Wesleyan University

From the SelectedWorks of Ruth Striegel Weissman

May, 1993

The Social Self in Bulimia Nervosa: Public Self-Consciousness, Social Anxiety, and PerceivedFraudulenceRuth H. Striegel-MooreLisa R. SilbersteinJudith Rodin

Available at: https://works.bepress.com/ruth_striegel/98/

Journal of Abnormal Psychology1993, Vol. 102, No. 2, 297-303

Copvrieht 1993 by the American Psychological Association, Inc.0021-843X/93/S3.00

The Social Self in Bulimia Nervosa: Public Self-Consciousness,Social Anxiety, and Perceived Fraudulence

Ruth H. Striegel-Moore, Lisa R. Silberstein, and Judith Rodin

Bulimic women appear preoccupied not only with their physical presentation but also with their"social self—how others perceive them in general. This study examined the relationship of thesocial self to body esteem and to bulimia nervosa. In Phase 1, in which 222 nonclinical women(aged 16 to 50) participated, the social-self measures of Perceived Fraudulence, Social Anxiety, andPublic Self-Consciousness were negatively associated with body esteem. In Phase 2, 34 bulimicwomen were compared with 33 Ss scoring high on the Eating Attitudes Test (EAT) and 67 matchedcontrol. Bulimic Ss, high-EAT Ss, and control Ss all differed on Perceived Fraudulence, and bulimicSs and high-EAT Ss scored higher than control Ss on Public Self-Consciousness and Social Anxiety.The findings strongly support the hypothesized link of social-self concerns to body dissatisfactionand bulimia nervosa.

A central symptom of eating disorders is preoccupation withappearance—a constant concern with how the physical self isviewed by others. Beneath this manifest symptom seems to lie apervasive concern with how others view the self in general.Eating disordered patients appear preoccupied with self-presen-tation and with how others perceive and evaluate them, a facetof the self first designated the "social self" by William James(1890/1983).'

Research suggests that women with bulimia nervosa and an-orexia nervosa experience significant difficulties in the area ofsocial adjustment, characterized by social anxiety, impover-ished relationships, and social isolation (Fairburn et al., 1990;Grissett & Norvell, 1992; Herzog, Pepose, Norman, & Rigotti,1985; Johnson & Berndt, 1983; Mizes, 1988; Norman & Herzog,1984; Rybicki, Lepkowsky, & Arndt, 1989; Tobin, Johnson,Steinberg, Staats, & Enright, 1991; Yager, Landsverk, & Edel-stein, 1987). The social disturbances do not seem to relate sim-ply to eating pathology: Even after the behavioral symptoms ofeating disorders remit, social maladjustment has been found topersist for many patients (Casper, 1990; Herzog, Keller, Lavori,Bradburn, & Ott, 1990; Mallik, Whipple, & Huerta, 1987; Nor-man & Herzog, 1986; Norman, Herzog, & Chauncey, 1986;Stonehill & Crisp, 1977; Yager et al., 1987). Related researchsuggests that disordered eating is linked to deficits in socialself-confidence and to preoccupation with self-presentation

Ruth H. Striegel-Moore, Department of Psychology, Wesleyan Uni-versity; Lisa R. Silberstein and Judith Rodin, Yale University.

We contributed equally to this study. This work was funded by agrant from the John D. and Catherine T. MacArthur Foundation Net-work on Health and Behavior.

We gratefully acknowledge the support of G. Terence Wilson, Timo-thy Walsh, and Kathleen Pike, who generously provided some of thepatients for thisstudy, and we thank Kelly Brownell for his feedback onan earlier draft of this article.

Correspondence concerning this article should be addressed toRuth H. Striegel-Moore, Department of Psychology, Wesleyan Univer-sity, Middletown, Connecticut 06459-0408.

(Blanchard & Frost, 1983; Gross & Rosen, 1988; McCoulay,Mintz, & Glenn, 1988; Mintz & Betz, 1986).

How might these difficulties in the social self be related tobody dissatisfaction and eating disorder? The body is a "socialobject," constantly accessible to the gaze of others, and women'sbodies are scrutinized by men and women alike with an atten-tiveness far surpassing that given to men's bodies (Silberstein,Striegel-Moore, & Rodin, 1987). If a woman feels inadequate inher physical appearance, then social anxiety seems a probableresult. Alternatively, if she feels insecure about herself in gen-eral, her worries about how others perceive her are likely toinclude, and perhaps become focused on, their evaluations ofher body. Hence, we would predict a relationship between bodyimage dissatisfaction and social anxiety.

Clinically, we have observed that eating disordered women,through their amplified attention to physical appearance, oftenseem to be struggling to construct an adequate social self. Theyseem exquisitely tuned to the evaluations others make of themand the expectations others hold for them, both in terms ofphysical appearance and at a far broader behavioral and psycho-logical level. As Johnson and Connors (1987) and Jones (1985)have summarized, these women seem to suffer from an overde-veloped "false self (Winnicott, 1965). In the absence of a "trueself," they become hypervigilant about public face and socialattitude. Although bulimic women may be successful at creat-ing an acceptable facade for the world, their concerns withfulfilling others' expectations at the expense of acknowledgingtheir own needs prevent them from developing a stable self-de-finition and leave them feeling fraudulent. According to clini-cal theorists, the false self originates in the early failure of signif-icant others to provide adequately an accurate, empathic re-

1 William James (1890/1983) proposed two additional componentsof the self: the spiritual self, reflecting the private, internal aspects ofones' experience, and the material self, involving material propertyand body and family (Baumgardner, Kaufman, & Cranford, 1990;Lamphere & Leary, 1990).

297

This

doc

umen

t is c

opyr

ight

ed b

y th

e A

mer

ican

Psy

chol

ogic

al A

ssoc

iatio

n or

one

of i

ts a

llied

pub

lishe

rs.

This

arti

cle

is in

tend

ed so

lely

for t

he p

erso

nal u

se o

f the

indi

vidu

al u

ser a

nd is

not

to b

e di

ssem

inat

ed b

road

ly.

298 R. STRIEGEL-MOORE, L. SILBERSTEIN, AND J. RODIN

sponse that would enable development of a secure sense of self(Bruch, 1985; Geist, 1989; Strober, 1990). The eating disorderedwoman's extreme efforts to achieve a socially desirable appear-ance represent her attempt to be accepted in the face of thissocial invalidation (Button, 1983; Chiodo, 1987). Hence,heightened public self-consciousness both results from andserves to perpetuate the sense of false self.

A small amount of research provides initial support for thehypothesized associations between these developmentallybased deficits in the social self and disturbed eating, althoughresearch has not yet used a clinical population. Utilizing theBell Object Relations Inventory, two studies of college womenfound relationships between disordered eating and the InsecureAttachment subscale, which purportedly reflects ambivalentand painful interpersonal relationships and fear of object loss(Becker, Bell, & Billington, 1987; Heesacker & Neimeyer,1990). Also in a college sample, Friedlander and Siegel (1990)found a relationship between a measure of perceived self-otherdifferentiation and disordered eating. However, research hasnot yet examined the basic clinical assumption that eating dis-ordered women are aware, to a heightened degree, of their self-presentation or the specific hypothesis that they experience asense of false self. We selected two measures that could enableus to examine the relationship of eating disorders to these twodimensions of the social self. The Self-Consciousness Scale(Fenigstein, Scheier, & Buss, 1975) assesses the concern withself-presentation and with others' view of the self. The Per-ceived Fraudulence Scale (Kolligan & Sternberg, 1991) mea-sures the degree to which the individual experiences the self asfalse, as competent on the exterior but inadequate on the inte-rior.

An initial question, addressed in Phase 1 of our research, ishow perceived fraudulence and public self-consciousness arerelated to body dissatisfaction in a nonclinical population. Al-though there is an extensive clinical literature concerning therole of the social self in bulimia nervosa, the empirical litera-ture on this issue has largely relied on nonclinical populations.Therefore, for Phase 2 we recruited a sample of bulimic patientsand compared them with control subjects and to a nonpatientsample of women scoring high on the EAT-26, a measure ofeating disorder symptoms (Garner, Olmsted, Bohr, & Garfin-kel, 1982). Inclusion of these groups enabled us to examine theextent to which social-self difficulties are evident in the clinicalsyndrome of bulimia nervosa, as compared with a nonclinicallevel of disregulated eating.

Phase 1

Method

Subjects

For Phase 1, a sample of 222 women between the ages of 16 and 50(M = 22.53, SD = 6.0) was recruited from two sources. Female studentsenrolled in introductory psychology courses at two private universitieswere recruited to participate in the study to fulfill a course require-ment. To overcome the pervasive problem in the body image and eat-ing disorders research field of limiting research efforts to college under-graduates, an additional group of women (n - 83) was recruited fromthe local community through a newspaper advertisement for paid re-

search subjects for studies on women's health. The majority of subjectswere Caucasian (84%). All subjects provided their height and weight,from which a body mass index (BM1) was computed. The mean BMIfor Phase 1 subjects was 22.22 (SD = 3.87).

Procedure

All subjects completed the measures in the order listed below.Perceived Fraudulence Scale. To assess the experience of a false

self, the Perceived Fraudulence Scale (PFS; Kolligian & Sternberg,1991) was administered. The PFS purports to assess the degree towhich one perceives oneself as an impostor or a phony. Subjects ratedeach of the 51 items (e.g., "I sometimes feel there's something false ormisleading about me that others don't notice"; "My private feelingsabout, and perception of, myself sometimes conflict with the impres-sions I give others through my public actions or behaviors"; and "Insome situations I feel like an impostor") on a 7-point scale rangingfrom 1 (strongly disagree) to 7 (strongly agree). The developers of thescale reported internal reliability of .90 and found perceived fraudu-lence to be substantially discriminable from depression, social-evalua-tive anxiety, and self-monitoring.

Self-Consciousness Scale. The Self-Consciousness Scale (SCS; Fen-igstein et al., 1975) was administered to measure how the subjects expe-rienced and reflected on the self. The SCS contains 23 items that arerated on a scale ranging from 0 (extremely uncharacteristic of self) to 4(extremely characteristic of self). The scale yields three factors: PublicSelf-Consciousness pertains to an awareness of the reaction of othersto the self (e.g., "I'm concerned about what other people think of me");Social Anxiety is defined as a discomfort in the presence of others (e.g.,"It takes me time to overcome my shyness in public situations"); andPrivate Self-Consciousness concerns attending to one's inner thoughtsand feelings (e.g., "I reflect about myself a lot"). Test-retest reliabilityat 2 weeks ranged from .73 to .84 (Fenigstein et al., 1975). More recentresearch has confirmed that the SCS is reliable (e.g., Mittal & Balasu-bramanian, 1987) and valid (e.g., Nasby, 1989; Schlenker & Weigold,1990).

Body Esteem Scale. Body dissatisfaction was assessed with theBody Esteem Scale (BES; Franzoi & Herzog, 1986; Franzoi & Shields,1984). The BES lists 35 aspects of physical appearance and physicalfunctioning. The BES has been found to correlate significantly with ameasure of body shape dissatisfaction and with self-esteem (Silber-stein, Striegel-Moore, Timko, & Rodin, 1988). Subjects rated satisfac-tion with each item on a 5-point scale ranging from 1 (very dissatisfied)to 5 (very satisfied). An overall body esteem score was calculated bysumming across all items, with higher scores indicating greater bodysatisfaction.

Eating Attitudes Test. The 26-item Eating Attitudes Test (EAT-26;Garner et al., 1982) measures thoughts, feelings, and behaviors asso-ciated with eating disorders such as anorexia nervosa and bulimia ner-vosa. Subjects rated each item on a 6-point scale ranging from never toalways. The three least pathological ratings receive a score of 0, and thethree responses indicating more disturbance are scored from 1 to 3. Anoverall score was obtained by summing across items, with higherscores suggesting more disordered eating. The scale developers reporthigh internal consistency (r = .90; Garner et al., 1982), and subsequentresearch has found the EAT-26 highly accurate in classifying eatingdisordered and non-eating disordered individuals (Gross, Rosen, Lei-tenberg, & Willmuth, 1986; Williams, Schaefer, Shisslak, Gronwaldt,&Comerci, 1986).

Symptom Distress Checklist-90R. The Symptom Checklist-90, Re-vised (SCL-90R; Derogatis, 1983) is a 90-item self-report inventorydesigned to measure a wide range of psychological symptoms. Exten-sive reliability and validity data have been published (Derogatis, 1983).On a 5-point scale ranging from 0 (not at all) to 4 (extremely), subjects

This

doc

umen

t is c

opyr

ight

ed b

y th

e A

mer

ican

Psy

chol

ogic

al A

ssoc

iatio

n or

one

of i

ts a

llied

pub

lishe

rs.

This

arti

cle

is in

tend

ed so

lely

for t

he p

erso

nal u

se o

f the

indi

vidu

al u

ser a

nd is

not

to b

e di

ssem

inat

ed b

road

ly.

SOCIAL SELF IN BULIMIA NERVOSA 299

rated the degree of discomfort they experienced during the past weekon each symptom. A Global Severity Index was derived by summingacross all items and dividing the sum by the total number of items.

Results

Phase 1 data analyses examined the relationship between thesocial-self variables and body dissatisfaction. Comparisons ofthe student sample and the community sample on all measuresof inquiry yielded no statistically significant differences, usingthe Bonferroni correction for multiple t tests. In our analyses,therefore, we considered the sample as a whole. Table 1 displaysmean scores for all measures utilized for Phase 1 analyses.2

We predicted an association of the social-self variables tobody dissatisfaction. As shown in Table 2, Perceived Fraudu-lence, Social Anxiety, and Public Self-Consciousness were nega-tively associated with body esteem, whereas Private Self-Con-sciousness was not significantly related to body esteem. To ex-amine the relationships between the social-self variables andbody esteem further, we calculated a stepwise regression. Be-cause of the significant negative association between BMI andbody esteem (r = -.27, p < .001), BMI was forced into theregression before entering the social-self variables as predictorsof the body esteem score. Controlling for BMI (b= -\ .21, SE =0.31, F= 16.42, p < .001, R2 = 0.07), Perceived Fraudulence(b = -.016, SE= 0.03, F= 61.72, p < .001, partial R2 =0.21)and Social Anxiety (b = -0.64, SE = 0.22, F= 11.37, p < .01,partial R2 = 0.04) emerged as significant predictors of bodyesteem (Final model: F= 32.75, p < .001, R2 = 0.33).

These findings provide strong support for the predicted asso-ciation of social-self concerns and body esteem in a nonclinicalpopulation. Phase 2 was then undertaken to ascertain how thesocial-self variables relate to the clinical syndrome of bulimianervosa.

Phase 2

Table 1Mean Scores for Phase 1 Subjects on Measures of Social SelfBody Esteem, and Eating Concern

Method

Subjects

Thirty-four patients were recruited as they presented for treatmentat two eating disorder clinics. Diagnosis of bulimia nervosa accordingto Diagnostic and Statistical Manual'of'Mental Disorders(Kv. 3rded., orDSM-HI-R; American Psychiatric Association, 1987) criteria was es-tablished by a trained clinician on the basis of an intake interview. Bydefinition, all patients engaged in binge eating on average at least twicea week during the past 3 months. Mean frequency of binge eatingepisodes per week was 7.46 (SD = 5.57). A majority of patients (76%)reported purging by means of vomiting, and an additional 12% indi-cated regular use of laxatives as a means of controlling their weight.Only 12% of the patients relied solely on dieting to counteract the ef-fects of binge eating. Of patients who vomited, 65% vomited at leastonce daily, and an additional 32% vomited at least twice a week. Laxa-tive abuse occurred less regularly: Only 15% of laxative abusers re-ported taking laxatives on a daily basis, and an additional 15% usedlaxatives at least twice a week. Mean duration of bulimia nervosa was6.80 years (SD = 4.15).

To compare the patient group with women evidencing elevated lev-els of disturbed eating, a group of Phase 1 subjects was identified whoscored above the clinical cutoff of 20 on the EAT (n = 38; "high-EAT"subjects). A group of matched control subjects was generated by select-

Measure

FraudulencePublic Self-ConsciousnessSocial AnxietyPrivate Self-ConsciousnessBody Esteem ScaleEating Attitudes Test-26

M(n = 222)

187.3526.1817.5337.41

114.2810.61

SD

42.435.305.286.74

18.309.60

ing Phase 1 subjects who scored below 20 on the EAT and who matchedthe bulimic patients or the high-EAT subjects on the criteria of race,age, and BMI. Thirty-four matched controls were found for the pa-tients. We were unable to find matches for 5 of the 38 high-EAT sub-jects owing to these 5 women's high body weights. These 5 women weretherefore dropped from the high-EAT sample. The final Phase 2 sam-ple thus included 134 women: 34 bulimic patients, 33 high-EAT sub-jects, and 67 matched controls. The majority of subjects (93%) wereCaucasian. The mean age of Phase 2 subjects was 23.36 years (SD =5.80 years).

Results

Mean scores on all measures for Phase 2 subjects are dis-played in Table 3. A multivariate analysis of variance compar-ing the three groups of subjects on the subscales pertaining toSelf-Consciousness found a significant overall group differ-ence, F(2,131) = 5.71, p < .001.

For between-group comparisons, the Bonferroni correction(p = .05/12) was applied, resulting in a significance level of.004. Planned contrasts comparing patients and controls weresignificant for Public Self-Consciousness, F(l, 93) = 15.35, p <.001, and for Social Anxiety, F(l, 99) = 12.58, p < .001, but notfor Private Self-Consciousness, F(l, 99) = 0.39, p = .39. Simi-larly, planned contrasts comparing high-EAT subjects and con-trol subjects found significant group differences on Public Self-Consciousness, F(\, 99) = 19.91, p < .001, and on Social Anxi-ety, F(l, 98) = 9.43, p < .001, but not on PrivateSelf-Consciousness, F(l, 98) = 5.28, p = .05. None of theplanned contrasts comparing patients and high-EAT subjectson the Self-Consciousness subscales was significant. An analy-sis of variance (ANOVA) yielded significant group differenceson Perceived Fraudulence, F(2,131) = 24.1, p < .001. Plannedpost hoc comparisons found all three groups differed signifi-cantly, whereby patients had significantly higher Perceived

2 Our community sample's mean score on Perceived Fraudulence isslightly lower than the mean for a college sample reported by Kolliganand Sternberg (1991; M= 199.5, SD = 52.30). A recent study of adultpatients and controls by Belfer and Glass (1992) reported mean scorescomparable to ours for Public Self-Consciousness (M = 24.97, SD =3.80), Social Anxiety (M = 16.27, SD = 4.17), and Private Self-Consciousness (M = 34.48, SD = 5.93). Our sample's mean scores onthe EAT-26 and the BES are also similar to those reported by Silber-stein and her colleagues, based on a college student sample (EAT: M =8.24, SD = 7.44; BES: M = 111.58, SD = 14.32; Silberstein et al., 1988).

This

doc

umen

t is c

opyr

ight

ed b

y th

e A

mer

ican

Psy

chol

ogic

al A

ssoc

iatio

n or

one

of i

ts a

llied

pub

lishe

rs.

This

arti

cle

is in

tend

ed so

lely

for t

he p

erso

nal u

se o

f the

indi

vidu

al u

ser a

nd is

not

to b

e di

ssem

inat

ed b

road

ly.

300 R. STRIEGEL-MOORE, L. SILBERSTEIN, AND J. RODIN

Table 2Correlations Between Measures of Social Self and Measures ofBody Esteem and Weight Concern in Phase 1 Subjects

Measure 1

1. BES2. EAT-263. PFS4. Public SCS5. Social Anxiety6. Private SCS

_-.30***-.45***-.23***-.35***-.08

—.15*.41***.15*.15*

—.32***.30***.17**

—.30*** —.43*** .16*

Note. BES = Body Esteem Scale; EAT-26 = Eating Attitudes Test;PFS = Perceived Fraudulence Scale; SCS = Self-Consciousness Scale.*p<.Q5. **p<.01. ***p< .001.

Fraudulence scores than high-EAT subjects who in turn scoredsignificantly higher than control subjects.

Table 3 also points to strong group differences on measuresof disordered eating (EAT-26), F(2,132) = 170.85, p < .0001,body esteem (BES), F(2,132) = 22.67, p < .0001, and generalpsychiatric distress (GSI), F(2,132) = 33.70, p < .0001. On eachof these measures, patients appeared significantly more dis-tressed than high-EAT women, who in turn obtained signifi-cantly higher scores than control subjects.

It seemed possible that the social-self variables differentiatedamong the groups in Phase 2 and were predictive of body es-teem in Phase 1 because they represent some basic level ofpsychiatric distress. Because all subjects had completed theSCL-90R, we were able to consider whether the social-self vari-ables made a specific contribution to body dissatisfactionabove and beyond the extent to which social-self dysfunctionreflects general psychological difficulty. To explore this ques-tion, we used all subjects from both phases. To represent groupmembership, two dummy variables were constructed (bulimiaand high-eat) and contrasted against control subjects as the ref-erence group. A hierarchical regression analysis was per-formed, entering the following variables (in order) to predictbody esteem: bulimia, high-eat, body mass index, global symp-tom index (SCL-90R), perceived fraudulence, and social anxi-ety, and interaction terms of Diagnostic Group X PerceivedFraudulence and Social Anxiety, respectively. None of the inter-actions was significant (data not shown).

After controlling for diagnostic group, BMI, and globalsymptom index,3 (F = 26.98, p < .001, R2 = .31), perceivedfraudulence (b = -0.18, SE = 0.03, partial R2 = 0.12, p < .001)and social anxiety (b = -0.67, SE = 0.27, partial R2 = 0.02, p <.05) remained significant predictors of body esteem (finalmodel: F= 33.17, p < .001, R2 = .45). Adding perceived fraudu-lence and social anxiety to the regression model increased theamount of variance explained from 31% to 45% (p < .01).

Discussion

The findings of this study strongly support the hypothesisthat social-self concerns are integrally linked to body dissatis-faction and eating disorder. In the nonclinical sample used inPhase 1, body dissatisfaction was related to Public Self-Con-sciousness, Social Anxiety, and Perceived Fraudulence, all of

which are dimensions of the social self. In contrast, body es-teem was not associated with Private Self-Consciousness,which is not a component of the social self. Similarly, Blanchardand Frost (1983) found that Public Self-Consciousness and So-cial Anxiety, but not Private Self-Consciousness, were asso-ciated with restrained eating in an undergraduate sample.

Phase 2 results offer empirical support for the clinical theorypostulating that social-self dysfunction plays an important rolein bulimia nervosa. Bulimic patients differed from normalcontrol women on Public Self-Consciousness, Social Anxiety,and Perceived Fraudulence but not on Private Self-Conscious-ness. High-EAT women also differed from normal controls onPublic Self-Consciousness and Social Anxiety, and on Per-ceived Fraudulence but not on Private Self-Consciousness. Ourfindings support Heatherton and Baumeister's (1992) assertionthat binge eaters possess heightened self-awareness of how theyare perceived and evaluated by others, but not of their intrapsy-chic experiences.

Whereas both the bulimic and high-EAT subjects had ele-vated scores on Public Self-Consciousness and Social Anxietycompared with the normal control subjects, the only social-selfmeasure that differentiated all three groups was the PerceivedFraudulence Scale. There may be ceiling effects on the PublicSelf-Consciousness and Social Anxiety scales that prevent de-tection of differences between the two groups with disregulatedeating. Alternatively, the results may suggest that the PerceivedFraudulence Scale addresses more precisely and sensitively anissue linked to the clinical syndrome of bulimia nervosa.

High-EAT women fell between the other two groups on mea-sures of body esteem, disordered eating, perceived fraudulenceand psychiatric distress. These findings support the view thateating disorders lie on a continuum ranging from unconcernwith weight and eating at one end, to weight dissatisfaction andmoderately disregulated eating in the middle, to the clinicalsyndromes of anorexia nervosa and bulimia nervosa at theother end (Rodin, Silberstein, & Striegel-Moore, 1985). High-EAT women may experience subclinical forms of eating dis-orders, or they may be at a preclinical stage of an eating dis-order. Given certain precipitants (such as stressful life events),these women may develop a full clinical syndrome. The factthat high-EAT women share similarities with, yet also differfrom, bulimic patients suggests that high-EAT women are aninteresting group worthy of further research. Careful investiga-tion of this population may offer a helpful look at what movessome women along the continuum of disordered eating to theextreme of a clinical syndrome.

Our analysis using SCL-90R scores of subjects from bothphases enabled us to ascertain that the relationship betweensocial-self deficits and body dissatisfaction is not simply a func-tion of general psychiatric distress: Perceived Fraudulence andSocial Anxiety remained significant predictors of body esteemwhen controlling for psychiatric symptoms. Initial research has

3 Bulimia: b = -24.07, SE= 3.55, T= -6.77, p < .001, partial R2 =0.16; high-EAT: b= -4.43, SE= 3.80, T= -2.89, p= .01, partial R2 =0.02; BMI: b = -1.41, SE = 0.53, T= -3.89, p < .01, partial R2 = 0.05;and global symptom index: b= -10.58, SE= 2.73, T= -5.13, p<. 001,partial R2 = 0.07.

This

doc

umen

t is c

opyr

ight

ed b

y th

e A

mer

ican

Psy

chol

ogic

al A

ssoc

iatio

n or

one

of i

ts a

llied

pub

lishe

rs.

This

arti

cle

is in

tend

ed so

lely

for t

he p

erso

nal u

se o

f the

indi

vidu

al u

ser a

nd is

not

to b

e di

ssem

inat

ed b

road

ly.

SOCIAL SELF IN BULIMIA NERVOSA 301

Table 3Mean Scores for Phase 2 Subjects on All Measures

Patients(» = 34)

High-eat subjects(n = 33)

Variable M SD M SD

Controls(n = 67)

M SD

FraudulencePublic SCSSocial AnxietyPrivate SCSBES totalEAT-26 totalGlobal symptoms indexBody mass index

235.94a29.00a19.71.36.8889.76a35.76a

1.45.21.64

44.865.605.396.98

21.1013.920.812.76

194.21b29.61.19.24a39.45

105.55b28.27b0.98b

22.47

39.063.924.496.33

14.537.390.502.92

176.58.24.97b15.01b35.94

114.70,5.76C0.51C

22.08

39.124.925.007.67

16.904.110.312.87

Note. Column means with different subscripts differ significantly (p < .001). SCS = Self-ConsciousnessScale; BES = Body Esteem Scale; EAT-26 = Eating Attitudes Test.

found that social anxiety is elevated in a variety of psychiatricdisorders suggesting that the Self-Consciousness Scale taps so-cial disturbance common to a variety of psychiatric disorders.A recent study reported significant differences between anxietydisorder patients and nonpsychiatric controls on Social Anxi-ety (Belfer & Glass, 1992). Furthermore, comparing agorapho-bic patients, patients with affective disorder (not further de-nned), anorexic patients, and bulimic patients, Belfer and col-leagues (Belfer, Crump, & Bradach, 1991) found no significantdifferences between the four patient groups on Public Self-Consciousness and on Social Anxiety. Hence, anorexics appearto be as socially anxious and as concerned about how others'perceive them as do bulimic patients. Moreover, the Self-Consciousness Scale does not discriminate between clinicalgroups; rather, it appears to measure a nonspecific pathologicalfeature (see also Ingram, 1990).

A limitation of our study is that it does not include otherclinical populations to examine the extent to which perceivedfraudulence plays a unique role in bulimia nervosa. Bruch's(1978, 1985) clinical writing has suggested that anorexic pa-tients are characterized by self-deficits likely to result in per-ceived fraudulence, yet anorexic patients need to be studied toconfirm Bruch's (1978, 1985) formulation. Interesting addi-tional comparison groups would include patients with personal-ity disorder suggestive of core social-self deficits, such as narcis-sistic personality disorder.

Our study does not permit us to identify the causal directionof the association between perceived fraudulence and bulimianervosa. The experience of a false self may constitute an impor-tant risk factor for bulimia nervosa: On the one hand, as anindividual struggles with a sense of fraudulence and inade-quacy, she focuses on creating a facade for the world, whichtranslates concretely into increased preoccupation with herbody and her eating. On the other hand, the experience of falseself may be a consequence of having an eating disorder, whichmay amplify perceived fraudulence. Binging and purging con-stitute a shameful secret hidden from others and provide aweight control method that feels fraudulent.

Prospective studies are needed to examine the etiologicalsignificance of Perceived Fraudulence, Public Self-Conscious-ness, and Social Anxiety in bulimia nervosa. Furthermore,

treatment outcome research should assess the degree to whichchanges in Perceived Fraudulence, Public Self-Consciousness,and Social Anxiety occur in treatment and how such changesrelate to prognosis. Recent treatment outcome studies havefound interpersonal psychotherapy to be as effective as cogni-tive-behavioral therapy in reducing bulimic symptoms (Fair-burn, Agras, & Wilson, in press; Fairburn et al., 1990; Wilfleyet al., in press), lending support to the view that social-selfconcerns are central to the bulimic syndrome and are vital toaddress in treatment.

Our findings support clinical theory and observation point-ing to the salience in eating disorders of preoccupation withself-presentation and an experience of the self as fraudulent.The impaired social self may contribute to the eating disorderin a variety of ways. A focus on weight and food may constitutean effort to confine the patient's anxiety to one sphere (Fried-lander & Siegel, 1990), as well as to gain social approval (But-ton, 1983). Furthermore, the disorder often keeps the womantied to her parents, thereby reducing her conflicts about depen-dency (Friedlander & Siegel, 1990). At a concrete level, patientsoften state that interpersonal incidents that focus them on theirsocial-self inadequacy constitute specific precipitants to bing-ing and purging. In a laboratory study of binge eaters, stress ofan interpersonal nature was found to be particularly likely toinduce the desire to binge (Cattanach, Malley, & Rodin, 1988).Social self-deficits are also likely to be reinforced and amplifiedas a consequence of the eating disorder, as the eating disorderboth increases the psychological sense of not fitting in andserves to isolate the woman further from the social world.

The findings bear important implications for clinical workwith eating disordered women. Typically, patients who feel likeimpostors have their feelings of insecurity and self-doubt dis-counted by significant others (Glance & Imes, 1978; Glance &O'Toole, 1988). This lack of validation of their feelings contrib-utes further to the sense of isolation and perceived fraudulence.Therapists are challenged to simultaneously affirm the pa-tient's experience of fraudulence and to increase her awarenessof cognitive distortions (Glance & O'Toole, 1988). Further-more, the bulimic patient's presentation of "pseudo self-suffi-ciency" (Goodsitt, 1985) and her hypervigilance about impres-sion management may make it difficult to establish a satis-

This

doc

umen

t is c

opyr

ight

ed b

y th

e A

mer

ican

Psy

chol

ogic

al A

ssoc

iatio

n or

one

of i

ts a

llied

pub

lishe

rs.

This

arti

cle

is in

tend

ed so

lely

for t

he p

erso

nal u

se o

f the

indi

vidu

al u

ser a

nd is

not

to b

e di

ssem

inat

ed b

road

ly.

302 R. STRIEGEL-MOORE, L. SILBERSTEIN, AND J. RODIN

factory therapeutic alliance. Last, the acute public self-consciousness of bulimic patients translates into an overridingconcern with the therapist's perception, and therapists mustremain aware of the patients' chameleonlike abilities to de-cipher and then match the needs and expectations of others,including the therapist.

ReferencesAmerican Psychiatric Association (1987). Diagnostic and statistical

manual of mental disorders (3rd ed., rev.). Washington, DC: Author.Baumgardner, A. H., Kaufman, C. M, & Cranford, J. A. (1990). To be

noticed favorably: Links between private self and public self. Person-ality and Social Psychology Bulletin, 16, 705-716.

Becker, B., Bell, M., & Billington, R. (1987). Object relations ego defi-cits in bulimic college women. Journal of Clinical Psychology, 43,92-95.

Belfer, P. L., Crump, S., & Bradach, K. M. (1991, November). Bodyawareness and self-consciousness in eating and anxiety disorderedindividuals. Paper presented at the 25th Annual Convention of theAssociation for Advancement of Behavior Therapy, New York, NY.

Belfer, P. L., & Glass, C. R. (1992). Agoraphobic anxiety and fear offear: Test of a cognitive-attentional model. Journal of Anxiety Dis-orders, 6, 133-146.

Blanchard, F. A., & Frost, R. O. (1983). Two factors of restraint: Con-cern for dieting and weight fluctuation. Behavioral Research andTherapy, 21, 259-267.

Bruch, H. (1978). The golden cage: The enigma of anorexia nervosa.Cambridge, MA: Harvard University Press.

Bruch, H. (1985). Four decades of eating disorders. In D. M. Garner &P. E. Garfinkel (Eds.), Handbook of psychotherapy for anorexia ner-vosa and bulimia (pp. 7-18). New York: Guilford Press.

Button, E. (1983). Personal construct theory and psychological well-being. British Journal of Medical Psychology, 56, 312-321.

Casper, R. C. (1990). Personality features of women with good out-come from restricting anorexia nervosa. Psychosomatic Medicine,52, 156-170.

Cattanach, L., Malley, R., & Rodin, J. (1988). Psychologic and physio-logic reactivity to stressors in eating disordered individuals. Psycho-somatic Medicine, 50, 591-599.

Chiodo, J. (1987). Bulimia: An individual behavioral analysis. Journalof Behavior Therapy and Experimental Psychiatry, 18, 41-49.

Glance, P. R., & Imes, S. A. (1978). The imposter phenomenon in highachieving women: Dynamics and therapeutic intervention. Psycho-therapy: Theory, Research, and Practice, 15, 241-247.

Glance, P. R., & O'Toole, M. A. (1988). The imposter phenomenon: Aninternal barrier to empowerment and achievement. Women andTherapy, 6, 51-64.

Derogatis, L. R. (1983). SCL-90-R: Administration manual-II for therevised version. Towson, MD: Clinical Psychometric Research.

Fairburn, C. G., Agras, W S., & Wilson, G. T. (in press). The research onthe treatment of bulimia nervosa: Practical and theoretical implica-tions. In G. H. Anderson & S. H. Kennedy (Eds.), The biology offcastand famine: Relevance to eating disorders. New \brk: AcademicPress.

Fairburn, C. G., Jones, R., Peveler, R. C., Carr, S. J., Solomon, R. A.,O'Connor, M. E., Burton, J., & Hope, R. A. (1990). Three psychologi-cal treatments for bulimia nervosa: A comparative trial. Archives ofGeneral Psychiatry, 48, 463-469.

Fenigstein, A., Scheier, M. F, & Buss, A. H. (1975). Public and privateself-consciousness: Assessment and theory. Journal of Consultingand Clinical Psychology, 43, 522-527.

Franzoi, S. L., & Herzog, M. E. (1986). The Body Esteem Scale: Aconvergent and discriminant validity study. Journal of PersonalityAssessment, 50, 24-31.

Franzoi, S. L., & Shields, S. A. (1984). The Body Esteem Scale: Multidi-mensional structure and sex differences in a college population.Journal of Personality Assessment, 48, 173-178.

Friedlander, M. L., & Siegel, S. M. (1990). Separation-individuationdifficulties and cognitive-behavioral indicators of eating disordersamong college women. Journal of Counseling Psychology, 37, 74-78.

Garner, D. M., Olmsted, M. P., Bohr, Y, & Garfinkel, P. E. (1982). TheEating Attitudes Test: Psychometric features and clinical correlates.Psychological Medicine, 12, 871-878.

Geist, R. A. (1989). Self-psychological reflection on the origins of eat-ing disorders. In J. R. Bemporad & D. B. Herzog (Eds.), Psychoanaly-sis and eating disorders (pp. 5-27). New York: American Academy ofPsychoanalysis.

Goodsitt, A. (1985). Self psychology and the treatment of anorexianervosa. In D. M. Garner & P. E. Garfinkel (Eds.), Handbook ofpsychotherapy for anorexia nervosa and bulimia (pp. 55-82). NewYork: Guilford Press.

Grissett, N. I., & Norvell, N. K. (1992). Perceived social support, socialskills, and quality of relationships in bulimic women. Journal ofConsulting and Clinical Psychology, 60, 293-299.

Gross, J., & Rosen, J. C. (1988). Bulimia in adolescents: Prevalence andpsychosocial correlates. International Journal of Eating Disorders, 7,51-61.

Gross, J., Rosen, J. C, Leitenberg, H., & Willmuth, M. E. (1986). Valid-ity of the Eating Attitudes Test and the Eating Disorders Inventoryin bulimia nervosa. Journal of Consulting and Clinical Psychology,54, 875-876.

Heatherton, T. F,& Baumeister, R. F. (1992). Binge eating as an escapefrom self-awareness. Psychological Bulletin, 110, 86-108.

Heesacker, R. S., & Neimeyer, G. J. (1990). Assessing object relationsand social cognitive correlates of eating disorder. Journal of Counsel-ing Psychology, 37, 419-426.

Herzog, D. B., Keller, M. B., Lavori, P. W, Bradburn, 1. S., & Ott, I. L.(1990). Course and outcome of bulimia nervosa. In M. M. Fichter(Ed.), Bulimia nervosa: Basic research, diagnosis, and therapy (pp.126-144). Chichester, England: Wiley.

Herzog, D. B., Pepose, M., Norman, D. K., & Rigotti, N. A. (1985).Eating disorders and social maladjustment in female medical stu-dents. The Journal of Nervous and Mental Disease, 173, 734-737.

Ingram, R. E. (1990). Self-focused attention in clinical disorders: Re-view and a conceptual model. Psychological Bulletin, 107, 156-176.

James, W (1983). The principles of psychology. Cambridge, MA: Har-vard University Press. (Original work published 1890)

Johnson, C, & Berndt, D. J. (1983). Preliminary investigation of bu-limia and life adjustment. American Journal of Psychiatry, 140, 774-777.

Johnson, C., & Connors, M. (1987). The etiology and treatment of bu-limia nervosa: A biopsychosocial perspective. New York: Basic Books.

Jones, D. M. (1985). Bulimia: A false self identity. Clinical Social WorkJournal, 13, 305-316.

Kolligan, J., & Sternberg, R. J. (1991): Perceived fraudulence in youngadults: Is there an "impostor syndrome?" Journal oj Personality As-sessment, 56, 308-326.

Lamphere, R. A., & Leary, M. R. (1990). Private and public self-pro-cesses: A return to James' constituents of the self. Personality andSocial Psychology Bulletin, 16, 717-725.

Mallick, M. J., Whipple, T. W, & Huerta, E. (1987). Behavioral andpsychological traits of weight-conscious teenagers: A comparison ofeating-disordered patients and high- and low-risk groups. Adoles-cence, 22, 157-168.

McCoulay, M., Mintz, L., & Glenn, A. A. (1988). Body image, self-es-teem, and depression-proneness: Closing the gender gap. Sex Roles,75,381-391.

This

doc

umen

t is c

opyr

ight

ed b

y th

e A

mer

ican

Psy

chol

ogic

al A

ssoc

iatio

n or

one

of i

ts a

llied

pub

lishe

rs.

This

arti

cle

is in

tend

ed so

lely

for t

he p

erso

nal u

se o

f the

indi

vidu

al u

ser a

nd is

not

to b

e di

ssem

inat

ed b

road

ly.

SOCIAL SELF IN BULIMIA NERVOSA 303

Mintz, L. B., & Betz, N. E. (1986). Sex differences in the nature, real-ism, and correlates of body image. Sex Roles, 15, 185-195.

Mittal, B., & Balasubramanian, S. K. (1987). Testing the dimensional-ity of the self-consciousness scales. Journal of Personality Assess-ment, 51, 53-68.

Mizes, J. S. (1988). Controlled comparison of bulimic and noneatingdisordered controls on the MMPI-168. International Journal of Eat-ing Disorders, 7, 425-428.

Nasby, W (1989). Private and public self-consciousness and articula-tion of the self-schema. Journal ofPersonality and Social Psychology,56, 117-123.

Norman, D. K., & Herzog, D. B. (1984). Persistent social maladjust-ment in bulimia: A 1-year follow-up. American Journal of Psychiatry,141, 444-446.

Norman, D. K., & Herzog, D. B. (1986). A 3-year outcome study ofnormal-weight bulimia: Assessment of psycho-social functioningand eating attitudes. Psychiatry Research, 19,199-205.

Norman, D. K., Herzog, D. B., & Chauncey, S. (1986). A one-year out-come study of bulimia: Psychological and eating symptom changesin a treatment and non-treatment group. International Journal ofEating Disorders, 5, 47-57.

Rodin, J., Silberstein, L. R., & Striegel-Moore, R. H. (1985). Womenand weight: A normative discontent. In T. B. Sonderegger (Ed.), Ne-braska Symposium on Motivation (pp. 267-308). Lincoln: Universityof Nebraska Press.

Rybicki, D. J., Lepkowsky, C. M., & Arndt, S. (1989). An empiricalassessment of bulimic patients using multiple measures. AddictiveBehaviors, 14, 249-260.

Schlenker, B. R., & Weigold, M. F. (1990). Self-consciousness and self-presentation: Being autonomous versus appearing autonomous.Journal of Personality and Social Psychology, 59, 820-828.

Silberstein, L. R., Striegel-Moore, R. H., & Rodin, J. (1987). Feeling fat:A woman's shame. In H. B. Lewis (Ed.), The role of shame in symptomformation (pp. 89-108). Hillsdale, NJ: Erlbaum.

Silberstein, L. R., Striegel-Moore, R. H., Timko, C., & Rodin, J. (1988).Behavioral and psychological implications of body image dissatis-faction: Do men and women differ? Sex Roles, 19, 219-232.

Stonehill, E., & Crisp, A. H. (1977). Psychoneurotic characteristics ofpatients with anorexia nervosa before and after treatment and atfollow-up 4-7 years later. Journal of Psychosomatic Research, 21,187-193.

Strober, M. (1990). Disorders of the self in anorexia nervosa: An organ-ismic-developmental paradigm. In C. Johnson (Ed.), Psychody-namic theory and treatment for eating disorders (pp. 354-373). NewYork: Guilford Press.

Tobin, D., Johnson, C., Steinberg, S., Staats, M., & Enright, A. B. (1991).Multi factorial assessment of bulimia nervosa. Journal of AbnormalPsychology, 100,14-21.

Wifley, D. E., Agras, W S., Telch, C. F., Rossiter, E. M., Schneider, J. A.,Cole, A. G., Sifford, L., & Raeburn, S. D. (in press). Group cognitive-behavioral treatment and interpersonal psychotherapy for the non-purging bulimic: A controlled comparison. Journal of Consultingand Clinical Psychology.

Williams, R., Schaefer, C., Shisslak, C, Gronwaldt, V, & Comerci, G.(1986). Eating attitudes and behaviors in adolescent women: Dis-crimination of normals, dieters, and suspected bulimics using theEating Attitudes Test and Eating Disorders Inventory. InternationalJournal of Eating Disorders, 5, 879-894.

Winnicot, D. W (1965). The maturational process and the facilitatingenvironment. New York: International Universities Press.

Yager, J., Landsverk, J., & Edelstein, C. K. (1987). A 20-month follow-up study of 628 women with eating disorders, I: Course and severity.American Journal of Psychiatry, 144, 1172-1177.

Received December 9,1991Revision received August 7,1992

Accepted August 7,1992 •

This

doc

umen

t is c

opyr

ight

ed b

y th

e A

mer

ican

Psy

chol

ogic

al A

ssoc

iatio

n or

one

of i

ts a

llied

pub

lishe

rs.

This

arti

cle

is in

tend

ed so

lely

for t

he p

erso

nal u

se o

f the

indi

vidu

al u

ser a

nd is

not

to b

e di

ssem

inat

ed b

road

ly.