Trapnell 1999

Journal of Personality and Social Psychology Copyright 1999 by the American Psychological Association, Inc.
1999, Vol. 76, No. 2, 284-304 0022-3514/99/S3.00
Private Self-Consciousness and the Five-Factor Model of Personality:

Distinguishing Rumination From Reflection
Paul D. Trapnell Jennifer D. Campbell

Ohio State University at Mansfield University of British Columbia
A distinction between ruminative and reflective types of private self-attentiveness is introduced and
evaluated with respect to L. R. Goldberg's (1982) list of 1,710 English trait adjectives (Study 1), the
five-factor model of personality (FFM) and A. Fenigstein, M. F. Scheier, and A. Buss's (1975)
Self-Consciousness Scales (Study 2), and previously reported correlates and effects of private self-
consciousness (PrSC; Studies 3 and 4). Results suggest that the PrSC scale confounds two unrelated,
motivationally distinct dispositions—rumination and reflection—and that this confounding may account
for the "self-absorption paradox" implicit in PrSC research findings: Higher PrSC scores are associated
with more accurate and extensive self-knowledge yet higher levels of psychological distress. The
potential of the FFM to provide a comprehensive framework for conceptualizing self-attentive disposi-
tions, and to order and integrate research findings within this domain, is discussed.
In this article we examine the private half of the popular public- The Self-Consciousness Scales
private taxonomy of trait self-consciousness (Fenigstein, Scheier,
Fenigstein et al. (1975) proposed a dispositional analogue of
& Buss, 1975) and propose a reconceptualization of private self-
Duval and Wicklund's (1972) concept of self-awareness, opera-
consciousness (PrSC) based on motivational distinctions implicit
tionalized by means of a self-report questionnaire. Factor analyses
in its relation to the five-factor model of personality (FFM). of their initial item pool did not, however, confirm their prelimi-
Although enthusiasm for the FFM is not universal (e.g., Block, nary hypothesis of a general disposition to be self-focused. On the
1995; Me Adams, 1992), for the present purposes we adopted the basis of those analyses, they concluded there were two psycholog-
FFM as an initial best approximation of a comprehensive taxon- ically distinct self-focusing tendencies: private self-consciousness
omy of personality trait descriptors and their superordinate pattern (PrSC—consciousness of one's inner feelings, thoughts, and phys-
of covariation.1 The potential of the FFM for organizing and ical sensations) and public self-consciousness (PbSC—conscious-
integrating dispositional research findings has been demonstrated ness of one's appearance to others). The public-private distinction
in several research domains (e.g., Marshall, Wortman, Vickers; was eventually extended to self-focus states (e.g., Buss, 1980) and
Kusulas, & Hervig, 1994; T. W. Smith & Williams, 1992; Tokar, was a key component of Carver and Scheier's influential research
Fischer, & Subich, 1998; Trull & McCrae, 1994). The PrSC program, providing "interlocking conceptual replications" (Carver
research literature is vast but has yet to be systematically related to & Scheier, 1981a, p. 40) of studies involving manipulated self-
the FFM. We propose in the present article a distinction between focus. Fenigstein et al.'s Self-Consciousness Scales (SCS) quickly
ruminative and reflective self-focus, derived from the FFM dis- became among the most influential trait measures in social psy-
chological research (Fenigstein, 1987).
tinction between Neuroticism and Openness to Experience, and
demonstrate how it may explain a paradox in the current research
literature on PrSC. The Problem of Motives in the SCS
Despite this popularity, the construct validity of the SCS has
been the object of considerable controversy (see Wicklund &
Gollwitzer, 1987, and the commentaries and rejoinders that follow
Paul D. Trapnell, Department of Psychology, Ohio State University at it). Researchers reasonably assumed that the PbSC and PrSC scales
Mansfield; Jennifer D. Campbell, Department of Psychology, University of
British Columbia, Vancouver, British Columbia, Canada.
We gratefully acknowledge a Social Sciences and Humanities Research
1
Council (SSHRC) Doctoral Fellowship awarded to Paul D. Trapnell and an Agreement on the optimal partitioning and rotation of the factor axes
SSHRC research grant, which supported this research. Thanks are also due within "a" five-factor model is also less than universal (John, 1990;
to Jerry Wiggins and Del Paulhus for their helpful advice throughout the McCrae, 1996a). The present studies derive from the five-factor concep-
project and to Ruth Hahn for innumerable contributions in the conduct of tions of Goldberg (1990) and Costa and McCrae (1992). For convenience,
the studies, data analysis, and preparation of this article. we use the terms Big Five and FFM interchangeably in the present article
Correspondence concerning this article should be addressed to Paul D. and refer to the five factors described by those models with the labels
Trapnell, Department of Psychology, Ohio State University at Mansfield, recommended by McCrae and John (1992): Extroversion, Agreeableness,
1680 University Drive, Mansfield, Ohio 44906. Electronic mail may be Conscientiousness, Neuroticism, and Openness to Experience, for Big Five
sent to trapnell.2@osu.edu. Factors I-V, respectively.
284
REFLECTION AND RUMINATION 285
primarily tap attentional differences between people and conse- attention, per se, not self-focus in the service of a specific psycho-
quently interpreted their findings in the context of attentional logical motive.
processes specified in self-awareness or self-regulation theories If the PrSC items do, however, measure particular motives for
(e.g., Nasby, 1985, 1989). Critics have noted, however, that the self-attending, the motivationally ambiguous wording of PrSC
scales do not simply index frequency of attending to the public or items that lends them their high face validity may importantly
the private selves; they index specific motives for doing so. The confound interpretation of their correlates and effects. Consider
confounding of motive with direction of attention in the content, two of the personality dimensions described by the FFM: Neurot-
correlates, and social effects of the PbSC scale, for example, led icism and Openness to Experience. People who score high in
Wicklund and Gollwitzer (1987) to suggest that it is better inter- Neuroticism describe themselves as prone to negative emotional
preted as a measure of "social dependency" than self-attention. states such as anxiety, depression, low self-esteem, pessimism,
Motivational confounds have also been cited with respect to the moodiness, irritability, and jealousy (e.g., Costa & McCrae, 1980;
PrSC scale, including negative affect (Ingram, 1989), need for Eysenck & Eysenck, 1985; Goldberg, 1990). People who score
self-knowledge (Franzoi, Davis, & Markweisse, 1990), and need high in Openness to Experience tend to seek out novel aesthetic,
for autonomy and uniqueness (Schlenker & Weigold, 1990). To emotional, and intellectual experiences and report a high frequency
the extent that the PbSC and PrSC scales measure differences in of, and particular interest in, imaginative and reflective thought
motives, needs, or values, as well as self-awareness, the possibility (e.g., McCrae & Costa, 1985, 1997a). Neuroticism and Openness
exists that effects of the scales thought to be due to self-attentional to Experience are essentially independent domains of individual
mechanisms (e.g., activation of a "comparitor" function in self- differences, a fact instantiated by the FFM. Both dimensions,
regulation) may instead be due to motivational ones (e.g., approach however, clearly imply chronically higher levels of PrSC—for
and avoidance motives related to interests, values, fears, expect- notably different reasons. Motivationally ambiguous items such as
ancies, etc.; cf. Trapnell, Meston, & Gorzalka, 1997). "I am always trying to figure myself out" can be readily interpreted
to mean either reason: psychological distress ("I am always rumi-
nating over or second guessing myself) or epistemic curiosity
Dispositional Self-Consciousness From the (e.g., "I love trying to figure myself out").
Perspective of the FFM A conventional assumption in the interpretation of PrSC scores
is that the motive for scoring high on the scale is not critical to the
We propose a reconceptualization of dispositional self- interpretation of PrSC effects. The philosopher's and the neurotic's
consciousness that acknowledges an inherent confounding of mo- equally high scores would imply the same cognitive state, private
tive and direction of attention in the dispositional domain. In self-awareness, which invokes the same processes (e.g., activation
contrast to the purely cognitive approach to trait self-focus pro- of a self-regulatory comparitor function), which precipitate the
posed by Fenigstein et al. (1975), we propose that self-attentive same basic psychological sequelae—in particular, heightened per-
dispositions likely require specification in terms of a particular ception (amplification) of affect and self-concept articulation. Al-
motive or value directing attention to the self. From this perspec- ternatively, we suggest that the motivational ambiguity of the PrSC
tive, a public-private distinction among self-attentive dispositions items causes them to capture both negative affective (neurotic) and
may not merit special status in self-awareness or self-regulation positive epistemic (intellective) sources of variance and that these
theories (cf. Carver & Scheier, 1985) because it represents merely two different determinants of PrSC variance independently deter-
one among many potentially useful superordinate classifications of mine neurotic and epistemic correlates and effects of the PrSC
motives and values relevant to dispositional self-awareness. In our scale. To explicate the PrSC's divergent motivational underpin-
view, the PrSC and PbSC scales do not operationalize the two nings, we describe below the factorial complexity of the scale and
superordinate categories of private and public self-attention; they an apparent contradiction in its psychological effects.
operationalize a specific motivational distinction within each of
these categories (e.g., epistemic curiosity and "appearance" con-
PrSC Factorial Complexity
cern associated with social conformity motives, respectively). We
assume other important motivational distinctions are possible, both PrSC item intercorrelations reliably reveal two factors (e.g.,
within and beyond those categories. Anderson & Bohon, 1996; Mittal & Balasubramanian, 1987;
We restrict our focus here to PrSC because detailed critiques of Piliavin & Charng, 1988) labeled Self-Reflectiveness (SR; six
PbSC have been published by others (Wicklund & Gollwitzer, items, e.g., "I'm always trying to figure myself out") and Internal
1987) and because the high face validity of the PrSC scale presents State Awareness (ISA; four items, e.g., "I'm alert to changes in my
a tougher challenge for our motivational confounding hypothesis mood") by Burnkrant and Page (1984). Although Bernstein, Teng,
than does the PbSC scale. The item content of the PrSC scale and Garbin (1986) argued that support for two factors is con-
seems remarkably free of motivational or affective connotations. founded by differences in the items' distributional properties,
Consider the wording of the items typically loading most highly on mounting evidence suggests that they are substantively, as well as
the PrSC factor in factor analytic studies of Fenigstein et al.'s statistically, different.
(1975) items: "I reflect about myself a lot," "I'm constantly ex- Foremost among this evidence is the fact that the SR factor
amining my motives," and "I am always trying to figure myself correlates positively with measures of negative affectivity or Neu-
out." It is not clear what motive, need, or value these items roticism, whereas the ISA factor correlates negatively or not at all
reference, and the definition of the construct suggests that the with such measures. This divergent pattern of findings has been
motivational ambiguity in the wording of PrSC items was inten- found for Neuroticism-related constructs such as anxiety and de-.
tional, as they were designed to index frequency of self-focused pression (P. J. Watson & Biderman, 1993; P. J. Watson, Morris, &
286 TRAPNELL AND CAMPBELL
Hood, 1988), self-esteem and identity seeking (Piliavin & Charng, These and other findings suggest that chronic attending to private
1988), and self-concept clarity (Campbell et al., 1996). The con- thoughts and feelings, per se, stimulates greater differentiation,
tradictory relation of ISA and SR with measures of identity seek- integration, accuracy, and cognitive accessibility of self-informa-
ing and self-concept clarity is particularly noteworthy because of a tion—that is, that it enhances self-knowledge.
postulated clarification effect of private self-attentiveness. PrSC If private self-focus increases self-knowledge, and if self-
"is assumed to make all private events, both affectively charged knowledge facilitates psychological adjustment, one would expect
and neutral, clearer and more distinct" (Buss, 1980, p. 14). The positive associations between PrSC and psychological health and
notion of self-concept articulation (i.e., self-clarification) is rou- adjustment. Instead, higher PrSC scores are reliably associated
tinely invoked to explain numerous PrSC information-processing with higher levels of psychological distress, a finding that may
effects (see below). If the ISA and SR factors both measure the reflect a broader phenomenon: Heightened self-focus appears to be
PrSC construct, it is puzzling why they should show opposite implicated in a remarkably broad range of psychopathology (In-
associations with self-reported clarity and certainty of self- gram, 1990).
perceptions (Campbell et al., 1996; Franzoi, 1983). Their divergent PrSC associations with psychological distress and pathology
relations with self-esteem, depression, and anxiety are important constitute a paradox for many traditional theories of adjustment:
for a similar reason. Higher self-esteem has been shown to predict Chronic, private self-attention appears to enhance self-knowledge
more certain self-knowledge (Campbell, 1990), and more resis- at the expense of psychological adjustment. Although one might be
tance to social influence attempts (Brockner, 1979, 1984), two tempted to interpret this relation in terms of a "sadder but wiser"
types of phenomena that figure prominently in efforts to test the phenomenon (e.g., Alloy & Abramson, 1979), note that self-
clarification hypothesis of PrSC. concept clarity typically shows a strong negative association with
neuroticism, depression, and poor self-esteem (Campbell et al.,
1996). This fact plainly contradicts any simple "sadder but wiser"
The "Self-Absorption Paradox"
assumption that PrSC-related improvements in self-clarification
In the article introducing the PbSC and PrSC scales, Fenigstein should have the effect of increasing psychological distress.
et al. (1975) noted the importance of self-awareness to many In addition, some of the traits associated with neuroticism pre-
schools of psychotherapy. Similarly, psychological-mindedness— dict the same self-knowledge-related behavioral phenomena as the
the disposition and ability to reflect on the meaning and motivation PrSC scale, but in an opposite direction. For example, PrSC has
of behavior, thoughts, and feelings in oneself and others (Appel- been associated with resistance to suggestibility and social con-
baum, 1973; Hall, 1992)—is highly prized by mental health pro- formity pressures (Froming & Carver, 1981; Scheier, Carver, &
fessionals (Farber, 1989), who often seek to promote it in patients Gibbons, 1979), whereas low self-esteem is generally implicated
as well as themselves and consider it important to therapeutic in behavioral plasticity effects such as suggestibility and compli-
progress, a viewpoint not without empirical foundation (Appel- ance (Brockner, 1984). More important is that both the PrSC and
baum, 1973; Conte et al., 1990; cf. Prochasta, DiClemente, & self-esteem literatures postulate the same underlying mediator of
Norcross, 1992). Farber (1989) considered PrSC "the construct these phenomena: self-knowledge. PrSC effects are typically at-
most closely related to psychological-mindedness" (p. 212) and tributed to more accurate knowledge of internal states and personal
consequently operationalized the latter with the PrSC scale. beliefs (e.g., Froming & Carver, 1981; Scheier et al., 1979); the
Psychotherapeutic advocacy of self-awareness assumes that behavioral plasticity effects of low self-esteem have been attrib-
veridical or authentic self-knowledge is critical for psychological uted to lack of certainty or confusion in the self-concept (Campbell
growth and maturity and that pondering the self improves the & Lavallee, 1993).
extent and accuracy of self-knowledge. The first assumption is
both ancient (e.g., the inscription "Know Thyself on Apollo's
Resolving the Paradox
shrine at Delphi) and central to many clinically derived theories of
adjustment (psychoanalytic, humanistic, existential), although its Why, then, does PrSC correlate positively with psychological
empirical status has been challenged in recent years (Taylor & distress? There are at least three different potential explanations
Brown, 1988). for this paradox. Two are cognitive explanations and are basic
The second assumption is a fundamental corollary of most predictions from self-attention and self-regulation theories. If self-
self-attention theories (e.g., Carver & Scheier, 1981a; Gibbons, attention increases awareness of shortcomings (Duval & Wick-
1990). Buss (1980) presented a particularly clear statement of this lund, 1972), habitual self-attending (PrSC) should precipitate
hypothesis: "Private self-conscious people regularly inspect their chronic negative affect. Constructs such as narcissism, however,
bodily processes and moods, reflect about their motives and goals, suggest that this is not necessarily true with respect to chronic or
and fantasize a lot about themselves. As a result of repeated dispositional self-attention. It is possible to be chronically self-
self-reflection [italics added], they know themselves very well" (p. attentive yet positively brimming with self-regard. A second in-
20). Self-knowledge, or self "clarification," is postulated to explain terpretation is that attention to inner states heightens awareness of
all three of the most widely researched psychological effects of those states (Scheier & Carver, 1977). PrSC should therefore
PrSC on the self: accuracy, for example, enhanced consistency and heighten awareness of one's mood in addition to other aspects of
validity of self-reports; articulation, for example, more extensive the self. If so, one would anticipate a positive relation between
and efficient processing of self-relevant information; and auton- PrSC and not only negative mood tendencies but also positive
omy, for example, enhanced reactance and reduced compliance ones. PrSC typically shows no between-subjects association, how-
and suggestibility (for reviews see Carver & Scheier, 1981b, 1985; ever, with measures of positive affectivity (Campbell, 1993) or
Gibbons, 1983, 1990; Nasby, 1985, 1989; Porterfield et al., 1988). traits related to Extraversion (e.g., Carver & Glass, 1976; McCrae,
1993), the personality dimension most closely associated with representative compilations of English trait adjectives currently available
positive affectivity (D. Watson & Clark, 1997). A third interpre- (John, 1990).
tation is implied by the fact that PrSC correlates about equally with
two unrelated personality dimensions: Neuroticism and Openness Procedure
to Experience (McCrae, 1993). The motivationally ambiguous
PrSC items may fail to differentiate neurotic motives for self- A convenient but reasonable assumption is that the 40 adjectives in
attending (e.g., anxiety) from epistemic ones (e.g., curiosity), Goldberg's 1,710 taxonomy containing the prefix self (e.g., self-assertive)
thereby confounding individual differences related to Openness to define the subset of terms most likely to include the exemplars of dispo-
sitional self-focus. Two of these 40 adjectives—self-conscious and self-
Experience with those related to Neuroticism.
examining—clearly stand apart in denoting self-attention without accom-
panying denotations of self-evaluation or interpersonal style. Self-
Overview conscious and self-examining clearly denote a type of self-attentive state
(consciousness of the self) or a type of self-attentive process (examining or
In the present studies we introduce and articulate a distinction inspecting the self) that directly implies a self-attentive state. The remain-
implicit in PrSC research findings: neurotic self-attentiveness, or ing adjectives refer to self-directed attitudes and evaluations (e.g., self-
rumination, and intellectual self-attentiveness, or reflection. We satisfied, self-pitying) and interpersonal tendencies (e.g., self-confident,
self-expressive, selfish).2 We therefore selected self-conscious and self-
first attempt to establish some generality of the proposed distinc-
examining as the most prototypic exemplars of the dispositional self-focus
tion across natural language descriptors of self-attention (Study 1)
construct (e.g., Fenigstein et al., 1975) within Goldberg's 1,710 taxonomy.
and the research scales proposed by Fenigstein et al. (1975; Study In Study 1, self-ratings on self-conscious and self-examining were corre-
2). We next introduce brief measures intended to maximally dif- lated with self-ratings on the remaining 1,708 trait adjectives. From the
ferentiate ruminative from reflective self-focus and evaluate their resulting pattern of correlations, we sought to explicate the differences in
relations with the original Fenigstein et al. (1975) scales, the meaning between these two natural language exemplars of trait self-focus.
proposed factors within the PrSC scale, and the FFM. Finally, we
test our principal hypothesis that the two motivationally distinct
traits of rumination and reflection independently explain PrSC Results and Discussion
correlates and effects. The 20 trait terms most highly correlated with self-conscious
and self-examining are presented in Table 1. With the exception of
Study 1 self-critical (shown in the last row of the table), traits strongly
correlated with self-conscious were not strongly correlated with
An underappreciated application of lexical trait taxonomies is self-examining, and vice-versa. Furthermore, the two adjectives
their use as empirical tools for mapping the potential nomological diverge in meaning in a way that corresponds to the motivational
terrain of a trait. The "lexical hypothesis" (Goldberg, 1993) holds distinction proposed here: Self-perceptions of being self-conscious
that all individual differences that are socially important enough are mainly related to negative self-perceptions and negative emo-
for people to notice and need to communicate about will tend to be tions, whereas self-perceptions of being self-examining are mainly
registered in their lexicon, typically in the form of a trait adjective related to epistemic interests and intellectual traits. These findings
(e.g., stupid) or noun (e.g., blockhead). To the extent that this is suggest that the English trait lexicon rather clearly distinguishes
true (cf. McCrae, 1990), archival lexical data provide an efficient neurotic from epistemically motivated self-attention. According to
means for surveying the psychological semantics of phenotypic the lexical hypothesis, this fact implies that such a distinction was
trait constructs (e.g., social evaluative anxiety) having natural a socially significant one to our English-speaking ancestors. The
language exemplars (e.g., shy, bashful). In Study 1 we exploited behavioral consequences of anxious versus epistemic self-focus
lexical data to comprehensively map the correlates of two natural are likely to vary in socially important ways.3
language exemplars of Fenigstein et al.'s (1975) trait conception of
self-awareness: self-conscious and self-examining.
2
The remaining 38 "self" hyphenated words contained in the Goldberg
Method list are: -assertive, -assured, -centered, -confident, -consistent, -controlled,
-critical, -deceiving, -defeating, -defensive, -deluding, -denying,
Participants -deprecating, -destructive, -disciplined, -disparaging, -doubting, -effacing,
-excusing, -expressive, -important, -indulgent, selfish, selfless, -pitying,
Participants were 187 university undergraduate volunteers (84 men, 103 -possessed,, -punishing, -reliant, -reproachful, -respecting, -restrained,
women) at the University of Oregon who completed self-reports on Gold- -revealing, -righteous, -sacrificing, -satisfied, -seeking, -sufficient, and
berg's (1982) "1,710" trait taxonomy. The data were generously made -willed. We would consider several additional terms frequently referred to
available to us by Lewis Goldberg (see Goldberg, 1982, for a more detailed in the PrSC literature as exemplars of dispositional self-focus (e.g., self-
description). aware, self-attentive, self-reflective). Their absence from Goldberg's
(1982) empirically derived list offered us no a priori rationale for consid-
eration in Study 1 and precluded their evaluation in these archival data.
Measures 3
Many of the correlates of self-examining reported in Table 1 bear more
Participants rated themselves on the list of 1,710 English trait adjectives than a passing resemblance to important PrSC research findings. For
assembled by Goldberg (1982) from a lengthier list of 2,800 trait terms example, the negative association with unamusable recalls associations
assembled by Norman (1967). The procedures used to construct these two reported between PrSC scores and intensity of laughter in response to
taxonomies of trait descriptors are detailed in Goldberg (1982) and John videotape presentations of broadcast media bloopers (Porterfield et al.,
(1990). Goldberg's 1,710 taxonomy is among the most comprehensive and 1988). Other correlates reported in Table 1 that resemble PrSC effects are
Study 2 Table 1
Top 20 Correlates of Self-Conscious and Self-Examining
Proponents of the FFM argue that the Big Five dimensions Among 1,708 English Trait Adjectives11
represent fundamental semantic axes of trait description (Gold-
berg, 1993; McCrae & Costa, 1997b). One implication of this view Adjective Self-conscious Self-examining
is that the Big Five will typically exert some pressure on the factor
structure of virtually any broadly defined pool of trait descriptors, Self-doubting .44 .19
Self-punishing .39 .16
regardless of its theoretical origins. The empirical evidence for
Self-disparaging .36 .12
these claims (for reviews see Goldberg, 1993; Wiggins & Trapnell, Pouty .35 -.04
1997) led us to consider whether the Big Five played an unrecog- Self-excusing .34 .13
nized role in Fenigstein et al.'s (1975) decision to abandon a Reclusive .32 .07
unifactor model of dispositional self-attention in favor a multifac- Self-deluding .31 .04
Seclusive .31 .17
tor one. Fenigstein et al.'s (1975) three scales were not constructed
Self-indulgent .31 .15
on the basis of an a priori, three-factor conception of self- Fretful .30 .14
attentiveness but unexpectedly emerged from a pool of 38 items Overnervous .30 .06
that had been written to measure "a unitary, homogeneous dispo- Self-deprecating .30 .01
sition to be self-attentive" (Fenigstein, 1987, p. 548). Is it possible Self-effacing .30 -.04
Oversensitive .29 .18
that this multifactor conception was, in part, produced by a Big
Bashful .29 -.08
Five "undertow" unknowingly unleashed beneath the surface of an Inward .29 .14
overly broad item pool? Self-reproachful .29 .17
Despondent .28 -.08
Consider, for example, the seven a priori categories Fenigstein Withdrawn .27 .06
et al. (1975) adopted to generate their item pool. Three catego- Anxious .27 .08
ries—concern over the appraisal of others; preoccupation with Reflective .04 .47
past, present, and future behavior; and recognition of one's posi- Philosophical -.02 .41
tive and negative attributes—clearly implicate general negative Introspective .13 .39
Sensitive .12 .39
affectivity or Neuroticism (e.g., anxiety, rumination, and [low]
Meditative .06 .38
self-esteem, respectively), whereas three others—sensitivity to in- High-principled .12 .35
ner feelings, introspective behavior, and a tendency to picture or Philosophizing -.01 .33
imagine oneself—clearly implicate Openness to Experience (e.g., Deep .02 .33
openness to feelings, ideas, and fantasy, respectively). Given these Unamusable -.02 -.32
Curious .01 .31
categories, one might anticipate at least two factors: a
Feelingful .02 .31
Neuroticism-related factor and an intellect/Openness-related fac- Progressive -.02 .31
tor. As it happened, three factors emerged (PbSC, PrSC, and Social Analytical .06 .30
Anxiety). The third factor label, Social Anxiety, implies that some Perceptive .01 .30
items were included that overlapped with a third Big Five dimen- Self-reliant -.02 .30
Free-minded -.02 .30
sion, Extraversion (or, more precisely, its negative pole, Introver-
Overthoughtful .16 .29
sion), a possibility that could readily account for their relative Self-willed .06 .29
dissociation from the other items (see Briggs, 1988). Contemplative .11 .28
Intellectual .08 .28
Although our foremost concern here is with PrSC, it is instruc-
tive to initially locate all three SCS within the FFM. Doing so Note. N = 187. Data are adapted from Goldberg (1982). Correlations
permits an evaluation of a Big Five explanation of Fenigstein et with an absolute value > .20 are presented in boldface type.
a
al.'s (1975) unexpected factor results and provides a common Omitted from the table is the only notable exception to the discriminant
pattern above we observed, the adjective self-critical, which correlated
moderately and equally with self-conscious (.36) and self-examining (.45).
feelingful and sensitive, which dispositionally correspond to the "affect
amplification" effects of PrSC (Scheier & Carver, 1977), and self-reliant
and self-willed, which appear to correspond to the autonomous identity perspective on the original SCS factors (PbSC, PrSC, and Social
orientation (Schlenker & Weigold, 1990) and greater resistance to com- Anxiety) and the PrSC subfactors (ISA and SR). In Study 2 we
pliance (Froming & Carver, 1981) and suggestibility (Scheier et al., 1979) therefore evaluated the Big Five profiles of the SCS and the PrSC
found to be associated with higher PrSC scores. The construct of Openness factors.
to Experience may offer useful insights into why these various phenomena
tend to be linked to epistemic, but not neurotic, self-attentiveness. See, for
example, Pinker's (1997, pp. 545-554) fascinating evolutionary analysis of Method
humor, which offers an explanation of why the antiauthoritarianism typical
of people who are dispositionally open to experience (Trapnell, 1994)
Participants
would tend to make them prone to bemusement. Perhaps it is more than a Five independent samples of university undergraduates (referred to as
historical accident that the notorious unamusability of prudes (e.g., Queen Samples A through E) were recruited from introductory psychology classes
Victoria of Britain) is a Western cultural cliche (e.g., comedian Dana during the early part of the fall semester, in four successive years. All of the
Carvey's character "The Church Lady"): Like authoritarianism, sexual participants completed questionnaires on a voluntary basis in exchange for
conservatism is associated with dispositionally low Openness to Experi- partial course credit. Between 5%-8% of participants from each sample
ence (Trapnell & Meston, 1996). were excluded because of missing data, leaving 555, 570, 441, 710, and
427 respondents in Samples A through E, respectively. Sex composition of ness (e.g., Cheek & Briggs, 1988) and social reticence (e.g.,
each sample was approximately 60% female. Ethnic composition of each D. Watson & Friend, 1969), Social Anxiety clearly represents
sample was approximately 60%-75% European ancestry and 25%-40% a strongly disaffiliative form of Neuroticism (Briggs, 1988;
East Asian ancestry (80% of the latter being Chinese ancestry). Because
Paulhus & Trapnell, 1998).
European ancestry participants did not differ meaningfully from Asian
ancestry participants on the self-attention measures in this study, the two The PbSC scale showed a strong positive relation to Neuroti-
ethnic groups were combined in each sample. Participants ranged in age cism and a very weak but positive association with Extraversion.
from 17 to 59, with 90% of them between 17 and 25. In contrast to social anxiety, PbSC represents a mildly affiliative
form of Neuroticism. This partly explains why Social Anxiety and
Measures PbSC are factorially distinct: Despite sharing strong associations
with Neuroticism, Social Anxiety and PbSC relate oppositely to a
Self-attentiveness. In each of the samples we administered Fenig- broad, factorially powerful dimension of trait description,
stein et al.'s (1975) SCS as the first measure in the assessment battery.
Extraversion.
As noted above, the SCS contains three scales: PbSC, PrSC, and Social
Anxiety. The PbSC scale consists of 7 statements (e.g., "I'm concerned Turning to the PrSC scale, the regressions indicate that PrSC
about what other people think of me") thought to measure chronic scores are moderately but equally associated with two unrelated
awareness and concern over the self as a social stimulus. The Social personality domains: Neuroticism and Openness to Experience. A
Anxiety scale consists of 6 statements thought to measure anxious representative sample of high PrSC scorers will tend to include
reactions to socially self-conscious states (e.g., "It takes me time to two different sorts of individuals: those inclined to neurotically
overcome my shyness in new situations"). The PrSC scale consists
ruminate about themselves, and those inclined to philosophical
of 10 statements (e.g., "I reflect about myself a lot") thought to measure
chronic attendance to one's thoughts and feelings. All three scales
curiosity (in addition, of course, to a third group of Woody
possess adequate internal consistency and test-retest reliability for Allenesque individuals who both ruminate and philosophize, or
scales of this length (Fenigstein et al., 1975). Because factor analyses of philosophically ruminate).
the PrSC items within each of our five samples consistently recovered The bottom half of Table 2 also shows important differences
the two factors noted earlier, we also calculated subscales representing between the SR and ISA subscales of the PrSC. SR and ISA are
SR and ISA by summing the constituent items.
both associated with Openness, but they differ in that SR alone is
The Big Five. Three different measures of the Big Five factors were related to Neuroticism and ISA to Conscientiousness. These dif-
administered. In Sample A, participants completed the Extended Inter-
ferences provide an explanation for why the PrSC items reliably
personal Adjective Scales (IASR-B5; Trapnell & Wiggins, 1990),
which measures the Big Five factors of Conscientiousness, Neuroti-
cleave into two factors. That ISA and SR items covary enough to
cism, and Openness with 20 adjectives each. The two remaining factors define a common factor may be attributed to their shared associ-
are measured by means of factor scores computed from the 64 adjec- ation with Openness to Experience.
tives of the Interpersonal Adjective Scales (Wiggins, 1994). Eight The substantial secondary relation of ISA and Conscientious-
8-item scales are scored from these adjectives; each scale represents a ness was not anticipated and merits brief comment. In their factor
different octant of the interpersonal circumplex (Wiggins, 1979). These
analytic effort to articulate content distinctions within the broad
eight scales are combined to yield two orthogonal factor scores, labeled
Dominance (DOM) and Love (LOV), which may be interpreted as
affective domain of Positive Affectivity, D. Watson and Clark
variants of the Big Five factors of Extraversion and Agreeableness, (1992) found "attentiveness" to be a replicable facet of Positive
respectively. DOM and LOV differ from conventional measures of Affectivity. They also reported strong positive associations be-
Extraversion and Agreeableness in that, although they define the same tween Conscientiousness and attentiveness in adjectival data. At-
two-factor plane as Extraversion and Agreeableness (McCrae & Costa, tentiveness was defined by self-ratings on four adjectives: atten-
1989), they represent slight rotations of these factors. DOM may be tive, alert, concentrating, and determined. In light of this, consider
interpreted as a slightly colder version of Extraversion and LOV as a
the four items that define the ISA factor of the PrSC: "I'm alert to
slightly bolder version of Agreeableness.
changes in my mood"; "I'm generally attentive to my inner feel-
The Big Five measure administered to Samples B, C, and D was the Five ings"; "I'm aware of the way my mind works when I work through
Factor Inventory (FFI; Costa & McCrae, 1989), a 60-item short form of the
a problem"; "Generally, I'm not very aware of myself (italics
NEO Personality Inventory (Costa & McCrae, 1985). Reliability and
validity evidence for the FFI is impressive (Costa & McCrae, 1992). In added). The trait words alert and attentive (and possibly aware)
Sample E, the Big Five instrument was the Revised NEO Personality apparently connote conscientious deployment of attention. The
Inventory (NEO-PIR; Costa & McCrae, 1992). The NEO-PIR is a 240-item robust association between ISA and Conscientiousness shown in
personality questionnaire designed to measure the FFM by means of six Table 2 (see also Realo & Allik, 1998) may be of some theoretical
brief (8-item) scales, each representing a content distinction or facet significance. Self-awareness arising directly from personality dif-
thought to be basic or fundamental for a particular Big Five dimension
ferences in self-control may obviously be of some relevance to
(e.g., the assertiveness facet of Extraversion).
self-regulation theory. Rather than pursue here the potentially rich
issue of Conscientiousness and self-attention, we focus on what we
Results and Discussion believe to be the more theoretically important outcome of this
study: the dual association of PrSC with Neuroticism and Open-
Scores on the Social Anxiety, PbSC, and PrSC scales, and
ness to Experience and the potential of that duality to explain the
the SR and ISA subscales, were each regressed on the various
measures of the FFM. Results of those regressions are pro- self-absorption paradox implicit in PrSC research findings. In the
vided in Table 2. Social Anxiety demonstrated strong positive subsequent studies reported here, we consider a reconceptualiza-
associations with Neuroticism and strong negative associations tion of the PrSC construct derived from an explicit consideration
with Extraversion. Like the closely related constructs of shy- of these two Bis Five dimensions.
Table 2
Multiple Regression of the Self-Consciousness Scales (Public, Private, and Social Anxiety) and
the ISA and SR Factors of the Private Self-Consciousness Scale (PrSC) on
Markers of the Big-Five Personality Factors
Big Five instrument
IASR-B5,a NEO-FFI, NEO-PIR,

Sample A Sample B + C + D Sample E
Factor (N = 555) (N = 1,721) (N = 427)
Criterion = Social Anxiety

Extraversion -.57*** -.38*** -.46***
Agreeableness -.14*** .09*** 17***
Conscientiousness .02 -.03 .05
Neuroticism .27*** .35*** .37***
Openness to Experience -.02 -.12*** -.06
R .73 .61 .67
Criterion = public: Self-Consciousness Scale
Extraversion .03 22*** .18***
Agreeableness -.01 -.03 .00
Conscientiousness .06 .08*** .07
Neuroticism .41*** .48*** gj**#
Openness to Experience -.11 -.03 -.09
R .41 .44 .47
Criterion = PrSC
Extraversion .04 .05 -.05
Agreeableness -.17*** _07*** .04
Conscientiousness .16*** .08*** .16***
Neuroticism .39*** .29*** .28***
Openness to Experience .48*** .33*** .36***
R .62 .43 .46
Criterion = SR factor of PrSC
Extraversion .01 .03 -.05
Agreeableness -.22*** __ j1*** .01
Conscientiousness .04 -.02 .03
Neuroticism .48*** .37*** .41***
Openness to Experience .39*** .27*** 29***
R .64 .48 .52
Criterion = ISA factor of PrSC
Extraversion .07 .07*** -.02
Agreeableness .00 .05 .04
Conscientiousness .27*** .21*** .23***
Neuroticism .04 -.01 .01
Openness to Experience .27*** .29***
R .48 .38 .38
Note. Table values are standardized regression coefficients (betas). Big Five factors having beta weights with
a magnitude > .20 across all samples are presented in boldface type. ISA = Internal State Awareness factor of
PrSC; SR = Self-Reflection factor of PrSC; IASR-B5 = Revised Interpersonal Adjective Scales; NEO-FFI =
NEO Five-Factor Inventory; NEO-PIR = Revised NEO Personality Inventory.
a
Extraversion and Agreeableness were represented by IASR-B5 Dominance and Love factor scores, respectively.
***/> < .001.
Study 3 sive attending to perceived threats, losses, and injustices to the self.
Neurotic and intellective forms of PrSC may both be construed Epistemic self-consciousness may chiefly involve playful explor-
as self-regulatory tendencies sharing the important characteristic ing of novel, unique, or alternative self-perceptions. The
of directing attention toward one's thoughts and feelings. The cognitive-affective processes attending these two self-regulatory
psychological motives, values, goals, and intentions that accom- functions—that is, safety versus exploration—are likely to differ in
pany each of these equally "private" traits are sufficiently differ- theoretically important ways (cf. Maslow, 1955). The broad psy-
ent, however, that their self-regulatory effects are likely to be chological differences that warrant separate Neuroticism and
distinct. Neurotic self-consciousness may chiefly involve compul- Openness factors within the FFM provide a reasonable basis for
specifying and separately measuring neurotic and intellective vari- interests (Carson & Mowsesian, 1993), and paranormal beliefs
ants of PrSC. The preliminary purpose of Study 3, therefore, was (Davies, 1985).
to construct efficient, reliable markers of neurotic and intellective In Study 3 we tested the hypothesis that scores on the rumina-
self-consciousness, dispositions we henceforth refer to as rumination and reflection scales would differentially explain a very broad
tion and reflection, respectively. range of PrSC correlates. We evaluated this hypothesis by exam-
Note that the rumination-reflection distinction being proposed ining the effects of partialing rumination and reflection scores
here is clearly not equivalent to the SR-ISA distinction among from PrSC's correlations with a broad range of traits relevant to
PrSC items. The SR items do not differentiate neurotic from Neuroticism and Openness. We also reexamined a reported asso-
intellective self-consciousness, a distinction we have argued is ciation between PrSC and self-other similarity judgments. Srull
fundamental. The ISA items do not differentiate intellective from and Gaelick (1983) reported that PrSC scores moderated a pre-
conscientious self-consciousness, a distinction the FFM suggests dicted asymmetry effect in self-other similarity judgments. On the
may be nontrivial. Note also that in advancing our own two-factor basis of theoretical models of similarity asymmetry phenomena
perspective on private self-attentiveness, we do not take issue with (e.g., Tversky, 1977), Srull and Gaelick interpreted that modera-
the psychometric integrity of the PrSC. There is ample common tion as evidence that PrSC enhances awareness of unique features
variance among SR and ISA items to justify scoring and interpret- of the self; that is, that it enhances self-knowledge. We hypothe-
ing the PrSC total score (with the caveat that substantive differ- sized that this PrSC effect on self-other similarity judgments
ences between SR and ISA may often require evaluation of their stems from PrSC's overlap with the epistemic trait of reflection
separate effects). Psychological interpretation of PrSC correlations rather than to private self-attending per se.
and effects depends, however, on the motive (e.g., neurotic or
intellective) directing attention toward the private self. That Method
knowledge cannot be inferred from the PrSC total score, or from Participants and Measures
the SR and ISA subscale scores, because all three confound diver-
Data were provided by Samples A through E, described above, and one
gent motives for looking inward.
additional sample, Sample F (N = 965), which was demographically
The primary goal of Study 3 was to evaluate whether the similar to the others but with a slightly greater proportion of Chinese
reflection-rumination distinction provides a useful conceptual ancestry participants (approximately 40%). Because the majority of the
framework for interpreting the self-absorption paradox. Study 2 latter were relatively recent immigrants who spoke English as a second
suggests that the PrSC scale items confound reflection with rumi- language, and because several correlations of interest significantly differed
nation, motivationally distinct tendencies that may be differentially in magnitude between the Chinese and European ancestry participants in
this sample (e.g., reflection with paranormal beliefs), we report Sample F
responsible for the two classes of PrSC effects defining the
findings only for the European ancestry participants who spoke English as
paradox. their firsf language (N = 551). We constructed two scales for this study to
The FFM profile of the PrSC demonstrated here and elsewhere distinguish the private self-attentive aspects of Neuroticism (rumination)
(e.g., McCrae, 1993) implies that the PrSC's dispositional correlates and Openness to Experience (reflection). Development of these scales is
will be those traits associated with either the Neuroticism or Openness detailed below. In addition to the Big Five inventories and the SCS,
dimensions. Although a comprehensive review of published corre- described in Study 2, participants completed measures of various traits
lates of the PrSC is beyond the scope of this article, we briefly note relevant to the PrSC research literature. Because all of these measures have
that the PrSC scale typically shows little or no association with been widely used in the personality research literature, and evidence of
their construct validity and reliability is well-known and may be readily
Extraversion (McCrae, 1993); traits closely associated with Extraver-
obtained from other sources, the measures are simply listed below with
sion, such as sociability and activity level (Carver & Glass, 1976);
citations to these sources.
traits related to Conscientiousness, such as impulsivity (Carver &
Glass, 1976), ego strength, and self-control (Davies, 1982); Agree- Measures relevant to Neuroticism included the short form of the Taylor
Manifest Anxiety scale (Bendig, 1956), the short form of Byrne's (1961)
ableness (McCrae, 1993); and traits closely associated with that do-
Repression-Sensitization scale (Paulhus & Levitt, 1983), the Rosenberg
main, such as forthrightness and trust (Davies, 1982). Self-Esteem Scale (Rosenberg, 1965), the Negative Affectivity scale of the
Significant positive correlations with PrSC are routinely reported, Positive and Negative Affect Schedule (PANAS; D. Watson, Clark, &
however, for traits associated with Neuroticism and Openness. Tellegen, 1988), the Beck Depression Inventory (Beck, 1967), and the
Neuroticism-related correlates include the core Neuroticism facets of Brief Symptom Index (Derogatis, 1975). Measures relevant to Openness to
anxiety (e.g., Flett & Blankstein, 1987; Hope & Heimberg, 1988; Experience included the short form of the Need for Cognition Scale
Matthews & Wells, 1988; Wells, 1991) and depression (Ingram, (Cacioppo, Petty, & Kao, 1984), the Need for Self-Knowledge Scale
(Franzoi et al., 1990), the Personal Identity subscale of the Aspects of
1989; Ingram & Smith, 1984; Larson & Cowan, 1988; T. W. Smith
Identity Scale (Cheek, 1989), the Right-Wing Authoritarianism Scale
& Greenberg, 1981) and other traits associated with Neuroticism,
(Altemeyer, 1981), and ad hoc short forms of two additional scales: the
such as fear of negative evaluation (Monfries & Kafer, 1994). Multidimensional Personality Questionnaire Absorption scale (Tellegen,
The PrSC scale also demonstrates positive correlations with a 1994) and the Personality Research Form Sentience scale (Jackson, 1984).
remarkably broad range of traits linked to Openness to Experience. To accommodate administrative constraints, we omitted one third of the
These include traits with an obvious theoretical relevance to PrSC, Sentience scale items and half of the Absorption scale items.4 Sample F
such as thoughtfulness (Turner, Scheier, Carver, & Ickes, 1978),
imaginativeness (Davies, 1982), and need for cognition (Reeves,
Watson, Ramsey, & Morris, 1995), and traits whose relation to 4
Exclusion criteria were item length and low face validity. In the
PrSC is less theoretically obvious, including personal identity context of the large number of other scales examined, bias in the results for
(Cheek & Briggs, 1982), need for uniqueness (Schlenker & these two scales as a result of item selection unwittingly favoring the
Weigold, 1990), radicalism (Davies, 1982), artistic occupational hypothesis was not considered consequential.
included a paranormal beliefs scale constructed by Paul D. Trapnell for this Experience (see especially Costa & McCrae, 1985; McCrae, 1993-1994,
sample for other research purposes. Its inclusion permitted re-evaluation of 1994, 1996b; McCrae & Costa, 1997a; Trapnell & Wiggins, 1990). Two
Davies's (1985) finding that paranormal belief is positively related to considerations were particularly influential. The first was the close empir-
PrSC. Cronbach's alpha estimate of reliability for the 10-item paranormal ical association of curiosity with Openness to Experience. Trait adjectives
belief scale was .88 for the Sample F participants of European ancestry such as curious, inquisitive, and inquiring tend to be among the best lexical
(N = 551). Typical of the paranormal belief scale content are the items markers of the Openness factor (e.g., Goldberg, 1990; Trapnell & Wiggins,
"Certain gifted psychics do possess extraordinary, unusual powers" and "I 1990), and Openness is correlated with trait curiosity (e.g., Costa &
believe certain people do have real memories of a past life lived before this McCrae, 1987), especially epistemic variants such as Murray's (1938)
one," which loaded most highly (.77 and .75, respectively) on the first "need for understanding" (e.g., Costa & McCrae, 1988). These associations
unrotated principal component underlying the 10 paranormal item inter- were influential in that they allow a basis for inferring a fundamental
correlations in this sample.5 motivational distinction underlying the proposed two forms of self-atten-
A random subset (n = 220) of participants in Sample A, and all Sample tion: anxiety/fear versus curiosity/exploration (for reviews of this distinc-
F participants, also completed a self-other similarity rating. In Sample A tion see Russell, 1973; Spielberger, Peters, & Frain, 1981).
the rating was done first, before the personality and attitude measures. In A second consideration was the consistently high loadings of trait
Sample F the rating was administered last, after the other measures. descriptors referring to philosophical reflectiveness on the Openness factor.
Randomly selected halves of these samples received one of two wordings For example, when Trapnell and Wiggins (1990) evaluated candidate items
of the similarity rating question. One half was asked "In general, how for an adjectival measure of Openness, adjectives associated with philo-
similar are you to others?"; the other half was asked "In general, how sophical reflectiveness (e.g., philosophical, reflective, contemplative, med-
similar are others to you"? In Sample A, responses were made on a 7-point itative, introspective) were found to have the strongest and most general
scale, ranging from 6 {extremely similar) to 0 (not at all similar). Re- relations with seven preliminary content facets of Openness. A scale
sponses in Sample F were made on a 5-point scale, ranging from 1 {very consisting of these and similar adjectives was the highest loading scale on
similar) to 5 (not at all similar). To simplify presentation of these data, we Goldberg's (1990) Intellect factor. These five adjectives are also among
reverse scored Sample F ratings. those most highly correlated with the self-focus exemplar self-examining in
Study 1 (see Table 1). Finally, note that a close connection between
self-attention and proclivity for abstract thinking has long been assumed in
Procedure multivariate trait psychology. In the 1930s, J. P. Guilford identified a
personality dimension unrelated to social introversion he first labeled
To operationalize the traits of rumination and reflection, we developed brief "Interest in the Self (italics added) and subsequently relabeled "Thought-
questionnaire measures of each construct by means of conventional construct- fulness" (see Guilford, 1975). This factor is conceptually and empirically
oriented methods (construct definition, rational item generation, empirical closely related to Openness to Experience (McCrae, 1993-1994; 1994).
checks on scale homogeneity, factor structure, and convergent and discrimi- These considerations together suggested a general, intellective category
nant associations with other traits). We inferred a preliminary conception of of self-attentiveness defined as reflections on the self motivated not by
dispositional self-attentiveness associated with Neuroticism (rumination) from distress.about the self but by epistemic curiosity, that is, pleasurable,
the research literature on the metacognitive and self-evaluative implications of intrinsic interest in abstract or philosophical thinking. We wrote 13 pre-
negative affectivity or Neuroticism, in particular its core traits of anxiety and liminary items to measure this intellective category of self-attentiveness.
depression (e.g., Costa & McCrae, 1980). Anxiety states and traits are asso- Most items were constructed by the simple method of translating adjective
ciated with worry (Borkovec, Shadick, & Hopkins, 1990); hypervigilent at- markers of reflectiveness (e.g., philosophical) into statements worded so as
tention to threat cues (Mathews, 1990); and chronic, intrusive thoughts (Rach- to ensure that each statement connoted positive, epistemic motivations
man & Hodgson, 1980). Depression is associated with attentional biases (e.g., "I love exploring my inner self," [italics added]).
favoring negative self-information (e.g., Greenberg & Pyszczynski, 1986), The items were assembled in a preliminary 28-item questionnaire called
cognitive undoing following negative life events (e.g., Horowitz, 1976), and the Rumination-Reflection Questionnaire (RRQ), with rumination items
ruminative tendencies (e.g., Nolen-Hoeksema, 1991; Nolen-Hoeksema, Mor- on the first page and reflection items on the second.7 Instructions read: "For
row, & Fredrickson, 1993). These considerations together suggest a general, each of the statements located on the next two pages, please indicate your
neurotic category of self-attentiveness defined as recurrent thinking or rumi- level of agreement or disagreement by circling one of the scale categories
nations about the self prompted by threats, losses, or injustices to the self (i.e., to the right of each statement. Use the scale as shown below." Responses
self-related recurrent thought associated with anxiety, depression, and anger, were given on a 5-point scale with the points strongly disagree (1),
respectively). We view this form of chronic serf-focus as primarily past disagree (2), neutral (3), agree (4), and strongly agree (5). Analyses based
oriented, in contrast to the closely related construct "worry," which tends to be on the first 129 participants of Sample A led us to drop 3 rumination items
future oriented (for a detailed theoretical discussion of recurrent thought in and 1 reflection item and to reword 2 rumination items and 1 reflection
general, see Wyer, 1996). Our conception of rumination would correspond to item. The remaining participants in Sample A completed the revised
what Martin and Tesser (1996) referred to as "working through, and regret" (p. version of the RRQ.8
9) or to what Gohm, Isbell, and Wyer (1997) referred to as "negative thinking
about the past" (p. 87), but it has a narrower focus in that it refers specifically
to the chronic self-attention accompanying these thoughts.
5
A fuller description of the item content and descriptive statistics of this
We wrote 15 preliminary questionnaire items to operationalize this
scale are available on request from Paul D. Trapnell.
conception of ruminative self-focus. We sought to ensure that each item 6
conveyed a suitably negative tone while still connoting a primarily atten- Dickman's (1990) operational distinction between negative and posi-
tional tendency rather than an affective one. To accomplish this, we tive forms of impulsivity appear to involve a similar item wording strategy
exploited the implicit undesirability of ruminative frequency. Thus, the for negative impulsivity.
7
majority of items contain explicitly or implicitly negative references to For arguments and data defending this "transparent" strategy of item
duration or frequency (e.g., long time afterward, often, always, great deal ordering in the questionnaire, see Goldberg (1992).
of time).6 8
We combined data across the two versions in subsequent analyses
A preliminary conception of dispositional self-attentiveness associated involving Sample A, because the means and standard deviations of the
with positive epistemic motives, reflection, was inferred from the literature preliminary and revised versions were comparable and because internal
specifying the cognitive and motivational characteristics of Openness to consistency of the preliminary version was good.
Results and Discussion correlated substantially and equivalently with reflection and rumi-
nation (mean rs = .53 and .53, respectively), whereas ISA corre-
The RRQ lated moderately with reflection (mean r = .39) but not with
Table 3 presents factor loadings and summary item statistics. As rumination (mean r = .05). Their divergent associations with
expected, the rumination and reflection items loaded highly and rumination imply that SR and ISA are psychologically as well as
uniquely on their corresponding factors. The correlation between psychorhetrically distinct (cf. Bissonette & Bernstein, 1990). SR
the rumination and reflection factors was minimal (r = .22, direct items appear to involve connotations of rumination (e.g., "I'm
oblimin rotation), less than that typically obtained between PrSC constantly examining my motives"; "I'm always trying to figure
and PbSC (e.g., across the 17 samples reviewed by Wicklund & myself out"; "I reflect about myself a lot" [italics added]), whereas
Gollwitzer, 1987, the median correlation between PrSC and PbSC ISA items apparently do not (e.g., "I'm generally attentive to my
scale scores was .42). Rumination and reflection appear to be inner feelings"). These differences in ruminative connotations may
essentially independent tendencies. Table 4 presents descriptive provide a sufficient explanation for the differential relation of SR
statistics and estimates of internal consistency for the RRQ scales. and ISA items with Rumination and the reliable identification of
Alpha estimates of reliability exceeded .90, and the mean interitem two PrSC factors.
correlation (r u ) exceeded .40 for both scales. Neither scale mean The fact that SR and ISA both demonstrate substantial associ-
differed significantly between women and men (both fs < 1). ations with reflection introduces the possibility that reflection may
fully account for the general component of PrSC item intercorre-
lations. If so, this would imply that the psychometric core of the
Relations Between the RRQ and the SCS
PrSC is, in fact, reflection. Shared variance between SR and ISA
Table 5 gives the correlations between the RRQ and SCS provides a convenient operationalization of this common variance
measures. As expected, Reflection and Rumination both correlated (the variance not specific to the SR or ISA factors—from a factor
substantially with PrSC (mean rs = .59 and .43, respectively) but analytic perspective, the construct of PrSC represents the higher
showed a divergent pattern of association with SR and ISA. SR order factor defined by common SR and ISA variance; residual
Table 3
Factor Structure and Descriptive Statistics of Rumination-Reflection Questionnaire (RRQ) Items
Factor Sample
loadings" statistics15
RRQ item M SD
Reflection
15. I love exploring my "inner" self. .81 -.04 3.10 1.06
23. I often love to look at my life in philosophical ways. .79 .01 2.84 1.11
22. I love to meditate on the nature and meaning of things. .76 -.06 3.14 1.14
17. I don't really care for introspective or self-reflective thinking. (—) .75 .00 3.51 0.96
16. My attitudes and feelings about things fascinate me. .72 -.03 3.20 1.00
18. I love analyzing why I do things. .71 .11 2.98 1.09
20. I don't care much for self-analysis. (—) .71 .06 3.46 0.99
14. I'm not really a meditative type of person. (—) .70 -.05 3.14 1.14
13. Philosophical or abstract thinking doesn't appeal to me that much. (—) .69 .09 3.42 1.17
24. Contemplating myself isn't my idea of fun. (—) .69 -.01 2.82 1.04
19. People often say I'm a "deep," introspective type of person. .67 .03 2.75 1.10
21. I'm very self-inquisitive by nature. .59 .11 3.34 0.93
Rumination
5. I tend to "ruminate" or dwell over things that happen to me for a really long time afterward. .03 .80 3.43 1.09
7. Often I'm playing back over in my mind how I acted in a past situation. .03 .78 3.72 0.90
2. I always seem to be rehashing in my mind recent things I've said or done. .02 .77 3.62 1.04
4. Long after an argument or disagreement is over with, my thoughts keep going back to what happened. .05 .71 3.84 1.02
6. I don't waste time rethinking things that are over and done with. (—) .02 .71 3.66 0.94
8. I often find myself reevaluating something I've done. .03 .70 3.74 0.88
11. I often reflect on episodes in my life that I should no longer concern myself with. .00 .70 3.29 1.03
12. I spend a great deal of time thinking back over my embarrassing or disappointing moments. .10 .69 3.10 1.15
9. I never ruminate or dwell on myself for very long. ( - ) .10 .65 3.32 1.00
10. It is easy for me to put unwanted thoughts out of my mind. (—) .06 .61 3.50 1.03
3. Sometimes it is hard for me to shut off thoughts about myself. .17 .59 3.18 1.10
1. My attention is often focused on aspects of myself I wish I'd stop thinking about. .10 .58 3.12 1.10
Note. Based on Samples D and E combined, N = 1,137. Reverse-scored items are indicated by (—). Factor loadings greater than 1.251 are shown in
boldface type.
a
Principal-components extraction, oblique rotation by means of direct oblimin method. b Means for reverse-scored items are means following reverse
scoring; item response options were strongly disagree (1), disagree (2), neutral (3), agree (4), and strongly agree (5).
Table 4
Psychometric Characteristics of Rumination-Reflection Questionnaire Scales
Combined sample Women Men

(N = 1,137) (n = 687) (n = 447)
Coefficient
Scale alpha '"ii M SD M SD M SD
Reflection .91 .48 3.14 0.76 3.17 0.76 3.09 0.76

Rumination .90 .43 3.46 0.71 3.48 0.71 3.42 0.71
Note. Tabled values are based on Samples D and E combined. Total sample includes cases with missing gender
information. Scale means are expressed as mean item responses, on the basis of a five-place response format.
r a = mean interitem correlation.
variance specific to SR and ISA defines those constructs but does Relations Among Rumination, Reflection, and the FFM
not define PrSC). We evaluated whether reflection explains co-
variation between SR and ISA by comparing the zero-order cor- The RRQ scales were designed explicitly to distinguish anxious
relation between the ISA and SR with their partial correlations, from inquisitive self-focus. We therefore anticipated a high, pos-
controlling for RRQ reflection. Results of this analysis are pre- itive association between the rumination scale and markers of
sented in Table 6. In light of the large sample sizes here, one can Neuroticism, and a high, positive association between the reflec-
confidently conclude that SR and ISA are associated with each tion scale and markers of Openness to Experience. We expected all
other because of their common relation to reflection. Taken to- other Big Five associations for the RRQ scales to be near zero. In
gether, the findings reported in Tables 5 and 6 suggest (a) that the all samples (A-F) this was essentially the case. For the rumination
psychometric core of PrSC is redundant to reflection and (b) that scale, convergent correlations with Neuroticism ranged from .57 to
PrSC total scores confound reflection variance with a second .64, and the largest divergent Big Five correlation was with Ex-
source of variance, neurotic rumination, that is specific to the traversion (—.15, Sample A). For the reflection scale, convergent
larger SR factor. correlations with Openness to Experience ranged from .61 to .68,
and the largest divergent Big Five correlation was with Neuroti-
Note, again, that these conclusions do not imply correspondence
cism (.15, Sample D). This pattern of association with the Big Five
of SR with rumination and ISA with reflection. SR and ISA are
supports our conception of rumination and reflection as the self-
both substantially correlated with reflection. We interpret SR as a
attentive aspects of Neuroticism and Openness to Experience,
blend of reflection and rumination and ISA as a blend of reflection
respectively.9
and something else (e.g., conscientious self-consciousness, or
alertness). SR and ISA have complex associations with rumination
and reflection and do not clearly correspond to either. PrSC Correlates
In Table 7 we provide a comprehensive set of dispositional
Table 5 correlates of PrSC grouped according to the Big Five factor they
Correlation of Private Self-Consciousness (PrSC) and PrSC are most associated with: Neuroticism or Openness to Experience.
Factors SR and ISA with Rumination and Reflection For each criterion measure, the table displays zero-order correla-
tions with the RRQ and PrSC scales, and partial correlations with
PrSC factor the PrSC scale controlling for the RRQ scales. Four general
Sample N PrSC ISA SR
9
RRQ rumination Additional component and regression analyses in Sample E (n = 427)
also supported this conclusion. In a conjoint principal-component analysis,
A 555 .44 .02 .53
570 .47 .07 .57 with varimax rotation, of NEO facets and all of the self-attention measures
B
C 441 .44 .10 .51 (reflection, rumination, SR, ISA, PbSC, and Social Anxiety), reflection
D 710 .41 .02 .53 obtained the highest absolute loading (.81) of the eight variables defining
E 427 .36 .02 .53 an Openness factor, and rumination obtained the third-highest absolute
M .43 .05 .53 loading (.76) of the 10 variables defining a Neuroticism factor. In contrast
to SR, ISA, PbSC, and Social Anxiety, neither rumination nor reflection
RRQ reflection
had any meaningful loading (e.g., >.2O) on a second Big Five factor (factor
A 555 .53 .31 .48 loading matrices from this analysis are available on request). In follow-up
B 570 .62 .43 .56 analyses, reflection scores were regressed onto the six NEO Openness
C 441 .63 .46 .54 facets, and rumination scores were regressed onto the six NEO Neuroticism
D 710 .56 .32 .52 facets. Significant beta weights on reflection included ideas (.40), feelings
E 427 .60 .42 .54 (.33), and aesthetics (.17). Significant beta weights on rumination included
M .59 .39 .53
self-consciousness (.25), depression (.21), and anxiety (.20). These find-
ings suggest that reflection and rumination, despite their narrow conceptual
Note. Tabled correlations greater than .08 are significantly different from
zero at p < .05. SR = Self-Reflection; ISA = Internal State Awareness; bandwidth, possess very general relations to Openness and Neuroticism,
RRQ = Rumination-Reflection Questionnaire. respectively.
Table 6 to ideas, absorption, sentience, right-wing authoritarianism, and

Correlation Between SR and ISA Factors of PrSC as a Function paranormal beliefs correlated more highly with RRQ reflection
of Rumination and Reflection than with PrSC, and in most cases all of the PrSC's covariation
with these Factor V traits was accounted for by reflection. Indeed,
Correlation between openness to ideas and openness to actions demonstrated significant
PrSCi'actors reversals in association with PrSC after reflection variance was
controlling for
controlled. The exceptions involved either inconsistent findings
Correlation between
Sample N PrSC factors Rum Ref (e.g., personal identity), joint association with rumination and
reflection (e.g., openness to feelings), or a relatively small sample
A 555 .21 .24 .09 size (e.g., need for self-knowledge and absorption).
B 570 .30 .31 .07
C 441 .27 .24 .04 Fourth, in both samples (A and F) PrSC associations with self-
D 710 .20 .25 .05 other similarity judgments were not attributable to private self-
E 427 .28 .32 .07 attentiveness per se but instead to the PrSC's reflection variance. As
M .25 .27 .06 was the case with all of the other correlates of PrSC presented in the
Note. Tabled correlations greater than .08 are significantly different from
bottom half of Table 7, perceptions of self-other similarity correlated
zero at p < .05. SR = Self-Reflection; ISA = Internal State Awareness; more highly with reflection than with PrSC, and partialing out reflec-
PrSC = Private Self-Consciousness Scale; Rum = rumination; Ref = tion eliminated the PrSC associations. Srull and Gaelick (1983) at-
reflection. tributed those associations to cognitive consequences of self-focus,
that is, heightened awareness of unique self-aspects. Note that chronic
ruminative self-focus does not show any consistent association with
conclusions may be drawn from these findings. First, the PrSC self-other similarity judgments. Perhaps ruminative self-focus does
scale is significantly correlated with every Neuroticism-related not lead to heightened awareness of unique self-aspects. The results
measure shown in this table, and every Openness-related measure, here introduce one alternative possibility: Greater perceived self-
with the exception of Openness to Values and Openness to Ac- other dissimilarity with higher levels of PrSC may be due, in part, to
tions, the NEO Openness facets that typically have the weakest openness-related motives to view oneself as unique (Schlenker &
association with the NEO Openness factor (e.g., Costa, McCrae, & Weigold, 1990). Motives to be different (see Snyder & Fromkin,
Dye, 1991). PrSC correlates appear to be quite predictable accord- 1980) are historically associated with artists and other social and
ing to their FFM status: Individual differences related to either the intellectual "rebels" within a given culture, that is, those most proto-
Neuroticism or Openness to Experience domains will—in gener- typically open to experience (McCrae, 1994; Sulloway, 1996). Given
al—tend to be associated with PrSC. the strength and generality of associations between reflection and
Openness-related dispositions (Table 7), one might anticipate a rela-
Second, in every instance rumination and reflection demon-
tively high "need for uniqueness" among highly reflective persons.10
strated stronger associations with the trait in question than did the
PrSC. This implies that PrSC correlates with all of these traits Taken together, the results reported in Table 7 offer a straightfor-
because its items indirectly capture elements of rumination and ward account of the self-absorption paradox implicit in PrSC research
reflection, constructs that have a more direct, distinct relation to findings. PrSC may not engender self-knowledge and psychological
these traits. distress by means of a unitary private self-attentional process such as
Third, rumination and reflection explained PrSC associations heightened awareness of private thoughts and feelings (Buss, 1980);
with all of these traits differentially (compare the two rightmost instead, PrSC self-knowledge effects may stem from epistemic pro-
columns of Table 7). This could be due, in part, to a "Big Five" cesses related to the Openness-related trait of reflection. In contrast to
bias in the sampling of PrSC correlates for this study. If these the predictions of self-awareness and self-regulation theories (Buss,
criteria are, however, fairly representative of the PrSC research 1980; Carver & Scheier, 1981a; Duval & Wicklund, 1972), the results
literature, the findings suggest that reflection and rumination will of Study 3 imply that PrSC self-knowledge effects may be statistically
differentially explain most, if not all, associations between PrSC and functionally independent of their association with negative affect.
and other traits. The widely discussed association of PrSC with PrSC associations with negative affect and psychological distress may
indexes of psychological distress (e.g., the Beck Depression In- stem from the dysfunctional cognitive processes implied by the
ventory) would appear to be an artifact of ruminative variance Neuroticism-related construct of rumination: uncontrollable problem
specific to the SR items. Partialing out rumination eliminated all of "scanning" and analysis that is unproductively repetitive (see Wyer,
PrSC's covariation with measures of psychological distress. Par- 1996).
tialing out reflection, on the other hand, left these correlations
unchanged or increased them (in the upper half of Table 7, com-
10
pare values in Column 3 with the corresponding values in Column Although these correlational findings support our conjecture that
7). Although chronic self-focused attention may indeed amplify reflection explains PrSC's association with self-other similarity judg-
the experience of emotions, including negative emotions (e.g., ments, in both samples we failed to replicate the previously reported
Nolen-Hoeksema, McBride, & Larson, 1997; Scheier & Carver, interaction between PrSC and the asymmetry effect (Srull & Gaelick,
1977), PrSC correlations with negative affect and psychopathology 1983). Additionally, self-other similarity ratings in Sample A were sig-
may have nothing to do with this process. nificantly higher in the others-to-self condition than in the self-to-others
condition, a finding that constitutes a reversal of the usual self-other
Results presented in the bottom half of Table 7 indicate that asymmetry effect. Failure to replicate the previously reported PrSC inter-
PrSC relations with cognitive traits and sociopolitical attitudes are action with the asymmetry effect may stem from procedural differences
due to PrSC overlap with reflection. Need for cognition, openness between our studies and Srull and Gaelick's (1983) study.
Table 7
PrSC Correlates as a Function of Rumination and Reflection
Zero-order correlation PrSC controlling for
PrSC Correlate Sample N Rum Ref PrSC Rum Ref
Neuroticism-related
IASR-B5 Neuroticism A 555 .60* .14* .32* .06 .32*
NEO-FFI Neuroticism D 710 .64* .15* .27* .00 .27*
NEO-PIR Neuroticism E 427 .63* .14* .25* .03 .21*
NEO-PIR Neuroticism: Anxiety E 427 .56* .18* .30* .10 .26*
NEO-PIR Neuroticism: Depression E 427 .58* .14* .28* .09 .29*
NEO-PIR Neuroticism: Vulnerability E 427 .50* .03 .16* -.04 .22*
NEO-PIR Neuroticism: Self-consciousness E 427 .58* .05 .25* .02 .32*
NEO-PIR Neuroticism: Hostility E 427 .37* .08 .19* .08 .22*
NEO-PIR Neuroticism: Impulsiveness E 427 .22* .11 .19* .10 .15*
Taylor Manifest Anxiety (s) C 441 .59* .07 .23* .03 .24*
Byrne Repression-Sensitization (s) C 441 .57* .04 .24* -.02 .27*
PANAS Negative affect D 710 .46* .16 .14* .03 .15*
PANAS Negative affect C 441 .43* .08 .26* .09 .26*
PANAS Negative affect B 570 .40* .05 .13* -.04 .12*
Beck Depression Inventory (s) C 441 .38* .04 .21* .06 .22*
Beck Depression Inventory (s) B 570 .36* .08 .16* .01 .13*
Derogatis Brief Symptom Index A 555 .43* .10 .29* .11 .29*
Openness-related
IASR-B5 Openness A 555 .10 .64* .43* .46* .03
NEO-FFI Openness D 710 .05 .61* .34* .37* .01
NEO-PIR Openness E 427 .09 .63* .37* .37* -.01
NEO-PIR Openness to Ideas E 427 .02 .54* .25* .27* -.14*
NEO-PIR Openness to Feelings E 427 .24* .54* .50* .46* .24*
NEO-PIR Openness to Fantasy E 427 .22* .40* .29* .24* .07
NEO-PIR Openness to Aesthetics E 427 .08 .54* .26* .26* -.11
NEO-PIR Openness to Actions E 427 -.18* .27* .05 .15* -.17*
NEO-PIR Openness to Values E 427 -.13 .16* .06 .14* -.04
Need for cognition (s) A 177a .01 .50* .33* .37* .02
Need for cognition (s) C 441 -.14 .48* .22* .31* -.12 .
Need for self-knowledge A 242a .10 .63* .51* .49* .17*
Absorption (s) A 242a .15 .50* .46* .43* .18*
PRF Sentience (s) E 427 .04 .42* .31* .32* .07
Personal identity A 242" -.02 .42* .30* .33* .05
Personal identity D 710 .08 .39* .35* .35* .17*
Right-wing authoritarianism E 427 .06 -.29* -.15* -.18* .03
Paranormal beliefs F 551b .04 .35* .24* .23* .00
"How similar are you to others?" A 109a -.25* -.18* -.13 -.03 .02
"How similar are you to others?" F 268" -.04 -.22* -.17* -.17* -.03
"How similar are others to you?" A 111° .03 -.41* -.23* -.30* -.01
"How similar are others to you?" 265b
F -.09 -.32* -.13* -.10 .10
Note. Short form or abbreviated version of a scale is indicated by (s). PrSC = Private Self-Consciousness Scale; Rum = rumination; Ref = reflection;
IASR-B5 = Revised Interpersonal Adjective Scales; NEO-FFI = NEO Five-Factor Inventory; NEO-PIR = Revised NEO Personality Inventory;
PANAS = Positive and Negative Affect Schedule; PRF = Personality Research Form.
a
Subsample of Sample A. b Participants of European ancestry.
* Experiment-wise p < .05 (p < .05/28 criterion scales, pairwise p < .002).
Study 4 bles Murray's (1938) "need for understanding," which in turn is

closely identified with Openness to Experience (Costa & Mc-
In Study 4 we evaluated the distinction between rumination
Crae, 1988). These identities imply that the PrSC effect on
and reflection self-focus in the context of a suggestion by
research volunteering may be attributable solely to the PrSC's
Franzoi et al. (1990), who argued that PrSC scores reflect a
overlap with Openness-related reflection.
motivational as well as an attentional tendency. In support of
this claim, they showed that higher PrSC scores are associated
with a greater likelihood of volunteering for psychology exper-
Method
iments. Franzoi et al. interpreted the underlying motive as Participants were those described in Sample B (see Study 2). The
"need for self-knowledge," an interpretation that closely resem- envelope of measures contained a signup form that stated:
Table 8
PrSC Effect on Volunteering for Psychology Experiments as a Function
of Openness to Experience and Reflection
ANOVA cell means and effects Ref Open PrSC SR ISA
Cell means
Signed form (n = 513) 3.2 39.6 23.9 13.1 10.8
Did not sign form (n = 61) 2.8 36.9 22.3 12.3 10.0
Between-groups effect
F(l, 573) 17.8*** 14.2*** 6.5* ns 9.0**
F(l, 573), Openness partialed 6.1* ns ns 4.6*
F(l, 573), reflection partialed ns ns ns ns
Note. Sample B, N = 574. PrSC = Private Self-Consciousness Scale; Ref = reflection; Open = Openness to
Experience; SR = Self-Reflection; ISA = Internal State Awareness.
* p < . 0 5 . **/?<.01. ***p<.001.
We are conducting a number of other studies later on in the school PrSC component carrying no rumination variance, accounted for
year on various topics, including personality dispositions and daily all of the PrSC association with research volunteering. General
life events. If you would like to be contacted about any of these Openness to Experience did not fully account for the ISA effect on
additional studies that you may wish to participate in (also in ex- volunteerism. The self-attentive aspect of Openness, reflection,
change for research credit points), please leave your name and phone
did.11
in the space provided below.
In concluding that "need for self-knowledge" underlies the
Of the 574 participants who returned completed personality questionnaires positive association between PrSC and volunteering for psychol-
in Sample B, 513 (89% of the total sample) provided their names and ogy experiments, Franzoi et al. (1990) developed their own 5-item
phone numbers on the form. Participants' scores on the PrSC, RRQ, and a Need for Self-Knowledge scale. This scale was included in the
global measure of Openness to Experience (from the FFI; see Study 2) present Study 3. Findings with this scale (reported near the bottom
were reanalyzed for the present study. of Table 7) were consistent with the results of this study: The Need
for Self-Knowledge scale was essentially unrelated to rumination
Results and Discussion (r = .16) and strongly related to reflection (r = .63), and reflection
accounted for the lion's share of this scale's association with PrSC.
Participants were designated as either "signers" (those who
completed the signup sheet) or "nonsigners" (those who did not
complete the signup sheet). We evaluated mean differences be- General Discussion
tween signers and nonsigners with one-way analyses of variance, In an exploratory factor analysis of the SCS items, Fenigstein et
and we examined mediation effects with two-group analyses of al. (1975) revealed two factors thought to reflect a fundamental
covariance. Results are presented in Table 8. Rumination means dichotomy in self-perception of long-standing utility in psychol-
did not differ between signers and nonsigners and are therefore not ogy: the public and private selves (e.g., James, 1890, cited in
included in the table. Means for PrSC were significantly lower Fenigstein, 1987). In this article we introduced and articulated a
among nonsigners than signers, F(l, 573) = 6.5, p < .05, a finding motivational distinction relevant to dispositional self-focus that
that replicates that of Franzoi et al. (1990). However, as antici- bears a family resemblance to another dichotomy of long-standing
pated, nonsigners tended to have even lower scores dian signers on utility in psychology: fear and curiosity. Rumination provides a
reflection, F(l, 573) = 17.8, p < .001, and Openness, F(l, 573) summary conception of self-attentiveness motivated by perceived
= 14.2, p < .001. Thus, as was the case in our earlier studies, threats, losses, or injustices to the self. Reflection provides a
effects for reflection tended to be stronger than for PrSC. summary conception of self-attentiveness motivated by curiosity
The principal hypothesis was that individual differences in or epistemic interest in the self. Rumination and reflection are
reflection would account for the motivational effect of PrSC on statistically and psychologically distinct: Rumination is uniquely
research volunteering. The F ratios reported in the last two rows of associated with the Neuroticism factor, and reflection is associated
Table 8 are most relevant to this hypothesis. Group differences in with the Openness factor of the FFM.
PrSC were completely redundant with the reflection scale. Con- We demonstrated the utility of the RRQ distinction by applying
trolling for group differences in Openness also eliminated the it to a paradox inherent in PrSC research findings. A large body of
PrSC effect. evidence suggests that higher PrSC scores are associated with
These findings imply that the PrSC scale predicts research greater self-knowledge yet higher levels of psychological distress.
volunteerism because it overlaps with the epistemic trait of reflec-
tion. The rumination component of PrSC variance, which accounts
for its association with various measures of psychopathology, is 1
' The reliable association between ISA and Conscientiousness (Study
not related to research volunteerism. This conclusion is further 2) might suggest other motives for research volunteerism: achievement
supported by the distinct pattern of effects for the PrSC factors, SR striving or dutifulness. In the present study, however, signers did not differ
and ISA. SR was not related to research volunteering. ISA, the from nonsigners on the FFI-Conscientiousness scale (p > .25).
We hypothesized that this paradox derives from motivational Implications for the PrSC Literature
ambiguity in the wording of many PrSC items that confounds two
unrelated dispositions, rumination and reflection. A series of stud- Factorial Complexity of the PrSC Scale
ies evaluating this hypothesis revealed the following findings. In light of these data, the view that the two recurrent factors
First, lay descriptors of self-attentiveness codified in the natural within the PrSC are due mainly to differences in item response
language {self-consciousness and self-examining), and the influen- distributions or item difficulties (Bernstein et al., 1986) does not
tial conceptions represented in the SCS scales, both exemplify, in appear tenable. The correlations for ISA and SR with the FFM are
some form, the motivational distinction articulated here. The So- clearly distinct, but, more important, the distinction is consistent
cial Anxiety and PbSC scales are primarily neurotically motivated with a natural language distinction between neurotic and intellec-
forms of self-attention, and the PrSC scale is primarily an intel- tive self-focus, and these and other findings reported here dovetail
lectually motivated form. However, all three SCS involve second- with other reports that the ISA and SR subscales show divergent
ary distinctions with respect to other broad personality differences: associations with theoretically relevant criteria of PrSC (e.g.,
Social Anxiety and PbSC relate oppositely to Extraversion, and Anderson & Bohon, 1996; Campbell et al., 1996; Lavallee &
different subsets of PrSC items have unique associations with Campbell, 1995; P. J. Watson et al., 1988).
Neuroticism and Conscientiousness.
Second, it is possible to distinguish neurotic self-attentiveness Reevaluation of Previous PrSC Research Findings
from Openness-related self-attentiveness by conventional ques-
In light of the psychological distinctions implied by the con-
tionnaire measures. Scales constructed to separately measure neu-
structs of rumination and reflection, independent mediation of
rotic and intellective self-focus had a minimal correlation with PrSC correlates and effects by the two RRQ scales raises important
each other, but both were substantially associated with PrSC. concerns about the routine cognitive (i.e., attentional) interpreta-
Mean rumination and reflection correlations with the larger PrSC tions of many PrSC findings. PrSC findings attributed to such
factor, SR, were, in fact, exactly equal (.53). cognitive mechanisms as "comparitor activation" (comparison of
Third, the rumination-reflection distinction sheds light on the self-aspects or one's behavior with a salient reference standard) or
issue of factorial complexity in the PrSC. The two factors that have "self-concept articulation" may reflect stable neurotic (proneness
been repeatedly recovered from the PrSC items can be attributed to to negative affect) or epistemic motivational differences between
item content differences related to rumination and reflection. What high and low scores on the PrSC. Differences in epistemic motives
primarily associates the factors is reflection (present in both SR and values, for example, may cause differences in task absorption
and ISA). What primarily distinguishes them is rumination or engagement, or intrinsic motivation, during experimental tests
(present in SR only). Analysis of ISA and SR relations with the of self-awareness and self-regulatory hypotheses. Task-relevant
FFM suggested a further distinction that we believe warrants motivational effects of both Openness and Neuroticism variance
future research attention: The ISA factor has a very reliable, tapped by the PrSC could differentially mimic the cognitive effects
secondary association with Conscientiousness. of self-attention postulated by self-regulation theories. This would
lead to the erroneous impression that particular state and trait
Fourth, rumination and reflection differentially explain disposi-
effects of "self-focus" were conceptual replications of the same
tional correlates of PrSC. Rumination connotations of a subset of
phenomena.
PrSC items explain PrSC associations with psychological distress
(e.g., depression, anxiety). Reflection explains PrSC associations Consider, for example, experiments reported by Scheier and
with intellectual traits (e.g., need for cognition, need for self- Carver (1983), who sought to test whether self-awareness prompts
knowledge) and a remarkably broad array of other traits known to comparison of one's behavior with salient behavioral reference
be associated with Openness to Experience (e.g., lack of authori- standards ("comparitor activation," in the language of self-
tarianism, paranormal beliefs) but that have no theoretically obvi- regulation theory). In one study, participants copied a complex
geometric figure with only intermittent visual access to the origi-
ous relation to self-focused attention. Because neurotic traits and
nal. The number of times they requested access to the original (i.e.,
intellective traits define major, uncorrelated dimensions of person-
the salient behavioral reference standard) operationalized the self-
ality, these data raise serious questions about the psychological
regulatory notion of comparitor activation. Results for manipulated
unity of PrSC correlates and effects.
self-attention paralleled those for the PrSC: As predicted, higher
Finally, results here suggest that PrSC effects on self-knowledge levels of PrSC were associated with more frequent requests to have
are due to PrSC overlap with reflection, not to private self- another peek at the original geometric figure. However, it is
attention per se. Reflection demonstrated stronger and more robust reasonable to ask whether people who are especially high in
associations than PrSC with self-other similarity judgments, re- intellectual curiosity or aesthetic sensitivity (the defining features
search volunteerism, and traits associated with enhanced informa- of Openness to Experience) are merely more interested in their
tion processing, such as need for cognition. PrSC associations with performance on such tasks. If such PrSC effects have more to do
all of these criteria were fully mediated by reflection. with the motive for self-attending (epistemic vs. neurotic) than the
These findings have important implications for theory and re- mere fact of it, we would expect reflection and rumination to
search on dispositional self-focus. We discuss below their impli- differentially explain PrSC comparitor activation effects. Reflec-
cations for the PrSC research literature, for the status of the tion should explain PrSC effects on tasks likely to motivate intel-
public-private distinction within trait conceptions of self-focused lectually curious people, and rumination should explain PrSC
attention, and for our general proposal to reconceptualize disposi- effects on tasks likely to trigger self-evaluative doubts and worries.
tional self-consciousness from the perspective of the FFM. - Note that some experimental tasks are likely to confound neurotic
and epistemic dispositional effects on comparitor activation (e.g., both costs and benefits. Removing the rose-colored coating from
performance tests of cognitive abilities). one's looking glass is unlikely to enhance self-confidence and
optimism (Taylor & Brown, 1988). The interpersonal benefits of
accurate self-perception may, however, be substantial. Reflective-
Self-Absorption Paradox
ness or psychological-mindedness appears to enhance observer
At one time, a burgeoning literature on the phenomenon of perceptions of mental health (Colvin, Block, & Funder, 1995) and
depressive realism appeared to provide firm empirical grounds for may buffer psychological distress within close relationships. In a
the proverbial lay notion of "sadder but wiser" (e.g., Alloy & recent report, for example, the depression risk of women who were
Abramson, 1979). More recently, those empirical grounds have dispositionally ruminative was significantly reduced if they had a
cracked and loosened, and depressed mood no longer appears to dispositionally reflective spouse (Preece, DeLongis, Campbell, &
promise as much cognitive clarity as it once did (e.g., Dobson & Trapnell, 1998). This intriguing "dyadic interaction" of reflection
Franche, 1989; Haaga & Beck, 1995). It now appears that de- and rumination within close relationships suggests that the
pressed mood can improve self-evaluative accuracy, and can im- reflection-rumination distinction may be especially relevant to the
pair it, depending on the match between valence of the feedback coping and adjustment literature. Rumination and reflection appear
cues and valence of the self-schema (e.g., Campbell & Fehr, 1990; to provide a useful 2 X 2 model of cognitive approach and
Dykman & Abramson, 1989). avoidance styles. In combination, they suggest an intriguing, al-
We noted that the repeated demonstration of two apparently ternative definition of four cognitive styles with deep roots in the
contradictory outcomes of PrSC, psychological distress and self- adjustment literature: sensitizing (high reflection, high rumina-
knowledge effects, seemed to suggest a sadder-but-wiser phenom- tion), repressive (low reflection, low rumination), vulnerable (low
enon we refer to as the self-absorption paradox: Frequent inspec- reflection, high rumination), and adaptable (high reflection, low
tion of one's feelings and thoughts seems to improve the accuracy rumination).
of self-knowledge, but at the cost of psychological well-being. The
present findings cast doubt on a sadder-but-wiser interpretation. A
The Public-Private Distinction Revisited
majority of high scorers on the PrSC scale are not in fact sadder
and wiser, rather they are sadder or wiser. Because rumination and According to self-awareness theory (Duval & Wicklund, 1972),
reflection are relatively independent dispositions, and the former directing attention to the self affects behavior by means of mech-
alone accounts for the "sadness" while the latter explains the anisms that ensue simply as a consequence of becoming self-
"wisdom" effects of the PrSC, there may be no psychological aware. Fenigstein et al. (1975) accordingly sought to operational-
paradox to resolve. ize a chronic self-focusing tendency by means of a self-report
The PrSC literature's self-absorption paradox may have little to questionnaire, but they unexpectedly discovered three theoretically
do with sadder-but-wiser phenomena. This conclusion clearly suggestive factors among their candidate items. One factor, Social
leaves open, however, the central question about these phenomena Anxiety, was theoretically downgraded relative to the others, and
that is so theoretically relevant to a construct such as PrSC: the the other two were thought to reflect a basic distinction among
relation of self-knowledge to mental health. What role, if any, does self-conscious processes: public and private ones. The nature of
veridicality or authenticity of self-perceptions play in psycholog- this process distinction amounted to a spatial metaphor: outside
ical health and adjustment? The relation of self-awareness and versus inside the self. The historically rich resonances of that self
mental health obviously depends first on one's definition of mental metaphor in psychology (for a review see Hogan & Cheek, 1982)
health and, consequently, the problem of values in that definition. may, however, have routinely distracted researchers from (a) con-
Those who place a high value on authenticity, uniqueness, and sidering the potential role of individual differences in motives and
autonomy of the self, or who especially value the epistemic activ- affect in PbSC and PrSC research findings and (b) exploring more
ity of reflection—people high in Openness to Experience—are fully the provocative new domain of individual differences that
likely to believe the emotional costs of dispassionate self- Fenigstein et al. (1975) had so importantly introduced to psycho-
perception are well worth it (e.g., Socrates, Sartre, Carl Rogers). logical research: self-attentive traits.
People with other value orientations, such as those who place a It is plausible the Big Five factors played a silent role in
higher value on authority, propriety, and tradition than on auton- Fenigstein et al.'s (1975) abandonment of a unifactor conception
omy and uniqueness—people low in Openness to Experience— of dispositional self-consciousness. The findings reported here are
may legitimately disagree (e.g., religious fundamentalists). The consistent with the hypothesis that three unanticipated categories
question of whether reflection fosters or hinders adjustment de- emerged from Fenigstein et al.'s item pool because it unwittingly
pends in part on the question of values (individual and cultural) in traversed a few factorial "black holes" of trait description: Neu-
conceptions of mental health (cf. M. B. Smith, 1961). roticism, Extraversion, and Openness to Experience (e.g., Table 2).
We believe the distinction proposed here between rumination Had it spanned additional Big Five dimensions (e.g., Agreeable-
and reflection will prove useful in future efforts to address dispo- ness), additional factors would have been likely. The plausible
sitional aspects of this question, although the relation of these two self-attention item "I love to admire how I look" has a narcissistic
traits to mental health is bound to be complex, especially for the flavor that would tend to associate it with the negative pole of
trait of reflection. Rumination may be a general risk factor for Agreeableness within the FFM. Items such as these would be
maladjustment (e.g., Nolen-Hoeksema et al., 1997), but reflection unlikely to correlate positively or substantially with PbSC items
is likely to both help and hinder psychological adjustment. Al- (e.g., "I'm self-conscious about the way I look"), or with social
though reflective self-focus may indeed foster articulation and anxiety, or with PrSC items. Had such items been included in their
veridicality of the self-schema, accurate self-perception involves preliminary item pool, and a "disagreeable PbSC" factor emerged
in addition to the "neurotic PbSC" factor they did obtain, it is between investigators owing to the lack of a common (trait de-
reasonable to suppose Fenigstein et al. would have found a simple scriptive) language.
inner-outer distinction among the factors less persuasive. In their The FFM suggests motivational distinctions relevant to trait
efforts to account for why a unitary, general self-awareness factor self-attention other than the two developed here. The disagreeable-
failed to emerge from their candidate item pool, they might alter- extroverted location within the FFM is associated with two widely
natively have centered on possible motivational and affective researched dispositions relevant to self-attention and self-regula-
distinctions among the factors that did emerge (prosocial, antiso- tion: narcissism (Raskin & Hall, 1979) and self-monitoring (Sny-
cial, epistemic, etc.). der, 1974). The interpersonal motives and values associated with
this region of the FFM suggest basic distinctions between these
Beyond Neuroticism and Openness traits and the self-attentive dispositions referred to in this article.
Raskin and Hall's (1979) construct of dispositional narcissism
Wicklund (1975) argued that an individual-difference approach describes dispositional self-focus in the service of a self-
to self-awareness is seriously hampered by inherent "third vari- enhancement motive (Robins & John, 1997) that is manifested in
able" problems: interpersonally disagreeable ways (e.g., superiority and entitle-
ment, both of which are implicit derogations of others, a reliable
In developing tests of the theory [of objective self-awareness] we have sign of hostility, lack of altruism, or both). Snyder's (1974) self-
attempted to avoid the potential ambiguities associated with measures
monitoring construct describes a self-regulatory trait in the service
of self-consciousness, since it is difficult to know whether [the dif-
ferences they measure] are relevant to actual differences in self-
of political motives, that is, social dominance, or the control of
focused attention, differences in types of personal standards or styles other people and resources (Robins & John, 1997).
of discrepancy reduction, differences in ability to avoid self-focusing The FFM Conscientiousness factor suggests another potentially
stimuli, or even theoretically irrelevant differences that would have a useful conception of self-attentiveness. Natural language trait
bearing on the results, (p. 268) terms such as alert, aware, and attentive are empirically identified
with this factor and appear in the items of the PrSC-ISA factor.
These remarks are noteworthy in that they appeared in the litera- The specification of a conscientious form of self-attentiveness, and
ture long before the empirical findings documenting potential the development of an appropriate measure, may prove useful in
motivational confounds in both the PbSC (reviewed by Wicklund applications of self-regulation theory to health issues (e.g., Mullen
& Gollwitzer, 1987) and PrSC scales (e.g., Franzoi et al., 1990; & Suls, 1982) and stress and coping (e.g., Frone & McFarlin,
Schlenker & Weigold, 1990). We may be more optimistic (or 1989).
foolish) than Wicklund about the potential scientific utility of
dispositional conceptions of self-attentiveness. We agree with References
Wicklund's implication that trait measures of self-focus will typ-
ically involve important individual differences other than atten- Alloy, L. Y., & Abramson, L. B. (1979). Judgment of contingency in
depressed and nondepressed students: Sadder but wiser? Journal of
tional ones, but, rather than ignore a trait approach, we suggest that
Experimental Psychology: General, 108, 441-485.
motivational "confounds" in self-attentive dispositions might be
Altemeyer, R. A. (1981). Right-wing authoritarianism. Winnipeg, Mani-
articulated in a systematic manner. One approach, the one we
toba, Canada: University of Manitoba Press.
propose here, is to explicitly reference self-attentive dispositions to Anderson, E. M., & Bohon, L. M. (1996). Factor structure of the Private
important motivational distinctions already summarized and or- Self-Consciousness Scale. Journal of Personality Assessment, 66, 144-
dered within comprehensive empirical trait taxonomies such as the 152.
FFM. Appelbaum, S. A. (1973). Psychological-mindedness: Word, concept, and
One benefit of such an approach is that the task of specifying essence. International Journal of Psychoanalysis, 54, 35-46.
candidate "third variables" is assumed by the taxonomy. We noted Beck, A. T. (1967). Depression: Clinical, experimental, and theoretical
aspects. New York: Hoeber.
that PrSC is associated (by means of reflection variance) with the
Bendig, A. W. (1956). The development of a short form of the Manifest
Openness factor of the FFM. There are other trait constructs
Anxiety Scale. Journal of Consulting Psychology, 20, 384.
closely associated with the Openness factor that have generated Bernstein, I., Teng, G., & Garbin, C. (1986). A confirmatory factoring of
research literatures as extensive as those of PrSC, including need the Self-Consciousness Scale. Multivariate Behavioral Research, 21,
for cognition (Cacioppo, Petty, Feinstein, & Jarvis, 1996), absorp- 459-475.
tion (Tellegen & Atkinson, 1974), and authoritarianism (Altem- Bissonette, V., & Bernstein, I. H. (1990). Artifacts can replicate: A reply
eyer, 1981). The FFM can readily alert investigators to the poten- to Piliavin & Chamg. Personality and Social Psychology Bulletin, 16,
tial for identities in research findings across such large, and largely 554-561.
independent, research literatures as these. In the absence of a Block, J. (1995). A contrarian view of the five-factor approach to person-
comprehensive trait taxonomy, rounding up the usual suspects to ality description. Psychological Bulletin, 117, 187—215.
make these identifications is a much more haphazard enterprise. Borkovec, T. D., Shadick, R., & Hopkins, M. (1990). The nature of normal
Routine reference to a well-structured taxonomy such as the FFM and pathological worry. In D. Barlow & R. Raper (Eds.), Chronic
anxiety and generalized anxiety disorder (pp. 29—51). New York: Guil-
also helps prevent the "jingle-jangle" fallacy (the same traits
ford Press.
secretly traveling under different names between investigators;
Briggs, S. R. (1988). Shyness: Introversion or neuroticism? Journal of
Block, 1995), which is responsible for trait psychology's enduring
Research in Personality, 22, 290-307.
"Tower of Babble" problem (Goldberg, 1971): incommunication Brockner, J. (1979). The effects of self-esteem, success—failure, and self-
consciousness on task performance. Journal of Personality and Social Costa, P. T., Jr., & McCrae, R. R. (1980). Still stable after all these years:
Psychology, 37, 1732-1741. Personality as a key to some issues in adulthood and old age. In P. G.
Brockner, J. (1984). Low self-esteem and behavioral plasticity: Some Baltes & O. G. Brim (Eds.), Life-span development and behavior
implications for personality and social psychology. In L. Wheeler (Ed.), (Vol. 3, pp. 65-102). New York: Academic Press.
Review of personality and social psychology (Vol. 4, pp. 237-271). Costa, P. T., Jr., & McCrae, R. R. (1985). The NEO Personality Inventory
Beverly Hills, CA: Sage. manual. Odessa, FL: Psychological Assessment Resources.
Burnkrant, R. R., & Page, T. (1984). A modification of the Fenigstein, Costa, P. T., Jr., & McCrae, R. R. (1987). Personality assessment in
Scheier, and Buss Self-Consciousness Scale. Journal of Personality psychosomatic medicine: Value of a trait taxonomy. In G. A. Fava and
Assessment, 48, 629-637. T. N. Wise (Eds.), Advances in psychosomatic medicine: Research
Buss, A. (1980). Self-consciousness and social anxiety. San Francisco: paradigms in psychosomatic medicine (Vol. 17, pp. 71-82). Basel,
W. H. Freeman. Switzerland: Karger.
Byrne, D. (1961). The Repression-Sensitization Scale: Rationale, reliabil- Costa, P. T., Jr., & McCrae, R. R. (1988). From catalogue to classification:
ity, and validity. Journal of Personality, 29, 334-349. Murray's needs and the five factor model. Journal of Personality and
Cacioppo, J. T., Petty, R. E., Feinstein, J. A., & Jarvis, W. B. G. (1996). Social Psychology, 55, 258-265.
Dispositional differences in cognitive motivation: The life and times of Costa, P. T., Jr., & McCrae, R. R. (1989). The NEO-Pl/FFI manual
individuals varying in need for cognition. Psychological Bulletin, 119, supplement. Odessa, FL: Psychological Assessment Resources.
197-253. Costa, P. T., Jr., & McCrae, R. R. (1992). Revised NEO Personality
Cacioppo, J. T., Petty, R. E., & Kao, C. F. (1984). The efficient assessment Inventory (NEO-PI-R) and NEO Five-Factor Inventory (NEO-FFI) pro-
of need for cognition. Journal of Personality Assessment, 48, 306-307. fessional manual. Odessa, FL: Psychological Assessment Resources.
Campbell, J. D. (1990). Self-esteem and clarity of the self-concept. Journal Costa, P. T., Jr., McCrae, R. R., & Dye, D. A. (1991). Facet scales for
of Personality and Social Psychology, 59, 538-549. Agreeableness and Conscientiousness: A revision of the NEO Person-
Campbell, J. D. (1993). [Annual personality and attitude survey, University ality Inventory. Personality and Individual Differences, 12, 887-898.
of British Columbia]. Unpublished raw data. Davies, M. F. (1982). Correlates of self-consciousness and the 16 Person-
Campbell, J. D., & Fehr, B. (1990). Self-esteem and perceptions of con- ality Factor Questionnaire. Journal of Psychology, 111, 123—128.
veyed impressions: Is negative affectivity associated with greater real- Davies, M. F. (1985). Self-consciousness and paranormal beliefs. Percep-
ism? Journal of Personality and Social Psychology, 58, 122-133. tual & Motor Skills, 60, 484-486.
Campbell, J. D., & Lavallee, L. F. (1993). Who am I? The role of Derogatis, L. R. (1975). The Brief Symptom Inventory. Baltimore: Clinical
self-concept confusion in understanding the behavior of people with low Psychometrics Research.
self-esteem. In R. F. Baumeister (Ed.), Self-esteem: The puzzle of low Dickman, S. J. (1990). Functional and dysfunctional impulsivity: Person-
self-regard (pp. 3-22). New York: Plenum. ality and cognitive correlates. Journal of Personality and Social Psy-
Campbell, J. D., Trapnell, P. D., Heine, S. J., Lavallee, L. F., Katz, I. M., chology, 58, 95-102.
& Lehman, D. R. (1996). Self-concept clarity: Measurement, personality Dobson, K., & Franche, R. (1989). A conceptual and empirical review of
correlates, and cultural boundaries. Journal of Personality and Social the depressive realism hypothesis. Canadian Journal of Behavioral
Psychology, 70, 141-156. Science, 21, 419-433.
Carson, A. D., & Mowsesian, R. (1993). Self-monitoring and private Duval, S., & Wicklund, R. A. (1972). A theory of objective awareness.
self-consciousness: Relations to Holland's vocational personality types. New York: Academic Press.
Journal of Vocational Behavior, 42, 212-222. Dykman, B. M., & Abramson, L. Y. (1989). Processing of ambiguous and
Carver, C. S., & Glass, D. C. (1976). The Self-Consciousness Scale: A unambiguous feedback by depressed and nondepressed college students:
discriminant validity study. Journal of Personality Assessment, 40, 169— Schematic biases and their implications for depressive realism. Journal
172. of Personality and Social Psychology, 56, 431-445.
Carver, C. S., & Scheier, M. F. (1981a). Attention and self-regulation: A Eysenck, H. J., & Eysenck, M. W. (1985). Personality and individual
control-theory approach to human behavior. New York: Springer- differences: A natural science approach. New York: Plenum.
Verlag. Farber, B. A. (1989). Psychological mindedness: Can there be too much of
Carver, C. S., & Scheier, M. F. (1981b). Self-consciousness and reactance. a good thing? Psychotherapy, 26, 210-216.
Journal of Research in Personality, 15, 16-29. Fenigstein, A. (1987). On the nature of public and private self-
Carver, C. S., & Scheier, M. F. (1985). Aspects of self, and the control of consciousness. Journal of Personality, 55, 543-553.
behavior. In B. R. Schlenker (Ed.), The self and social life (pp. 146- Fenigstein, A., Scheier, M. F., & Buss, A. (1975). Public and private
174). New York: McGraw-Hill. self-consciousness. Journal of Consulting and Clinical Psychology, 43,
Cheek, J. M. (1989). Aspects of identity. In D. Buss & N. Cantor (Eds.), 522-527.
Personality psychology: Recent trends and emerging directions (pp. Flett, G. L., & Blankstein, K. R. (1987). Self-focused attention in test
275-285). New York: Springer-Verlag. anxiety and depression. Journal of Social Behavior and Personality, 2,
Cheek, J. M., & Briggs, S. R. (1982). Self-consciousness and aspects of 259-266.
identity. Journal of Research in Personality, 16, 401-408. Franzoi, S. L. (1983). Self-concept differences as a function of private
Cheek, J. M., & Briggs, S. R. (1988). Shyness as a personality trait. In self-consciousness and social anxiety. Journal of Research in Person-
W. R. Crozier (Ed.), Shyness and embarrassment: Perspectives in social ality, 17, 275-287.
psychology (pp. 315-337). Cambridge, England: Cambridge University Franzoi, S. L., Davis, M. H., & Markweisse, B. (1990). A motivational
Press. explanation for the existence of private self-consciousness differences.
Colvin, C. R., Block, J., & Funder, J. (1995). Overly positive self- Journal of Personality, 58, 641-659.
evaluations and personality: Negative implications for mental health. Froming, W. J., & Carver, C. S. (1981). Divergent influences of private and
Journal of Personality and Social Psychology, 68, 1152-1162. public self-consciousness in a compliance paradigm. Journal of Re-
Conte, H. R., Plutchik, R., Jung, B. B., Picard, S., Karasu, T. B., & search in Personality, 15, 159—171.
Lotterman, A. (1990). Psychological mindedness as a predictor of psy- Frone, M. R., & McFarlin, D. B. (1989). Chronic occupational stressors,
chotherapy outcome: A preliminary report. Comprehensive Psychia- self-focused attention, and well-being: Testing a cybernetic model of
try, 31, 426-431. stress. Journal of Applied Psychology, 74, 876-883.
Gibbons, F. X. (1983). Self-focused attention and self-report validity: The Wyer (Ed.), Ruminative thoughts: Advances in social cognition (Vol. 9,
"veridicality" hypothesis. Journal of Personality, 51, 517-542. pp. 1-47). Hillsdale, NJ: Erlbaum.
Gibbons, F. X. (1990). Self-attention and behavior: A review and theoret- Maslow, A. (1955). Deficiency motivation and growth motivation. In
ical update. In M. P. Zanna (Ed.), Advances in experimental social M. R. Jones (Ed.), Nebraska Symposium on Motivation. Lincoln: Uni-
psychology (Vol. 23, pp. 249-295). New York: Academic Press. versity of Nebraska Press.
Gohm, C. L., Isbell, L. M., & Wyer, R. S., Jr. (1997). Some thoughts about Mathews, A. (1990). Why worry? The cognitive function of anxiety.
thinking. In R. S. Wyer (Ed.), Ruminative thoughts: Advances in social Behavior Research and Therapy, 28, 455-468.
cognition (Vol. 9, pp. 81-95). Hillsdale, NJ: Erlbaum. Matthews, G., & Wells, A. (1988). Relationships between anxiety, self
Goldberg, L. R. (1971). A historical survey of personality scales and consciousness, and cognitive failure. Cognition and Emotion, 2, 123—
inventories. In P. McReynolds (Ed.), Advances of psychological assess- 132.
ment (Vol. 2, pp. 293-336). Palo Alto, CA: Science and Behavior McAdams, D. P. (1992). The five-factor model in personality: A critical
Books. appraisal. Journal of Personality, 60, 329-361.
Goldberg, L. R. (1982). From Ace to Zombie: Some explorations in the McCrae, R. R. (1990). Traits and trait names: How well is openness
language of personality. In C. D. Spielberger & J. N. Butcher (Eds.), represented in natural languages? European Journal of Personality, 4,
Advances in personality assessment (Vol. 1, pp. 203-234). Hillsdale, NJ: 119-129.
Erlbaum. McCrae, R. R. (1993). Moderated analyses of longitudinal personality
Goldberg, L. R. (1990). An alternative "description of personality:" The stability. Journal of Personality and Social Psychology, 65, 577-585.
Big Five factor structure. Journal of Personality and Social Psychol- McCrae, R. R. (1993-1994). Openness to Experience as a basic dimension
ogy, 59, 1216-1229. of personality. Imagination, Cognition, and Personality, 13, 39-55.
Goldberg, L. R. (1992). The development of markers for the Big Five McCrae, R. R. (1994). Openness to Experience: Expanding the boundaries
factor structure. Psychological Assessment, 4, 26-42. of Factor V. European Journal of Personality, 8, 251-272.
Goldberg, L. R. (1993). The structure of phenotypic personality traits. McCrae, R. R. (1996a). Integrating the levels of personality. Psychological
American Psychologist, 48, 26-34. Inquiry, 7, 353-356.
Greenberg, J., & Pyszczynski, T. (1986). Persistent high self-focus after McCrae, R. R. (1996b). Social consequences of experiential openness.
failure and low self-focus after success. The depressive self-focusing Psychological Bulletin, 120, 323-337.
style. Journal of Personality and Social Psychology, 50, 1039-1044. McCrae, R. R., & Costa, P. T., Jr. (1985). Openness to Experience. In R.
Hogan & W. H. Jones (Eds.), Perspectives in personality (Vol. 1, pp.
Guilford, J. P. (1975). Factors and factors of personality. Psychological
145-172). Greenwich, CT: JAI Press.
Bulletin, 82, 802-814.
McCrae, R. R., & Costa, P. T., Jr. (1989). The structure of interpersonal
Haaga, D. A. F., & Beck, A. T. (1995). Perspectives on depressive realism:
traits: Wiggins' circumplex and the five factor model. Journal of Per-
Implications for cognitive theory of depression. Behavior Research and
sonality and Social Psychology, 56, 586-595.
Therapy, 33, 41-48.
McCrae, R. R., & Costa, P. T., Jr. (1997a). Conceptions and correlates of
Hall, J. A. (1992). Psychological-mindedness: A conceptual model. Amer-
Openness to Experience. In R. Hogan, J. A. Johnson, & S. R. Briggs
ican Journal of Psychotherapy, 46, 131-140.
(Eds.), Handbook of personality psychology (pp. 825-847). Orlando,
Hogan, R., & Cheek, J. M. (1982). Identity, authenticity, and maturity. In
FL: Academic Press.
T. R. Sarbin & K. E. Scheibe (Eds.), Studies in social identity (pp.
McCrae, R. R., & Costa, P. T., Jr. (1997b). Personality trait structure as a
339-357). New York: Praeger.
human universal. American Psychologist, 52, 509-516.
Hope, D. A., & Heimberg, R. G. (1988). Public and private self conscious-
McCrae, R. R., & John, O. P. (1992). An introduction to the five-factor
ness and social phobia. Journal of Personality Assessment, 52, 626-639.
model and its applications. Journal of Personality, 60, 175—215.
Horowitz, M. J. (1976). Stress response syndromes. New York: Aronson.
Mittal, B., & Balasubramanian, S. (1987). Testing the dimensionality of the
Ingram, R. (1989). Affective confounds in social-cognitive research. Jour- Self-Consciousness Scales. Journal of Personality Assessment, 51, 5 3 -
nal of Personality and Social Psychology, 57, 715-722.
68.
Ingram, R. (1990). Self-focused attention in clinical disorders: Review and Monfries, M. M., & Kafer, N. F. (1994). Private self-consciousness and
a conceptual model. Psychological Bulletin, 107, 156-176. fear of negative evaluation. Journal of Psychology, 128, 447-454.
Ingram, R. E., & Smith, T. W. (1984). Depression and internal versus Mullen, B., & Suls, J. (1982). "Know thyself:" Stressful life changes and
external focus of attention. Cognitive Therapy and Research, 8, 139- the ameliorative effect of private self-consciousness. Journal of Exper-
152. imental Social Psychology, 18, 43-55.
Jackson, D. N. (1984). Personality Research Form manual (3rd ed.). Port Murray, H. A. (1938). Explorations in personality. New York: Oxford
Huron, MI: Research Psychologists Press. University Press.
John, O. (1990). The Big Five factor taxonomy: Dimensions of personality Nasby, W. (1985). Private self-consciousness, articulation of the self-
in the natural language and in questionnaires. In L. A. Pervin (Ed.), schema, and recognition memory of trait adjectives. Journal of Person-
Handbook of personality, theory and research (pp. 66-100). New York: ality and Social Psychology, 49, 704-709.
Guilford Press. Nasby, W. (1989). Private and public self-consciousness and articulation of
Larson, R. J., & Cowan, G. S. (1988). Internal focus of attention and the self-schema. Journal of Personality and Social Psychology, 56,
depression: A study of daily experience. Motivation and Emotion, 12, 117-123.
237-249. Nolen-Hoeksema, S. (1991). Responses to depression and their effects on
Lavallee, L. F., & Campbell, J. D. (1995). Impact of personal goals on the duration of depressive episodes. Journal of Abnormal Psychology,
self-regulation processes elicited by daily negative events. Journal of 100, 569-582.
Personality and Social Psychology, 69, 341-452. Nolen-Hoeksema, S., McBride, A., & Larson, J. (1997). Rumination and
Marshall,.G. N., Wortman, C. B., Vickers, R. R., Jr., Kusulas, J. W., & psychological distress among bereaved partners. Journal of Personality
Hervig, L. K. (1994). The five-factor model of personality as a frame- and Social Psychology, 72, 855-862.
work for personality-health research. Journal of Personality and Social Nolen-Hoeksema, S., Morrow, J., & Fredrickson, B. L. (1993). Response
Psychology, 67, 278-286. styles and the duration of episodes of depressed mood. Journal of
Martin, L. L., & Tesser, A. (1996). Some ruminative thoughts. In R. S. Abnormal Psychology, 102, 20-28.
Norman, W. T. (1967). 2,800 Personality trait descriptors: Normative In H. G. Voss & H. Keller (Eds.), Neugierforschung (pp. 197-225).
operating characteristics for a university population. Ann Arbor, MI: Weinheim, Germany: Beltz.
Department of Psychology, University of Michigan. Srull, T. K., & Gaelick, L. (1983). General principles and individual
Paulhus, D. L., & Levitt, K. (1983). A short version of the Repression- differences in the self as a habitual reference point: An examination of
Sensitization Scale: The Mini-RS. Unpublished manuscript, University self-other judgments of similarity. Social Cognition, 2, 108-121.
of British Columbia. Sulloway, F. J. (1996). Born to rebel: Birth order, family dynamics, and
Paulhus, D. L., & Trapnell, P. D. (1998). Typological measures of shyness. creative lives. New York: Pantheon Books.
Journal of Research in Personality, 32, 182-201. Taylor, S. E., & Brown, J. D. (1988). Illusion and well-being: A social
Piliavin, J., & Charng, H. (1988). What is the factorial structure of the psychological perspective on mental health. Psychological Bulletin, 103,
Private and Public Self-Consciousness Scales? Personality and Social 193-210.
Psychology Bulletin, 14, 587-595. Tellegen, A. (1994). Manual for the Multidimensional Personality Ques-
Pinker, S. ,{1997). How the mind works. New York: W. W. Norton. tionnaire. Minneapolis: University of Minnesota Press.
Porterfield, A. L., Mayer, F. S., Dougherty, K. G., Kredich, K. E., Kron- Tellegen, A., & Atkinson, G. (1974). Openness to absorbing and self-
berg, M. M., Marsee, K. M., & Okazaki, Y. (1988). Journal of Research altering experiences ("absorption"), a trait related to hypnotic suscepti-
in Personality, 22, 409-424. bility. Journal of Abnormal Psychology, 83, 268-277.
Preece, M., DeLongis, A., Campbell, J. D., & Trapnell, P. D. (1998, May). Tokar, D. M., Fischer, A. R., & Subich, L. M. (1998). Personality and
Husbands' reflection buffers the long-term effects of wives' rumination vocational behavior: A selective review of the literature, 1993-1997.
on depressed mood. Poster presented at the annual meeting of the Journal of Vocational Behavior, 53, 115-153.
American Psychological Society, Washington, DC. Trapnell, P. D. (1994). Openness vs. intellect: A lexical left turn. European
Prochasta, J. O., DiClemente, C. C , & Norcross, J. C. (1992). In search of Journal of Personality, 8, 273-290.
how people change: Applications to addictive behavior. American Psy- Trapnell, P. D., & Meston, C. M. (1996, July). Sex and the five factor
chologist, 47, 1102-1114. model of personality: Nice guys finish last. Poster presented at the 104th
Rachman, S., & Hodgson, R. (1980). Obsessions and compulsions. Engle- Annual Convention of the American Psychological Association, To-
wood Cliffs, NJ: Prentice Hall. ronto, Ontario, Canada.
Raskin, R. N., & Hall, C. S. (1979). A narcissistic personality inventory. Trapnell, P. D., Meston, C. M., & Gorzalka, B. (1997). Spectatoring and
Psychological Reports, 45, 590. the relationship between body image and sexual experience: Self-focus
Realo, A., & Allik, J. (1998). The Estonian Self-Consciousness Scale and or self-valence? Journal of Sex Research, 34, 267-278.
its relation to the five-factor model of personality. Journal of Personality Trapnell, P. D., & Wiggins, J. S. (1990). Extension of the Interpersonal
Assessment, 70, 109-124. Adjective Scales to include the Big Five dimensions of personality.
Reeves, A. L., Watson, P. J., Ramsey, A., & Morris, R. J. (1995). Private Journal of Personality and Social Psychology, 59, 781-790.
self-consciousness factors, need for cognition, and depression. Journal Trull, T. J., & McCrae, R. R. (1994). A five-factor perspective on person-
of Social Behavior and Personality, 10, 431-443. ality disorder research. In P. T. Costa, Jr., & T. A. Widiger (Eds.),
Robins, R. W., & John, O. P. (1997). The quest for self-insight: Theory and Personality disorders and the five-factor model of personality (pp.
research in accuracy and bias in self-perception. In R. Hogan, J. A. 59-71). Washington, DC: American Psychological Association.
Johnson, & S. R. Briggs (Eds.), Handbook of personality psychology Turner, R. G., Scheier, M. F., Carver, C. S., & Ickes, W. (1978). Correlates
(pp. 649-680). Orlando, FL: Academic Press. of self-consciousness. Journal of Personality Assessment, 42, 285-289.
Rosenberg, M. (1965). Society and the adolescent self-image. Princeton, Tversky, A. (1977). Features of similarity. Psychological Review, 84,
NJ: Princeton University Press. 327-352.
Russell, P. A. (1973). Relationships between exploratory behavior and fear. Watson, D., & Clark, L. A. (1992). Traits and temperament: General and
British Journal of Psychology, 64, 417-433. specific factors of emotional experience and their relation to the five
Scheier, M. F., & Carver, C. S. (1977). Self-focused attention and the factor model. Journal of Personality, 60, 441-476.
experience of emotion: Attraction, repulsion, and elation. Journal of Watson, D., & Clark, L. A. (1997). Extraversion and its positive emotional
Personality and Social Psychology, 35, 625-636. core. In R. Hogan, J. A. Johnson, & S. R. Briggs (Eds.), Handbook of
Scheier, M. F., & Carver, C. S. (1983). Self-directed attention and the personality psychology (pp. 767-793). Orlando, FL: Academic Press.
comparison of self with standards. Journal of Experimental Social Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and vali-
Psychology, 19, 205-222. dation of brief measures of positive and negative affect: The PANAS
Scheier, M. F., Carver, C. S., & Gibbons, F. X. (1979). Self-directed Scales. Journal of Personality and Social Psychology, 54, 1063—1070.
attention and awareness of bodily states, and suggestibility. Journal of Watson, D., & Friend, R. (1969). Measurement of social-evaluative anxi-
Personality and Social Psychology, 37, 1576-1588. ety. Journal of Consulting and Clinical Psychology, 33, 448-457.
Schlenker, B. R., & Weigold, M. F. (1990). Self-consciousness and self Watson, P. J., & Biderman, M. D. (1993). Narcissistic Personality Inven-
presentation: Being autonomous versus appearing autonomous. Journal tory factors, splitting, and self-consciousness. Journal of Personality
of Personality and Social Psychology, 59, 820-828. Assessment, 6, 41-57.
Smith, M. B. (1961). Mental health reconsidered: A special case of the Watson, P. J., Morris, R. J., & Hood, R. W., Jr. (1988). Sin and self-
problem of values in psychology. American Psychologist, 16, 299-306. functioning, Part 1: Grace, guilt, and self-consciousness. Journal of
Smith, T. W., & Greenberg, J. (1981). Depression and self-focused atten- Psychological Theology, 16, 254-269.
tion. Motivation and Emotion, 5, 323-331. Wells, A. (1991). Effects of dispositional self-focus, appraisal, and atten-
Smith, T. W., & Williams, P. G. (1992). Personality and health: Advan- tion instructions on responses to a threatening stimulus. Anxiety Re-
tages and limitations of the five-factor model. Journal of Personality, 60, search, 3, 291-301.
395-423. Wicklund, R. A. (1975). Objective self-awareness. In L. Berkowitz (Ed.),
Snyder, C. R. (1974). Self monitoring of expressive behavior. Journal of Advances in experimental social psychology (Vol. 8, pp. 233-275). New
Personality and Social Psychology, 30, 526-537. York: Academic Press.
Snyder, C. R., & Fromkin, H. L. (1980). Uniqueness: The human pursuit Wicklund, R. A., & Gollwitzer, P. M. (1987). The fallacy of the private-
of difference. New York: Plenum. public self-focus distinction. Journal of Personality, 55, 491-523.
Spielberger, C. D., Peters, R., & Frain, F. J. (1981). Anxiety and curiosity. Wiggins, J. S. (1979). A psychological taxonomy of trait descriptive terms:
The interpersonal domain. Journal of Personality and Social Psychol- Wyer, R. S. (Ed.). (1996). Ruminative thoughts: Advances in social cog-
ogy, 37, 395-412. nition (Vol. 9). Hillsdale, NJ: Erlbaum.
Wiggins, J. S. (1994). The Interpersonal Adjective Scales (IAS) manual.
Odessa, FL: Psychological Assessment Resources.
Wiggins, J. S., & Trapnell, P. D. (1997). Personality structure: The return
of the Big Five. In R. Hogan, J. A. Johnson, & S. R. Briggs (Eds.), Received June 10, 1997
Handbook of personality psychology (pp. 737-765). Orlando, FL: Aca- Revision received August 3, 1998
demic Press. Accepted August 27, 1998 •

Trapnell 1999

Uploaded by

Copyright:

Available Formats

Trapnell 1999

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Trapnell 1999

Uploaded by

Copyright:

Available Formats

Journal of Personality and Social Psychology Copyright 1999 by the American Psychological Association, Inc.

1999, Vol. 76, No. 2, 284-304 0022-3514/99/S3.00

Private Self-Consciousness and the Five-Factor Model of Personality:

Paul D. Trapnell Jennifer D. Campbell

Big Five instrument

IASR-B5,a NEO-FFI, NEO-PIR,

Criterion = Social Anxiety

Combined sample Women Men

Reflection .91 .48 3.14 0.76 3.17 0.76 3.09 0.76

Table 6 to ideas, absorption, sentience, right-wing authoritarianism, and

Zero-order correlation PrSC controlling for

PrSC Correlate Sample N Rum Ref PrSC Rum Ref

Study 4 bles Murray's (1938) "need for understanding," which in turn is

ANOVA cell means and effects Ref Open PrSC SR ISA

You might also like