Social
Cognition,
Vol. 17, No. 2, 1999, pp. 186-208
INTERNAL REPRESENTATIONS OF OTHERS
IN SELF-REGULATION: A NEW LOOK AT A
CLASSIC ISSUE
MARLENE M. MORETTI
Simon Fraser
University
E.TORY HIGGINS
Columbia
University
impact of internal representations of others on self-regulation has received lit
empirical attention. The current study measured people's own guides for their
behavior and the guides they believed their parents held for them and distinguished
between: (a) guides perceived as shared between oneself and parents (i.e., identi
fied guides); (b) perceived parental guides not adopted as one's own (i.e.,
introjected guides); and (c) self-guides independent from one's parents. As hypoth
esized, only identified and independent guides significantly predicted emotional
and interpersonal functioning. Introjected guides, the "felt presence" of parents
within the self-system, did not predict functioning. Significant sex differences were
found: Independent self-guides predicted emotional and interpersonal problems in
men but not women; identified self-guides predicted
functioning in women but not
men. Results are discussed in relation to psychodynamic and relational theories of
self-development and research on gender differences in socialization. The benefits
and liabilities of a self-regulatory focus that emphasizes self-other relatedness are
The
tle
discussed.
representations of ourselves and
interpersonal context. Over time, experiences
Our mental
others coalesce to form
minant of
an
others
develop
within
an
of ourselves in relation to
internal structure that is
an
important
deter
of ourselves and others, and of
psychological experiences
strategies that we employ in interpersonal relationships. Thus, our
internal representations of self and of others are intimately connected,
both developmentally and structurally. In this article we examine the
our
the
This research
supported by SSHRC grant 410-92-1620. The authors thank Douglas
preparation of this article.
Address correspondence to Dr. Marlene M. Moretti,
Department of Psychology, Simon
Fraser University, 8888 University Drive,
Bumaby, B.C., Canada, V5A 1S6; e-mail:
was
Scholar for his assistance in the
[email protected].
186
INTERNAL REPRESENTATIONS OF OTHERS
187
self-regulatory implications of significant-other representations within
the self-system. The approach that we have adopted is guided by
self-discrepancy theory (Higgins, 1987; 1989), a model that describes the
structure of self-representation and the implications of
discrepancy be
tween self-guides.
SELF-DISCREPANCY THEORY
Self-discrepancy theory (Higgins, 1987; 1989) identifies three domains of
self-representation: the actual-self which is comprised of the attributes
that individuals believe they actually possess; the ideal-self which repre
sents the traits or characteristics that individuals wish or hope to pos
sess; and the ought-self which represents traits or characteristics that
individuals believe they have a duty or obligation to possess. Each of
these domains of the self can be considered from several perspectives:
one's own perspective on the self (i.e., the attributes one believes one re
ally does possess, the attributes one wishes to possess, and the attributes
one believes one should
possess) or the inferred perspectives ofsignificant
others (i.e., the attributes that
other individual believes one
or
believes
one
one assumes
one's
mother, father,
or an
actually possesses, wishes one possessed,
should possess). One's own perspective on the ac
analogous to what is commonly referred to as the self-concept.
Other self-state representations, such as one's own hopes and wishes for
the self (ideal: own self-state) or the duties and obligations that one be
lieves significant others hold for the self (ought: other self-state), offer
important standards, or self-guides for the regulation and evaluation of
the actual-self. Self-state representations related to individuals own per
spectives and the inferred perspectives of others exist together within
the self-system. This self-system guides and in turn is shaped by inter
personal experiences.
To date, the bulk of research has focused on examining the emotional
and two
consequences of discrepancy between the actual-self
self-evaluative guides the ideal-self and the ought-self. Several stud
ies have confirmed the unique relationship of actual-ideal discrepancy
with dejection related emotions and actual-ought discrepancy with agi
tation related emotions (Higgins, Klein, Strauman, 1985; Scott &
tual-self is
O'Hara, 1993; Strauman, 1989). Research also indicates that
self-discrepancies are important vulnerability factors for depression
and anxiety (Strauman, 1992; Strauman & Higgins, 1987, 1988). It has
also been shown that the emotional consequences of activating
self-discrepancies are due to the structural relationships between
self -state representations rather than the content of representations per
se (Moretti & Higgins, 1990a).
MORETTI AND HIGGINS
188
STANDPOINTS ON THE SELF
Little is known about the
of different
standpoints (e.g., own
representations. To date, most
collapsed across self-guides repre
importance
embodied within self-state
parental)
self-discrepancy studies have either
senting own and other standpoints on the self, considered only one
standpoint (e.g., ideal: own or ought: own), or have contrasted ac
tual-ideal: own discrepancy with actual-oughtother discrepancy. Thus,
the effect of standpoint per se has received little empirical attention.
Fundamental questions regarding the development and consequences
of actual-own and actual-other discrepancy have not yet been ad
dressed. For example, does it matter whether the actual-self is discrep
ant from one's own hopes and wishes versus the hopes and wishes one
vs.
believes others hold for the self?
concept of internalization may be useful
The
in
understanding the
process by
self-regulatory perspectives are inte
into
the
Even
grated
self-system.
though theorists define internalization
in slightly different ways, there is general agreement that it involves the
gradual transformation of regulatory functions provided by significant
others and society into inner regulatory mechanisms (Blatt & Behrends,
1987; Hartmann, Kris, & Lowenstein, 1946; Meissner, 1980, 1981; Ryan,
which
own
and other
Deci, & Grolnick, 1995; Schafer, 1968). Schafer (1968) argues that "inter
nalization refers to all those processes by which the subject transforms
real or imagined regulatory interactions with his environment, and real
or
imagined
characteristics of his environment, into inner
(pg. 9)." Through this process we adopt,
and characteristics
regulations
as our
own,
self-regulatory guides or standards that once existed external to the self.
A common assumption held by
psychodynamic and object relations
theorists is that the process of internalization is present
throughout the
lifespan but that it takes different forms at different levels of develop
ment (Blatt & Behrends, 1987; Meissner, 1981).
Psychological maturation
is accompanied by increasing differentiation between the self and others
and this is reflected in the internalization process. In its earliest form, in
ternalization involves little or no self-other differentiation. This form of
internalization, sometimes referred to as "incorporation" (Schafer, 1968;
Meissner, 1981), is assumed
to be common in
young children and in
process by which "object representations completely lose their
character and are merged or fused with
with
volves
object
a
self-representations
distinction" (Meissner, 1981, p. 18).
With greater psychological maturation,
out
regulatory functions are iden
being outside of the self (e.g., parental). These functions may be
taken into the self, or "introjected," in an
attempt to provide inner regu
lation. The inrrojection of external
regulatory functions leads to greater
internal self-regulation, but these functions continue to be
experienced
tified
as
INTERNAL REPRESENTATIONS OF OTHERS
as
the "felt
presence's of
1968). It is
189
others" (Sandler &
Rosenblatt, 1962; Schafer,
the process of
identification, which involves the
only through
restructuring of preexisting self-regulatory guides and structures that
some
aspects of self-regulation come to be adopted and integrated as
one's own. In contrast to introjection, identification marks the true trans
formation of external regulatory guides to shared
self-guides that are
identified as part of the "true" or core self (Deci &
Ryan, 1991). Respect
and support for autonomy, clear structure, and
positive affective in
volvement facilitate the process of internalization or
integration" (Deci,
Eghrari, Patrick, & Leone, 1994; Grolnick & Ryan, 1989; Ryan, Deci &
"
Grolnick, 1995).
With the
introjection
exception
of
incorporation,
an
ongoing dynamic process of
development. This process
and identification characterizes
is not necessarily restricted to childhood but rather
occurs across the life
span (Blatt & Behrend, 1987). As the self becomes reorganized and
self-regulatory guides are adopted as one's own, the capacity for inner
regulation increases and dependence on the environment for
self-regulation decreases (Hartmann, Kris, & Lowenstein, 1946). Thus,
the internalization of self-regulatory mechanisms is an adaptive devel
opmental process because it reduces dependency on external factors for
self-regulation and ensures effective regulation across varying environ
ments
and life crises.
Despite the importance of the internalization process in the develop
ment of the self-system, there has been no research directly examining
the structure and psychological implications of introjected versus
identified self-guides. The lack of research in this area reflects, in part,
the difficulty of developing a methodology to examine these issues. If
internalization is important to self-regulation, one would predict that
identified self-guides would have a more powerful impact on func
tioning than would introjected self-guides, because identified guides
represent the "true self" (Deci & Ryan, 1991) and the acceptance of a
shared reality about the self with significant others (Hardin & Higgins,
1996). Thus, our perception of the relationship between our behavior
or
performance and our identified self-guides is likely to have a pow
erful impact on our feelings. When we believe that our behavior or per
formance is congruent with identified self-guides, this is likely to
promote feelings of satisfaction and contentment. In contrast, when
we believe that our behavior or
performance is discrepant from our
identified self-guides, this is likely to evoke strong negative feelings
about the self. This is not to say that perceiving our behavior as incon
gruent with the introjected self-guides of significant others will not
also have negative repercussions. These discrepancies, however,
strike further from the
core
of ourselves, and to
some
extent we can
MORETTI AND HIGGINS
190
FIGURE 1. Identified, introjected parental guides and independent self-guides in
relation to the actual-self. Areas of overlap between the actual-self and self-guides
active in self-regulation (i.e., attribute matches or
non-overlap between the actual-self and self-guides represent
currently active in self-regulation (i.e., attribute nonmatches).
represent self-guides currently
mismatches). Areas of
self-guides
dismiss
not
discrepancies
accept them
with
introjected self-guides
because
we
do not
as our own.
self-guides are identified or introjected parental standards.
Self-regulatory guides may also develop in the context of other impor
tant relationships, or as a function of exposure to general social norms,
prescriptions, and standards. In addition, important historical figures or
social virtues may provide a source of information that is internalized
into self-regulatory guides, particularly if these guides have relevance in
Not all
of individual life
experiences, talents, or interests (Moretti & Hig
gins, press). Self-guides that originate from these diverse sources may
be equally important to identified parental guides in self-regulation if
they are adopted as one's own and represent an aspect of the true self. In
this sense, these independent self-guides share important characteristics
of identified guides and therefore are likely to be of equal significance in
self-regulation.
In summary, we distinguished between three types of self-guides:
identified, introjected, and independent self-guides (see Figure 1). Iden
tified self-guides are adopted as one's own and shared by ourselves and
others. These self-guides are integrated as part of the true self and repre
sent a perceived shared
perspective on the self with significant others. In
contrast, introjected self-guides are those self-guides that represent
terms
in
other's standards for the self, but
jected guides
are
not
are
fully integrated
not
adopted
into the
as
one's
own.
Intro
self, but represent the felt
INTERNAL REPRESENTATIONS OF OTHERS
191
presence of others in the self.
Finally, independent self-guides are those
guides that are uniquely one's own.
Following self-discrepancy theory, we also distinguished self-guides
in terms of their relationship with the actual-self.
Self-guides may either
be congruent (match) or discrepant (mismatch) with the actual-self. In
addition, there are self-guides not currently active in self-regulation but
that potentially could become active (i.e., nonmatches).
In the current study, we focused on internalization of parental guides.
To reiterate our predictions, we hypothesized that discrepancy between
the actual-self and identified (parental guides) and independent
self-guides would be more predictive of emotional and interpersonal
functioning than would discrepancy between the actual-self and
introjected (parental) guides. We tested this hypothesis by comparing the
unique predictive relationship between identified, introjected, and inde
pendent self-guides, and feelings of distress and interpersonal difficulty.
More specifically, we hypothesized that when the actual-self is perceived
as
discrepant from identified or independent self-guides, participants
would report significantly elevated levels of overall psychological dis
tress, interpersonal sensitivity and interpersonal problems. Conversely,
when the actual-self is perceived as congruent with identified or inde
pendent self -guides, we predicted that participants would report signifi
cantly lower levels of psychological distress, interpersonal sensitivity,
and interpersonal problems. In contrast, we predicted that discrepancy or
congruency between the actual-self and introjected self-guides would not
be predictive of participants' reports on these dependent measures.
Our selection of dependent variables was based on the basic premise
of self -discrepancy theory that increased discrepancy is related to psy
chological distress. Because the current study did not focus on the
unique relationship between ideal and ought self-discrepancy and types
of psychological distress (i.e., dejection and agitation respectively), we
selected a measure of overall psychological distress. In addition, we in
cluded measures of interpersonal functioning (interpersonal sensitivity,
interpersonal problems) because of the current focus on the interper
sonal dimension within self-discrepancy theory (i.e., own versus other
self-state representations). We reasoned that discrepancies involving
different standpoints on the self may play a particularly important role
in interpersonal problems.
GENDER DIFFERENCES IN REGULATORY SIGNIFICANCE OF
SELF-GUIDES
There is
gesting
a
substantial literature
that the central
on sex
predictions
differences in socialization sug
study should be considered
of this
MORETTI AND HIGGINS
192
separately
for males and females.
male and female socialization has
(1991) suggests that
one common
Although research on differences in
produced mixed findings, Higgins
theme
across
many studies is that par
ents tend to monitor and control the behavior of their
daughters
more
(see Block, 1983; Fagot, 1978; Huston, 1983; Parke &
RadkeYarrow, Zahn-Waxler, & Chapman, 1983; Rothbart
1983;
Slaby,
& Maccoby, 1966; Rothbart & Rothbart, 1976). Higgins (1991) argues that
one consequence of these socialization differences is that females are
than their
sons
likely to develop stronger self-other contingencies, or significant other
guides than males. Our recent research (Moretti, Rein, & Wiebe, 1998) on
the regulatory significance of own versus other standpoint guides sup
ports this view and is consistent with Cross and Madson's (1997)
review
gender differences in cognition, motivation, emo
tion, and social behavior. Cross and Madson (1997) concluded that
women in North America are socialized to construct and maintain an
of research examining
"interdependent" self-system; in contrast, men are socialized to develop
an
"independent" self-system.
Baumeister and Sommer (1997) argue that the sex differences noted by
Cross and Madson (1997) do not imply that males are any less motivated
toward ensuring relatedness than are females. Rather they suggest that
and maintain
there is fundamental difference in how males and females maintain
re
specifically, they suggest that males are socialized to es
tablish and maintain relatedness with a broader social sphere than are
latedness. More
of power and dominance by males is directed
Although the views of Cross and Madson (1997) and
females, and that the
toward this
goal.
use
Baumeister and Sommer (1997) differ in terms of the extent to which
they assume relatedness is a primary motivation for males and females,
they are not necessarily incompatible in terms of how they understand
gender differences in self-construal. Both agree that females tend to con
strue the self in terms of experiences in intimate relationships. Both also
agree that males construct the self in terms of experiences that differenti
ate or distinguish themselves from others. For females, relatedness is
maintained through the communality between self and other; for males,
it is maintained through differentiation and distinctiveness.
Cross and Madson (1997) and Baumeister and Sommer's (1997) posi
tions lead to a similar prediction regarding gender differences: The psy
chological relevance of identified and independent self-guides is likely
different for males and females. More
specifically, females may be more
strongly
by
self-guides that represent a perceived
shared or "interdependent" view of the self than by independent
self-guides. In contrast, males may be more strongly influenced by inde
pendent self-guides that represent differentiation from others than by
identified self-guides. To test these predictions, we completed separate
influenced
identified
INTERNAL REPRESENTATIONS OF OTHERS
193
for males and females. For females, we
hypothe
relationship between the actual-self and identified
(i.e., discrepancy and congruency) would be significantly
regression analyses
sized that the
self-guides
more
predictive of functioning than would the relationship between the
actual-self and independent self-guides.
Conversely, for males we hy
that
the
between
the actual-self and independ
pothesized
relationship
ent self-guides would be
more
significantly
predictive of functioning
than would the relationship between the actual-self and identified
self-guides.
METHOD
PARTICIPANTS
Participants
were
males) enrolled
slightly older
24.88 years).
110
at
than
psychology undergraduates (74 females and 36
University. Females tended to be
males (25.57 vs. 23.53 years respectively; Mean
Simon Fraser
=
MEASURES
Selves
1986).
Questionnaire (adapted from Higgins, Bond, Klein,
Participants
were
asked to
spontaneously generate
&
Strauman,
sets
of up to 10
attributes that describe their actual-self (i.e., attributes they be
lieve they actually possess), and their self-guides, including their
traits
or
ideal-self (i.e., attributes they ideally wish or hope to possess), and
ought-self (i.e., attributes they believe they should or ought to possess).
In addition, they generate separate lists of attributes that they believe
their mother and father hope they possess (ideal: other) or feel they
should possess (ought: other). For each list of attributes, participants rate
the extent to which they believe they possess each attribute (or wish to
possess; or should possess) on a 4-point scale ranging from 1 (slightly) to
4 (extremely).
The Selves was scored to distinguish identified, introjected, and inde
pendent guides. Identified or perceived shared guides were scored as
those attributes that participants listed in their own-ideal or ought
self-guides, as well as in their parental ideal or parental ought guides. In
trojected guides were scored as attributes that participants listed in their
parental guides (ideal or ought) but did not list in their own self-guides.
Maternal and paternal introjected and identified self-guides were
scored separately and then averaged. Independent self-guides were
scored as ideal or ought self-guide attributes, which were listed only in
participant's own self-guides and not in parental guides.
MORETTI AND HIGGINS
194
Self-guides
were
nonmatches, based
Self-guide
matches
then classified
on
as
matches, mismatches,
or
their relation to attributes of the actual-self.
were
defined
as
attributes
present
in both the
ac
self-guide that were synonymous in meaning, with ex
tent ratings that did not vary by more than 1 point (e.g., actual-self:
friendly 3; own-ideal: friendly 4). Self-guide mismatches were defined
tual-self and the
either synonymous attributes present in both the actual-self and
self-guide, with extent ratings that varied by 2 or more points (e.g., ac
as
friendly 1; own-ideal: friendly 4); or antonymous attributes in
self-guide (e.g., actual-self: unfriendly; own-ideal:
were defined as attributes that were
nonmatches
pres
friendly). Finally,
ent only in actual-self or the self-guide, but not in both.
Combining across these two dimensions of coding produced the fol
lowing self-guide variables: identified matches and identified mis
matches; introjected matches and introjected mismatches; and
independent matches and mismatches. Nonmatches were not included
in the present analysis.
The Symptom Checklist-90-Revised (SCL-90-R; Derogatis, 1983). This
90-item self-report measure asked participants to indicate the frequency
of a wide range of psychological distress symptoms on a 5-point scale
ranging from 0 (not at all) to 4 (extremely). This measure yields a global
severity index of psychological distress and several specific distress
subscales. For the purpose of this study, we focused only on the global
index of distress and the interpersonal sensitivity subscale. Derogatis
(1977) reported adequate test-retest and internal consistency reliability,
and concurrent and discriminant validity for the SCL-90-R.
Revised Inventory of Interpersonal Problems (IIP; Alden, Wiggins, &
Pincus, 1990). This 64-item self-report measure assesses a wide range of
interpersonal problems. Participants were asked to respond to each item
on a
5-point scale ranging from 0 (not at all) to 4 (very often). Eight inter
personal problem scales can be derived, corresponding to the
circumplex model of interpersonal functioning. In addition, an overall
score can be
computed to provide an index of overall interpersonal diffi
culties. Adequate internal consistency, structural
stability in independ
ent samples, and convergent
has
been
established (Alden,
validity
Wiggins, & Pincus, 1990). In the current study, we examined the overall
level of interpersonal problems.
tual-self:
the actual-self and
PROCEDURE
Participants were invited to complete the study to earn partial credit to
requirements for Introductory Psychology. Participants completed
consent forms that described the
study as investigating feeling, behaviors,
ward
INTERNAL REPRESENTATIONS OF OTHERS
and
self-concept.
Measures
were
195
completed
in random order in
a
small
group setting. Upon completion of the measures, participants were pro
vided with a written description of the research and the opportunity to
contact the researcher
regarding questions
about the
study and results.
RESULTS
COMPONENTS OF THE SELF-SYSTEM
Proportion of Parental
Guide that is
Identified
versus
Introjected.
We first
examined the percentage of self-guides that were classified into each of
the six categories as described above. Forty percent of parental guides
classified as identified guides (i.e., shared between own and paren
perspective) and 60% were classified as introjected guides (i.e., only
represented in parental guide). Identified parental guides were signifi
cantly more likely to be maternal than paternal (M .43 vs. M .36; t (107)
2.28, p < .05), and were significantly more likely to match than mis
.19 vs. M
match the actual-self (M
.09 for maternal guides respec
tively, f(107) 3.15, p < .01; M .18 vs. M .06 for paternal guides
respectively, t (107) 4.25, p < .001). A relatively small proportion of
identified guides were nonmatches with the actual-self (15% and 12% of
maternal and paternal identified guides respectively).
Introjected guides were significantly less likely to be maternal than pa
ternal (M
.57 vs. M
.64, f(107) 2.28, p < .05), and were significantly
.13 vs. M .04
more likely to be match than mismatch the actual-self (M
for maternal guides respectively, f(107)
4.23, p < .001; M .14 vs. M
.04 for paternal guides respectively, f(107)
3.72, p < .001). However, in
contrast to identified guides, the majority of introjected guides were
nonmatches with the actual-self (39% and 47% of maternal and paternal
introjected guides respectively).
Proportion of Own Guide Comprised ofIdentified Parental Guides. Although
less than half of all available parental guides were classified as identified,
these guides nonetheless constituted 44% of own guides. The remaining
56% of own self-guides were independent and did not overlap with pa
rental guides. Parental guides shared with the own standpoint were more
likely to be maternal than paternal (M .36 vs. M .27, f(107) 4.23, p <
.001),1 and they were more likely to be those that matched than mis
.07 for maternal guides respec.16 vs. M
matched the actual-self (M
were
tal
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
1. Not surprisingly, there is some degree of overlap between maternal and paternal
self-guides (19%). Thus, the sum of parental guides identified as within the own regulatory
the average percent
system (i.e., sum of maternal and paternal guides: 63%) is greater than
identified within the own self-regulatory system (44%).
of
guides
parental
age
MORETTI AND HIGGINS
196
tively, f(107) 3.56, p < .001; M .14 vs. M= .04 for paternal guides
respectively, f(107) 4.98, p < .001). Identified guides also included a pro
portion of guides not currently active in regulation of the actual-self (i.e.,
nonmatches, 13% and 9% for maternal and paternal identified guides re
spectively). These results indicate that there is a considerable overlap be
tween own and parental self-guides, and that this overlap represents a
substantial portion of all guides identified as one's own.
=
=
=
AFFECTIVE AND INTERPERSONAL IMPORTANCE OF
SELF-REGULATORY GUIDES
hypothesized that identified and independent self-guides would be
predictive of emotional distress and interpersonal functioning than
would introjected guides. Zero order correlations between self-guide vari
ables (independent matches; independent mismatches; identified
matches, identified mismatches; introjected matches, introjected mis
matches) and dependent measures are presented in Table 1.
Results support our prediction that identified and introjected guides
are more
strongly related to functioning than are independent guides.
Specifically, congruency (i.e., matches) between the actual-self and iden
tified guides predicted marginally lower levels of distress, interpersonal
sensitivity, and interpersonal problems. In contrast, discrepancy (mis
matches) between the actual-self and identified guides predicted signifi
cantly higher levels of distress, interpersonal sensitivity and
interpersonal problems. Similarly, congruency between the actual-self
and independent guides predicted significantly lower levels of distress,
interpersonal sensitivity and interpersonal problems while discrepancy
predicted significantly higher levels of distress, interpersonal sensitiv
ity, and interpersonal difficulty. Congruency and discrepancy between
the actual-self and introjected guides was not significantly related to the
dependent variables.
To further test the specificity of the relationships between identified,
introjected, and independent guide discrepancy with dependent mea
sures, self-guides were simultaneously regressed on to SCL-90 psycho
logical distress, interpersonal sensitivity, and IIP total interpersonal
problems scores.
The pattern of results again confirmed our hypotheses: The only signif
icant predictors to emerge in the analysis were identified and
independ
ent guides (see Table 2).
Specifically, discrepancy between the actual-self
and identified guides predicted significantly higher levels of
psychologi
cal distress and interpersonal sensitivity. Congruency between the ac
tual-self and identified guides predicted significantly lower levels of
interpersonal sensitivity. Similarly, congruency between the actual-self
We
more
en
C
to
9
a
<
On \D
u
E
^
TO
C-
o
|
3
;
o
<
1-
(X
-a
c
"to
to
c
o
33
en
JfU
"to
CLl
U-,
o
CD
o
ll,
e
to
V
1TO
!3
2
33
2
,o
(O
cu
X.
TO
Ui
to
>
T3
C
o>
ex
Cj
Q
-o
g
t
a.
3
Q
(J
to
C
TO
ai
TO
C
CO
>
CLl
TO
c
o
en
"
e
01
PL,
cu
0)
o
c
"O
c
u
o
in
a
5
>
-O
c
a.
<
5o
c
tn
o
D
TO
s
c
ai
5
c
o
U
0)
ca
en
'3
u
<T>
c
s
cu
CD
I
cu
X-
<
ii
"to
'Sb
_o
tn
01
"m
"3
01
O
TO
cu
CO
P-
"to
o
u
c
L-
o
CJ
U
-a
o
J
"to
en n
2
o
|H
0>
N
I
I
01
TJ
-J
BO
fS
0)
'3
A
-d
-C
entife atches ismatc depen atches ismatc
U 222
=
197
22
MORETTI AND HIGGINS
198
TABLE 2.
Summary
Regression Analyses for Independent Self-Guides,
Introjected Self-Guides Predicting SCL-90 and UP Scales (All
of Simultaneous
Identified Self-Guides and
Participants)
Variable
Significant
Predictor
SCL-90 Global Distress
Identified Matches
Identified Mismatches
Independent Matches
Independent Mismatches
Introjected Matches
Introjected Mismatches
SCL-90
pr_
Beta
/
.09
.09
-.96
.25
.24
2.68**
-2.56**
.24
.24
.09
.08
.95
.05
.05
-.56
.09
.08
.93
Interpersonal Sensitivity
Identified Matches
Identified Mismatches
Independent Matches
Independent Mismatches
Introjected Matches
Introjected Mismatches
-2.35*
.22
.21
.30
.29
.21
.20
-2.22*
3.36***
.14
.13
1.46
-.05
.05
-.53
.09
.08
.92
-1.33
UP Total Problems
Identified Matches
.13
.13
Identified Mismatches
Independent Matches
.15
.15
1.63
.09
.10
-1.01
Independent Mismatches
Introjected Matches
Introjected Mismatches
.20
.20
.03
.03
.33
-.01
.01
-.10
Note.p< .05,*p<
2.11*
.01.
independent guides predicted significantly lower levels of psycho
logical distress and interpersonal sensitivity, and discrepancy between
the actual-self and independent guides predicted significantly higher
levels of interpersonal problems. Introjected guides were not significant
predictors.
and
GENDER DIFFERENCES IN REGULATORY SIGNIFICANCE OF
SELF-GUIDES
predicted that females are more strongly influenced by identified
guides than by independent guides, and that males are more strongly in
fluenced by independent guides than by identified guides. Before we ex
amined sex differences in the psychological impact of self-guides, we
first examined sex differences in the proportion of identified, introjected
and independent guides. Beginning with an assessment of all available
parental guides we found that comparable proportions were classified
as identified and introjected for males and females (identified
guides:
39% vs. 43% for males and females respectively; introjected guides: 57%
vs. 61% for males and females
respectively). No significant sex differWe
INTERNAL REPRESENTATIONS OF OTHERS
ences were
found in the
(match, mismatch,
or
proportion
199
of maternal
nonmatch) classified
as
or
paternal guides
or
introjected.
identified
Identified parental guides constituted a substantial proportion of own
guides for both males and females (45% vs. 46% for males and females re
spectively). No significant sex differences were found in the proportion
of the own guides that was comprised of maternal or paternal guides
(match, mismatch, or nonmatch). Thus, females did not identify or introject a greater or lesser percentage of parental guides than did males, nor
did females' versus males' own guides contain different proportions of
parental guides. This does not preclude the possibility, however, that the
psychological relevance of identified guides is different for males and fe
males. Zero-order correlations for males and females separately (see Ta
ble 1) suggest that the relevance of identified and independent guides
was not equivalent for males and females. For females, identified
matches were marginally significantly correlated with lower levels of dis
tress, interpersonal sensitivity, and interpersonal problems, while identi
fied mismatches were significantly correlated with increased levels of
distress, interpersonal sensitivity and interpersonal problems. In con
trast, for males identified matches were not significantly correlated with
dependent variables. Significance testing indicated that the correlation
between identified mismatches and interpersonal sensitivity was signifi
cantly larger for females (r .43) as compared to males (r .02), (z 2.10, p
.01, two-tailed). Similarly, the correlation between identified mis
=
=
=
=
matches and distress, and the correlation between identified mismatches
and interpersonal problems was marginally significantly larger for fe
males (r .25 and r .40 respectively) as compared to males (r -.04 and r
=
=
=
respectively), (z 1.62, p .06; two-tailed and z 1.40, p .10;
two- tailed, respectively).
Turning to results for independent guides, a similar pattern of correla
tions between independent matches, mismatches and dependent vari
ables was observed for males and females. However, the magnitude of
correlations was generally larger for males; the correlation between in
dependent matches and interpersonal problems was marginally signifi
cantly larger for males (r -.44) than for females (r -.15) (z 1.53, p .06;
=
.09
=
=
=
=
=
=
=
=
two- tailed).
predicted gender differences in guide relevance,
separate regression analyses were completed for male and female par
ticipants. In each analysis, all guide types were entered simultaneously
to identify the unique contribution of each guide type. Results for males
versus females supported the hypothesis of gender-specific guide rele
vance. For males, congruency between the actual-self and independent
guides predicted marginally significantly lower levels of psychological
distress and interpersonal problems (see Table 3). Neither identified
nor introjected guides significantly predicted functioning. For females,
To further examine
MORETTI AND HIGGINS
200
TABLE 3.
Summary
(Male
Regression Analyses for Independent Self-Guides,
Introjected Self-Guides Predicting SCL-90 and IIP Scales
of Simultaneous
Identified Self-Guides and
Participants)
Variable
Predictor
Significant
pr
Beta
t
SCL-90 Global Distress
Identified Matches
.02
.03
.13
Identified Mismatches
.20
.21
1.09
-.34
-.44
-1.91
.06
.06
.31
-.09
-.02
-.08
-.45
-.02
-.11
-.04
-.04
-.19
.13
.13
.69
-.28
-.36
-1.54
Independent Matches
Independent Mismatches
Introjected Matches
Introjected Mismatches
SCL-90
Interpersonal Sensitivity
Identified Matches
Identified Mismatches
Independent Matches
Independent Mismatches
Introjected Matches
Introjected Mismatches
.12
.13
.64
-.11
-.11
-.59
-.02
-.02
-.10
Identified Matches
Identified Mismatches
-.01
-.01
-.07
.06
.05
.29
Independent Matches
Independent Mismatches
Introjected Matches
Introjected Mismatches
-.31
-.36
-1.67
IIP Total Problems
Note, p
<
.27
.26
1.43
-.13
-.11
-.66
.13
.12
.66
.10.
discrepancy between the actual-self and identified guides predicted
significantly higher levels of distress, interpersonal sensitivity, and
marginally higher levels of interpersonal problems (see Table 4). Con
gruency between the actual-self and identified guides predicted signif
icantly lower levels of interpersonal sensitivity. In contrast to the result
for males, independent guides were not a significant predictor of func
tioning in females. Introjected guides also did not emerge as a signifi
cant predictor for females.
Further analyses were completed for males and females, differentiat
ing between identified and introjected maternal versus paternal guides.
Again, results for males confirmed the significance of independent
guides in predicting functioning. Regression analyses entering identi
fied maternal, introjected maternal, and independent guides revealed
that independent matches predicted
significantly lower levels of dis
tress (pr
-.41, p < .05), interpersonal sensitivity (pr
.38, p < .05), and
marginally fewer interpersonal problems (pr -.34, p .08). Similarly,
analyses entering identified paternal, introjected paternal, and inde
pendent guides showed that independent matches predicted margin=
=
=
=
INTERNAL REPRESENTATIONS OF OTHERS
201
TABLE 4.
Summary of Simultaneous Regression Analyses for Independent Self-Guides,
Identified Self-Guides and Introjected Self-Guides Predicting SCL-90 and IIP Scales
(Female
Participants)
Variable
Significant
Predictor
pr
Beta
t
SCL-90 Global Distress
Identified Matches
Identified Mismatches
Independent Matches
Independent Mismatches
Introjected Matches
Introjected
SCL-90
Mismatches
.18
-.16
.38
.38
.20
-.19
-1.43
3.25*'
-1.60
.06
.06
.47
.04
-.03
-.30
.16
.15
1.30
-2.39*
Interpersonal Sensitivity
Identified Matches
Identified Mismatches
.29
-.26
.43
.42
.19
-.17
-1.52
.10
.09
.79
.07
.06
.58
.15
.12
1.19
Identified Matches
.19
-.19
-1.53
Identified Mismatches
.23
.23
1.86
Independent Matches
Independent Mismatches
Introjected Matches
Introjected Mismatches
.02
-.02
-.14
.13
.13
1.00
Independent Matches
Independent Mismatches
Introjected Matches
Introjected Mismatches
3.82*
IIP Total Problems
Note, p
<
.10,
*p
<
.05, p
<
.01,
*p
<
.08
.08
.68
.01
-.01
-.11
.001.
.10). Neither maternal nor
ally lower level of distress (pr -.32, p
paternal identified or introjected guides were significant predictors.
For females, both maternal and paternal identified guides were pre
dictive of functioning, replicating the pattern of results found when
these guides were collapsed. Contrary to the overall pattern of results,
however, introjected maternal guides were also found to be a significant
predictor. Specifically, regression analyses entering maternal identified,
maternal introjected and independent guides revealed that identified
and introjected mismatches predicted significantly higher levels of dis
tress (pr
.31, p < .01 and pr .28, p < .05 respectively) and interpersonal
sensitivity (pr .50, p < .0001 and pr .29, p < .05 respectively). Identified
guides that matched the actual-self predicted marginally lower levels of
interpersonal sensitivity (pr -.27, p < .05). Regression analyses entering
paternal identified, paternal introjected, and independent guides re
vealed that identified matches predicted significantly lower levels of
distress (pr -.27, p < .05) and interpersonal sensitivity (pr -.29, p < .05),
while paternal mismatches predicted significantly higher levels of dis
tress (pr
.38, p < .01) and interpersonal sensitivity (pr .38, p < .01). Nei=
-
=
=
=
-
=
=
=
-
=
MORETTI AND HIGGINS
202
independent nor introjected paternal guides
predicting functioning.
ther
were
significant
in
DISCUSSION
The overall
pattern of findings show that perceived parental guides play
significant role in self-regulation when these guides are shared as one's
Specifically, discrepancy between the actual-self and perceived
shared or identified guides predicted higher levels of psychological dis
tress and interpersonal sensitivity; congruency between the actual-self
and perceived shared guides predicted significantly lower levels of in
terpersonal sensitivity. In contrast, the relationship between the ac
tual-self and nonshared or introjected guides was not significant in
predicting psychological or interpersonal functioning. These results
suggest that parental guides not fully integrated into the self are not par
ticularly important in self-regulation. Independent self-guides, those
guides nominated as one's own but not perceived as shared with par
ents, were also significant in predicting emotional and interpersonal
a
own.
functioning. We found that congruency between the actual-self and in
dependent guides predicted lower psychological distress and interper
sonal sensitivity, while discrepancy predicted increased interpersonal
problems. Independent guides likely originate from diverse sources,
such as peer relationships, romantic relationships, and general social
standards. Thus, it is important to note that the guides we have classified
as independent in this study may only be "independent" relative to per
ceived parental guides. It is possible that independent guides in the cur
rent study were important predictors of functioning because they
represent a perceived shared definition of self, with significant individu
als other than one's parents. It is important that future research investi
gate the diverse interpersonal and social contexts that provide standards
which may be internalized into the self.
Some evidence of gender differences emerged in the
study. Although
proportions
parental guides,
identified guides did not hold equal relevance for self-regulation. Males
were influenced
by the relation between their actual-self and their own
rather
than identified guides. When males perceived their
independent
actual-self as congruent with their independent guides, they reported
less distress and fewer interpersonal problems. These results were con
sistent, regardless of whether our analyses focused on maternal or pater
nal guides. In contrast, females were influenced by identified rather than
independent guides. When females perceived their actual-self as dis
crepant from identified guides, they reported higher levels of distress,
males and females identified similar
of
INTERNAL REPRESENTATIONS OF OTHERS
203
interpersonal sensitivity, and more interpersonal problems. When they
perceived their actual-self as congruent with identified guides, they re
ported less interpersonal sensitivity. Identified guides were significant
in female self-regulation, regardless of whether they were maternal or
paternal in origin. Interestingly, maternal but not paternal introjected
guides that mismatched the actual-self also predicted distress and inter
personal sensitivity in females.
It is important to view our gender difference findings as preliminary,
particularly given the small sample size of males in the study. Although
we did not find a gender difference in the extent to which
independent
and identified guides were represented in the self-concept, we did find
some evidence that these
guides were differentially important for males
and females. Our results
are
consistent with theories that stress the rela
tional context in which females
develop a sense of self (Jordan et al.,
According
self-development emerges within
a context that
emphasizes communality with others, particularly moth
to this
1991).
view, female
ers, rather than distinctiveness. As a consequence, women tend to expe
rience and define themselves in terms of their relationships with others,
closely connected with their experiences
relationships. Our finding regarding the relevance of
maternal guides, regardless of whether they were identified or
introjected, is particularly intriguing in pointing to the special signifi
cance of
mother-daughter relationships in women's self-development.
We found that maternal guides influenced how women evaluated them
selves, regardless of whether they shared or accepted these guides as
their own. This was not the case for paternal guides; only paternal
guides that were perceived as shared or identified were significant pre
dictors. The findings are consistent with Surrey's (1991) view that
mother-daughter relationships involve a process of "interactive valida
tion" in which mothers and daughters move from their own to each
other's perspectives depending on relational needs. If this process in
deed occurs, it would increase the relevance of all maternal guides, re
gardless of whether or not they are adopted as one's own. This is
precisely the pattern of results we found for women in relation to mater
nal guides.
The tendency to self-regulate in terms of close relationships likely car
ries benefits and liabilities. Close relationships can be a source of sup
port during times of stress. But there are risks. If shared self-guides
become overly influential in self-regulation, self-worth may rest on per
ceived approval from others, and more generally on the quality of close
relationships. It is possible that a gender difference self-regulatory style
contributes to women's greater vulnerability to depression. The tenand their
sense
of self- worth is
within intimate
MORETTI AND HIGGINS
204
dency of women to self-regulate in terms of intimate relationships, com
coping strategies (Nolen-Hoeksema, 1993, 1996),
for
increase
risk
depression. The risk may be particularly acute if
may
women become overly focused on how to reconstruct themselves to be
congruent with their relational context at the cost of employing other ef
fective means of coping. Understanding the tendency of women to
self-regulate in relational terms has implications for developing specific
types of therapeutic interventions that may be helpful in alleviating de
pression (Moretti, Higgins, & Feldman, 1990).
Our findings are also consistent with Cross and Madson's (1997)
view that males are socialized to construe the self in terms of independ
bined with ruminative
ence
and autonomy, whereas females
are
socialized to construe the self
relationships. But, previously noted, our defini
independent guides rested on participants' perception that their
parents did not share these guides. This does not rule out the possibility
that "independent" guides were shared with a broader social sphere, as
suggested by Baumeister and Sommer (1997). This possibility is sup
ported by research investigating guide differentiation and conflict dur
ing adolescence (Higgins & Loeb, 1995; Higgins, Loeb & Moretti, 1995).
Specifically, this research showed that discrepancy between the ac
tual-self and peer self-guides decreased more for boys than girls be
tween Grades 8 and 9, while discrepancy between the actual-self and
parental guides increased more for boys than girls during this period.
In other words, boys more than girls moved away from parental guides
and toward peer guides. Thus, our finding that males are more highly
influenced than females by "independent" guides may simply be a re
flection of a gender difference in the importance of self-regulatory
guides based on a broader social sphere. It is also important to empha
size that our findings do not suggest that relationships are important
in terms of intimate
as
tion of
for females but
unimportant
for males. As Baumeister and Sommer
(1997), the self-related meaning of relationships
males and females without
implying gender
a
can
be different for
difference in need for
re
latedness.
The
methodology outlined in this article provides a new approach for
studying the idiographic significance of internal self-other representa
tions. It adds to the growing research on interpersonal aspects of
self-representation (Aron, Aron, Tudor & Nelson, 1991; Davis, Conklin,
Smith & Luce, 1996; Moretti et al., 1996) by differentiating the distinct
ways in which internal
representations of others influence
self-regulation.
integrates important concepts from
and
relational,
psychodynamic,
social-cognitive theories, and provides
an innovative
for
approach
understanding the self. There are several
This work
INTERNAL REPRESENTATIONS OF OTHERS
limitations inherent in the current
205
approach that should be considered in
future research. First, the current methodology relies on participant's
self-reports to assess the self-system, and emotional and interpersonal
It is
unlikely that participants are aware of the relations be
representations that underlie the structure of the
self-regulatory system because participants are never asked to directly
compare or contrast attributes of one self-state representation with an
other. Research on the effects of priming self-discrepancies shows the
impact of discrepancies on affect occurs in the absence of awareness of
the discrepancies (Strauman, 1989; Strauman & Higgins, 1987). None
theless, even though it is difficult to stipulate precisely how a self-report
bias could account for the findings, such a bias cannot be completely
ruled out. Second, we measured perceived "shared reality" rather than
objective shared reality. Clearly, the perceived reality that we hold of
ourselves in relation to others has a profound impact on how we feel and
behave, but the "real" aspects of interpersonal relationships also play an
important role. Thus, it would be important to measure self-guides as re
ported by individuals and their significant others in future studies.
Finally, we have focused on gender differences in the psychological rele
vance of internal
representations of others, but it is likely that these dif
ferences represent individual differences in socialization experiences.
Thus, it is important that future researchers directly examine the social
ization experiences that give rise to individual differences in
self-regulatory orientation. Longitudinal studies provide a more sensi
tive method for detecting the effects of socialization experiences on
self-regulation than do cross-sectional designs, as implemented in the
current study.
to dif
As previously noted, there are trade-offs costs and benefits
ferent regulatory styles. North American society and traditional
psychodynamic models tend to equate psychological maturity and
health with greater independence. One might argue, however, that both
a clear differentiation of self from others, as well as an appreciation of the
relatedness between self and others, characterize psychological matu
rity. This process is a dynamic interaction between the individual and
interpersonal and social ecology which involves both the integration of
existing self-regulatory possibilities, and the exploration of new
self-regulatory possibilities. Ultimately, the emergence of a
self-regulatory system that represents a fluid but integrated shared real
ity about the self with significant others, including parents, peers and
partners and aspects of society in general is most likely beneficial to
problems.
tween
self-state
psychological health.
MORETTI AND HIGGINS
206
REFERENCES
Alden, L. E., Wiggins, J. S., & Pincus, A. L. (1990). Construction of circumplex scales for the
Inventory of Interpersonal Problems. Journal of Personality Assessment, 55(3&4),
521-536.
Aron, A., Aron, E. N., Tudor, M., & Nelson, G. (1991). Close
other in the self.
relationships
Journal of Personality and Social Psychology, 60,
as
including
241-253.
Baumeister, R. F., & Sommer, K. L. (1997). What do men want? Gender differences and two
spheres of belongingness: Comment on Cross and Madson (1997). Psychological Bul
letin, 122, 38-44.
Blatt, S. J., & Behrends, R. S. (1987). Internalization,
ture
Block, J. H.
separation-individuation, and the na
therapeutic action. International Journal of Psycho- Analysis, 68, 279-297.
(1983). Differential premises arising from differential socialization of the sexes:
of
Some
conjectures.
Child
Development, 54,
1335-1354.
Cross, S. E., & Madson, L. (1997). Models of the self: Self-construals and gender.
cal Bulletin, 122, 5-37.
Davis, M. H., Conklin, L., Smith, A., & Luce, C (1996). Effect of perspective
cognitive representation of persons:
ity and Social Psychology, 70,
A
merging of self and
other.
Psychologi
taking on
the
Journal of Personal
713-726.
Deci, E. L.,
Eghrari, H., Patrick, B. C, & Leone, D. R. (1994). Facilitating internalization: The
self-determination theory perspective. Journal of Personality, 62, 119-142.
Deci, E. L., & Ryan, R. M. (1991). A motivational approach to self: Integration in personal
ity. In R. A. Dienstbier (Ed.), Nebraska Symposium on Motivation: Perspectives on moti
vation. Vol. 38 (pp. 237-288). Lincoln, NE: University of Nebraska Press.
Derogatis, L. R. (1977). SCL-90: Administration, scoring and procedure manual-I for the R (re
vised) version. Baltimore: John Hopkins University School of Medicine.
Derogatis, L. R. (1983). SCL-90-R: Administration, scoring and procedures manual-Ilfor the re
vised version. Towson, MD: Clinical Psychometric Research.
Fagot, B. I. (1978). The influence of sex of child on parental reactions to toddler children.
Child Development, 49, 459-465.
Grolnick, W. S., & Ryan, R. M. (1989). Parent styles associated with children's
self-regulation and competence in school. Journal of Educational Psychology, 81,
143-154.
Grusec, J. E., & Goodnow, J. J. (1994). Impact of parental discipline methods
internalization of values: A
reconceptualization of current points
on
the child's
of view.
Develop
Psychology, 30(1), 4-19.
Hardin, C. & Higgins, E. T. (1996). Shared reality: How social verification makes the subjec
tive objective. In R. M. Sorrentino & E. T. Higgins (Eds), Handbook
of motivation and
cognition, Vol. 3: The interpersonal context (pp. 28-84). New York: Guilford.
Hartmann, H., Kris, E., & Lowenstein, R. M. (1946). Comments on the formation of psychic
structure. In Papers on psychoanalytic
psychology: Psychological issues (1964) (pp.
27-55). New York: International University Press.
Higgins, E. T. (1987). Self-discrepancy: A theory relating self and affect. Psychological Re
mental
view, 94, 319-340.
Higgins,
E. T. (1989). Continuities and discontinuities in
self-evaluative processes: A
self-regulatory and
developmental theory relating self and affect. Journal of
Personality, 57, 407-444.
Higgins, E. T. (1991). Development of self-regulatory and self-evaluative processes: Costs,
benefits, and tradeoffs. In M. R. Gunnar & L. A. Sroufe (Eds.), Selfprocesses and devel
opment: The Minnesota Symposia on child development (Vol. 23, pp. 125-165). Hillsdale,
NJ: Erlbaum.
INTERNAL REPRESENTATIONS OF OTHERS
207
E. T., Bond, R. N., Klein, R., & Strauman, T. (1986).
Self-discrepancies and emo
vulnerability: How magnitude, accessibility, and type of discrepancy influ
ence affect. Journal
of Personality and Social Psychology, 51, 5-15.
Higgins, E. T., Klein, R., & Strauman, T. (1985). Self-concept discrepancy theory: A psycho
logical model for distinguishing among different aspects of depression and anxiety.
Social Cognition, 3, 51-76.
Higgins, E. T. & Loeb, I. (1995). The four A's of life transition effects: Attention, accessibil
ity, adaptation and adjustment. Social Cognition, 13, 215-242.
Higgins, E. T., Loeb, I., & Moretti, M. M. (1995). Self-discrepancies and developmental
shifts in vulnerability: Life transitions in the regulatory significance of others. In D.
Cicchetti (Ed.), Rochester symposium on dez>elopmental psychopathology (pp. 190-230).
Rochester, NY: University of Rochester Press.
Huston, A. (1983). Sex-typing. In P. H. Mussen (Series Ed.) & E. M. Hetherington (Vol. Ed.),
Handbook of child psychology: Vol. 4. Socialization, personality and social development
(pp. 387-468). New York: Wiley.
Jordan, J. V., Kaplan, A. G., Miller, J. B., Stiver, I. P., & Surrey, J. L. (1991). Women's growth in
connection: Writings from the Stone Center. New York: Guilford.
Meissner, W. W. (1980). The problem of internalization and structure formation. Interna
tional Journal of Psychoanalysis, 61, 237-248.
Meissner, W. W. (1981). Internalization in psychoanalysis. New York: International Univer
sity Press.
Moretti, M. M., & Higgins, E. T. (1990a). Relating self-discrepancy to self-esteem: The con
tribution of discrepancy beyond actual-self ratings. Journal of Experimental Social
Psychology, 26, 108-123.
Moretti, M. M., & Higgins, E. T. (1990b). The development of self-system vulnerabilities:
Social and cognitive factors in developmental psychopathology. In R. J. Sternberg &
J. Kolligian, Jr. (Eds.), Competence considered (pp. 286-314). New Haven, CT: Yale
University Press.
Moretti, M. M., & Higgins, E. T. (in press). Own versus other standpoints in self-regulation:
Developmental antecedents and functional consequences. Review of General Psy
chology.
Moretti, M. M., Higgins, E. T., & Feldman, L. A. (1990). The self-system in depression: Con
ceptualization and treatment. In C. D. McCann & N. Endler (Eds.), New directions in
depression research, theory and practice (pp. 127-156). Toronto: Wall & Thompson.
Moretti, M. M., Rein, A. S., & Wiebe, V. J. (1998). Relational self-regulation: Gender differ
ences in risk for dysphoria. Canadian Journal of Behavioural Science, 30, 243-252.
Moretti, M. M., Segal, Z., McCann, C. D., Shaw, B. F., Miller, D. T., & Vella, D. (1996).
Self-referent versus other-referent information processing in dysphoric, clinically
and Social Psychology
depressed and remitted depressed individuals. Personality
Higgins,
tional
Bulletin, 22, 68-80.
Nolen-Hoeksema, S. (1993). Sex differences in control of depression. In D. M. Wegner & J.
W. Pennebaker (Eds.), Handbook of mental control (pp. 306-324). Englewood Cliffs,
NJ: Prentice Hall.
Nolen-Hoeksema, S. (1996). Chewing the cud and other ruminations. Ruminative
135-144).
thoughts. In R. S. Wyer Jr. (Ed)., Advances in social cognition, Vol. 9 (pp.
Mahwah, NJ: Erlbaum.
Parke, R. D., & Slaby, R. G. (1983). The development of aggression. In P. H. Mussen (Ed.),
Handbook
of child psychology,
Vol. 4: Socialization,
personality,
and social
(pp. 547-641). New York: Wiley.
Radke-Yarrow, M., Zahn- Waxier, C, & Chapman, M. (1983). Children's
development
prosocial disposi-
MORETTI AND HIGGINS
208
tions and behavior. In P. H. Mussen
(Ed.), Handbook of child psychology, Vol. 4: Social
personality, and social development (pp. 643-691). New York: Wiley.
Rothbart, M. K., & Maccoby, E. E. (1966). Parents' differential reactions to sons and daugh
ters. Journal of Personality and Social Psychology, 4, 237-243.
Rothbart, M. K., & Rothbart, M. (1976). Birth order, sex of child, and maternal help giving.
ization,
Sex Roles, 2, 39-46.
Ryan, R. M., Deci, E. L., & Grolnick, W. S. (1995). Autonomy, relatedness, and the self: Their
relation to development and psychopathology. In D. Cicchetti & D. J. Cohen (Eds),
Dei'elopmental Psychopathology, Volume 1: Theory & Methods (pp. 618-655). New
York: Wiley.
Sandler, J., & Rosenblatt, B. (1962). The concept of the representational world. The Psycho
analytic Study of the Child, 27, 128-145.
Schafer, R. (1968). Aspects of internalization. New York: International University Press.
Scott, L., & O'Hara, M. W. (1993). Self-discrepancies in clinically anxious and depressed
university students. Journal of Abnormal Psychology, 102, 282-287.
Strauman, T. J. (1989). Self-discrepancies in clinical depression and social phobia: Cogni
tive structures that underlie emotional disorders? Journal of Abnormal Psychology,
98, 14-22.
Strauman, T. J. (1992). Self-guides, autobiographical memory, and anxiety and dysphoria:
Toward
a cognitive model of
vulnerability to emotional distress. Journal of Abnormal
Psychology, 101, 87-95.
Strauman, T. J., & Higgins, E. T. (1987). Automatic activation of self-discrepancies and
emotional syndromes: Cognitive structures influence affect. Journal of Personality
and Social Psychology, 53, 1004-1014.
Strauman, T. J., & Higgins, E. T. (1988). Self-discrepancies as predictors of vulnerability to
distinct syndromes of chronic emotional distress. Journal ofPersonality, 56, 685-707.
Surrey, J. L. (1991). Theself-in-relation: A theory of women's development. In Jordan, J. V.,
Kaplan, AG., Miller, J. B., Stiver, I. P., & Surrey, J. L. (1991). Women's growth in con
nection: Writings from the Stone Center (pp. 51-66). New York: Guilford.