of their lives,
little is known about how people evaluate and think about these changes.
This seems important given that when people change, those changes have
the potential to threaten our desires for consistency in the self and oth-
ers (Festinger, 1957; Fromm, 1941; Hardin & Higgins, 1996; Landau et al.,
2004). Personality changes are also at odds with our motivation to believe
that our understanding of the world is correct (Festinger, 1957; Fromm,
1941; Hardin & Higgins, 1996). In this way, people may employ metaphors
to help them “make sense” of personality changes (e.g., Landau, Robinson,
& Meier, 2013).
diSCovery in The Self and oTherS 179
Two studies were conducted to examine our hypothesis that people are
likely to engage self-discovery metaphors when they perceive a change in
the self or another person as being positive. Across both studies, results re-
vealed a clear and strong relationship between the judgment that someone
has changed for the “better” and the endorsement of the idea that the change
was an example of self-discovery. Speciﬁcally, in Study 1, participants were
asked to write about any change they had observed within themselves or
a close other and respond to questions about that change. Results revealed
a strong positive correlation between evaluative judgments and discovery
endorsement. In Study 2, we manipulated evaluative judgments by explic-
itly asking participants to write about changes that they believed were for
the better or the worse. Results revealed that, regardless of whether the
target was the self or close other, positive changes were far more likely to
be endorsed as self-discoveries than negative changes.
Taken together, these studies are the ﬁrst to suggest that people employ
self-discovery metaphors when they think about real changes in the self
and others. These studies are also the ﬁrst to identify a variable (i.e., per-
ceived positivity of the change) that predicts when self-discovery meta-
phors are likely to be employed. Given that people think of discovered
self-aspects as representative of a person’s “true self” (Schlegel, Vess, &
Arndt, 2012), the results provide converging evidence with Newman et
al.’s (2014) contention that people believe all true selves are fundamentally
good. That is, when people perceive someone as making a change that is
“for the better,” they perceive that person as becoming more like their true
self. The current results build upon those of Newman and colleagues by
demonstrating the power of these lay beliefs about the true self beyond the
domain of morality and beyond the judgment of hypothetical targets, thus
further suggesting this powerful lay belief has meaningful implications in
WHY SELF-DISCOVERY METAPHORS MATTER
The current ﬁndings speak to how the process of identity development
is perceived and the ways in which people make sense of their own (and
others’) life stories. One purpose of a life story is to create a narrative that
provides one’s life with a sense of coherence and purpose (e.g., McAdams,
1995; McAdams, Diamond, de St. Aubin, & Mansﬁeld, 1997). Self-discov-
ery metaphors may serve an important role in promoting a sense of coher-
ence in a life story by promoting perceptions of consistency even in the face
of self-concept change. That is, when people feel that they have discovered
a “new aspect” of themselves, they can reﬂect back on the past to search
for supportive evidence that this was indeed who they have always been
and more fully integrate this evidence into their life story. In this way, self-
180 benCh eT al.
discovery metaphors may be an important vehicle that people use to ﬁnd
evidence that, at their core, they are the same person over time, despite the
changes they see in themselves. This might help them establish a coher-
ent life story and reap the psychological and physical health beneﬁts that
are associated with life story coherence (Baerger & McAdams, 1999; Pen-
nebaker, 1997; White & Epston, 1990). In this way, discovery themes may
constitute a part of the cultural “expectations of what makes a healthy nar-
rative and a healthy self” (McLean, Pasupathi, & Pals, 2007, p. 262). Consis-
tent with this possibility, previous work on self-discovery metaphors has
found that endorsing the idea that the true self was discovered positively
predicts meaning in life judgments (Schlegel et al., 2012). Our exploratory
correlation analyses form Study 1 similarly revealed that discovery beliefs
(speciﬁc to the change participants wrote about) positively correlated with
meaning in life, self-esteem, and satisfaction with life. Further, those cor-
relations did not seem to be explained simply by the positivity of “discov-
ered” changes, as evaluation scores were not signiﬁcantly correlated with
any of these three well-being outcomes. Further, these relationships were
not moderated by target (i.e., self vs. other), suggesting that there is some-
thing about self-discovery metaphors, even when applied to close others,
that relates to well-being. Though not tested directly, we suspect that this is
because discovery metaphors help people derive meaning from the chang-
es. Research suggests that even positive personality changes (e.g., decreas-
ing neuroticism, increasing conscientiousness) are associated with poorer
subjective well-being and health (Eizenman, Nesselroade, Featherman,
& Rowe, 1997; Human, Biesanz, Miller, Chen, Lachman, & Seeman, 2013;
Mroczek & Spiro, 2007). However, when self-discovery metaphors are ap-
plied to the self, this should promote life story coherence (as discussed in
the previous paragraph; Bearger & McAdams, 1999) and potentially “dif-
fuse” the threat aroused by changes in the self. Similarly, self-discovery
metaphors applied to others may “diffuse” the threat that changes in close
others pose to our desires for others to be consistent (e.g., Landau et al.,
2004) and thus protect our desired belief that one’s understanding of the
world is correct (Festinger, 1957; Fromm, 1941; Hardin & Higgins, 1996). Of
course, these analyses are preliminary and should certainly be interpreted
with some caution. However, these results provide at least some indication
that the use of self-discovery metaphors is either a marker of (or can help
foster) positive psychological functioning. Future research should directly
examine how the presence of discovery sequences in life stories may re-
late to the perceived continuity of one’s life story and how this, in turn,
may inﬂuence other outcomes such as well-being, self-concept clarity, and
meaning in life.
diSCovery in The Self and oTherS 181
Future research should examine the potential bi-directionality of the effects
observed in these studies. Perhaps prompting someone to think about a
change as an example of self-discovery would positively impact the per-
son’s evaluation of that change. This could be useful for people struggling
to accept change in the self or a close other.
It would also be interesting to examine whether these effects are unique
to self-discovery metaphors or would be found for other metaphors. Are
there other metaphors that may be applied to a change and equally effective
in helping people make sense of it? Waterman (1984) proposed that self-
creation is the alternative metaphor to self-discovery. Might self-creation
metaphors be equally useful in making sense of changes? We suspect not.
As Waterman notes, self-creation metaphors suggest there is no true self
and subscribing to such a view can actually result in “existential dread.”
Consistent with Waterman’s contentions, previous research has found that
the true self is far more likely to be considered discovered than created,
and that, unlike self-discovery metaphors, the endorsement of self-creation
metaphors fails to consistently relate to meaning in life (Schlegel, Vess, &
Arndt, 2012). However, there may be other meaning making metaphors
that can be identiﬁed through further research.
A related issue is how self-discovery metaphors may or may not be ap-
plied in different situations. For example there are certain times in life
when personality change is less likely to be threatening because changes
are normative and even expected (e.g., after getting a job or having chil-
dren). What metaphors might be applied in these situations? Even if they
aren’t threatening, we suspect that self-discovery metaphors may still be
applied. Other research has shown that behaving in normative ways feels
authentic (e.g., Fleeson & Wilt, 2010; Sherman, Nave, & Funder, 2012). As
such, we expect normative changes might be even more likely to be seen
as movements toward the true self and that people may think of such life
transitions as moments in their life when they “found themselves.” Simi-
larly, we have found that when people think about how they have changed
over the course of their lives, they tend to think they have become increas-
ingly authentic over time and that they will continue to become more au-
thentic in the future (Seto & Schlegel, 2015), suggesting that people may see
movements toward the true self as a normative part of maturation.
182 benCh eT al.
The current ﬁndings are limited by the experimental methods employed
and do not include a strong test of the causal effect of change valence on
self-discovery beliefs. That is, participants were given instructions to self-
select examples of changes and were not assigned to directly experience a
manipulated feeling of change. The strictest test of the key causal pathway
would require randomly assigning participants to experience or observe
a positive or negative personality change and measuring the difference in
self-discovery beliefs that follow. While our methods do rule out several al-
ternative explanations, they cannot directly speak to the strictest interpre-
tation of causality. An additional limitation may be the relative homogene-
ity of the samples in terms of age and culture. Previous studies with adult
samples have revealed evidence suggestive of a shared belief that true
selves are fundamentally good (Newman et al., 2014), but future studies
should consider the potential role of age and culture in the present effects.
Indeed, research has demonstrated that dialectical cultures are more ac-
cepting of self-concept inconsistencies than non-dialectical cultures (Spen-
cer-Rodgers, Boucher, Mori, Wang, & Peng, 2009), suggesting that culture
could have an interesting impact on the perception of change.
A growing body of research has demonstrated that people change over
time (e.g., Hopwood et al., 2011; Roberts et al., 2006; Robins et al., 2005)
and that people believe they have changed over the course of their lives
(Heckhausen, Dixon, & Baltes, 1989; Robins et al., 2005; Quoidbach, Gil-
bert, & Wilson, 2013). Our ﬁndings are, to our knowledge, the ﬁrst to exam-
ine how people evaluate these changes within the self and close others in
terms of self-discovery metaphors. The ﬁndings build on previous research
(Schlegel et al., 2012) that suggests self-discovery beliefs play an important
role in the way we think about the self and others as well as research on
people’s beliefs that the true selves are fundamentally good (Newman et
al., 2014). Our hope is that this work will provide a fertile foundation for
future research on the nature of identity development, the construction of
life stories, and person perception.
diSCovery in The Self and oTherS 183
Andersen, S. M., & Williams, M. (1985).
Cognitive/affective reactions in
the improvement of self-esteem:
When thoughts and feelings make a
difference. Journal of Personality and
Social Psychology, 49, 1086-1097.
Arndt, J., Schimel, J., Greenberg, J., & Pyszc-
zynski, T. (2002). The intrinsic self and
defensiveness: Evidence that activating
the intrinsic self reduces self-handicap-
ping and conformity. Personality and So-
cial Psychology Bulletin, 28, 671-683.
Baerger, D. R., & McAdams, D. P. (1999). Life
story coherence and its relation to psy-
chological well-being. Narrative Inquiry,
Baumeister, R. F. (1998). The self. In D. T. Gil-
bert, S. T. Fiske, & G. Lindzey (Eds.),
Handbook of social psychology (4th ed., pp.
680-740). New York: McGraw-Hill.
Brown, J. D. (1986). Evaluations of self and
others: Self-enhancement biases in social
judgments. Social Cognition, 4, 353-376.
Brown, J. D. (2012). Evaluations of self and oth-
ers: Self-enhancement biases in social
judgment. Social Cognition, 4, 353-376.
Chambers, J. R., & Windschitl, P. D. (2004). Bi-
ases in social comparative judgments:
The role of nonmotivated factors in
above-average and comparative-opti-
mism effects. Psychological Bulletin, 130,
Diener, E., Emmons, R. A., Larsen, R. J., & Grif-
ﬁn, S. (1985). The Satisfaction with Life
Scale (SWLS). Journal of Personality As-
sessment, 49, 71-75.
Eizenman, D. R., Nesselroade, J. R., Feather-
man, D. L, & Rowe, J. W. (1997). Intra-
individual variability in perceived con-
trol in an older sample: The MacArthur
successful aging studies. Psychology and
Aging, 12, 489-502.
Festinger, L. (1957). A theory of cognitive dis-
sonance. Stanford: Stanford University
Fleeson, W., & Wilt, J. (2010). The relevance
of Big Five trait content in behavior to
subjective authenticity: Do high levels
of within-person behavioral variabil-
ity undermine or enable authenticity
achievement? Journal of Personality, 78,
Fromm, E. (1941). Escape from freedom. New
York: Holt, Rinehart and Winston.
Greenwald, A. G. (1980). The totalitarian ego:
Fabrication and revision of personal his-
tory. American Psychologist, 35, 603-618.
Hardin, C. D., & Higgins, E., T. (1996). Shared
reality: How social veriﬁcation makes
the subjective objective. In E. T. Higgins
& R. M. Sorrentino (Eds.), Handbook of
motivation and cognition: Vol. 3. The inter-
personal context (pp. 28-84). New York:
Harter, S. (2002). Authenticity. In C. R. Snyder
& S. J. Lopez (Eds.), Handbook of positive
psychology (pp. 382-394). New York: Ox-
ford University Press.
Harter, S., & Monsour, A. (1992). Development
analysis of conﬂict caused by opposing
attributes in the adolescent self-portrait.
Developmental Psychology, 28, 251-260.
Heatherton, T. F., & Nichols, P. A. (1994). Per-
sonal accounts of successful versus
failed attempts at life change. Personali-
ty and Social Psychology Bulletin, 20, 664-
Heckhausen, J., Dixon, R. A., & Baltes, P. B.
(1989). Gains and losses in development
throughout adulthood as perceived by
different adult age groups. Developmen-
tal Psychology, 25, 109-121.
Hopwwod, C. J., Donnellan, M. B., Blonigen,
D. M., Krueger, R. F., McGue, M., Iacono,
W. G., & Burt, A. S. (2011). Genetic and
environmental inﬂuences on personal-
ity trait stability and growth during the
transition to adulthood: A three-wave
longitudinal study. Journal of Personality
and Social Psychology, 100, 545-556. doi:
Human, L. J., Biesanz, J. C., Miller, G. E.,
Chen, E., Lachman, M. E., & Seeman,
T. E. (2013). Is change bad? Personality
change is associated with poorer psy-
chological health and greater metabolic
syndrome in midlife. Journal of Persona-
lity, 81, 249-260.
Johnson, J. T., Robinson, M. D., & Mitchell, E.
B. (2004). Inferences about the authen-
tic self: When do actions say more than
mental states. Journal of Personality and
Social Psychology, 87, 615-630.
Kernis, M. H., & Goldman, B. M. (2006). A
multi-component conceptualization of
184 benCh eT al.
authenticity: Theory and research. In M.
P. Zanna (Ed.), Social motivation: Cons-
cious and unconscious processes (pp. 210-
227). New York: Academic Press.
Landau, M. J., Johns, M., Greenberg, J., Pyszc-
zynski, T., Martens, A., Goldenberg, J.
L., & Solomon, S. (2004). A Function of
form: Terror management and structur-
ing of the social world. Journal of Perso-
nality and Social Psychology, 87, 190-210.
Landau, M. J., Robinson, M. D., & Meier, B. P.
(2013). The power of metaphor: Examining
its inﬂuence on social life. Washington, DC:
American Psychological Association.
Lench, H. C., Smallman, R., Darbor, K. E., &
Bench, S. W. (2015). Motivated percep-
tion of probabilistic information. Cogni-
tion, 133, 429-442.
McAdams, D. P. (1995). What do we know
when we know a person? Journal of Per-
sonality, 63, 363-396.
McAdams, D. P. (2005). What psycho-
biographers might learn from
personality psychology. In W. T. Schultz
(Ed.), Handbook of psychobiography (pp.
64-83). New York: Oxford University
McAdams, D. P., Diamond, A., de St. Aubin, E.,
& Mansﬁeld, E. D. (1997). Stories of com-
mitment: The psychosocial construction
of generative lives. Journal of Personality
and Social Psychology, 72, 678-694.
McLean, K. C., Pasupathi, M., & Pals, J. L.
(2007). Selves creating stories creating
selves: A process model of self-develop-
ment. Personality and Social Psychology
Review, 11, 262-278.
Mroczek, D. K., & Spiro, A. III (2007). Personal-
ity change inﬂuences mortality in older
men. Psychological Science, 18, 371-376.
Newman, G. E., Bloom, P., & Knobe, J. (2014).
Value judgments and the true self. Per-
sonality and Social Psychology Bulletin, 40,
Pahl, S., Eiser, J. R., & White, M. P. (2009).
Boundaries of self-positivity: The effect
of comparison focus in self-friend com-
parisons. The Journal of Social Psychology,
Pennebaker, J. W. (1997). Writing about
emotional experiences as a therapeutic
process. Psychological Science, 8, 162-166.
Quoidbach, J., Gilbert, D. T., & Wilson, T. D.
(2013). The end of history illusion. Sci-
ence, 339, 96-98.
Roberts, B. W., Walton, K. E., & Viechtbauer, W.
(2006). Patterns of mean-level change in
personality traits across the life course:
A meta-analysis of longitudinal stud-
ies. Psychological Bulletin, 132, 1-25. doi:
Robins, R. W., Noftle, E. E., Trzesniewski, K.
H., & Roberts, B. W. (2005). Do peo-
ple know how their personality has
changed? Correlates of perceived and
actual change in young adulthood.
Journal of Personality, 73, 489-522. doi:
Rosenberg, M. (1965). Society and the adolescent
self-image. Princeton, NJ: Princeton Uni-
Schlegel, R. J., & Hicks, J. A. (2011). The true
self and psychological health: Emerging
evidence and future directions. Social
and Personality Psychology Compass, 5,
Schlegel, R. J., Hicks, J. A., Arndt, J., & King,
L. A. (2009). Thine own self: True self-
concept accessibility and meaning in
life. Journal of Personality and Social Psy-
chology, 96, 473-490.
Schlegel, R. J., Hicks, J. A., Davis, W. E., Hirsch,
K. A., & Smith, C. E. (2013). The dynam-
ic interplay between perceived true self-
knowledge and decision satisfaction.
Journal of Personality and Social Psycholo-
gy, 104, 542-558.
Schlegel, R. J., Hicks, J. A., King, L. A., & Arndt
J. (2011). Feeling like you know who you
are: Perceived true self-knowledge and
meaning in life. Personality and Social
Psychology Bulletin, 37, 745-756.
Schlegel, R. J., Vess, M., & Arndt, J. (2012). To
discover or to create: Metaphors and the
true self. Journal of Personality, 80, 969-
Sedikides, C., Gaertner, L., & Toguchi, Y.
(2003). Pancultural self-enhancement.
Journal of Personality and Social Psycholo-
gy, 84, 60-79.
Seto, E., & Schlegel, R. J. (2015). Perception of au-
thenticity across the lifespan. Manuscript
Sherman, R. A., Nave, C. S., & Funder, D. C.
(2012). Properties of persons and situa-
tions related to overall and distinctive
personality-behavior congruence. Jour-
nal of Research in Personality, 46, 87-101.
Simmons, J. P., Nelson, L. D., & Simonsohn, U.
(2011). False-positive psychology: Un-
diSCovery in The Self and oTherS 185
disclosed ﬂexibility in data collection
and analysis allows presenting anything
as signiﬁcant. Psychological Science, 22,
Spencer-Rodgers, J., Boucher, H. C., Mori,
S., Wang, L., & Peng, K. (2009). The
dialectical self-concept: Contradiction,
change, and holism in East Asian cul-
tures. Personality and Social Psychology
Bulletin, 35, 29-44.
Steger, M. F., Frazier, P., Oishi, S., & Kaler, M.
(2006). The Meaning in Life Question-
naire (MLQ): Assessing the presence of
and search for meaning in life. Journal of
Counseling Psychology, 53, 80-93.
Swann, W. B. (1990). To be adored or to be
known: The interplay of self-enhance-
ment and self-veriﬁcation. In R. M. Sor-
rentino & E. T. Higgins (Eds.), Foundati-
ons of social behavior (Vol. 2, pp. 408-448).
New York: Guilford.
Vess, M., Schlegel, R. J., Hicks, J. A., & Arndt, J.
(2014). Guilty, but not ashamed: “True”
self-conceptions inﬂuence affective re-
sponses to personal shortcomings. Jour-
nal of Personality, 82, 213-224.
Waterman, A. (1984). Identity formation: Dis-
covery or creation? Journal of Early Ado-
lescence, 4, 329-341.
White, M., & Epston, D. (1990). Narrative means
to therapeutic ends. New York: Norton.