USF Libraries
USF Digital Collections

The effects of ambiguous appearance-related feedback on body image, mood states, and intentions to use body change strat...

MISSING IMAGE

Material Information

Title:
The effects of ambiguous appearance-related feedback on body image, mood states, and intentions to use body change strategies in college women
Physical Description:
Book
Language:
English
Creator:
Herbozo, Sylvia
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla
Publication Date:

Subjects

Subjects / Keywords:
Commentary
Weight
Size
Physical
Eating disorders
Dissertations, Academic -- Psychology -- Doctoral -- USF   ( lcsh )
Genre:
non-fiction   ( marcgt )

Notes

Summary:
ABSTRACT: Previous research has demonstrated the influential role of physical appearance-related feedback in the development of body image and eating disturbances. Teasing and negative feedback have been established as strong correlates and predictors of body dissatisfaction, maladaptive eating behaviors, and psychological distress. However, very little is known about ambiguous appearance-related feedback and its impact on others. The current study sought to explore this area with an experimental study to examine the effects of ambiguous appearance-related feedback on body image, mood states, and intentions to use body change strategies. Undergraduate women (N=146) were randomly assigned to an ambiguous appearance-related or ambiguous nonappearance-related feedback condition. Body image, mood states, and intentions to diet, exercise, and use unhealthy weight control methods were assessed before and after feedback was provided by a confederate.Results indicated no significant differences between feedback conditions in body image and mood states. The mean trends for all mood state, with the exception of anger, indicated better mood states after ambiguous appearance-related feedback compared to after ambiguous nonappearance-related feedback. State anger was greater in the ambiguous appearance-related feedback condition suggesting that this particular type of feedback was interpreted in a negative manner. Further, there was a significant difference between feedback conditions for intentions to diet and use bulimic behaviors, with lower levels in the ambiguous appearance-related feedback condition. No significant differences were found for intentions to exercise. State appearance comparison was not shown to mediate the relationship between ambiguous feedback and body image, mood states, or intentions to use body change strategies.Trait appearance satisfaction, appearance comparison, appearance schematicity, and thin ideal internalization were found to moderate the relationship between ambiguous feedback and state depression. Trait appearance comparison moderated the relationship between ambiguous feedback and intentions to use bulimic behaviors. Exploratory analyses conducted with subsamples developed using high versus low levels of trait disturbance showed significant results for the subsample based on trait appearance comparison levels. The findings are discussed in the context of possible reasons for the unexpected responses to the ambiguous appearance-related versus nonappearance-related feedback. The limitations of the study and directions for future research are also noted.
Thesis:
Dissertation (Ph.D.)--University of South Florida, 2008.
Bibliography:
Includes bibliographical references.
System Details:
Mode of access: World Wide Web.
System Details:
System requirements: World Wide Web browser and PDF reader.
Statement of Responsibility:
by Sylvia Herbozo.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 147 pages.
General Note:
Includes vita.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 002019012
oclc - 427370365
usfldc doi - E14-SFE0002487
usfldc handle - e14.2487
System ID:
SFS0026804:00001


This item is only available as the following downloads:


Full Text
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 002019012
003 fts
005 20090722090728.0
006 m||||e|||d||||||||
007 cr mnu|||uuuuu
008 090722s2008 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0002487
035
(OCoLC)427370365
040
FHM
c FHM
049
FHMM
090
BF121 (Online)
1 100
Herbozo, Sylvia.
4 245
The effects of ambiguous appearance-related feedback on body image, mood states, and intentions to use body change strategies in college women
h [electronic resource] /
by Sylvia Herbozo.
260
[Tampa, Fla] :
b University of South Florida,
2008.
500
Title from PDF of title page.
Document formatted into pages; contains 147 pages.
Includes vita.
502
Dissertation (Ph.D.)--University of South Florida, 2008.
504
Includes bibliographical references.
516
Text (Electronic dissertation) in PDF format.
520
ABSTRACT: Previous research has demonstrated the influential role of physical appearance-related feedback in the development of body image and eating disturbances. Teasing and negative feedback have been established as strong correlates and predictors of body dissatisfaction, maladaptive eating behaviors, and psychological distress. However, very little is known about ambiguous appearance-related feedback and its impact on others. The current study sought to explore this area with an experimental study to examine the effects of ambiguous appearance-related feedback on body image, mood states, and intentions to use body change strategies. Undergraduate women (N=146) were randomly assigned to an ambiguous appearance-related or ambiguous nonappearance-related feedback condition. Body image, mood states, and intentions to diet, exercise, and use unhealthy weight control methods were assessed before and after feedback was provided by a confederate.Results indicated no significant differences between feedback conditions in body image and mood states. The mean trends for all mood state, with the exception of anger, indicated better mood states after ambiguous appearance-related feedback compared to after ambiguous nonappearance-related feedback. State anger was greater in the ambiguous appearance-related feedback condition suggesting that this particular type of feedback was interpreted in a negative manner. Further, there was a significant difference between feedback conditions for intentions to diet and use bulimic behaviors, with lower levels in the ambiguous appearance-related feedback condition. No significant differences were found for intentions to exercise. State appearance comparison was not shown to mediate the relationship between ambiguous feedback and body image, mood states, or intentions to use body change strategies.Trait appearance satisfaction, appearance comparison, appearance schematicity, and thin ideal internalization were found to moderate the relationship between ambiguous feedback and state depression. Trait appearance comparison moderated the relationship between ambiguous feedback and intentions to use bulimic behaviors. Exploratory analyses conducted with subsamples developed using high versus low levels of trait disturbance showed significant results for the subsample based on trait appearance comparison levels. The findings are discussed in the context of possible reasons for the unexpected responses to the ambiguous appearance-related versus nonappearance-related feedback. The limitations of the study and directions for future research are also noted.
538
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
590
Advisor: J. Kevin Thompson, Ph.D.
653
Commentary
Weight
Size
Physical
Eating disorders
0 690
Dissertations, Academic
z USF
x Psychology
Doctoral.
773
t USF Electronic Theses and Dissertations.
856
u http://digital.lib.usf.edu/?e14.2487



PAGE 1

The Effects of Ambiguous A ppearance-related Feedback on Body Image, Mood States, and Intentions to Use Body Changes Strategies in College Women by Sylvia Herbozo A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Psychology College of Arts and Sciences University of South Florida Major Professor: J. Kevin Thompson, Ph.D. Michael Brannick, Ph.D. Jonathon Rottenberg, Ph.D. William Sacco, Ph.D. Joseph Vandello, Ph.D. Date of Approval: May 24, 2007 Keywords: commentary, weight, si ze, physical, eating disorders Copyright 2007, Sylvia Herbozo

PAGE 2

Acknowledgements This dissertation is dedicated to my parents, Mauro and Gloria Herbozo, who have given me unconditional love and supported me in all my efforts to pursue my goals. I would like to also thank my sister, Beve rly, whose guidance during my undergraduate years led me to pursue a doctoral degree. Sh e has not only inspired me to strive for academic excellence but also continuously provided me with words of encouragement that helped me achieve my goals. In additi on to my family, I woul d like to thank a close friend and colleague, Sheri Jacobs, who ha s supported me in so many ways as I completed my dissertation. I am also especially grateful to my majo r professor, J. Kevin Thompson, Ph.D. for his guidance and patience throughout this process. Thanks also to all of the members of the Body Image Res earch Group for being great colleagues and friends and creating a wonderful work environment.

PAGE 3

i Table of Contents List of Tables iv List of Figures v Abstract vi Chapter One. Introduction 1 Overview 1 Negative Appearance-related Feedback, Body Image, and Eating Disturbance 6 The Cognitive Processing Model 18 The Social Comparison Model 27 Pilot Study 32 Chapter Two. Method 53 Participants 53 Measures 53 Demographic Information 54 Body Mass Index 54 Body Image 54 Body Image Mood 55 Appearance Schematicity 56 Thin-Ideal Internalization 57 Appearance Comparison 57 Appearance-related Teasing 59 Dieting 59 Bulimic Symptoms 60 Exercise 61 Eating Disorder Screening 61 Message Source Rating Form 62 Distraction Task 62 Experimental Stimuli 63 Confederates 63 Procedure 63 Design and Analyses 66 Chapter Three. Results 71 Preliminary Analyses 71 Planned Analyses 77 ANCOVAs and MANCOVAs on State Measures 77

PAGE 4

ii ANCOVAs and MANCOVAs on Intention Measures 77 Mediation Analyses 79 Moderation Analyses 81 Exploratory Analyses 86 ANCOVAs and MANCOVAs for Appear ance Comparison Subsample 87 Chapter Four. Discussion 90 References 99 Appendices 109 Appendix A: Demographic Information 110 Appendix B: The Multidimensional Body Self-Relations Questionnaire 111 Appendix C: Body Image States Scale 112 Appendix D: Visual Analogue Scales 114 Appendix E: Appearance Schema Inventory-Revised Short Form 115 Appendix F: Sociocultural Att itudes Towards Appearance Questionnaire-3 Internalization-General subscale 117 Appendix G: Physical App earance Comparison Scale 118 Appendix H: State Comparison Scale 119 Appendix I: Physical Appearan ce-related Teasing Scale 120 Appendix J: Dutch Eating Behavior Questionnaire-Restraint Scale 122 Appendix K: Modified Dutch Eatin g Behavior QuestionnaireRestraint Scale 123 Appendix L: Eating Disorder Examination-Questionnaire-Bulimia subscale 124 Appendix M: Modified Eating Di sorder Examination-Questionnaire Bulimia subscale 126 Appendix N: Multidimensional Heal th Behavior Inventory Exercise subscale 127 Appendix O: Modified Multidimensional Health Behavior Inventory Exercise subscale 128 Appendix P: Message Source Rating Form for the Pilot Study 129 Appendix Q: Flyer for Ambi guous Appearance-related Feedback Condition of the Pilot Study 130 Appendix R: Coupon for Am biguous Appearance-related Feedback Condition of the Pilot Study 131 Appendix S: Script for Ambi guous Appearance-related Feedback Condition of the Pilot Study 132 Appendix T: First Debriefing Fo rm for Ambiguous Appearance-related Feedback Condition of the Pilot Study 133 Appendix U: Second Debriefing Form for Ambiguous A ppearance-related Feedback Condition of the Pilot Study 134 Appendix V: Revised Message Source Rating Form for the Current Study 135 Appendix W: Flyer for Ambi guous Appearance-related Feedback Condition of the Current Study 136 Appendix X: Flyer for Am biguous Nonappearance-related Feedback Condition of the Current Study 137

PAGE 5

iii Appendix Y: Coupon for Am biguous Appearance-related Feedback Condition of the Current Study 138 Appendix Z: Coupon for Am biguous Nonappearance-related Feedback Condition of the Current Study 139 Appendix AA: Script for Ambiguous Appearance-related Feedback Condition of the Current Study 140 Appendix BB: Script for Ambi guous Nonappearance-related Feedback Condition of the Current Study 141 Appendix CC: First Debriefing Form for Ambi guous Appearance-related Feedback Condition of the Current Study 142 Appendix DD: First Debriefing Form fo r Ambiguous Nonappearance-related Feedback Condition of the Current Study 143 Appendix EE: Standardized Beta Weights and R2 Values for Moderation Analyses with Trait Appearance Satisfaction 144 Appendix FF: Sta ndardized Beta Weights and R2 Values for Moderation Analyses with Trait Appearance Comparison 145 Appendix GG: Sta ndardized Beta Weights and R2 Values for Moderation Analyses with Trait Appearance Schematicity 146 Appendix HH: Sta ndardized Beta Weights and R2 Values for Moderation Analyses with Trait Thin-Ideal Internalization 147 About the Author End Page

PAGE 6

iv List of Tables Table 1 Means and Standard Deviati ons for Pre-test State Measures by Condition 46 Table 2 Significance Levels for Univariate and Multivariate Analyses by Condition 48 Table 3 Means, Standard Deviations, and Signifi cance Levels of Univariate Analyses by Conf ederate 49 Table 4 Means and Standard Deviations for Pre-test Trait and State Measures and Message Source Rating Form by Condition 72 Table 5 Significance Leve ls for Univariate Analyses by Condition 74 Table 6 Means, Standard Deviations, and Significan ce Levels for Univariate Analyses by Conf ederate 75 Table 7 Correlations am ong Pre-test Trait and State Variables 76 Table 8 Means, Standa rd Errors, Significance Levels and Partial 2 Values for Planned Univariate and Multivariate Analyses by Condition 78 Table 9 Correlations among Ambiguous Feedback, State Appearance Comparison, and Outcome Variables 80 Table 10 Correlations among Ambiguous Feedback, Trait, and Outcome Variables 82 Table 11 Standardized Beta Weights and R2 Values for Significant 85 Moderation Analyses

PAGE 7

v List of Figures Figure 1. General Moderational Model 84

PAGE 8

vi The Effects of Ambiguous A ppearance-related Feedback on Body Image, Mood States, and Intentions to Use Body Change Strategies in College Women Sylvia Herbozo ABSTRACT Previous research has demonstrated the infl uential role of physical appearance-related feedback in the development of body imag e and eating disturbances. Teasing and negative feedback have been established as strong correlates and predictors of body dissatisfaction, maladaptive eati ng behaviors, and psychological distress. However, very little is known about ambiguous appearance-rel ated feedback and its impact on others. The current study sought to explore this area with an experimental study to examine the effects of ambiguous appearance-related feedback on body image, mood states, and intentions to use body change strategies. Undergraduate women (N=146) were randomly assigned to an ambiguous appearance-re lated or ambiguous nonappearance-related feedback condition. Body image, mood states, and intentions to diet, exercise, and use unhealthy weight control methods were assesse d before and after feedback was provided by a confederate. Results indicated no significant differences between feedback conditions in body image and mood states. The m ean trends for all mood state, with the exception of anger, indicated better mood states after ambiguous appearance-related feedback compared to after ambiguous nonappe arance-related feedback. State anger was greater in the ambiguous app earance-related feed back condition suggesting that this

PAGE 9

vii particular type of feedback was interpreted in a negative manner. Further, there was a significant difference between feedback condition s for intentions to diet and use bulimic behaviors, with lower levels in the ambi guous appearance-related feedback condition. No significant differences were found for in tentions to exercise. State appearance comparison was not shown to mediate the rela tionship between ambiguous feedback and body image, mood states, or intentions to use body change strategies. Trait appearance satisfaction, appearance comparison, app earance schematicity, and thin ideal internalization were found to moderate the relationship between ambiguous feedback and state depression. Trait appearance comp arison moderated the relationship between ambiguous feedback and intentions to use bu limic behaviors. E xploratory analyses conducted with subsamples developed using high versus low levels of trait disturbance showed significant results for the subsample based on trait appearance comparison levels. The findings are discussed in the contex t of possible reasons for the unexpected responses to the ambiguous appearance-related versus nonappearance-related feedback. The limitations of the study and directions for future research are also noted.

PAGE 10

1 Chapter 1 Introduction Overview The past decade has seen a heightened research interest in body image disturbance, specifically as it relates to the development and ma intenance of unhealthy weight loss practices. Body image is most commonly defined as a subjective evaluation of ones physical appearance which includes va rious perceptual, aff ective, cognitive, and behavioral components (Offman & Bradle y, 1990; Thompson, 1990). Any type of maladaptive response that is related to c oncerns about body size or shape is known as body image disturbance (Thompson, 1995). Most prevalence rates of body image disturbance have focused on body dissatisfaction. It has become evident that over the past 25 years, body dissatisfaction has signifi cantly increased for both females and males, with females reporting greater body dissatisf action about most physical attributes (Thompson et al., 1999). The pe rvasiveness of body dissatisfac tion in our society is an issue of concern given its association with numerous health concerns, including eating disturbances (Thompson et al., 1999; Cash & Deagle, 1997), social anxiety (Cash & Flemming, 2002), depression (Denniston, Roth, & Gilroy, 1992; Noles, Cash, & Winstead, 1986), sexual difficulties (Wierder man, 2002), and poor self-esteem (Powell & Hendricks, 1999; Thompson & Altabe, 1991). B ody dissatisfaction is also recognized as a precursor to eating distur bances (Thompson et al. 1999) The potential negative

PAGE 11

2 consequences of body dissatisfaction have le d to a greater emphasis on factors that influence body image development. Interpersonal experiences, such as social interactions and feedback about ones physical appearance, have emerged as signi ficant contributors to body image (Thompson et al., 1999). An individuals body image is st rongly affected by others reactions to their physical appearance (Thompson et al., 1999). Appearance-related feedback, in particular, provides individuals with a great deal of inform ation about how others view them and the acceptability of their physical at tributes. It often reflects others opinions and expectations regarding physical attractiven ess. This type of feedback includes both verbal and nonverbal messages about ones physical appearance, ranging from direct comments to more ambiguous comments or subtle body language. Within the category of appearance-related commentary, there are even different types that seem to have distinct meani ngs and implications. Compliments about a physical attribute can be interpreted in a positive manner and enhance body image. On the contrary, appearance-related teasing a nd criticism have a negative connation and seem to contribute to poor body image (Thomps on et al., 1999). Interestingly, positive appearance-related commentary has also been as sociated with levels of distress that are similar to negative appearance-relate d commentary (Herbozo & Thompson, 2006a, 2006b). In regards to less explicit appear ance-related feedback, the meaning of ambiguous comments and subtle body language are not always as evident. Ambiguous comments are those that may be interpreted in a negative, neutral, or positive manner. Examples of ambiguous appearance-related co mment are as follows: Have you been to the gym lately? or You look different since I last saw you. In a similar way, body

PAGE 12

3 language, such as facial expr essions and gestures (e.g., poi nting), may be viewed as indicating approval or disapprova l of a physical attribute. Given the high value our society places on female physical attractiveness, it is not surprising that females are frequent target s of appearance-related feedback. Both adolescent and adult females receive appear ance-related feedback from family, peers, friends, romantic partners, and even strangers (Thompson, 1992; Tantleff-Dunn & Thompson, 1995). Retrospective studies have shown the prevalence of appearancerelated feedback in childhood and adolescence. In a sample of college women, Cash (1995) found that 72% of the women had been te ased or criticized about their appearance for an average duration of 5.8 years. Peers in general (60%) were most often identified as perpetrators of teasing or criticism about physical appear ance. Also, among the most frequent perpetrators, the worst perpetrators were collectively peers in general, specific peer(s), and friends (62%). Family members (35%), including brothers, sisters, mothers, fathers, and other relatives, were also menti oned as the worst perpetrator. Rieves and Cash (1996) also examined negative appear ance-related feedback among college females and found results consistent with those of Cash (1995). Schwartz et al. (1999) found that both college females and males receive verbal and nonverbal appearance-related feedback from their parents. Ninety -six of the females were exposed to less direct, and often nonverb al, appearance-related feedback from their mother and 94% from their father. Similar rates were found for the males (mother, 93%; father, 92%). To a lesser extent, parental f eedback in the form of weight-related teasing comments was also reported. Twenty-one percen t of females were recipients of weightrelated teasing comments made by their mo ther and 28% by their father. Among the

PAGE 13

4 males, 15% received weight-related teasing comments from their mother and 18% from their father. In general, there was a slight trend for females to receive more appearancerelated feedback compared to males which co incides with the greater emphasis on female physical appearance in Western cultures (Thom pson et al., 1999). It was also found that indirect and subtle appearance -related feedback (e.g., facial expressions) occurred more frequently than overt teasing comments. More recently, researchers have exam ined a range of appearance-related feedback, including positive verbal commentary. Her bozo and Thompson (2006a) developed the Verbal Commentary on Physi cal Appearance Scale (VCOPAS) which has two subscales that assess positive appearance-related comments regarding body shape and overall physical appearance, in additi on to a subscale that measures negative appearance-related comments. They found similar frequency rates of positive and negative comments using the VCOPAS with a college female sample. Positive appearance-related feed back has also been studied in older females. McLaren, Kuh, Hardy, & Gauvin (2004) investigated both pos itive and negative comments recalled across the lifespan among middle-aged wome n. Over half the women received positive comments from their partners. In contrast about one quarter of the women received negative comments from this same source. Also, while growing up, approximately one quarter of the women received positive comm ents primarily from their mothers (64%) whereas one third of them received negative comments mostly from peers (62%). These findings indicate the importance of evaluati ng different types of appearance-related feedback from various sources across age groups.

PAGE 14

5 The possible detrimental effects of ne gative appearance-rela ted feedback have been demonstrated in many nationwide-wide surveys and empirically-based studies (Thompson et al., 1999). Ther e is overwhelming evidence su ggesting that appearancerelated teasing is sign ificantly related to body image and eating problems as well as poor psychological functioning in adolescent and adult females (e.g., Fabian & Thompson, 1989; Thompson, 1991; Thompson et al. 1995 ; Gleason, Alexander, & Somers, 2000; van den Berg, Wertheim, Thompson, & Paxton, 2002). Although weight-related teasing has received the most attention in previous research, recent studies have indicated that even nonverbal and subtle negative feedback is also associated with negative outcomes, such as body dissatisfaction, eating disturbance, and low self-esteem (Thompson et al., 1999; Tantleff-Dunn & Gokee, 2004). However, a review of the literature indicates that ve ry little is known about how less direct, ambiguous forms of appearance-re lated feedback, in particular, lead to problematic consequences for some individuals but not others. It is evident that the link between ambiguous appearance-related feedback and negative outcomes lacks theoretical conceptualization regard ing the processing and acceptability of that type of feedback. The individual factors associated with proces sing such appearance-related feedback in a negative manner have not been thoroughly expl ored. Further, only a few studies have examined appearance-related feedback usi ng an experimental design (e.g., Tantleff-Dunn & Thompson, 1998; Furman & Thompson, 2002) and no studies have done so with ambiguous appearance-related commentary in an in vivo context. The noted limitations of the current research on appearance-rela ted feedback are worth addressing given the well-documented prevalence and effects of negative appearance-related feedback among

PAGE 15

6 females. Therefore, the current study on the eff ects of ambiguous appearance-related feedback will use a cognitive processing model as a conceptual framework. In this study, ambiguous appearance-related feedback or ambiguous nonappearance-related feedback was provided to undergraduate college females in order to evaluate the immediate effects on body image, mood states, and intentions to diet, use unhealthy weight control methods, and exercise. State appearance comparison was examined as a potential mediator. In addition, the trait variab les of appearance sa tisfaction, appearance schematicity, thin-ideal internalization, and a ppearance comparison as well as history of appearance-related teasing were assessed as potential moderators. The first section of this paper will review the literature examining the influence of negative appearance-related feedback on body image, eating disturbance, and psychological functioning. Next, the cognitive processing and social comparison models of body image disturbance will be described. The application of these models to eating disorders will also be discussed. Finally, a pilot study will be presented in detail followed by the hypotheses for the current study. Negative Appearance-related Feedback, Body Image, and Eating Disturbance In the first nationwid e survey conducted in Psychology Today Cash, Winstead, and Janda (1986) found that women who expe rienced appearance-related teasing during childhood were more likely to be dissatisfied with their appearance as adults. Using a subsample from the survey by Cash and colleagues (1986), Brown, Cash, & Lewis (1989) found that compared to adolescent fe male controls, adolescent females with binge-purge behaviors had a greater history of appearance-related teasing, were less satisfied with

PAGE 16

7 their weight, experienced more anxiety about becoming fat, and had poorer psychosocial adjustment. A more recent nationwide survey conducted in Psychology Today (Garner, 1997) further supports the notion that appearan ce-related feedback in fluences ones body satisfaction. Of the 4,000 respondents, 44% of the women and 35% of the men reported that their body image was shaped by b eing teased by others during childhood and adolescence. Many of the respondents co mments demonstrate the extent to which previous appearance-related teasing has a ffected their body image. For example, a 37year-old woman wrote, No ma tter how thin I become, I al ways feel like the fat kid everyone made fun of (p. 42, Garner, 1997). In addition, several respondents mentioned other interpersonal factors that have influenced their current body image. Forty percent of the women and 29% of the men reported that their partner s opinion about their appearance makes them feel unhappy about their body. Thirty percent of the women and 19% of the men said that being around some one critical also ma kes them feel bad about their body. Numerous early research studies also provide evidence for the negative outcomes associated with teasing during childhood a nd adolescence. In a sample of college women, Thompson and Psaltis (1988) found that both teasing frequency and effect were strongly related to overall phys ical appearance satisfacti on and eating disturbance. Fabian and Thompson (1989) reported that frequency of teasing was significantly associated with body satisfaction and effect of teasing with eating disturbance and depression in adolescent females. In regards to females with unhealthy eating patterns, Thompson (1991) found that eating disturbed co llege females experienced more teasing and negative effects of teasing compared to asymptomatic college females. The eating

PAGE 17

8 disturbed college females also reported less general appearan ce satisfaction. As a result of these initial findings, th e negative verbal commentary model of body image and eating disturbances wa s proposed (Thompson, 1992). Negative appearance-related commentary, es pecially teasing, has been recognized as an important developmental factor in the formation of ones body image. The evident role of appearance-related teasing in the context of body image and eating disturbances led researchers to further investigate teasing a nd negative appearance-related feedback as well as other types of feedback using more-empirically supported measures. In a 3-year longitudinal study, Cattarin and Thompson (1994) found that teasing history predicted later body dissa tisfaction in adolescent females. A covariance structure modeling (CSM) study by Thompson et al. (1995) indicated that teasing history directly influenced the development of body image and ea ting disturbances in a dolescent females. Similar findings have been reported in cro ss-cultural studies w ith adolescents (e.g., Lunner et al., 2000; van den Berg et al ., 2002). For instance, in a CSM study on Australian adolescent girls, van den Berg and colleagues (2002) found that teasing history was the strongest predictor for body dissatisfac tion. These studies with adolescents suggest that teas ing may be related to the onset of body image and eating disturbances. The negative correlates of appearance-rela ted teasing are also evident in studies with adult females. Thompson and Heinberg (1993) reported that weight and size teasing among college females uniquely predicted body di ssatisfaction and eating disturbance. In a previously mentioned study w ith college females, Cash (1995) found that being teased or criticized about ones phys ical appearance was moderate ly upsetting or more upsetting

PAGE 18

9 for about 71% of the women who experienced these events. Approxi mately 70% of the teased or criticized women also said they still think about these events and that they influence their current body image to some extent. Interestingly, both prevalence and emotional impact (e.g., level of distress) of a ppearance-related teasi ng or criticism were strongly related to appearan ce evaluation and s ituational body-image dysphoria. This latter finding highlights the need to examin e the emotional impact of appearance-related teasing or criticism experiences, in addition to the frequency of such experiences. The findings of Cash (1995) were replic ated and extended in a similar study by Rieves and Cash (1996) which was also desc ribed earlier. They reported that of the women who were teased or criticized about their physical appearance, 70% said these experiences were moderately to extremely upsetting and 38% felt they had a negative impact on their body image development. It was also found that being teased or criticized about ones physical appearan ce seemed to have a harmful effect on body image during adulthood. These events were associated with curre nt negative appearance evaluation, maladaptive appearance assump tions, body image dysphoria, and overweight preoccupation. In a study focusing on ethnic differences among college women, Akan and Grilo (1995) found that frequency of weight and size-related teasing was associated with problematic eating behaviors and attitudes and body dissati sfaction in African-Americans and Caucasians. These associations were not found among Asian-American females; however, this ethnic group repor ted significantly less exposure to weight and size-related teasing. Stormer and Thompson (1996) found that history of weight-r elated teasing was a significant predictor of body image disturbance for a sample of college females. In a

PAGE 19

10 subsequent CSM study with college wome n, Thompson, Coovert, and Stormer (1999) reported that the effect of appearance-rela ted teasing history on body image and eating disturbance was mediated by appearance-ba sed comparisons. Body image was also identified as a mediational link between appearance-related teasing and eating disturbance. These findings coincide with other CSM studies on adolescent females (e.g., Thompson et al. 1995, van den Berg, Werthe im, Thompson, 2002) that indicate teasing regarding physical appearance may contribu te to the development of body image and eating disturbances. Cross-cultural studies further demonstrat e the negative correlat es of appearancerelated teasing on adult females. In a study intended to replicate and extend the findings of Thompson and Stromer (1996), Mautne r, Owen, and Furnham (2000) examined appearance-related teasing and body image di sturbance in college females from the United States, Italy, and England. Consiste nt with the study by Stormer and Thompson (1996), history of weight-rela ted teasing predicted body imag e disturbance in all three Western cultures. There were no cultural diffe rences in the relationship between teasing history and body image disturbance. Mo re recently, Shroff and Thompson (2004) evaluated the relationships between body mass index (BMI), history of weight-related teasing, media internalization, and body imag e and eating disturbance in a sample of Indian adolescent and adult females. For bot h female samples, weight-related teasing mediated the relationship between BMI and body dissatisfaction. Th is finding indicates that the occurrence of weight-r elated teasing, not weight (B MI) per se, may lead to body dissatisfaction. It also supports previous re search (e.g., Lunner et al., 2000, van den Berg et al., 2002) with adolescent fema les in other Western cultures.

PAGE 20

11 Unlike the research on appearance-relat ed feedback among females, only a few studies have examined the potential negative im pact of such interpersonal experiences for males. Gleason, Alexander, and Somers (2000) evaluated the influenc e of three types of childhood teasing (competency, weight, and appearance) on self-esteem and body image in a sample of college females and males. In general, males were negatively affected by fewer forms of teasing compared to female s. Nevertheless, more frequent childhood teasing was a significant predictor of lo wer self-esteem and poorer body image for both females and males. An interesting finding wa s that females and males were affected by certain types of teasing in different ways. Competency-rela ted teasing predicted selfesteem in males whereas appearanceand co mpetency-related teasing predicted selfesteem in females. The only predictor of body image for both females and males was weight-related teasing. Different effects of appearance-related f eedback on females and males have also been found for less direct forms of feedbac k. Tantleff-Dunn et al. (1995) reported that appearance-related feed back, both verbal and nonverbal, was associated with poorer body image and more eating disturbance for both fe males and males. However, appearancerelated feedback was also related to lower se lf-esteem and greater depression in females. Schwartz et al. (1999) found that appearance-related feedback, including nonverbal feedback, was associated with and predictiv e of overall physical appearance satisfaction in females, but not males. Yet, appearance -related feedback was a correlate and predictor of psychological functioning for both fema les and males. The appearance-related feedback studies conducted thus far with females and males illustrate the possible harmful effects of such feedback for both genders.

PAGE 21

12 Studies with obese individuals suggest th at negative appearance-related feedback may be particularly problematic for this population, especially adolescents. A population-based study of eating patterns and weight concerns among 4,746 adolescents (Project Eating Among Teens) demonstrates the potential negative consequences of weight-related teasing for overweight adol escents (Neumark-Sztainer et al., 2002). Neumark-Sztainer and colleagues (2002) re ported numerous associations between weight-related teasing and unhea lthy weight control methods, such as diet pills, laxatives, diuretics, and self-induced vomiting. The occurr ence of teasing, rather than weight status (BMI), seemed to contribute to greater use of unhealthy weight control methods. In addition, a study by Eisenberg, Neumark-Sztainer, and Story (2003) found that more frequent exposure to weight-related teas ing was significantly related to less body satisfaction, poor self-esteem, more depressi ve symptoms, and higher rates of suicide ideation and attempts in adolescent females and males across weight groups. As in the study by Neumark-Sztainer et al. (2002), weight -related teasing, not weight status (BMI) per se, influenced the negative outcomes. The potential negative effects of weight -related teasing are evident in research with obese adults as well. Grilo et al. (1994) reported that greate r frequency of weight and size teasing was related to more negative appearance evaluation and body dissatisfaction in obese wome n during adulthood. Wardle, Walter, and Fox (2002) found that women with a childhood onset of obesity (i.e., reported being overweight by age 16) had a higher BMI, greater body dissatisfaction, greater history of childhood teasing, and lower self-esteem compared to women with an adult-onset. It was al so found that greater childhood teasing was associated with earlier age of obesity onset, higher BMI, higher

PAGE 22

13 body dissatisfaction, and lower self-esteem in th e entire sample. In a sample of obese women seeking weight loss, Matz, Foster Faith, and Wadden (2002) identified adult teasing, but not youth teasing, as a significant predictor of body image dissatisfaction. Greater exposure to teasing during adulthood was also associ ated with higher levels of body image dissatisfaction. As noted by Matz Foster, Faith, and Wadden (2002), these latter findings indicate the value of assessing obese fema les current interpersonal experiences. Previous research has primar ily focused on teasing and feedback during childhood and adolescence and its negative conse quences in later years, with little attention given to such inci dents occurring in adulthood. Furthermore, there is preliminary evid ence that indicates appearance-related teasing is associated with binge eating a nd poor psychological functioning. In a study (Project Eating Among Teens) noted earlier, Ne umark-Sztainer et al (2002) found that overweight adolescents who e xperienced frequent weight-re lated teasing were more likely to engage in binge eating than overwei ght adolescents who were not teased. This relationship between weight-re lated teasing and binge eati ng remained statistically significant even after BMI and demographic va riables were controlled for. In addition, Jackson, Grilo, and Masheb (2000) reported that both weight and size teasing and general appearance teasing were signifi cantly associated with poor se lf-esteem and depression in a clinical sample of women w ith binge eating disorder. Only general appearance teasing was associated with current body dissatisfa ction. In a study comparing females with bulimia nervosa and females with binge-eati ng disorder, Jackman a nd colleagues (2002) found that weight and size teas ing was associated with lowe r self-esteem whereas general appearance teasing was related to lower se lf-esteem and more depression among the

PAGE 23

14 females with bulimia nervosa. For the females with binge eating disorder, general appearance teasing was associated with more dietary restraint and depression. Thus, these initial studies indicate th at teasing may play a role in the development of an eating disorder characterized by binge eating a nd affect certain areas of psychological functioning. Anecdotal evidence also illustrates how individuals with eating disorders can be affected by receiving feedback about their phys ical appearance. Many eating-disordered patients report that receiving negative appear ance-related feedback led them to believe they are physically defective in some way (Rosen, 1992, p. 169). Body dysmorphic disordered patients similarly recall appearan ce-related comments, which triggered or worsened their preoccupation with an app earance defect (Rosen, 1992, p. 169). These case histories suggest that ne gative appearance-related f eedback may have an enduring impact on ones body image. It is likely th at certain individuals who are frequently exposed to appearance-related feedback beco me sensitive to this issue and react more negatively to future incidents of su ch feedback (Thomspon, et al., 1999). Although numerous studies support the noti on that appearance-related feedback may contribute to the onset or maintenance of body image and eating problems, most of these studies are correlational in nature and do not allow for causal explanations. To date, only four studies have investigated ne gative appearance-related feedback in an experimental setting. Heinberg and Thomps on (1992) examined the effects of body size feedback and target comparison group on colle ge females overall body dissatisfaction. The females were provided feedback in which their body size was identified as smaller or larger (positive or negative feedback) when compared to the average USF student or

PAGE 24

15 average USA citizen (particularistic or universalistic group). Body dissatisfaction, weight anxiety (state), mood, and self-est eem were assessed before and after the feedback. The degree to which the females fe lt their feedback was negative or positive and identified their specific comparison group as an important comparison group was also evaluated. The results indicated the type of feedback did not significantly influence the females body image, mood, or self-esteem. However, body image disturbance was found in females who received body size feedb ack in comparison to a particularistic group but not those given feedback in reference to a universalistic group. The females with a particularistic target group reported greater anxi ety and distress about their bodies regardless of the type of feedback given to them. Based on these findings, Heinberg and Thompson (1992) emphasize the importance of examining different target comparison groups and the threatening nature of comparisons with similar others. Tantleff-Dunn and Thompson (1998) studied the effects of body image anxiety and appearance-related feedback on recall, j udgment, and affective responses using two videotaped vignettes with college women. Each vignette consisted of a social interaction between a male and female acquaintance, with the male providing subtle appearancerelated feedback (verbal or nonverbal) or non-appearance-related feedback to the female. After watching the videos, free recall of th e social interaction and perceived reaction (ranging from negative to positive) of the woman in the video were assessed. Mood reactions to the vignettes were also examined. The findings i ndicated that free recall of the appearance-related feedback was not significa ntly different for females with high or low levels of body image anxiety. Howeve r, high body image anxiety females found incidents of appearance-related feedback to be more negative for the female recipient

PAGE 25

16 than the non-appearance-related feedback. This difference was not found for the low body image anxiety females. Also, compared to females with low body image anxiety, those with high body image anxiety experienced higher levels of anger after viewing the appearance-related feed back video. Tantlefff-Dunn and Thompson (1998) concluded that the reactions of the high body image anxiety females might have been influenced by a cognitive bias, leading them to perceive cert ain social interactions in a more negative manner. More recently, in a sample of college women, Furman & Thompson (2002) examined the influence of teasing hist ory on ones mood and body satisfaction after reading vignettes in which another female is the target of teasing. The female in the vignette either received a t easing comment regarding her physical appearance or her abilities during a social intera ction. Unexpectedly, the result s indicated that a history of teasing was not a significant predictor for mood responses in the negative appearance or abilities scenarios. Only eating disturban ce uniquely predicted mood reactions for both scenarios. Furman and Thompson (2002) noted that the failure of teasing history to significantly affect mood res ponses might be due to the fe w women who reported teasing experiences. Another possibility is that teasing history might have influenced the onset of eating disturbance without affecting psyc hological responses to weight and shaperelated experiences (Furman & Thopmson, 2002). Befort and Rickard (2003) investigated the effect of figure-size feedback on body image, self-esteem, and negative mood states of college men and women. This was a selected nonclinical sample that had a normal body weight range and did not report any symptoms of eating disorders. Both men a nd women were given posit ive, negative, or no

PAGE 26

17 feedback about their figure-size. The f eedback involved describing their body figure as ideal (positive feedback) or somewhat overweight or disproportionate (negative feedback) according to the opi nions of their classmates. Body esteem, weight and appearance satisfaction, mood st ates, self-esteem, and feedback on physical appearance were measured before and after the fee dback was provided. Gender differences in response to the figure-size fee dback were of particular inte rest. Men were expected to show a positive response bias whereas women were expected to respond in line with the positive or negative valence of the feedback. The findings showed no significant differences between men and women. As noted by Befort and Rickard (2003), it is likely that the selection criteria for the study cont ributed to the lack of gender differences in response to the feedback. There was minima l gender differences in body image a priori due to the criteria used. Given the lim ited research involvi ng manipulation of appearance-related feedback, it is evident that more experimental studies are needed to test the negative verbal commen tary model of body image and eating disturbances. The extant literature on physical appe arance-related feedback, particularly negative feedback, clearly demonstrates its potential harmful effects on body image, eating patterns, and psychological functioning. Yet, there are still many unanswered questions about the processing of appearance-related feedback and the manner in which it leads to negative outcomes among only a subset of individuals. This is especially true with ambiguous-related feedback. Can ambiguous appearance-related feedback be processed in a way that produces effects similar to those of teasing and negative appearance-related feedback? What individual factors influence how message recipients process ambiguous appearance-related feedback ? Specifically, what individual factors

PAGE 27

18 influence whether or not they accept this fee dback and subsequently modify attitudes and behaviors related to their physic al appearance? The current study examines the impact of several individual factors (e.g., message reci pients appearance sa tisfaction, appearanceschema activation) on responses to ambiguous appearance-related feedback provided in an experimental setting. As highlighted earl ier, studies on appearance-related feedback have not used an in vivo experimental desi gn to investigate the pr ocessing of ambiguous appearance-related feedback. Previous studies also have not examined individual factors in the context of processing ambiguous appearance-related f eedback. The novel application of a cognitive processing para digm to the area of ambiguous appearancerelated feedback is likely to increase our understanding of how this feedback may contribute to body image and eating disturbances. The Cognitive Processing Model Within the past few years, researchers have developed a cognitive, or information processing, model for the body image disturba nce associated with eating disorders (Vitousek and Hollon, 1990; Thompson, et al. 1999). The most recent model by Williamson et al. (2004) integrates previous rese arch in this area to provide an extensive framework for understanding the role of info rmation processing in the development of body image disturbance and the differing levels of this disturbance. As with other cognitive models, the foundation of Williams on et al.s model (2004) is schemas and schema-driven processing of information. Schemas are generally described as cognitive structures or mental represen tations that influence the proc essing of information. Selfschemas, in particular, have been defined as cognitive generalizations about the self, derived from past experiences, that organize and guide the processing of the self-related

PAGE 28

19 information contained in an individuals social experience (p. 64, Markus, 1977). The cognitive processing model by Williamson et al. (2004) proposes a body self-schema which consists of memory and knowledge stor es about the self and body size/shape and eating issues. As suggested by memory theories (B ower, 1981; Lang, 1984), it is argued that memories related to the body are associat ed with emotional memories related to the body. An implication of this assertion is th at if a body memory is activated, then the corresponding emotional body memory will be activated as well (Williamson et al., 1999). The reverse is also expected to occur. Furthermore, it is suggested that the activation of the body self-schema is dete rmined by the relevance of body-related information to environmental events. As the body self-schema is activated more frequently, it develops into a more dense network of associations that becomes easily activated and accessible from memory. This body self-schema is hypothesized to affect how much an individual pays attention to body and food-rela ted stimuli and interprets self-relevant events. Consistent with research on cognitive bi as, Williamson et al.s model (2004) is also based on the notion that an individuals psychologica l concerns influence their schemas and bias the processing of information relevant to those conc erns. It is argued that the body self-schema of individuals who are overly conc erned with body size/shape, food, and/or eating issues biases the manner in which information relate d to such issues is processed. These errors in information processing are not limited to individuals diagnosed with eating disorders, but rath er, are also common among normal-weight and underweight individuals with particular tr aits (Williamson et al., 2004). Individuals characterized by a fear of fatness, an excessi ve concern with body size/shape, thin ideal

PAGE 29

20 internalization, and perfectionism/obsessiona lity are considered to be the most susceptible to cognitive body-image related biases (Williamson, et al, 2004). Of these traits, thin ideal internalization has receive d the most empirical s upport as a risk factor promoting the development of body image and eating disturbances. Thin ideal internalization is defined as the extent to which an individual cognitively buys into socially defined ideals of at tractiveness and engages in beha viors designed to produce an approximation of these ideals (Thompson & Stice, 2001). In a recent meta-analysis study of risk and maintenance factors for eating pathology, Stice (2002) found that thinideal internalization was a cau sal risk factor for body di ssatisfaction, dieting, negative affect, binge eating, and bulimic symptoms. Gi ven that this particular trait is based on prior processing of sociocultural messages re garding the ideal body shape, it is likely to have the strongest influence on the manne r in which body-related information is processed. Further, different types of cognitive biases have been identified in Williamson et al.s model (2004), including attentional bias, memory bias, judgment bias (or selective interpretational bias), body size overestima tion, and preference for extreme thinness. This model also hypothesizes that among the susceptible individuals, cognitive biases are activated by exposure to body or food-re lated information, ambiguous stimuli, and self-reflection tasks, which are expected to activate the body self-schema. Finally, Williamson et al.s model (2004) proposes va rious interactions between the body selfschema, negative emotion, and cognitive biases all of which suggest a feedback loop where one component can potentially activate th e other. For instance, it is hypothesized that activation of cognitive biases also ac tivates negative emotion and negative emotion

PAGE 30

21 interacts with the body self-schema to elicit cognitive biases. The model argues that the feedback loop may be experienced as an obsession and/or overwhelming anxiety that must be escaped or avoided. This aver sive experience may lead to the use of compensatory behaviors or other behaviors (e.g., body checking) whic h reduces negative emotions but at the same time, reinforces the behavior (Williamson, 1990). There is strong evidence for the concep tual foundations and hypotheses of Williamson et al.s model (2004). In particul ar, previous research supports the models cognitive body image-related biases and de monstrates the manner in which they contribute to body image and eating problems. The attentional bias, which is defined as increased attention towards stimuli relate d to body size/shape and food, has been investigated in a number of studies with eating disordered and non-eating disordered samples. These studies have used differen t tasks, including the modified Stroop Color Naming test (Long, Hinton, & Gillespie, 1994; Fairburn et al., 1991; Perprina et al., 1993), dichotic listening task (Schotte, McNa ly, & Turner, 1990), and lexical decision task (Fuller, Williamson, & Anderson, 1995). It has been found that depending on the type of laboratory task, the selective processing of information associated with body size/shape and food (e.g., greater attention for such stimuli) seems to either impair or enhance task performance. St udies utilizing the emotional St roop task have demonstrated an interaction effect character ized by slower reaction times for the color naming of eating disorder salient words (e.g., fat). For instance, Long, Hinton, and Gillespie (1994) found that anorexic and obese restrained eat ers took longer than a control group, to colorname bodyand food-related words. Perprina et al. (1993) found a similar interference effect in anorexics, bulimics, and control with restrained eating behaviors. Studies

PAGE 31

22 focusing exclusively on bulimics have also no ted an interference in naming food, shape, and weight words (Cooper et al., 1992; Cooper & Fairburn, 1992) as well as body words (Davidson & Wright, 2002) in comparison to controls. In contrast to emotional St roop task studies, studies usi ng a dichotic listening task or a lexical decision task have demonstrated that performance is e nhanced by attentional bias. Schotte, McNally, and Tu rner (1990) conducted a dichotic listening task study and found that bulimics detected body-related words in the unattended passage more frequently than normal controls. Similarly, in a lexical decision task study, Fuller and colleagues (1995) found that normal weight-w omen with high body dysphoria were more accurate in detecting body and food words compared to women with low body dysphoria; however, there were no group differences in de tecting control words. The findings of these studies indicate that individuals highly preoccupied with body size/shape and food, not only those with eating disorders, direct greater attention towards information related to their concerns. The memory bias of Williamson et al.s model (2004) has also been examined. It refers to a recall bias in which informa tion related body size/shape and food is more readily encoded in memory and easily recalled than other types of information. Previous studies have used self-referent encodi ng tasks (Baker, Williamson, & Sylve, 1995; Sebastian, Williamson, & Blouin, 1996) as well as naturalistic memory recall tasks (Watkins et. al, 1995) to eval uate a memory bias for body-related stimuli. Baker, Williamson, and Sylve (1995) found that wome n with high body dysphoria recalled more fat words and fewer thin words than the women with low body dysphoria. There were also significant differences between the recall of body-relate d words and depressive and

PAGE 32

23 neutral words. No significant recall differences indicative of a recall bias were reported for the low body dysphoric group. With regards to negative mood, the results showed that current body size estimation and body dys phoria increased following negative mood induction in a subset of the sample. In c ontrast, recall bias for fatness stimuli was not affected by this procedure. In a similar study, Sebastian, Williamson, and Blouin (1996) found support for a memory bias in women with an eating disorder They reported an increased recall for fat body words in an eating disorder group, but not the high body dysphoric group or control group. The lack of a memory bias for fat body words in the high body dysphoric group is likely due to use of a normal control gr oup rather than a lo w body dysphoric group and the subsequent reduction in the power to detect group differences (Williamson, 1996). No significant group differences were eviden t in recall of nonfat body words or neutral words. Furthermore, Watkins and colleague s (1995) found that compared to individuals with low body dysphoria, those with high body dysphoria recalled more body-related items than other items (e.g., office items, non-o ffice related items, food-related items) in an office setting. There were no significant differences between the groups for recall of other items. These studies on memory bias sugg est the presence of an enhanced recall of fat body words in individuals with an eating disorder and those overly concerned with body size/shape and food issues. The st udy by Baker and colleagues (1995) also indicated that this recall bias for fatness stimuli is not influenced by negative mood states whereas body size estimation and body dysphoria are reactive to su ch mood states. Previous studies have also investigated the judgment (o r selective interpretation) bias in the Williamson et al. model (2004). This bias is defined as selective interpretation

PAGE 33

24 of information in ambiguous situations that is consistent with one s body size and shape concerns. Watkins et al. (1995) found that compared to the low body dysphoric group, the high body dysphoric group had a significan tly higher frequency of interpreting ambiguous words (e.g., polysemous such as ch est or homophones such as waste or waist) with a body-shape meaning than a non-body shape meaning. There were no significant differences between the groups rega rding the interpretati ons of neutral (e.g., non-body shape) ambiguous words. Jackman and colleagues (1995) found that athletes with high body dysphoria applied a fatness interpretation to b ody-related ambiguous situations whereas the athletes with low body dysphoria applied a thinness interpretation to the same ambiguous situations. No signifi cant differences were found in terms of how they interpreted ambiguous situ ations pertaining to health and performance concerns. Consistent with the findings of Jack man et al. (1995), a similar study by Perrin (1995) indicated that women with body dys phoria and those with eating disorders recalled body-related situati ons with a fatness interpre tation. The nonsymptomatic women used a thinness interpretation for the sa me ambiguous situations. The results also showed that both body dysphoria and eating disorder women were able to modify their cognitions when given instructions (e.g., im agine the scene with either a positive or negative meaning) to guide their interpretations. Although th is effect was small, this latter finding has important implications for th e treatment of eating disorders. It suggests that the negative manner in which eati ng disorder patients process body-related ambiguous information pertaining to the self can be altered with th erapeutic intervention. In addition to the attentional, memory, and judgment biases, Williamson (1996) has argued for recognizing body size overestim ation and preference for thinness as other

PAGE 34

25 forms of body image cognitive biases. He noted that a majority of the body estimation tasks involve making judgments about an ambiguous stimulus without any guidance; therefore, these tasks ca n be viewed as ambiguous situations that are susceptible to biased judgments. For example, some tasks require individuals to estimate the size of a particular body part or select a silhouette figure that most resemble s their body shape. Tasks with self-referent instructions are hypothesized to trigger the body self-schema, including negative emotions and memories a ssociated with body size /shape, and to most likely result in biased judgments of body si ze (e.g., a fatness interpretation) among high dsyphoric individuals. These bias ed estimations are predicted to be consistent with the negative memories and emotions that are part of the body self-schema. Williamson (1996) also noted that such judgment biases are probably influenced by the body-related atten tional and memory biases discus sed earlier. Individuals who easily recall and selectively pr ocess body-related information, such as those with high body dysphoria, may be more likely to misinterpret body-related stimuli in ambiguous situations. Furthermore, body size estimation has been shown to be labile in high dysphoric individuals experiencing negative em otion (Baker, Williamson, & Sylve, 1995; McKenzie, Williamson, & Cubic, 1993). As previously mentioned, the activation of body memories and of related emotional me mories should co-occur, according to memory theories (Bower, 1981). The activati on of negative emotion, in particular, is expected to make high body dys phoric individuals sensitive to body-related stimuli and in turn, lead to greater body si ze over estimation. This activa tion seems to determine the extent to which body size ove restimation is labile.

PAGE 35

26 In regards to the preference for thinness, Williamson (1996) noted that it should be viewed as a persons standard or exp ectation for perfection of physical appearance (p. 55). This preference for thinness (or idea l body size) has been shown to be distinct from actual perceived body size. Studies have indicated that ideal and current body size estimates are separate constructs (Gleaves et al., 1995; Williamson et. al, 1993), with ideal body size being the construc t that is more stable and unaffected by negative emotion (Baker, Williamson, & Sylve, 1995; McKenzie et al., 1993). Given these findings, Williamson (1996) proposed that the preference for extreme thinness might be due to an anchoring bias in which the idea l body size standard is anchored at a very thin level. This anchor of the thin ideal body size is hypot hesized to shift to lower body weights over time in further efforts to reduce body dysphoria (Williamson, 1996). It is also predicted to motivate individuals to lose even more we ight and result in a drive to attain an extremely thin body. The reviewed studies on cognitive biases involving attention, memory, judgment, and preference for extreme thinness support th e theoretical basis of the information processing model for body image disturban ce (Williamson et al., 2004). The findings from these studies suggest that the cognitive biases are specific to bodyand eatingrelated information and are common in indi viduals who are preoccupied with their body size and shape, in addition to t hose with eating disorders. Ho wever, it is evident that only a limited number of studies have examined each cognitive bias, many of which are not recent studies. Additional research on the c ognitive biases that serve to maintain body image and eating problems is warranted. The current study was intended to contribute to this area by further exploring one of these bi ases, the judgment bias, specifically in the

PAGE 36

27 context of ambiguous appearance-related f eedback. Based on Williamson et al.s (2004) cognitive processing model, this type of investigation may provide insight on the processing of ambiguous appearance-related feedback that can result in negative outcomes. The Social Comparison Model The social comparison theory (Festinger, 1954) is another relevant model that provides a good framework for further unders tanding why some women might respond to ambiguous appearance-related feedback in a more negative manner than other women. According to the social comparison model (Festinger, 1954), individu als have an innate drive to self-evaluate themselves on numer ous attributes, one of which is physical appearance. It is proposed that individuals who are uncer tain about an attribute will determine their standing on that particular attribute by comparing themselves to objective sources of information or to others in the social environment. In addition to selfevaluation, self-improvement and self-enhancem ent have been identified as motives for engaging in social comparison that differentially influence the target selected for comparison (Wood, 1989). Individuals seeki ng self-improvement select a comparison target that is inferior on the attrib ute of interest, which is known as a downward comparison (Kruglanski & Mayseless, 1990). In contrast, those who desire selfenhancement select a comparison target that is superior on the at tribute of interest, resulting in an upward comparison (Kruglanski & Mayseless, 1990). This latter type of comparison has been associated with negative outcomes, such as increases in emotional distress and decreases in self-esteem (Major, Testa, & Bylsma, 1991).

PAGE 37

28 Social comparisons related to physical appearance, in particular, have been examined in the area of body image and eating dist urbances. In this context, it is argued that individuals with a tende ncy to compare their physical appearance to others and engage in upward comparisons are more likely to be negati vely influenced by sociocultural messages on the th in ideal body for females. Se veral correla tional studies have consistently shown that higher levels of social comparison tendencies are associated with greater with body dissati sfaction (Thompson et al., 1999). Experimental studies (e.g., Heinberg & Thompson, 1992, Cattarin et al., 2000) in this area have further indicated that an overall tende ncy to engage in social co mparisons, not the type of comparison per se, plays an important role in body image disturbance. Given findings related to social comp arison, researchers ha ve proposed social comparison as a possible mediator linking va rious factors to body image disturbance (Heinberg & Thompson, 1995; Irving, 1990; Richins, 1991). In a covariance structure modeling study noted earlier, Thompson, Coovert, & Stromer (1999) found that appearance-based social comparison mediated the relationship of early appearancerelated teasing to body image and eating disturbance in a sample of college women. More recent studies have examined the act ual appearance comparison processing. The findings from these studies support the mediationa l role of this specif ic type of social comparison processing. Tiggemann and Slater (2003) reported that the link between exposure to thin ideal body images in mu sic videos and body satisfaction for college women was mediated by state appearan ce comparison. Similarly, Tiggemann and McGill (2004) found that state appearan ce comparison was a mediator for the relationship between exposure to thin id eal body images in magazines and mood and

PAGE 38

29 body dissatisfaction among college women. These fi ndings suggest that state social comparison is one mechanism by which sociocultural pressures regarding female physical attractiveness might re sult in negative effects. Previous research has not focused specifically on state appearance comparison as a mediator linking appearance-related feedback, such as ambiguous feedback, with body image and eating disturbances. It is likely that the effect of ambiguous appearancerelated feedback may depend on whether or not the message recipient engages in comparisons with the message source in terms of physical appearance. As with thin ideal body images in the media, women who compare themselves to a thin, attractive female message source may be those that respond negatively to ambiguous appearance-related feedback from that source. For instance, a woman who receives am biguous appearance-related feedback and then compares herself to a thin, attractive female message source, may be more vulnerable to experiencing body dissatisfaction a nd negative emotions. It is possible that the message recipient may place higher value on feedback coming from a thin, attractive message source who fits with the sociocultural ideal body shap e. This message recipient may also be more likely to interpret the indirect feedback in a more negative manner. She may assume that the thin, attractive me ssage source is informing her that she does not meet current social standards of physical attractiveness and in turn, feel bad about her own physical appearance. The subsequent body dissatisfaction may even lead her to engage in or increase the use of body change strategies, such as di eting, unhealthy weight loss methods, or exercising, to enhance her physical appearance. Thus, appearance comparison processing following the occu rrence of ambiguous appearance-related

PAGE 39

30 feedback is an area worth exploring in th e current study given the potential detrimental effects. The processing of appearance compar isons may be an important target for body image and eating disturbance interventions. In sum, previous research has shown that the development of body image and eating disturbance is greatly influenced by physical appearance-related feedback. Most studies in this area have primarily focuse d on teasing and negativ e appearance-related feedback. There is strong evidence that su ch feedback is often associated with body dissatisfaction, maladaptive ea ting behaviors, and psycholog ical distress (Thompson et al., 1999). However, very little is known about ambiguous appearance-related feedback and its impact on others. Specifically, th e processing of ambiguous appearance-related feedback has not been investigated. The literature on the cognitive model proposed by Williamson et al. (2004) provides support for examining schematic processing and cognitive biases in an ambiguous context. Of the cognitive biases described in Williamson et al.s model, the judgment bias seems the most relevant to the potential errors in the processing of ambiguous appearance-related feedback. Willia mson et al.s (2004) model also identifies several individual factors, such as appearance satisfaction, appearance schematicity, and thin-ideal internalization that might play an important role in the processing of such information. Further, studies related to the social comparison model of body image disturbance suggest that the te ndency to engage in social comparisons as well as actual appearance comparisons are individual factor s that are also like ly to influence how ambiguous appearance-related feedback is pr ocessed. Also, given the prevalence and effects of appearance-related teasing, histor y of appearance-related teasing is another

PAGE 40

31 individual factor that should be considered in the processing of ambiguous appearancerelated feedback. The current study examines these individual factors in the context of ambiguous appearance-related feedback usi ng a cognitive processing framework. This study is expected to have important clini cal implications by indicating how ambiguous appearance-related feedback can potentia lly contribute to body image and eating disturbance. More specifically, it is expect ed that the findings will help identify factors to consider in interventions focusing on errors in processing ambiguous information related to physical appearance. The main purpose of the current study is to examine the influence of ambiguous appearance-related feedback on the message recipients body image, mood states, and intentions to use body change strategies using an in vivo exposure design. The message recipients are undergraduate female students who will receive either ambiguous appearance-related or non-a ppearance-related feedback. Body image and mood states will be assessed before and after the feedb ack is provided. Intentions to diet, use unhealthy weight control methods, and exercise will be assessed following the feedback. State appearance comparison will be studied as a potential mediator. Trait appearance satisfaction, trait appearance sc hematicity, trait thin-ideal in ternalization, trait appearance comparison, and history of appearance-related teasing will be examined as potential moderators influencing the messa ge recipients response to the feedback. It is expected that some message recipients will process th e ambiguous appearance-r elated feedback in a more negative manner and experience worse outcomes

PAGE 41

32 The following hypotheses are examined: 1. Body image and mood states will be more negative in the ambiguous appearancerelated feedback condition than the ambiguous nonappearance-related feedback condition after receiving the feedback. 2. Intentions to diet, use unhealthy weight control methods, and exercise will be greater in the ambiguous appearance-related f eedback condition than the ambiguous nonappearance-related feedback conditi on after receiving the feedback. 3. State appearance comparison will mediate the effect of ambiguous appearance-related feedback on body image, mood states, and in tentions to diet, use unhealthy weight control methods, and exercise. 4. Trait appearance sati sfaction, trait appearance sc hematicity, trait appearance comparison, trait thin-ideal internalization, a nd history of appearance -related teasing will moderate the effect of ambiguous appearan ce-related feedback on body image, mood states, and intentions to diet, use unhea lthy weight control me thods, and exercise. Pilot Study Prior to the current st udy, a pilot study was conduc ted with a sample of undergraduate females. Only ambiguous app earance-related feedback was examined. The primary goals of this pilot study were to determine if: 1) the message recipients viewed the confederates employment status, feedback, and flyers as credible; 2) the message recipients viewed the confederates as thin, attractive, and warm; 3) the ambiguous appearance-related feedback i nduced negative body image and moods among the message recipients. The secondary goa l of the pilot study was to examine a combination preto post-test design and post-test only design, with a prime and no prime.

PAGE 42

33 The prime was the administration of the trait measures at the beginning of the study. It was anticipated that the pilot study would indica te if: 1) the pre-test scores for the prime condition are different from the pre-test scores for the no prime condition; 2) the pre-test scores (prime or no prime c ondition) affect posttest scores (prime, post-test condition and no prime, post-test condition). The findings related to the study design were used to select the most appropriate order for administering measures in the current study. The experimental stimuli and procedures were also slightly modified based on the pilot study findings. Method Participants The participants consisted of 51 female undergraduate students recruited from the University of South Floridas participant pool The age of the sample ranged between 18 and 30 ( M = 20.25, SD = 1.77). The sample consisted of 56.9% Caucasian ( N = 29), 13.7% African American (N = 7), 11.8% Hispanic (N = 6), 7.8% Asian-American ( N = 4), and 9.7% other ( N = 5). Based on self-reported height and weight, the average body mass index (BMI) was in the normal range ( M = 23.60, SD = 4.95), with scores ranging from 16.64 to 43.07. Approximately 6% were underweight ( N = 3), 58.8% were normal weight ( N = 30), 25.5% were overweight ( N = 13), and 9.8% were obese ( N = 5). None of the participants had a current or past history of an eating disorder diagnosis nor received treatment for an eating disorder. A ll participants received extra credit points in a psychology course as compensation for participating in the study. Participants were randomly assigned to one of four conditions varying in the measures and order of measures administere d. The following four conditions were used

PAGE 43

34 with the order of measures listed: (1) admi nistration of the trait measures, pre-test measures, and post-test measures, (2) administ ration of the trait measures and post-test measures, (3) administration of the pre-test measures and post-test measures, or (4) administration of the post-test measures only. The measures administered to each participant depended on which group she had been randomly assigned to. Some measures were not administered to participants in certain groups (see Procedure section). All measures will be briefly described in this section and with greater detail in the measures section of the current study. Measures Demographic Information Participants were asked to complete a form with demographic information including age, height, weight, race/ethnicity and year in school (see Appendix A). Selfreported height in inches and weight in pounds were used to calculate body mass index (BMI) for all participants. The standard fo rmula was utilized: [(weight in pounds/(height in inches)2] X 703. Body Image The Multidimensional Body-Self Relations Questionnaire-Appearance Evaluation subscale (MBSRQ-AE; see Appendix B, Brow n, Cash, & Mikulka, 1990) was used to measure the respondents satisfaction with her physical appearance. Reliability was acceptable (Cronbachs alpha = .84) in this sample. The MBSRQ-AE was administered as a trait measure before the feedback was given to the participants. The Body Image States Scale (BISS; see Appendix C, Cash, Fleming, Alindogan, Steadman, & Whitehead, 2002) was used to measure stat e body dissatisfaction. Reliability was

PAGE 44

35 acceptable for the pre-test (Cronbachs alpha = .86) and post-test (Cronbachs alpha = .80) use of the BISS in this sample. The BISS was administered as a state measure before and after the feedback. Body Image and Mood The Visual Analogue Scales (VAS; A ppendix D, Heinberg & Thompson, 1995) were used to examine a variety of subjectiv e states and conditions. Only the following VAS items were assessed: satisfaction w ith overall appearance, anger, anxiety, depression, and self-confidence. The VAS wa s administered as a state measure before and after the feedback was given to the participants. Appearance Schematicity The Appearance Schema Inventory-Revised (ASI-R; Appendix E, Cash, Melnyk, & Hrabosky, 2004) was used to assess body image investment with regard to particular beliefs and assumptions about the importance, meaning, and influence of appearance in ones life. Reliability was acceptable (Cronbachs alpha = .86) in this sample. The ASIR was administered as a trait measure before the feedback. Thin Ideal Internalization The Sociocultural Attitudes Towards App earance Scale-3-Internalization-General subscale (SATAQ-3-I-G; see Appendix F, Thompson et al., 2004) was used as a measure of social comparisons with and desires to l ook like models and stars in various media. Reliability was excellent (Cronbach's alpha = .94) in this sample. The SATAQ-3-I-G was administered as a trait measure before the feedback.

PAGE 45

36 Appearance Comparison The Physical Appearance Comparison Scale (PACS; see Appendix G, Thompson, Heinberg, & Tantleff-Dunn, 1991) was used to measure the tendency to compare oneself to others on different aspects of physical appearance. Reliability was acceptable (Cronbach's alpha = .70) in this sample. Th e PACS was administered as a trait measure before the feedback. The State Appearance Comparison Scale (see Appendix H) was used to assess comparison engendered by exposure to the expe rimental manipulation. Reliability was acceptable (Cronbachs alpha = .76) in this samp le. This state measure was administered after the feedback following the post-test measures. Appearance-related Teasing The Physical Appearance-Related T easing Scale (PARTS; see Appendix I, Thompson, Fabian, Moulton, Dunn, & Altabe, 1 991) was used to assess teasing history and consists of the Weight/Siz e Teasing and the General App earance Teasing subscales. Reliability was excellent (Cronbach's alpha = .92) in this sample. The PARTS was administered as a trait measure before the feedback. Dieting The Dutch Eating Behavior Questionnaire-R estraint Scale (DEBQ-RS; van Strien, Frijters, Bergers, & Defares, 1986) was us ed to measure the frequency of dieting behaviors. The directions of this scale were modified to assess usual and intended dieting behaviors (see Appendix J and K). Relia bility was acceptable for the DEBQ-RS (Cronbachs alpha = .78) and excellent for the modified DEBQ-RS (Cronbachs alpha =

PAGE 46

37 .93) in this sample. The DEBQ-RS assessing current dieting behaviors was administered as a trait measure before the feedback. The modified DEBQ-RS assessing intended dieting behaviors was administered as a post-test measure after the feedback. Bulimic Symptoms The Eating Disorder Examination-Questionnaire-Bulimia subscale (EDE-Q-B; see Appendix L, Fairburn & Beglin, 1994) was us ed to assesses the frequency of binge eating and purging (e.g., vomiting, laxative and diuretic use, excessive exercising) over the past week. A modified version of the EDE-Q-B was also utilized to assess intentions to use unhealthy weight control behaviors (see Appendix M). Reliability was acceptable for the EDE-Q-B (Cronbachs alpha = .85) and low for the modified EDE-Q-B (Cronbachs alpha = .62) in this sample. The EDE-Q-B was administered as a trait measure before the feedback. The modified EDE-Q-B was administered as a post-test measure after the feedback. Exercise The Multidimensional Health Behavior Inventory-Exercise subscale (MHBI-E; Kulbok, et al., 1999) was used to measure th e frequency of physical activity such as vigorous exercise for at least 20 minutes a day, three times a week. The directions of the MHBI-E were modified to assess usual and intended exercise be haviors (see Appendix N and O). Reliability was acceptable for th e MHBI-E (Cronbachs alpha = .85) and modified MHBI-E (Cronbachs alpha = .84) in this sample. The MHBI-E assessing current exercise behaviors was administered as a trait measure before the feedback. The

PAGE 47

38 modified MHBI-E assessing intended exercise behaviors was administered as a post-test measure after the feedback. Eating Disorder Screening Potential participants were prescreened via USF Experimentrak in order to minimize any risk associated with recei ving ambiguous appearan ce-related feedback during the study. The prescreening consisted of a yes/no question asking respondents if they have ever been diagnosed or treate d for an eating disorder. A yes response excluded potential partic ipants from the study. They were not allowed to sign up for the study via Experimentrak. Message Source Rating Form The Message Source Rating Form (MSRF) was created to assess the participants opinion about the message source (e.g., confed erate) on various domains (see Appendix P). These questions were related to the message sources employment status, feedback, and flyers (i.e., level of credibility) as well as he r physical appearance (i.e., level of thinness, attractiveness) and warmth. A com posite score for the three items on credibility was developed. Reliability for this compos ite was acceptable (Cronbachs alpha = .80). This form was administered to all participants after they were fully debriefed at the end of the study. Distraction Task A distraction task was used after the ad ministration of the trait measures as a washout period prior to the administration of the pre-test measures, experimental manipulation, and post-test measures. This distraction task involved asking participants

PAGE 48

39 to think about the countries of the world and to list their t op ten destinations. Previous research has shown that brief (5-8 minutes ), externally-focuse d, active tasks bring experimentally-induced dysphoric mood states back to baseline levels (Lyubomirsky & Nolen-Hoeksema, 1993, 1995; Morrow & Nolen-Hoeksema, 1990). Experimental Stimuli The experimental stimuli for the pilot (appearance condition only) and main study (appearance and nonappearance conditions) were developed at the same time. The feedback topics for each condition include d physical appearance focusing on cosmetic procedures (appearance condition) and academic competence focusing on academic tutoring services (nonappearance condition). The latter topic was selected as a more neutral topic that still reflected a personal attr ibute. A script for the feedback interaction as well as a flyer and coupon were made for each condition. These materials were based on information compiled from websites of numerous sites offering the services of interest, cosmetic surgery procedures and tu toring services. Specifically, websites for cosmetic surgery offices and centers and those for tutoring services were reviewed. The terms and format used by these sources were included in the experimental stimuli. The noted websites were examined in order to provide accurate information about the procedures/services being offered to the public and to increase the exte rnal validity of the study by including the type of information available to the public via websites. An effort was also made to develop a script, flyer, and coupon that were very similar for each condition with the exception of the topic bei ng addressed. The same template was used for all materials.

PAGE 49

40 After the flyers and coupons were developed, each flyer and coupon was reviewed by an expert panel consisting of one licensed clinical psychologist and five doctoral students in clinical psychology speci alizing in the area of body image and eating disturbance. This expert panel was asked to examine each flyer and coupon in terms of its content, readability, and face validity. They were also asked to evaluate the similarity between the flyers with regard to wordi ng, content coverage, and length. The coupons were examined in the same manner. Minor modifications in wording were made to the flyers and coupons based on the feedback receiv ed from the research lab members. Only the flyer and coupon for a cosmetic surgery center (South Tampa Center for Cosmetic Surgery) were used in the pilo t study (see Appendix Q and R). Confederates Two female research assistants who are id entical twins served as the confederates for the study. These twins had a slender body type. Their body mass index (BMI) was about 19, which is at the lower end of the normal range (18.5-24.5). The underweight range is less than 18.5. As the confederates, these twins wore a fitted polo shirt with jeans and dress shoes. They wore the same polo shirt that matched the company that they were pretending to work for (South Tampa Cent er for Cosmetic Surgery). Each twin had an approximately equal number of assignments as confederate to th e four conditions Procedure Undergraduate female students were recruited from psychology courses to participate in a study called mood and body sa tisfaction. Each pa rticipant completed the experimental protocol on an individual ba sis. This study only included an ambiguous

PAGE 50

41 appearance-related feed back condition. The measures and order of the measures were manipulated. When the participant arrived at the laborat ory, the researcher told her that she was running two participants at a time and was still waiting for the second participant to arrive. She then asked her to please take a seat outside of the experimental room and closed the door behind her. A female confed erate arrived shortly after the participant and knocked on the door to the experimental r oom. The researcher opened the door and asked the confederate if she wa s here for a research study. Af ter the confederate said yes, the researcher asked her to come in as well as the participant sitt ing outside the room. She then asked them to take a seat in either chair at the table. Since the confederate came in first, she could choose where to sit and alwa ys sat in the chair cl osest to the door. For the remainder of the study, the female confeder ate and the actual part icipant were treated the same. After entering the experimental room, the participants were informed that the study is examining the relationship betw een mood and body satisfaction which was really party of the cover stor y. All participants were first asked to provide informed consent by reading and signing the informed consent document. This document varied depending on which condition the participant wa s assigned to. Partic ipants in condition 1 (trait, pre-test, post-test) and 2 (trait, post-test) conditions were then asked to complete a packet of trait measures. This packet included the MBSRQ-AE, ASI-R, SATAQ-3-I-G, PACS, PARTS, DEBQ-RS, EDE-Q-B, and MHB I-E. The brief, five to eight minute distraction task only followed the completion of the trait measures in conditions 1 and 2. This task was intended to decrease the possibility of the trait meas ures having an effect

PAGE 51

42 on the subsequent measures. Participants in condition 3 (pre-test, post-test) and 4 (posttest) were not administered the packet of tr ait measures nor the di straction task. Next, participants in condition 1 (tra it, pre-test, post-test) and 3 (pre-test, post-test) completed the pre-test BISS and VAS measures. Particip ants in condition 2 (t rait, post-test) and 4 (post-test only) were not administ ered any pre-test measures. The participants (in all conditions) then watched a neutral video clip (a 5-minute nature video). The researcher left the room after this video clip was started. Once the video clip was finished, the confederate gave the participant a flyer (see Appendix Q) and provided ambiguous appearance-related feedback This feedback was ambiguous in that it could be interpreted in a neutral or negativ e manner. It was based on a 3to 4-minute prescripted dialogue (see Appendix S) relevant to the condition. Following the feedback, the confederate ex cused herself from the experimental room to use the restroom and asked the participant to let the researcher know where she went if she returned before she came back. Before leaving, the c onfederate pulled out a stack of discount coupons (see Appendix R) fo r the South Tampa Center for Cosmetic Surgery. While placing this stack on the table in front of them, she sa id Feel free to take a coupon if you would like a student discount fo r our center, and then left the room. This away time was intended to give the participant time to process the feedback that she was given and to decide if she wanted to ta ke advantage of the st udent discount without feeling pressured by the presence of the conf ederate. While the confederate was in the restroom, the researcher intentionally reentere d the experimental room and acted surprise to see that the confederate was not there. This was done to further disguise the confederates role in the st udy. The confederate came back to the experimental room

PAGE 52

43 shortly after the res earcher returned. The participants (in all conditions) were then asked to complete the post-test BISS and VAS as well as the modified DEBQ -RS, EDE-Q-B, and MHBI-E behavioral intention questionnaires. The last measure completed by the participant (in all conditions) was the State Appearance Comp arison measure. After the post-test assessment, all participants were asked to write their opinion about the purpose of the study in order to assess whether the ma nipulation was adequately concealed. Specifically, they were asked What was the purpose of this study? Finally, participants were fully debriefed about the real purpos e of the study and the rationale for using deception. They also received a debriefing form (see Appendix T) that described the purpose of the study in greater detail. They we re asked to read this form while in the experimental room and were not allowed to keep this form. They were then given a form (see Appendix U) with contact information fo r therapy services and suggested readings. They were allowed to take th is second form with them. Af ter the debriefing process, the participants were asked to complete the Me ssage Source Rating form. They were then awarded their extra credit point s and asked not to discuss the study with anyone. Analyses The pilot study used a combination preto post-test design and post-test only design, with a prime and no prime. The pr ime was the administration of the trait measures at the beginning of the study. The pre-test scores were compared to determine the effect of the prime, and post-test scores were compared to determine the effect of the prime and pre-testing.

PAGE 53

44 Skewness and kurtosis values were examined for all outcome variables. Log transformations were conducted on four va riables that had valu es outside of the acceptable range. The transformed variables were used in subsequent analyses. Only the primary variables of interest, specifically all state measures, were examined. Preliminary analyses were conduc ted to identify any initial di fferences among the groups. The demographic variables and BMI were examined by condition. One-way ANOVAs were used for continuous variables and 2 were used for categorical variables. Each item of the MSRF and the MSRF cred ibility composite scor e were evaluated. Separate one-way ANOVAs were performed on the message source form to test for differences by condition across all four c onditions. Separate one-way ANOVAS were also conducted on each item of the MSRF to assess for differences by confederate across all four conditions. Separate analyses were conducted on the state measur es to assess for differences by condition. One-way ANOVAS were performed on the pre-test scores of the BISS for condition 1 (administration of the trait m easures, pre-test measures, and post-test measures) and condition 3 (administration of the pre-test measures and post-test measures). One-way MANOVAS were perfor med on the pre-test scores of the VAS measures for the condition 1 (administration of the trait measures, pre-test measures, and post-test measures) and condition 3 (administra tion of the pre-test measures and post-test measures). Next, one-way ANCOVAS were run on the post-test scores of the BISS for condition 1 (administration of the trait m easures, pre-test measures, and post-test measures) and condition 3 (administration of the pre-test measures and post-test measures) using the pre-test scores as the covariate. One-way MANCOVAS were run

PAGE 54

45 on the post-test scores of the VAS measures for condition 1 (administration of the trait measures, pre-test measures, and post-test measures) and condition 3 (administration of the pre-test measures and post-te st measures) with pre-test scores as covariates. Finally, one-way ANOVAS were also conducted on the pos t-test scores of the BISS for all four conditions which administered post-test measures. One-way MANOVAS were run on the post-test scores of the VAS measures for all four conditions which administered posttest measures. In order to assess for state changes, pair ed sample t-tests were conducted on the preand post-test scores of the BISS and VAS for condition 1. The same t-tests were run on these preand post-test scores for condition 3. The other conditions did not administer both preand post-test measures. All an alyses were performed with SPSS 15.0. Results Preliminary Analyses Analyses were conducted to assess for any initial differences among the groups on the demographic variables. No significant differences were found across the groups on race, 2 (12)=8.27, p>.05., age, F (3,47)=1.22, p >.05, year in school, F (3,47)=.50, p>.05, and BMI (3,47)=.65, p>.05. The pre-test measures were not examined given that these measures were not administered in all conditions. Based on these preliminary analyses, it can be assumed that random assignment was successfully used. The data from one participant in condition 1 (adm inistration of the trait meas ures, pre-test measures, and post-test measures) was not included in the anal yses due to an error in the order of the measures. All other particip ant data was included in the subsequent analyses.

PAGE 55

46 Analyses Descriptive Statistics Table 1 lists the measures administered to each condition and the descriptives for those measures. As noted earlier, the MSRF was administered to all participants across conditions. Higher mean scores on the MSRF reflect a greater level of a particular feature (e.g. credibility) on a scal e of 1 to 5. An examination of the means for the MSRF items focusing on credibility indicates that on average, the message source (confederate) was viewed by all groups as credible in te rms of her employment status, feedback, and flyer. The credibility composite means (gener ally 11 out of 15) further suggest adequate credibility across all groups. The means for the re maining items indicate that on average, the message source was perceived by all groups as thin, attractive, and warm. Table 1 Means and Standard Deviations for Pre-test State Measures by Condition Measure Condition 1 trait, pre-test, post-test N = 11 Condition 2 trait, post-test N = 13 Condition 3 pre-test, post-test N=13 Condition 4 pre-test, post-test N = 14 Pre-test BISS 36.36 (8.88) 33.23 (8.11) Pre-test VAS Appearance 64.55 (16.77) 59.23 (18.56) Pre-test VAS Anger 7.82 (10.05) *2.08 (1.96) 10.92 (19.10) *2.20 (2.57) Pre-test VAS Self-Confidence 66.55 (19.34) 62.00 (18.49) Pre-test VAS Anxiety 24.82 (25.58) 22.62 (26.08) Pre-test VAS Depression 15.82 (14.85) *3.39 (2.17) 14.54 (22.67) Note : = transformed value; BISS = Body Image St ates Scale; VAS = Visu al Analogue Scale.

PAGE 56

47Table 1 (Continued) Measure Condition 1 trait, pre-test, post-test N = 11 Condition 2 trait, post-test N = 13 Condition 3 pre-test, post-test N=13 Condition 4 pre-test, post-test N = 14 Post-test BISS 34.91 (7.78) 33.77 (9.42) 34.23 (7.56) 34.50 (7.48) Post-test VAS Appearance 61.46 (19.97) 58.69 (25.08) 57.85 (19.00) 65.29 (16.59) Post-test VAS Anger 8.37 (12.22) *2.11 (2.07) 1.85 (3.16) *.81 (1.14) 17.15 (24.61) *2.88 (3.10) 12.29 (20.79) *2.40 (2.65) Post-test VAS Self-Confidence 67.55 (19.61) 64.31(23.24) 60.00 (19.51) 65.86 (19.22) Post-test VAS Anxiety 22.09 (29.49) 22.69 (27.74) 14.15 (20.66) 23.93 (26.53) Post-test VAS Depression 15.45 (16.63) *3.18 (2.43) 15.62 (19.73) *3.07 (2.59) 15.31 (23.52) *2.67 (2.97) 23.00 (25.19) *4.04 (2.69) MSRF Employment Credibility 3.91 (.70) 3.77 (.83) 3.92 (.86) 3.69 (1.18) MSRF Feedback Credibility 3.91 (.70) 3.85(1.07) 3.75 (.87) 3.54 (.78) MSRF Flyer Credibility 3.91 (.94) 3.92 (.51) 3.67 (.98) 3.92 (.95) MSRF Thinness 3.45 (.52) 3.62 (.77) 3.54 (.51) 3.62 (.51) MSRF Attractiveness 3.27 (.47) 2.84 (.80) 3.15 (.55) 3.46 (.66) MSRF Warmth 3.27 (1.01) 3.38 (.96) 3.15 (.90) 2.85 (1.14) MSRF Credibility Composite 11.72 (1.84) 11.33 (2.02) 11.33 (2.57) 11.15 (2.48) Note : = transformed value; BISS = Body Image St ates Scale; VAS = Visu al Analogue Scale. MSRF = Message Source Rating Form. ANOVAs on Message Source Rating Form The credibility and other features of the message source were evaluated across conditions. Separate one-way ANOVAs on the each of the credibility scores as well as the credibility composite score indicated no si gnificant group differences in the ratings of the message sources employment status, fee dback, and flyer (see Table 2). Separate

PAGE 57

48 one-way ANOVAs on each of the remaining me ssage source scores also showed no significant group differences in the ratings of her thinness, attractiveness, and warmth. Finally, separate one-way ANOVAS on each it em of the MSRF generally indicated no significant differences in the ratings of the message source when comparing both message sources (see Table 3). Both sources seemed to be equated on credibility, thinness, attractiveness, a nd warmth. The only exception was on the credibility item assessing employment status, wi th one confederate rated as s lighter more credible than the other confederate ( M = 4.14 versus M = 3.59). Table 2 Significance Levels for Univariate and Multivariate Analyses by Condition Measure Conditions 1 & 3 Conditions 1, 2, 3, & 4 Pre-BISS F(1,22)=.82, p>.05 a Pre-VAS F(4, 20) = .34, p>.05b Post-BISS F(1,22)=.05, p>.05 c F(1,22)=.04, p>.05 a Post-VAS F(4,20)=1.42, p>.05 d F(12,47)=1.25, p>.05 b MSRF Employment Credibility F(3,46)=.19, p >.05 a MSRF Feedback Credibility F(3,45)=.43, p >.05 a MSRF Flyer Credibility F(3,44)=.25, p >.05 a MSRF Thinness F(3,46)=.20, p >.05 a MSRF Attractiveness F (3,46)=2.11, p >.05 a MSRF Warmth F(3,46)=.68, p >.05 a MSRF Credibility Composite F(3,44)=.13, p >.05 a Note : Condition 1= administration of trait, pr e-test, and post-test measures; Condition 2 = administration of trait and post-test measures; Condition 3 = administration of pre-test and post-test measures; Condition 4 = administration of post-test measures; BISS = Body Im age States Scale; VAS = Vi sual Analogue Scales Appearance Satisfaction, Anger, Self-Confidence, Anxiety, and Depression; MSRF = Message Source Rating Form. Letter a subscript = F test for ANOVA; Letter b subscript = F test for MANOVA; Letter c subscript = F test for ANCOVA; Letter d subscript = F test for MANCOVA.

PAGE 58

49Table 3 Means, Standard Deviations, and Significance Levels for Un ivariate Analyses by Confederate MSRF Item Confederate 1 Confederate 2 F value Employment Credibility 3.59 (.91) 1 4.14 (.79) 2 F(1,48)=5.09, p <.05 Feedback Credibility 3.57 (.96) 4.00 (.63) F(1,47)=3.15, p >.05 Flyer Credibility 3.85 (.95 ) 3.86 (.73) F(1,46)=.00, p >.05 Thinness 3.55 (.63) 3.57 (.51) F(1,48)=.01, p >.05 Attractiveness 3.21 (.49) 3.14 (.85) F(1,48)=.11, p >.05 Warmth 3.10 (1.14) 3.24 (.77) F(1,48)=.22, p >.05 Note. Number subscripts denote signific ant differences across confederates. ANOVAs and MANOVAs on State Measures Pre-test analyses were conducted to examine initial differences in state body image among conditions 1 and 3 using a one-way ANOVA. Pre-test analyses were also performed using a one-way MANOVA to examine initial differences in mood states among conditions 1 and 3. Post-t est analyses were then performed for conditions 1 and 3 using a one-way ANCOVA to assess for differences in state body image after the ambiguous appearance-related feedback was given with pre-test scores as the covariate. Post-test analyses were also conducted for conditions 1 and 3 using a one-way MANCOVA to assess for differences in m ood states after the ambiguous appearancerelated feedback with pre-test scores as the covariates. No significant group differences were found on the pre-test measures nor on the post-test state measur es (see Table 2). Additionally, post-test analys es were conducted using a one -way ANOVA to assess for differences in state body image across all condi tions. Post-analyses were also run to examine differences in multiple mood stat es among all conditions using a one-way MANOVA. No significant gr oup differences were found on the post-test state measures

PAGE 59

50 among all conditions (see Table 2 above). Thes e findings indicate that conditions 1 and 3 did not differ in state body image and mood st ates before or after ambiguous appearancerelated feedback. When examined together, th e four conditions also did not differ in state body image and mood state following the feedbac k. Overall, the administration of trait measures prior to preand post-test measures did not significantly influence the scores on these measures. Paired Sample t-tests on State Measures Pre-post test analyses were then conduc ted using paired sample t-tests to assess for state changes in body image and moods in conditions 1 and 3. These analyses were not performed for the other conditions (2 and 4) that did not incl ude both preand posttest measures. No significant state changes were found. However, it is important to note that the N for each condition was very small (N = 11 for condition 1, N = 13 for condition 3). There were mean trends in the expected direc tions for two variables of particular interest, state body image and a ppearance satisfaction. In condition 1, state body image scores decreased from pre-test ( M = 36.36, SD = 8.89) to post-test ( M = 34.91, SD = 7.78). State appearance satisfac tion decreased from pre-test ( M = 64.55, SD = 16.77) to post-test ( M = 61.45, SD = 19.97) in condition 1. State appearance satisfaction also decreased pre-test ( M = 59.23, SD = 18.56) to post-test ( M = 57.85, SD = 19.00) in conditions 3. These changes in body image suggest that participants in conditions 1 and 3 felt wo rse about their physi cal appearance after receiving ambiguous appearance-related feedb ack. There were also increases in state

PAGE 60

51 anger in conditions 1 (Pre-test: M = 7.82, SD = 10.05, Post-test: M = 8.37, SD = 12.22) and condition 3 (Pre-test: M = 10.92, SD = 19.10, Post-test: M = 17.15, SD = 24.61), indicating that participants in both conditions were more upset following the feedback. The transformed anger variables were us ed in these comparisons, although the M s and SD s for the non-transformed variables were reported. Therefore, the ambiguous appearance-related feedback likely contributed to the negative changes in body image and anger at post-test. Overall, findings from the pilot study related to the incl usion and order of measures showed that post-test responses following the ambiguous appearance-related feedback did not differ significantly across c onditions. For the two conditions (1 and 3) using both preand post-test measures, there were generally higher pr e-test and post-test scores in the condition (1) th at first administer ed a trait measure. However, the administration of a trait measure versus no tr ait measure prior to the preand post-test measures did not result in significantly different post-test responses. Given that the examined traits were of interest as potent ial moderators, a decision was made to use the measures and order of these measures in condi tion 1 (administration of trait, pre-test, and post-test measures) for the main study. Mo st importantly, the findings regarding the message sources suggested that the ambiguous appearance-related feedback was accepted as real feedback. There was evidence that th is feedback induced lo wer levels of positive body image and higher levels of anger. The experimental manipulation seemed to be effective. Both message sources were also si milarly perceived as thin and attractive, two characteristics relevant to the state comparison hypotheses for the current study. These

PAGE 61

52 findings led the same cover story and confeder ates to be used for the current study which will be described in the next chapter.

PAGE 62

53 Chapter 2 Method Participants The particip ants for the current study were 146 female undergraduate students recruited from the University of South Floridas participant pool. The PASS Power Analysis program (Hintze, 2001) was used to determine the samp le size required to achieve sufficient power (.80) at an alpha level of .05 for the analyses. The age of the sample ranged between 18 and 29 ( M = 20.28, SD = 2.21). The sample was racially diverse consisting of 56.8 % Caucasian ( N = 83), 19.9% African American (N = 29), 13.7%, Hispanic ( N = 20), 6.2% Asian-American ( N = 29), and 3.4% other ( N = 5). Based on self-reported height and weight, th e average body mass index (BMI) was in the normal range ( M = 23.80, SD = 4.89), with scores ranging from 15.30 to 41.60. Approximately 7% (6.8%) of th e sample were underweight ( N = 10), 64.4% were normal weight ( N = 94), 17.8% were overweight ( N = 26), and 11% were obese ( N = 16). None of the participants had a current or past history of an eating disorder diagnosis nor received treatment for an eating disorder. All participants received extra credit points in a psychology course as compensation for participating in the study. Participants were randomly assigned to one of two conditions varying in the type of ambiguous feedback that they received. The following two conditions were used: (1) ambiguous appearance-related feedback or (2) ambiguous nonappearance-related

PAGE 63

54 feedback. Unlike the pilot st udy, all participants in the cu rrent study were administered the same questionnaires and in the same order. The administered of measures were not different per condition. Measures The measures used in the pilot study we re also used in the current study. The only exception was the Message Source Ra ting Form (see Appendix V) which was revised based on findings from the pilot study. Demographic Information Participants were asked to complete a form with demographic information including age, height, weight, race/ethnicity and year in school (see Appendix A). Selfreported height in inches and weight in pounds were used to calculate the body mass index (BMI) for all participants. Body Mass Index (BMI) BMI is a measure of weight for height. It is often used as a variable in body image and eating disturbance research to acc ount for the effects of body mass. Higher BMI values represent higher levels of body mass (Garrow & Webster, 1985). The standard formula was utilized to compute BMI: [(weight in pounds/(height in inches)2] X 703. Body Image The Multidimensional Body-Self Relations Questionnaire-Appearance Evaluation subscale (MBSRQ-AE; see Appendix B, Br own, Cash, & Mikulka, 1990) is a 7-item questionnaire that measures the respondents satisfaction with her phys ical appearance. It

PAGE 64

55 uses a 5-point scale ranging from definitely disagree to definitely agree This subscale has shown good internal consistency (Cronb achs alpha = .88) in a sample of 1,070 women (Brown et al., 1990). Reliability wa s good (Cronbachs alpha = .85) in this sample. The MBSRQ-AE was used as a trait measure before the feedback was given to the participants. The Body Image States Scale (BISS; see Appendix C, Cash, Fleming, Alindogan, Steadman, & Whitehead, 2002). The BISS is a 6-item measure that assesses state body dissatisfaction. It uses a 9-poi nt, bipolar Likert response fo rmat. This scale has shown adequate 2-3 week test-retest reliability (.69) and internal consistency (Cronbachs alpha = .77) in a sample of undergraduate women (Cas h, et al., 2002). It has also demonstrated good convergent validity with tr ait body image measures (Cash, et al., 2002). The BISS was administered as a state measure before and after the feedback. Reliability was good for the pretest (Cronbachs alpha = .85) and posttest (Cronbachs alpha = .84) use of the BISS in this sample. Body Image and Mood Visual Analogue Scales (VAS; Appendix D, He inberg & Thompson, 1995). The VAS are brief, nonverbal measures for ev aluating a variety of subjective states and conditions. Only the following VAS items were examined: satisfaction with overall appearance, anger, anxiety, depression, and self-confidence. For each VAS item, respondents are asked to indicate their cu rrent position on the named mood state or construct by placing a verti cal mark on a 100 mm horizontal line where the two endpoints represent the extreme paramete rs of that particular stat e or construct (e.g., none to

PAGE 65

56 extreme). The distance from the farthest left point on the line (none) measured in millimeters reflects the respondents position (Thompson et al., 1999). The VAS has been widely-used because it is brief and can be repeated within a short time period. Additionally, given that participants do not choose a specific number (as with Likert formats), the VAS has been noted as resistant to the possible carryover effects wherein participants might repeat prior responses at post-testing. (Thompson, 2004). Various VAS have shown good convergen ce validity with longer measures of similar constructs (Heinberg & Thompson, 1995). Previous research has reported correlations above .60 between the VAS mood m easures and similar scales on the Profile of Mood States, and that the VAS Appearance Satisfaction co rrelates .68 with the Eating Disorders Inventory-Body Dissatisfaction s ubscale (Heinberg & Thompson, 1995). The VAS has also demonstrated good test-retest reli abilities (rs =.70.93) (Birkeland et. al, 2004). The VAS was administered as a state measure before and after the feedback was given to the participants. The order of the VAS items was varied between participants for both the pre-test and post-test Four different random or ders of the VAS items were created. One of the four orders was ra ndomly selected for each administration. Appearance Schematicity Appearance Schema Inventory-Revised (ASI-R; Appendix E, Cash, Melnyk, & Hrabosky, 2004). The ASI-R is a 20-item meas ure that assesses body image investment with regard to particular beliefs and assumptions about the importance, meaning, and influence of appearance in ones life. It uses a 5-point Likert scale ranging from strongly

PAGE 66

57 disagree to strongly agree This measure has shown good internal consistency (Cronbachs alphas > .80) as well as good convergent validit y with other measures of body image dimensions and psychosocial functioning (Cash, Melnyk, & Hrabosky, 2004). Reliability was good (Cronbachs alph a = .87) in this sample. The ASI-R was used as a trait measure before the feedback. Thin Ideal Internalization Sociocultural Attitudes Towards Appearance Scale-3-Internalization-General subscale (SATAQ-3-I-G; Appendix F, Thompson et al., 2004) is a 9-item measure of social comparisons with and desires to look lik e models and stars in various media. This measure uses a five-point Likert scale ranging from definitely agree to definitely disagree. It has shown excellent reliability (Cronbachs alpha = .96) (Thompson et al., 2004). Reliability was also excellent (Cronbachs alpha = .95) in this sample. The SATAQ-3-Internalization-General subscale wa s used as a trait measure before the feedback. Appearance Comparison Physical Appearance Comparison Sc ale (PACS; Appendix G, Thompson, Heinberg, & Tantleff-Dunn, 1991). The PACS is a 5-item scale that measures the tendency to compare oneself to others on diffe rent aspects of physical appearance. This scale uses a 5-point Like rt scale ranging from never to always It has demonstrated adequate internal reliability and test-retes t reliability as well as moderate convergent validity with measures of body image dissatisfaction, eati ng disturbance, and self-esteem (Thompson, Heinberg, & Tantleff, 1991). Reliability was acceptable (Cronbachs alpha

PAGE 67

58 = .77) in this sample. The PACS was admi nistered as a trait measure before the feedback. State Appearance Comparison Scale. The State Appearance Comparison Scale (see Appendix H) is a 3-item scale de signed to index comparison engendered by exposure to the experimental manipulation. These items are very similar to the items used in previous experimental studies examining state appearance comparison (e.g., Tiggeman & Slater, 2003; Tiggemann & McGill 2004), which have demonstrated high internal consistency (Cronbachs al pha = .91; Tiggemann & McGill, 2004). In those studies, Tiggemann and colleagues (2003, 2004) developed three items to assess participants appearan ce-related thoughts ( no thought to a lot of thought) and comparisons ( no comparison to a lot of comparison) while viewing video clips or magazine advertisements. For the current study, appearance processing was measured by asking respondents to indicate the extent to which they thought about their own appearance over the past fifteen minutes (which is the time pe riod when they will receive the ambiguous feedback). This item used a 7-point Likert s cale ranging from no thought about my appearance to a lot of thought about my appearance Similarly, appearance comparison was measured by asking respondents to indicate the extent to which they compared their overall appearance to that of the research participant (e.g., confederate) sitting next to them. They were also asked to indicate the extent to which they compared specific body parts. A 7-point Like rt scale ranging from no comparison to a lot of comparison was used for both comparison items. As in pr evious studies by Tiggemann and colleagues (2003, 2004), a composite measure of state appearance comparison was obtained by

PAGE 68

59 averaging the scores for all th ree items described above. Th e ratings on these items have been shown to be highly correlated (Ti ggemann & McGill, 2004). Reliability was acceptable (Cronbachs alpha = .71) for the compos ite measure in this sample. This state measure was administered after the fee dback following the post-test measures. Appearance-related Teasing The Physical Appearance-Related T easing Scale (PARTS; see Appendix I, Thompson, Fabian, Moulton, Dunn, & Altabe, 1991). The PARTS is an 18-item measure that assesses teasing history and consists of the Weight/Size Teasing and the General Appearance Teasing subscales. The Weigh t/Size Teasing and General Appearance Teasing subscales have demonstrated adequate internal consistency (Cronbachs alpha = .91 and .71, respectively), and te st-retest reliability (r = .86 and .87, respectively) for a sample of college females. The PARTS ha s also shown moderate convergent validity with measures of eating disturbance, body dissatisfaction, social comparison, depression, and self-esteem (Thompson et al., 1991). Re liability was good (Cronbachs alpha = .90) in this sample. The PARTS was administered as a trait measures before the feedback. Dieting The Dutch Eating Behavior Questionnaire-R estraint Scale (DEBQ-RS; van Strien, Frijters, Bergers, & Defares, 1986). The DE BQ-RS is a 10-item scale that measures the frequency of dieting behavior s. It uses a 5-point Like rt scale that ranges from never to always. The directions of this scale were m odified to assess usua l and intended dieting behaviors (see Appendix J and K). The DE BQ has shown good internal consistency (Cronbachs alpha = .95) and test-retest reliability (r = .92) (Allison, Kalinsky, &

PAGE 69

60 Gorman, 1992). Reliability was excellent for the DEBQ-RS (Cronbachs alpha = .95) and modified DEBQ-RS (Cronbachs alpha = .92) in this sample. The DEBQ-RS assessing current dieting behaviors was admi nistered as a trait measure before the feedback. The modified DEBQ-RS a ssessing intended dieting behaviors was administered as a post-test measure after the feedback. Bulimic Symptoms The Eating Disorder Examination-Questionnaire-Bulimia subscale (EDE-Q-B; see Appendix L, Fairburn & Beglin, 1994). The EDE-Q is derived from the Eating Disorder Examination (EDE; Fairburn & Cooper, 1993), which is a widely used semistructured interview. The EDE-Q-Bu limia subscale is a 12-item measure that assesses the frequency of binge eating and purging (e.g., vomiting, laxative and diuretic use, excessive exercising) over the past w eek. The frequency of these behaviors is measured in terms of the number of days that they occurred. The EDE-Q has shown adequate internal consistency (Cronbachs alpha=.84) (Fairburn & Beglin, 1994) as well as acceptable criterion validity and convergent validity (Black & Wilson, 1996). Reliability was acceptable (Cr onbachs alpha = .80) in this sample. The EDE-Q-B was administered as a trait measure before the feedback. A modified version of the EDE-Q-B was also utilized to assess intentions to use unhealthy weight control behaviors (see Appe ndix M). The items (10-12) that measure the frequency of vomiting, laxatives/diuretics use, and excessive exercising to control weight were modified to assess intentions to engage in these compensatory behaviors using a 5-point Likert scale. Items related to intentions to use diet pills, fasting, and meal

PAGE 70

61 skipping as weight control methods were also added to the scale. Reliability for the modified version was somewhat low (Cronbachs alpha = .65) in this sample. An item analysis did not reveal improvements in reliabi lity if any item was deleted. The modified EDE-Q-B was administered as a post-test measure after the feedback. Exercise The Multidimensional Health Behavior Inventory-Exercise subscale (MHBI-E; Kulbok, et al., 1999). The MHBI-Exercise subscale is a 4-item measur e that assesses the frequency of physical activity such as vigorous exercise for at least 20 minutes a day, three times a week. This scale uses a 5-point Likert sc ale ranging from never to always The directions of the MHBI-E were modifi ed to assess usual and intended exercise behaviors (see Appendix N and O). Participants were as ked How often do you. to measure usual exercise habits and also la ter asked How often do you intend to. to measure exercise intentions. The MHBI has demonstrated acceptable internal consistency (Cronbachs alpha = .80) and cont ent and convergent validity (Kulbok et al., 1999). Test-retest reliability has not been assessed. Reliability was acceptable for the MHBI-E (Cronbachs alpha = .84) and modifi ed MHBI-E (Cronbachs alpha = .83) in this sample. The MHBI-E assessing current exercise behaviors was administered as a trait measure before the feedback. The m odified MHBI-E assessing intended exercise behaviors was administered as a posttest measure after the feedback. Eating Disorder Screening Potential participants were prescreened via USF Experimentrak in order to minimize any risk associated with recei ving ambiguous appearan ce-related feedback

PAGE 71

62 during the study. The prescreening consisted of a yes/no question asking respondents if they have ever been diagnosed or treate d for an eating disorder. A yes response excluded potential partic ipants from the study. They were not allowed to sign up for the study via Experimentrak. Message Source Rating Form The Message Source Rating Form (MSRF) c onsists of six questions that assess the participants opinion about the message so urce (e.g., confederate) on various domains (see Appendix V). The questions were rela ted to the message sources employment status, feedback, and flyers (e.g., level of cr edibility) as well as he r physical appearance (e.g., level of thinness, attractiveness) and warm th. This was the same form used for the pilot study, with the exception of the first item. It was changed to more directly ask if the message source was viewed as an actual employee of the company that she was supposedly working for. The MSRF was us ed to determine if the confederates employment status, feedback, and flyers are cr edible and the confederates are viewed as thin, attractive, and warm. A composite sc ore for the three items on credibility was developed. Reliability for this composite wa s acceptable (Cronbachs alpha = .88). This form was administered to all participants afte r they were fully debrie fed at the end of the study. Distraction Task A distraction task was used after the ad ministration of the trait measures as a washout period prior to the administration of the pre-test measures, experimental manipulation, and post-test measures. This distraction task involved asking participants

PAGE 72

63 to think about the countries of the world and to list their top ten dest inations. It was the same task that was used in the pilot study. Experimental Stimuli The flyers used in the pilot study were slightly modified for the current study. The modifications consisted of changes to the wo rding used in the cosmetic surgery flyer. These changes were also made for the academic services flyer to keep them similar in terms of content and format (see Appendix W and X). No changes were made to the coupons (see Appendix Y and Z). Confederates The two confederates used for the pilot study were the same confederates used for the current study. As noted earlier, the conf ederates are identical twins with a BMI of about 19 falling at the lower end of the norma l range (18.5-24.5). The dress code used in the pilot study was also replicated. The conf ederates wore a fitted polo shirt with jeans and dress shoes. The only difference in their appearance was the company logo on the polo shirt being worn. Each confederate wo re the polo shirt that matched the company that they were pretending to work for (eit her South Tampa Center for Cosmetic Surgery or South Tampa Center for Academic Enhan cement). Each twin had an approximately equal number of assignments as c onfederate to the two conditions. Procedure The procedure for the current study was very similar to the procedure used in the pilot study. Undergraduate female students were recruited from psychology courses to participate in a study called mood and body sa tisfaction. Each pa rticipant completed

PAGE 73

64 the experimental protocol on an individual ba sis. The main difference between the pilot study and current study is that the pilot study only consisted of an ambiguous appearancerelated feedback condition whereas the curr ent study included an ambiguous appearancerelated feedback condition and an ambiguous nonappearance-related feedback condition. When the participant arrived at the laborat ory, the researcher told her that she was running two participants at a time and was still waiting for the second participant to arrive. She then asked her to please take a seat outside of the experimental room and closed the door behind her. A female confed erate arrived shortly after the participant and knocked on the door to the experimental r oom. The researcher opened the door and asked the confederate if she wa s here for a research study. Af ter the confederate said yes, the researcher asked her to come in as well as the participant sitt ing outside the room. She then asked them to take a seat in either chair at the table. Since the confederate came in first, she could choose where to sit and alwa ys sat in the chair by th e door. For the rest of the study, the female confederate and the ac tual participant were treated the same. After sitting down at the table, the participant and confederate were informed that the study was examining the relationship between mood and body satisfaction which was really party of the cover story. The partic ipants were first asked to provide informed consent by reading and signing the informed cons ent document. They were then asked to complete a packet of trait measures wh ich included the MBSRQ-AE, ASI-R, SATAQ-3I-G, PACS, PARTS, DEBQ-RS, EDE-Q-B, and MH BI-E. The brief, five to eight minute distraction task followed the completion of the trait measures in or der to decrease the possibility of the trait measures having an effect on the subsequent measures. Participants then completed the pre-test BISS and VAS.

PAGE 74

65 Next, the participants watched a neutral vi deo clip (a five minute nature video). The researcher left the room after this vide o clip was started. Once the video clip was finished, the confederate gave the partic ipant a flyer (see Appendix W and X) and provided ambiguous feedback which varied (e .g., appearance-relate d or nonappearancerelated) depending on which condition the par ticipant was assigned to. This feedback was ambiguous in that it could be interpreted in a neutral or negative manner. It was based on a 3to 4-minute prescripted dial ogue (see Appendix AA and BB) relevant to each condition. The dialogue for the ambiguous appearance-related condition was the same one used in the pilot study with one minor modification. Permanent hair removal was removed from the list of procedures noted by the confederate because a few participants in the pilot study said that this component inte rrupted the procedure. Following the feedback, the confederate ex cused herself from the experimental room to use the restroom and asked the participant to let the researcher know where she went if she returned before she came back. Before leaving, the c onfederate pulled out a stack of discount coupons (see Appendix Y and Z) for the center she was supposedly working for (either South Tampa Center for Cosmetic Surgery or South Tampa Center for Academic Services). While placing this stack on the table in front of them, she said Feel free to take a coupon if you would like a student discount for our center, and then left the room. This away time was intended to give the participant time to process the feedback that she was given and to decide if she wanted to take advantage of the student discount without feeling pressured by the pr esence of the confederate. While the confederate was in the restroom, the research er intentionally reentered the experimental room and acted surprise to s ee that the confederate was not there. This was done to

PAGE 75

66 further disguise the confederates role in the study. The confederate came back to the laboratory room shortly after the researcher returned. The participants were then asked to complete the post-test BISS and VAS as well as the modified DEBQ-RS, EDE-Q-B, and MHB I-E behavioral inten tion questionnaires. The last measure completed by the participants was the State Appearance Comparison measure. After the post-test assessment, all participants were asked to write their opinion about the purpose of the study in order to assess whether th e manipulation was adequately concealed. Specifically, they we re asked What was the purpose of this study? Finally, participants were fully debriefed about the real purpose of the study and the rationale for using decepti on. They also received a debriefing form (see Appendix CC and DD) that described the purpose of the study in greater detail. This was generally the same form used in the pilot study with additio nal information about each condition. The participants were asked to read this fo rm while in the experimental room and were not allowed to keep this form. They were then given a form (see Appendix U) with contact information for therapy services and suggested readings. They were allowed to take this second form with them. After th e debriefing process, th e participants were asked to complete the Message Source Rating fo rm. They were then awarded their extra credit points and asked not to discuss th e study with anyone. Design and Analyses The current study used a preto po st-test design for both conditions with ambiguous appearance-related feedback or nonappearance related feedback. Skewness and kurtosis values were examined for all outco me variables. Descri ptive statistics were

PAGE 76

67 computed for all variables. Pearson Produc t Moment and Point-Biserial correlations were calculated for all continuous and categor ical variables, respectively, examined as dependent variables. Preliminary analyses were conducted to assess for any initial differences among the groups. The demographic variables, BMI, pre-test trait vari ables (MBSRQ-AE, ASIR, SATAQ-3-I-G, PACS, PART S, DEBQ-RS, EDE-Q-B, a nd MHBI-E), and pre-test state variables (BISS and VAS) were exam ined by condition. One-way ANOVAS were used for continuous variables and 2 was used for categorical variables. Each MSRF item and the MSRF credibility composite score were examined. Separate one-way ANOVAs were performed on the message source form to assess for differences by condition. Separate one-way ANOVAS were also conducted on each item of the MSRF to assess for differences by confederate across both conditions. The hypotheses focusing on group diffe rences were tested by conducting ANCOVAs and MANCOVAs with relevant pre-test measures as the covariate. Hypotheses 1 stated that ther e would be more negative changes in state body image and mood states for the ambiguous appearance -related feedback condition than the ambiguous nonappearance-related feedback c ondition following the feedback. This hypothesis was tested by using a on e-way ANCOVA on the BISS and a one-way MANCOVA on the VAS (appearance satisfacti on, anger, self-confidence, anxiety, and depression) with the pre-test scor es as the covariates. Similarly, hypotheses 2 stated that there w ould be greater intentions to diet, use unhealthy weight control methods, and exer cise in the ambiguous appearance-related

PAGE 77

68 feedback condition than the ambiguous nonappe arance-related feedb ack condition after receiving the feedback. To examine this hypothesis, separate one -way ANCOVAS were run on the modified intentions versions of the DEBQ-RS, EDE-Q-B, and MHBI-E using the pre-test trait scores as the covariate. Hypothesis 3 stated that state appearan ce comparison would mediate the effect of ambiguous appearance-related feedback on state body image, mood states, and intentions to diet, use unhealthy weight control met hods, and exercise. Th e guidelines provided by Baron and Kenny (1986) for testing mediati on were used. For hypothesis 3, several preconditions for establishing mediation, as outlined by Baron and Kenny (1986), were examined. First, all three correlations among the independent, dependent, and mediator variables in question must be st atistically significant. If one of these correlations is not significant, then significant mediation ca nnot be found. The other preconditions for identifying mediation are as follows: (1) th e independent variable (ambiguous feedback) must affect the mediator variable (state appearance comparison), (2) the independent variable (ambiguous feedback) must affect the dependent variable (state body image, mood states, intentions to di et, intentions to use unhea lthy weight control methods, intentions to exercise), (3) the mediator variable (state appearance comparison) must affect the dependent variable (s tate body image, mood states, in tentions to diet, intentions to use unhealthy weight control methods, inten tions to exercise), (4) when controlling for the mediator variable (state appearance co mparison), the effect of the independent variable (ambiguous feedback) on the dependent variable (state body image, mood states, intentions to diet, intentions to use unhealthy weight co ntrol methods, intentions to exercise) should be close to ze ro. Mediation effects were only tested for the variables

PAGE 78

69 that met all preconditions. The scores of de pendent variables assesse d at pre-test were used as a covariate in each regressi on equation for that variable. Hypothesis 4 stated that trait appe arance satisfaction, trait appearance schematicity, trait appearance co mparison, trait thin-ideal inte rnalization, and history of appearance-related teasing w ould moderate the effect of ambiguous appearance-related feedback on state body image, mood states, a nd intentions to diet use unhealthy weight control methods, and exercise. The guidel ines provided by Baron and Kenny (1986) for testing moderation were also utilized. The correlation between the independent variable (ambiguous feedback) and mode rator variable (trait) s hould reflect no significant relationship. Ideally, correlations between th e moderator variable and dependent variable should also indicate no significant relations hip, allowing for a more interpretable moderation effect (interaction te rm). A moderation effect is established if the interaction between the independent variable (ambiguous feedback) and the moderator variable (trait) is significant in a regression analys es after controlling for the effects of the independent variable (ambiguous feedback) and the moderator (trait), according to Baron and Kenney (1986). The interaction has to offer additional prediction beyond that accounted for by the other variables in the re gression equation. As with the mediation tests, the scores of dependent variables assessed at pre-test were used as a covariate in each regression equation for that variable. Additionally, exploratory analyses were performed with subsamples of the original sample. These subsamples were developed by computing the median score for each trait measure (MBSRQ-AE, PACS, ASI-R, PARTS, SAT AQ-3-I-G) administered at pre-test. This score was then used as a ma rking point to identify a subsample based on

PAGE 79

70 that trait measure. Depending on the trait m easure, trait scores at or above/below the median score were considered to fall in a clinical range. This led to a clinical group and non-clinical group. Separate 2 (Feedback Condition: appearance, nonappearance) X 2 (Disturbance Level: high trait level, low trait level) ANCOVAs were computed using the new group condition (Disturbance Level) as an additional between subjects factor. A 2 (Feedback Condition: appearance, nonappear ance) X 2 (Disturbance Level: high trait level, low trait level) M ANCOVA was also computed us ing the new group condition (Clinical/Non-Clinical) as an additional between subjects factor. The ANCOVAs and MANCOVAs run for each subsample were a replication of the planned ANCOVAs and MANCOVAs on the measures described above. Relevant pre-test scor es were also used as covariates in these anal yses. All analyses were performed with SPSS 15.0.

PAGE 80

71 Chapter 3 Results Preliminary Analyses Ten participants were omitted from all anal yses resulting in a sample size of 146. The ambiguous appearance-related conditi on had an N of 80. The ambiguous nonappearance-related condition had an N of 66. Reasons for exclusion included one of the following: (1) the pre-tests were not fu lly completed, (2) the study procedures were interrupted by the participant, or (3) there was an error in the administration of measures. An examination of skewness and kurtosis values for all outcome variables indicated that nine variables had values outside of the acc eptable range. Log transformations were conducted for these variables. Subsequent anal yses used the transformed variables. The descriptive statistics for the pre-test trait and state measures and the message source form by condition are presented in Table 4. The pos t-test state measures will be reported in a later section.

PAGE 81

72Table 4 Means and Standard Deviations for Pre-test Trait and State Measures and Message Source Rating Form by Condition Measure Ambiguous Appearance-related Feedback (N=80) Ambiguous Nonappearance-related Feedback (N=66) MBSRQ-AE 25.39 (4.22) 23.94 (5.50) PACS 13.65 (3.76) 14.55 (3.88) ASI-R 66. 13 (11.36) 69.00 (11.78) PARTS 28.18 (11.08)a *5.22 (.95) a 32.14 (12.59) *5.57 (1.06) SATAG-I-G 24.66 (10.29) 24.79 (9.27) DEBQ-RS 23.06 (9.41) a 26.97 (10.42) EDE-Q-B 9.31 (8.18) a *2.69 (1.45) 12.28 (9.52) *3.15 (1.55) MHBI-E 11.04 (3.40) 11.82 (4.28) BISS 34.13 (8.8) a 30.41 (9.33) b VAS Appearance 62.90 (22.20) 56.65 (24.29) VAS Anger 9.94 (18.13) *2.19 (2.29) 13.68 (20.12) *2.65 (2.60) VAS Self-Confidence 62.56 (24.01) 60.95 (22.69) VAS Anxiety 25.65 (26.21) *4.15 (2.92) 28.47 (30.39) *4.21 (3.31) VAS Depression 14.43 (21.52) *2.78 (2.61) 22.06 (27.87) *3.53 (3.12) SCS Appearance Thoughts 4.32 (1.71) 4.88 (1.46) SCS Overall Appearance 2.75 (1.63) 3.04 (1.60) SCS Specific Body Parts 2.43 (1.48) 2.39 (1.60) SCS Total 9.50 (3.96) 10.31 (3.58) MSRF Employment Credibility 3.49 (1.00) a 4.21 (.71) Note: = transformed value; MBSRQ-AE = Multidimensional Body-Self Relations QuestionnaireAppearance Evaluation; Appearance Schematicity Inventory-Revised; PARTS = Physical Appearancerelated Teasing Scale; SATAQ-3-I-G = Sociocu ltural Attitudes Towards Appearance QuestionnaireInternalization General subscale; DEBQ-RS = Dutch Eating Behavior Questionnaire-Restraint Scale; Eating Disorder Examination-Questionnaire-Bulimi a subscale; MHBI-E = Multidimensional Health Behavior Inventory-Exercise subscale; SCS=State Comparison Scale; MSRF = Message Source Rating Form. Letter subscripts indicate signifi cant differences across conditions.

PAGE 82

73Table 4 (Continued) Measure Ambiguous Appearance-related Feedback (N=80) Ambiguous Nonappearance-related Feedback (N=66) MSRF Employment Credibility 3.49 (1.00) a 4.21 (.71) MSRF Feedback Credibility 3.56 (1.00) a 4.32 (.66) MSRF Flyer Credibility 3.68 (1.01) a 4.38 (.70) MSRF Thinness 3.58 (.59) 3.58 (.70) MSRF Attractiveness 3.32 (.67) 3.41 (.52) MSRF Warmth 3.30 (.72) 3.48 (.71) MSRF Credibility Composite 3.58 (.89) a 4.30 (.59) Note: = transformed value; MSRF = Message Source Ra ting Form. Letter subscripts indicate significant differences across conditions. Analyses were conducted to assess for any initial differences among the groups on the demographic variables. No significant differences were found across the groups on race, 2 (4)=1.66, p>.05., age, F (1,44)=1.74, p>.05, year in school, F (1,144)=2.19, p>.05, and BMI (1,144)=.42, p>.05. Significant differences were found among the groups for two pre-test trait variables a nd one pre-test state variable. There were significant group differences on history of teasing, F (1,44)=4.35, p =.04 and dieting behaviors, F (1,44)=5.66, p=.02. A significant group difference was also found on state body image, F (1,44)=6.57, p=.01. These significant findings were addressed by covarying out each of these variables in pl anned analyses. The items and credibility composite score of the MSRF were examined separately using one-way ANOVAs to assess for differe nces by condition. These items measured the degree to which the message source (r andomly assigned to each participants condition) was rated as credible in terms of their employment status, feedback, and flyer

PAGE 83

74 as well as thin, attrac tive, and warm. Significant group di fferences in ratings were found for the three items focusing on credibility and th e credibility composite (see Table 5). An examination of these three item means for each group showed that the participants in the ambiguous appearance-related condition rated th e message source as more credible than those in the ambiguous nonappearance-related c ondition (see Table 4 above). This issue was addressed by considering the use of the credibility composite as a covariate in subsequent analyses. The option of omitting participants with a credibility composite score of less than 3 (reflec ting average credibility) from subsequent analyses was also explored. This omission created an N of 126. A comparison of the original results with the results based on using each of the noted me thods indicated that there were only minor changes in the findings. Therefore, the credib ility composite was not used as a covariate and no participants were omitted from subsequent analyses. Table 5 Significance Levels for Univaria te Analyses by Condition MSRF Item F value Employment Credibility F(1,143)=23.96, p <.05 Feedback Credibility F(1,143)=28.16, p <.05 Flyer Credibility F(1,143)22.07, p <.05 Thinness F(1,143)=.00, p >.05 Attractiveness F(1,143)=.83, p >.05 Warmth F(1,143)=2.30, p >.05 Credibility Composite F(1,143)=32.14, p <.05

PAGE 84

75 One-way ANOVAs conducted separately on each MSRF item to assess for differences by confederate across conditions al so showed no significant differences (see Table 6). When comparing both message sources, they seemed to be rated as equivalent in their level of credibility, thinness, attractiv eness, and warmth. It was assumed that the message sources did not significan tly differ in key features re lated to the cover story and hypotheses. Table 6 Means, Standard Deviations, and Significance Levels for Un ivariate Analyses by Confederate MSRF Item Confeder ate 1 Confederate 2 F value Confederate Employment Credibility 3.68 (.98) 3.95 (.90) F(1,143)=3.01, p >.05 Confederate Feedback Credibility 3.76 (1.04) 4.03 (.83) F(1,143)=2.84, p >.05 Confederate Flyer Credibility 3.91 (1.06) 4.08 (.84) F(1,143)=1.11, p >.05 Confederate Thinness 3.53 (.61) 3.62 (.67) F(1,143)=.77, p >.05 Confederate Attrac tiveness 3.32 (.58) 3.39 (.63) F(1,143)=.42, p >.05 Confederate Warmth 3.26 (.70) 3.49 (.72) F(1,143)=3.73, p >.05 Correlations among the pre-test trait and state measures were examined to ensure relationships coincided with research on body im age. As expected, a majority of the pretest variables were significantly correlated in the predicted directions (see Table 7). Exercise behaviors was the only variable that was correlated with few pre-test trait and state measures. However, it was significan tly correlated with dieting and bulimic behaviors which has been shown in previous research studies.

PAGE 85

76 Correlations between pre-test and posttest state score (state body image, appearance satisfaction, anger, self-confidence, anxiety, and depression) for each variable were evaluated. Correlations between pre-te st trait scores and corresponding intention scores (dieting, bulimic, and exercise behavior s) for each variable were also examined. Significant correlations were f ound for all of th e variables ( r s ranging from .54 to .90). The pre-test scores were used as covariates in subsequent an alyses to decrease the chance of within-group error variance, given the correlations between these scores were highly correlated.

PAGE 86

76Table 7. Correlations among Pre-test Trait and State Variables 1 2 3 4 5 6 7 8 9 10 11 12 13 14 1. MBSRQ-AE 2. PACS -.43** 3. ASI-R -.36** .67** 4. PARTS -.33** .27** .27** 5. SATAQ-I-G -.30** .63** .60** .08 6. DEBQ-RS -.53** .47** .42** .35** .39** 7. EDE-Q -.50** .54* .54** .39** .49** .68** 8. MHBI-E -.13 .10** .16 .17* .08 .44** .26** 9. BISS .71** -.42** -.36** -.36** -.36* -.61** -.64* -.19* 10. VAS Appear .61** -.38** -.35** -.29** -.42* -.48** -.59* -.11* .77** 11. VAS Anger -.22** .37** .24** .16 .33** .25** .28** .10 -.31** -.19* 12. VAS SelfCo .61** -.39** -.34** -.33** -.40** -.45** -.52** -.15 .67** .72** -.28** 13. VAS Anxiety .19* .29** .20** .15 .19* .19* .33** -.05 -.27** -.20* .49** -.34** 14. VAS Depress -.38** .43** .34** .25** .33** .34** .44** .02 -.47** -.38** .56** -.42** .62** Note : MBSRQ-AE = Multidimensional Body-Self Relations Questio nnaire-Appearance Evaluation subscale; AS I-R = Appearance Schematicity InventoryRevised; PARTS = Physical Appearance-related Teasing Scale; SA TAQ-I-G = Sociocultural Attitudes Towards Appearance Questionnair e-3-Internalization General subscale; DEBQ-RS = Dutch Eating Behavior Questionnaire-R estraint Scale; EDE-Q-B = Eatin g Disorder Examination-Question naire-Bulimia subscale; MHBI-E = Multidimensional Health Behavior Inventory-Exer cise subscale; BISS = Body Image States Scale; VAS Appear = Visual Analogue Scale Appearance; VAS Anger = Visual Analogue Scal e Anger; VAS Self-Con = Visu al Analogue Scale Self-Confidence; VAS Anxiety = Visual Analogue Scale Anxiety; VAS Depress = VAS Depression. *p<.05 **p<.01

PAGE 87

77 Planned Analyses ANCOVAs and MANCOVAs on State Measures A one-way ANCOVA was performed to ex amine group differences in post-test state body image using pre-test scores as the covariate (Hypothesis 1). A one-way MANCOVA was also conducted to examine group differences in multiple post-test state moods using pre-test scores as the covariates (Hypothesis 1). As shown in Table 8, no significant group differences were found in the ANCOVA for state body image covarying out the pre-test score. An examination of the mean trends suggested that the ambiguous appearance-related feedb ack resulted in changes that were in the opposite direction (more positive) of those produced by the ambiguous nonappearance-related feedback. Similar results were found for the various mood states, with the exception of anger. There was a marginally significant condition effect in the MANCOVA for state moods, F (4, 142)=2.23, p=.055, partial 2= .076. An evaluation of the un ivariate analyses revealed a significant condition eff ect for state anger, F (1, 141)=6.08, p<.05, partial 2= .042. The ambiguous appearance-related feedback condition (adjusted M = 2.09) had greater levels of anger after the feedback than the ambiguous nonappearan ce-related feedback condition (adjusted M = 1.56). ANCOVAs on Intention Measures Additionally, separate one-way ANCOVAs were computed to examine group differences in intentions to use body change st rategies with corresponding pre-test scores as the covariate (Hypothesis 2). As shown in Table 9, a marginally significant condition effect was found for dieting intentions in the ANCOVA, F (1,144)=3.77, p=.054, partial

PAGE 88

78 2=.026. The ambiguous nonappearance-related feedback condition (adjusted M = 19.07) resulted in greater dieting intentions than the ambiguous appearance-related feedback (adjusted M = 17.74). A significant c ondition effect was also found in the ANCOVA for intentions to use bulimic behaviors, F (1,144) = 9.67, p<.05, partial 2=.063. The ambiguous nonappearance-related feedback condition (adjusted M = 2.97) elicited greater bulimic intentions than the ambiguous appearance-related feedback (adjusted M = 2.77). The ANCOVA analyses revealed no signifi cant group differences for intentions to exercise when pre-test scores were controlled. Table 8 Means, Standard Errors, Significance L evels, and Partial 2 Values for Planned Univariate and Multivariate Analyses by Condition Ambiguous Appearance-related Feedback Adjusted means & SE Ambiguous Nonappearancerelated Feedback Adjusted means & SE F and p values partial 2 values ANCOVA State Body Image 34.04 (.41) 33.55 (.45) F(1, 143)=.64, p >.05 partial 2=.004 MANCOVA All Mood States F(4, 142)=2.23, p =.055 partial 2= .076 ANCOVA State Appearance Satisfaction 64.81 (1.57) 60.93 (1.73) F(1, 144)=2.71, p >.05 a partial 2=.019 State Anger *2.09 (.14) 1 *1.56 (.16) 1 F(1, 144)=6.08, p <.05a partial 2=.042 State SelfConfidence 65.01 (1.25) *63.36 (1.38) F(1, 144)=.77, p >.05 a partial 2=.005 State Anxiety *3.08 (.21) *3.10 (.23) F(1, 144)=.00, p >.05 a partial 2=.000 Note. = transformed value. Number subscripts indicate significant differences ac ross conditions. Letter a subscript = Univariate F tests from MANCOVA.

PAGE 89

79Table 8 (Continued) Ambiguous Appearance-related Feedback Adjusted means & SE Ambiguous Nonappearancerelated Feedback Adjusted means & SE F and p values partial 2 values ANCOVA State Depression *2.41 (.16) *2.47 (.18) F(1, 144)=.07, p >.05 a partial 2=.000 Diet Intentions 17.74 (.46) 19.07 (.50) F(1,144)=3.77, p =.054 partial 2=.026 Bulimic Intentions 2.77 (.04) 1 2.97 (.05) 1 F(1,144)=9.67, p <.05 partial 2=.063 Exercise Intentions 12.25 (.29) 13.06 (.32) F(1,143)=3.54, p >.05 partial 2=.024 Note. = transformed value. Number subscripts indicate significant di fferences across conditions. Mediational Analyses State appearance comparison was hypothesized to mediate the effect of ambiguous appearance-related feedback on state body image, mood, and intentions to diet, use unhealthy weight cont rol methods, and exercise (Hy pothesis 3). As indicated by Baron and Kennys (1986) guidelines for establishing mediation, the correlations among the predictor, mediator, and each outcome va riable were first examined. All three correlations among the independent, dependent, and mediator variables in question were not statistically significant for the planned me diation analyses (see Ta ble 9). Specifically, the mediator (state appearance comparison) was not significantly correlated with the independent variable (ambiguous feedback) ( rpb = -.11, p>.05), which should be the case if the independent variable affects the medi ator. Given that this precondition was not met, the planned mediational an alyses were not performed.

PAGE 90

80Table 9 Correlations among Ambiguous Feedback, State Appearance Comparison, and Outcome Variables 1 2 3 4 5 6 7 8 9 10 11 1. Feedback Cond 2. State App Comp -.11 3. PT BISS 22** -.28** 4. PT VAS Appear .17* -.35** .81** 5. PT VAS Anger .04 .22** -.18* -.18* 6. PT VAS Self-Con .07 -.28** .71** .77** -.21* 7. PT VAS Anxiety -.02 .22** -.22** -.25** .48** -.32** 8. PT VAS Depress -.13 .24** -.45** -.40* .49* -.43** .66** 9. DEBQ-RS Inten -.25** .20* -.50 -.42** .11 -.33** .18* .27** 10. EDE-Q-B Inten -.29** .19* -.43 -.41** .15 -.33** .18* .25** .62** 11. MHBI-E Inten -.18** -.13* -.22 -.18 .00 -.08 .06** .03 .48** .38** Note : Feedback Cond = Feedback Condition; State App Comp = State Ap pearance Comparison; PT BISS = Post-test Body Image States Scale ; PT VAS Appear = Post-test Visual Analogue Scale A ppearance; PT VAS Anger = Post-test Visual An alogue Scale Anger; PT VAS Self-Con = Po st-test Visual Analogue Scale Self-Confidence; PT VAS Anxi ety = Post-test Visual Analogue Scale Anxiety; PT VAS Depress = Post-test VAS Depres sion; DEBQ-RS Inten = Dutch Eating Behavior Questionnaire-Restraint Scale Inte ntions; EDE-Q-B Inten = Eating Disorder Examination-Questionnai re-Bulimia subscale Intentions; MHBI-E = Multidimensional Health Beha vior Inventory-Exercise subscale Intentions. *p<.05 **p<.01

PAGE 91

81 Moderational Analyses Trait appearance satisfac tion, trait appearance schematicity, trait appearance comparison, trait thin-ideal in ternalization, and history of appearance-related teasing were each hypothesized to moderate the effect of ambiguous appearance-related feedback on state body image, mood, and intentions to diet, use unhealthy weight control methods, and exercise (Hypothesis 4). Using the gui delines provided by Baron and Kenny (1986) for testing moderation, the correlations among the independent variable, each moderator variable, and each dependent variable were first examined. The correlations among ambiguous feedback and each trait variable in question were not statistically significant, with the exception of teasing history ( r = -.17, p<.05). Given th at there should be no relationship between the independent variab le and moderator variable, moderational analyses were not performed for testing teasi ng history as a moderator. Additionally, the correlations between each modera tor variable and almost all dependent variables were significant (see Table 10). According to the Baron and Kennys (1986) guidelines, however, it is not required that the moderato r and dependent variable be uncorrelated in order to perform moderation analyses. It is also important to not e that all moderator variables were measured prior to the expe rimental manipulation, an ideal temporal sequence in moderation analyses (Kenny, 2006; Kraemer, Wilson, Fairburn, & Agras, 2002). Based on these findings, moderation an alyses were run on all the hypothesized moderator variables exce pt teasing history.

PAGE 92

82Table 10 Correlations among Ambiguous Feedback, Trait, and Outcome Variables 1 2 3 4 5 6 7 8 9 10 11 12 13 1. Feedback Cond 2. MBSRQAE .15 3. PACS -.12 -.43** 4. ASI-R -.12 -.36** .67** 5. PARTS -.17* -.33** .27** .27** 6. SATAQ I-G -.01 -.30** .63** .60** .08 7. PT BISS .22** .67** -.34** -.30** -.34** -.32** 8. PT VAS Appear .17* .60** -.35** -.34** -.35** -.41** .81** 9. PT VAS Anger .04 -.09 .31** .16 .07 .26** -.18** -.18* 10. PT VAS Self-Con .07 .54** -.36** -.35** -.40** -.40** .71** .78* -.21* 11. PT VAS Anxiety -.02 -.12 .26** .20* .08 .22* -.22** -.25** .48** -.32** 12. PT VAS Depress -.13 -.33** .42** .36** .21* .36** -.45** -.40** .49** -.43** .66** 13. DEBQ-RS Inten -.25** -.41** .43** .46** .32** .37** -.50** -.42** .11 -.33** .18* .27** Note : Feedback Cond = Feedback Condition; MBSRQ-AE = Multidimensional Body-Self Relations Questionnaire-Appearance Evaluation subscale; ASI-R = Appearance Schematicity Inventory-Revised; PARTS = Physical Appearance-related Teasing Scale; SATAQ-I-G = Sociocultural Attitud es Towards Appearance Questionnaire-3-Internalization General subscale; PT BISS = Post-t est Body Image States Scale; PT VAS Appear = Post-test Visual Analogue Scale Appearance; PT VAS Anger = Post-test Visual Analogue Scale Anger; PT VAS Self-Con = Post-test Visual Anal ogue Scale Self-Confidence; PT VAS Anxiety = Post-test Visual Analogue Scale Anxiety; PT VAS Depress = Post-test VAS Depression; DEBQ-RS Inten = Dutch Eating Behavior Que stionnaire-Restraint Scale Intentions. *p<.05, **p<.01

PAGE 93

83Table 10 (Continued) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 14. EDE-QB Inten -.29** -.35** .31** .35** .31** .31** -. 43** -.41** .15 -.33** .18* .25** .62** 15. MHBI-E Inten -.18* -.22** .12** .22** .24** .10 -.22** -.18* -.00 -.08 -.06** .03 .48** .38** Note : EDE-Q-B Inten = Eating Disorder Examination-Questionnaire-Bulim ia subscale Intentions; MHBI-E = Multidimensional Health Behav ior InventoryExercise subscale Intentions. *p<.05, **p<.01

PAGE 94

84 Trait appearance satisfacti on, trait appearance schematicity, trait appearance comparison, and trait thin-ideal internalization were each tested as a moderator of state body image, mood, and intentions to diet, use unhealthy weight control methods, and exercise. The general model for all tests of moderation is illustrated in figure 1. Regression analyses were performed with am biguous feedback as the predictor (path a), each trait variable as a moderator, and the in teraction of the predictor and the moderator (path c). Centered variables were created a nd used instead of the original variables to reduce the likelihood of multicollinearity probl ems. For each regression, the pre-test score of the outcome variable (paths w, x, y, z) was also used as a covariate for that variable. If the interaction (p ath c) is significant, then the moderator being examined is established as a moderator. Figure 1. General Moderational Model Nine separate regression equations were computed for each of the four trait moderators noted above (see Appendix EE to HH). Only six of the regressions had Moderator (Trait Variable) Predictor X Moderator Interaction Outcome (State Body Image, State Mood, & Intentions to use Body Change Strategies) a b c Pre-test Score Predictor (Ambiguous Feedback) x y z w

PAGE 95

85 significant interactions, indica ting a moderation effect (see Ta ble 11). Trait appearance satisfaction, trait appearance comparison, tra it appearance schematic ity, and trait thin ideal internalization were es tablished as a moderator be tween ambiguous feedback and post-test state depression. All four hypot hesized trait moderato rs had significant interaction terms (path c) with magnitudes of standardized s ranging from .15 to 20, after controlling for pre-test state depression score, ambiguous feedbac k, and pre-test trait score. The R2s for the models ranged from .59 to .81. Table 11 Standardized Beta Weights and R2 Values for Significant Moderation Analyses Trait Post-test Variable Pre-Test Variable Ambiguous Feedback Pre-test Trait Variable Interaction R2 Appearance Satisfaction State SelfConfidence .91** .05 .05 -.10* .81 State Depressison .84** -.02 -.11 .15** .74 Appearance Comparison State Depression .81** -.10 .20** -.18** .74 Bulimic Intentions .74** -.10 -.12 .20* .59 Appearance Schematicity State Depression .83** -.01 .22** -.19** .75 Thin Ideal Internalization State Depression .82** -.02 .24* -.20* .75 Note : Post-test BISS = Post-test Body Image States S cale; Post-test VAS Appe ar = Post-test Visual Analogue Scale Appearance; Post-tes t VAS Anger = Post-test Visual Anal ogue Scale Anger; Post-test VAS Self-Con = Post-test Visual Analo gue Scale Self-Confidence; Post-tes t VAS Anxiety = Post-test Visual Analogue Scale Anxiety; PT VAS Depress = Post-test VAS Depression; DEBQ-RS Inten = Dutch Eating Behavior Questionnaire-Restraint Scale Intentions; EDE-Q-B Inten = Eating Disorder ExaminationQuestionnaire-Bulimia subscale Intentions; MHB I-E = Multidimensional Health Behavior InventoryExercise subscale Intentions. *p<.05 **p<.01

PAGE 96

86 Additionally, trait appearance satisfaction modera ted the effect of ambiguous feedback on state self-confidence. The inte raction term (path c) for this model was significant, standardized = -.10, p <.05, when the pre-test self-confidence score, ambiguous feedback, and pre-test trait appearan ce satisfaction were controlled for. The R2 for the entire model was .81. Trait app earance comparison also moderated the relationship between ambiguous f eedback and intentions to use bulimic behaviors. This model had a significant interacti on term (path c), standardized = .20, p <.05, after controlling for pre-test bulimic behaviors, ambiguous feedback, and pre-test trait appearance comparison. All ot her tests for moderation resulted in non-significant interaction terms (see Appendix EE to HH). Overall, the findings from the regression analyses suggest that trait appearance satisfaction, trait appearance comparison, tra it appearance schematic ity, and trait thin ideal internalization all moderated the relationship between ambiguous feedback and post-test state depression. Appearance sa tisfaction and appearance comparison also each moderated an additional relationship betw een ambiguous feedback and state selfconfidence and between this feedback and bulim ic intentions, respectively. None of the hypothesized trait variab les moderated the effect of am biguous feedback and post-test variables reflecting body image (BISS and VAS a ppearance satisfaction). Exploratory Analyses Exploratory analyses were conducted on subsamples of individuals based on a median split on trait measures. This was done because of the unexpected findings using the full sample. As noted earlier, the ambi guous appearance-related condition had more positive outcomes (e.g., better body image, highe r self-confidence) compared to the

PAGE 97

87 ambiguous nonappearance-related condition. Th e initial hypotheses focusing on group differences (Hypothesis 1 and 2) were tested using these two subsamples. The following analyses were conducted for each subsample. A 2 (Feedback Condition: appearance, nonappear ance) X 2 (Disturbance Level: high trait level, low trait level) ANCOVA was run on state body image using the pre-test scores as the covariate. A 2 (Feedback Condition: appearance, nonappearance) X 2 (Disturbance Level: high trait level, low trait level) MANC OVA was performed on multiple state moods with pre-test scores as the covariates. Separate 2 (Feedback Condition: appearance, nonappearance) X 2 (Disturbance Level: high trait level, low trait level) ANCOVAs were run on intention behaviors using relevant pr e-test scores as the cova riate. The MANCOVAs and ANCOVAs indicated significant interacti ons between feedback condition and the subsample split into high versus low leve ls of trait appearance comparison. The subsamples based on high versus low levels of appearance satisfaction, appearance schematicity, teasing history, a nd thin-ideal internalization did not have any significant interaction effects. Only the findings fo r appearance comparison subsample will be reported. ANCOVAs and MANCOVAs for App earance Comparison Subsample There was a significant interaction betw een feedback condition and dispositional level of appearance comparison in the ANCOVA for state body image, F (1,141) = 5.39, p<.05, 2 = .037. Females with high trait a ppearance comparison had a more positive body image after receiving the ambiguous a ppearance-related feedback (adjusted M = 36.00) compared to those with high tra it appearance comparison in the ambiguous nonappearance-related feedback (adjusted M = 32.89) and those with low trait appearance

PAGE 98

88 comparison after receiving the ambiguous appearance-related condition (adjusted M = 33.58) and ambiguous nonappearance-related condition (adjusted M = 33.78). There was also a significant inte raction between fee dback condition and dispositional level of appearan ce comparison in the MANCOVA, F (5, 133) = 4.22, p=.001, 2 = 14. An examination of the ANCOV As indicated a significant condition effect for state anger, F (1, 137) = 9.88, p<.05, 2 = .067. The ambiguous appearancerelated feedback condition (adjusted M = 2.32) resulted in greater levels of anger than the ambiguous nonappearance-relate d feedback (adjusted M = 1.52). The mean trends for the interaction effect indicated that females w ith high trait appearance comparison had the greater levels of state ange r after receiving the ambiguous appearance-related feedback (adjusted M = 2.69) in comparison to those with high trait appearance comparison after receiving the ambiguous nonappearance-related feedback (adjusted M = 1.42) and those with low trait appearance comparison afte r receiving the ambiguous appearance-related feedback (adjusted M = 1.95) and nonappearance-related feedback (adjusted M = 1.62). A significant interaction between feedb ack condition and dispositional level of appearance comparison was found for state depression, F (1,137) = 8.64, p<.05, 2 = .059. Females with high trait appearance comparison had greater levels of state depression in the ambiguous nonappearance-related condition (adjusted M = 3.45) compared to those with high trait appearance comparison in the ambiguous appearance-related condition (adjusted M = 2.15) and those with low trait app earance comparison in the ambiguous appearance-related condition (adjusted M = 2.46) and in the ambiguous nonappearancerelated condition (adjusted M = 2.14).

PAGE 99

89 In addition, a significant interaction be tween feedback condition and dispositional level of appearance comparison was found in the ANCOVA for intentions to use bulimic behaviors, F (1, 141) = 6.51, p<.05, 2 = .044. Females with high trait appearance comparison had greater bulimic intentions in the ambiguous nonappearance-related condition (adjusted M = 3.19) than those with high tr ait appearance comparison in the ambiguous appearance-related condition (adjusted M = 2.69) and those with low trait appearance comparison in the ambiguous appearance-related condition (adjusted M = 2.90) and in the ambiguous nonappear ance-related condition (adjusted M = 2.78). There were no other significant interactions between feedback condition and dispositional level of appearance comparison for intentions to use body change strategies.

PAGE 100

90 Chapter 4 Discussion The main objective of the current study was to examine the immediate effects of ambiguous appearance-related feedback on state body image, mood, and intentions to use body change strategies. The ambiguous appearance-related condition was hypothesized to elicit more negative state body image and mood than the ambiguous nonappearancerelated feedback condition. Similarly, it was expected th at the ambiguous appearancerelated condition would produce greater intentions to use body change strategies, including dieting, bulimic, and exercise behaviors compared to the ambiguous nonappearance-related feedback condition. Additional objectives of the current study were to examine state appearance comparison as a mediator as well as a number of traits as moderators influencing the immediate responses to the feedback. Stat e appearance comparison was hypothesized to mediate the relationship between ambiguous feedback and state body image, mood, and intentions to use th e noted body change strategies. Tr ait appearance satisfaction, trait appearance schematicity, trait a ppearance comparison, trait thin -ideal internalization, and appearance-related teasing history were e xpected to each moderate the relationship among the ambiguous feedback and state body imag e, mood, and intentions to use certain body change strategies. Initial tr ait levels were also used to identify subsamples in order to explore the possibility of differences in responses to the ambiguous feedback as

PAGE 101

91 dependent not only on feedback condition but also on trait level of disturbance (high versus low). Most of the hypotheses were not suppor ted by the findings of the current study. In terms of body image, there was no eviden ce of significant diffe rences in state body image at post-test as expected. Post-test differences were marginally significant for a set of mood states (appearance sa tisfaction, anger, self-confidence, anxiety, depression). However, the mean trends for each mood state, with the exception of anger, indicated better mood states after the ambiguous appear ance-related feedback in comparison to after the ambiguous nonappearan ce-related feedback. In co ntrast to the other mood states, the finding for state anger at post-t est was in the predicted direction. As hypothesized, state anger was greater in the ambiguous a ppearance-related feedback condition than in the ambiguous nonappearan ce-related feedback condition. A trend reflecting this post-test difference in state anger was also found in the pilot study, suggesting distinct anger responses followi ng each type of feedback. The current findings are consistent with those of Tantleff-Dunn and Thompson (1998) who also found greater anger at post-test for participants exposed to a videotape of male to female interactions containing appear ance-related comments and behaviors compared to those exposed to a videotape with the same type of interactions but w ithout the appearancerelated feedback. In line w ith the current study, they we re attempting to demonstrate biased interpretation of app earance-related information. Appearance-related feedback seemed to have a more negative effect on state anger in comparison to the nonappearance-related feedback for both studies.

PAGE 102

92 The current study also found significant differences for intentions to diet and intentions to use bulimic be haviors at post-test. Contrary to hypotheses regarding intention behaviors, the ambiguous nonappear ance-related feedback elicited greater dieting and bulimic intentions compar ed to the ambiguous nonappearance-related feedback. Dieting and bulimic intentions were significantly lower in the ambiguous appearance-related condition. There were no significant differences for intentions to exercise at post-test. The examination of the ambiguous appear ance-related and nonappearance-related feedback in mediational and moderational an alyses did not fully support the hypotheses. Mediational analyses could not be performed testing state appearance comparison as a mediator for the effect of ambiguous feedb ack on state body image, mood, and intentions to diet, use unhealthy weight control me thods, and exercise. State appearance comparison was not significantly corr elated with ambiguous feedback (rpb = -.11, p>.05), which is a prerequisite for conducting medi ational analyses. The correlations among the predictor, mediator, and each outcome variable must all be significant, according to Baron and Kennys (1986) guidelines for es tablishing mediation. The lack of a relationship between state app earance comparison and ambiguous feedback indicated that mediational analyses should not be conducted. Moderational analyses were conducted for f our of the five proposed moderators. Appearance-related teasing history was not examined as moderator because it was not significantly correlated w ith ambiguous feedback (r = -.17, p<.05). A significant correlation between the predictor and moderato r is required for tests of moderation as noted by Baron and Kenny (1986). Modera tional analyses examined appearance

PAGE 103

93 satisfaction, appearance comparison, app earance schematicity, and thin-ideal internalization as potential m oderators of state body image, various moods, and intentions to use body change strategies (dieting, bulimic, and exercise behaviors). No moderators were established for state body image. Howeve r, all four traits were identified as moderators for the same mood state, depr ession. Trait appearan ce satisfaction, trait appearance comparison, trait appearance schematicity, and trait thin ideal internalization each moderated the relationship between ambiguous feedback and post-test state depression. Trait appearance satisfaction was also a moderator for ambiguous feedback and post-test self-confidence. In addition, trait appearance comparison moderated the relationship between ambiguous feedback and in tentions to use bulimic behaviors. All other moderational tests indicate d that other mood states or intention behaviors were not moderated by any traits.. Exploratory analyses were conducted on s ubsamples developed using high versus low levels of trait disturbance. It was hypot hesized that females with more problematic dispositional levels (high or low dependi ng on the trait) would respond to ambiguous appearance-related feed back in a more negative manner in terms of state body image, various moods, and intentions to use body ch ange strategies. Th ere were significant findings for the subsample based on dispositi onal levels of appearance comparison (high versus low trait levels). A number of these findings support and expand upon findings reported on the original sample. Most notabl y, there were post-test differences in state anger in the predicted direc tion. As in the original sample, the ambiguous appearancerelated condition elicited greater anger than the ambiguous nonappearance-related condition in this subsample. There were also post-test differences in state depression,

PAGE 104

94 state body image, and intentions to use bulim ic behaviors which were influenced by both feedback condition and dispositional level. Contrary to expe ctations, the levels of state depression and body image as well as intention behaviors were highest for females with high trait comparison in the ambiguous nonappearance-related condition. Nevertheless, these findings highlight appearance comparison as an important trait to consider and further investigate in the context of appearance-related feedback. It is important to try to understand th e possible reasons for the failure to find support for the bulk of hypotheses. The overa ll lack of significant findings in the expected directions, particularly for st ate body image and mood changes after the feedback, may reflect limitations of the feedb ack stimuli. Feedback regarding academic tutoring services was the control feedback selected to compare to appearance-related feedback on cosmetic surgery procedures because of its similar focus on a personal attribute (academic competen ce and physical appearance). The nonappearance-related feedback condition was con ceptualized as a neutral comparison group and was hypothesized to elicit little, if any, changes in state body image, moods, or intention behaviors. However, it appears that the nonappearance-related feedback was not received as lightly as expected. The feedback on tutoring services may have induced negative responses for individuals concerned with th eir academic performance esp ecially given that a school setting was used. For instance, the mean tre nds for the mood states indicated slightly lower self-confidence and greater anxiety and depression in the nonappearance-related condition. With regard to state body image, the mean trends for the two mood states specific to physical appearance showed lowe r body image and appearance satisfaction in

PAGE 105

95 this same condition. Although this finding may s eem counterintuitive, it is may related to the mood states that were found to be mo re negative for the nonappearance-related feedback condition. Feelings of depression, for instance, have been associated with poor body image in previous research (e.g., Mo ri & Morey, 1991, Noles, Cash, & Winstead, 1986). It has been argued that females with greater feelings of depression are often more likely to view themselves as less physically a ttractive than females with lower feelings of depression. The same may be true for females with lower feelings of self-confidence given the current societal em phasis on physical attractiveness as a determinant of female beauty. Females with more negative feelings in general may have a body image that is more vulnerable to external feedback from ot hers. In addition, the traits identified as moderators suggest other possible explanations for the unexpected findings. Trait levels of appearance satisfaction, appearance comp arison, appearance schematicity, and thin ideal internalization were each moderators fo r the ambiguous feedback and depression. It is likely that whether or not individuals react negatively to ambiguous feedback depends on particular dispositional levels. This seem s to be most relevant to trait appearance comparison as indicated by the significant find ings from the subsample with high versus low levels of trait appearance comparison. Addi tional research is need ed to elucidate the role of the identified moderators in this cont ext using feedback stimuli that has addressed the noted limitations. Furthermore, the appearance-related and nonappearance-related feedback was intended to be ambiguous in nature to allow for a neutral or negative (biased) interpretation. However, it is likely that the feedback was not similar in its level of ambiguity. In comparison to the appearance -related feedback, the nonappearance-related

PAGE 106

96 feedback may have been perceived as more relevant to the partic ipants (e.g., student status), possibly allowing for more nega tive interpretations and responses. The appearance-related feedback focused on procedures to alte r ones physical appearance which may not have been of any interest for so me participants. If so, this feedback may have served as a reminder of the extreme meas ures that some individuals take to address their appearance concerns. Rather than induce negative responses, the appearance-related feedback may have unintentionally made so me participants feel better on various affective domains, including body image. There are also other limitations that warra nt attention. The sample size for the study was relatively small and may have infl uenced the power to detect interaction effects. In regards to the sample itself, only female undergraduate students were included in the study with a majority of them of bei ng Caucasian. Future research should examine the hypotheses addressed in this study with samples consisting of males and non-college students as well as a proportionate number of ethnic groups. Additionally, an exclusion criterion related to previous eating disorder diagnosis or treatment was used to minimize any risk associated with receiving ambiguous appearance-related feedback during the study. However, this criterion may have restri cted the range of the sample in regard to disordered eating, and in turn, affected th e results by reducing a ny potential negative effects of the appearance-related feedback on individuals with higher levels of disordered eating. Research on more eating disturbed samples should be conducted. Finally, the study only used self-report measures which ma y not accurately reflec t actual behaviors, especially for eating behaviors. A further lim itation with regard to self-report measures is that a social desirability scale was not included in the study. Given the deception

PAGE 107

97 component and relatively sensit ive nature of the appearancerelated feedback, there is a strong possibility that participants responde d in a manner that they felt was socially acceptable. Future research should aim to more accurately assess actual behavioral change by using a social desirability scale in addition to the selected measures. In addition, the time period from when the confederate left the room after providing the feedback to when she returned was approximately five minutes which may have been too long and revealed the cove r story. Although an examination of the confederates credibility sugge sted high credibility, it is no t known whether participants were reporting their true beliefs regarding the cover story, esp ecially since credibility was evaluated after the debriefing. As state earlier, a social desira bility scale was not utilized and such a measure would be useful in dete rmining the extent to which the cover story, specifically the confederates role and feedback, was accepte d as genuine. It is also important to note that the script used by the confederates was devel oped to be brief and straightforward with little to no discussion between the confed erate and the participant. There is a possibility that th e appearance-related feedback, in particular, was too short or not detailed enough to allow for negative in terpretations and effects on body image and mood states. Therefore, it is unclear whether the length of the feedback affected the results. Despite the minimal support for the hypotheses, future research should further investigate the effects of appearancerelated feedback on body image and eating behaviors in an ambiguous setting. Previous research has shown that both verbal and more subtle forms of appearance-related f eedback are common and associated with body image problems and disordered eating (Thom pson et al., 1999; Tantleff-Dunn & Gokee,

PAGE 108

98 2004). An examination of this type of f eedback in an ambiguous context can provide useful information with regard to errors in processing appearance-related information that likely exist among individuals with poor body image as argued by Williamson et al. (2004). A better understanding of the different ways that ambiguous appearance-related information, such as feedback provided by othe rs, can be interpreted is likely to have important clinical implications. Additional research in this area may potentially indicate which cognitive strategies and approaches should be utilized in treatments for body image and eating disturbances allowing fo r more empirically-based treatments.

PAGE 109

99 References Allison, D.B., Kalinsky, L.B., & Gorman, B.S. (1992). A comparison of the psychometric properties of three m easures of dietary restraint. Psychological Assessment, 4, 391-198. Akan, G. E., & Grilo, C. M. (1995). Sociocul tural influences on eating attitudes and behaviors, body image, and psychological functioning: A comparison of AfricanAmerican, Asian-American, and Caucasian college women. International Journal of Eating Disorders 18 181-187. Black, C.M.D. & Wilson, G.T. (1996). Assessmen t of eating disorders: Interview versus questionnaire. International Journal of Eating Disorders, 20 43-50. Baker, J. D., Williamson, D. A., & Sylve, C. (1995). Body-image disturbance, memory bias and body dysphoric: Effects of a negative mood induction. Behavior Therapy 26, 747-759. Baron, R. M., & Kenny, D. A. (1986). The Moderator-Mediator variab le distinction in social psychological research: Conceptual, strategic, and statis tical considerations. Journal of Personality and Social Psychology 51 1173-1182. Befort, C. A., & Rickard, K. M. (2003). Gende r similarities in response to figure-size feedback in a selected nonclinical population. Sex Roles 49, 71-80. Birkeland, R., Thompson, J. K., Herbozo, S ., Roehrig, M., & Cafri, G. (2005). Media exposure, mood, and body image dissatisfactio n: A test of social comparison and information processing paradigms. Body Image: An International Journal of Research 2, 53-61. Bower, G. H. (1981). Mood and memory. American Psychologist 36, 129-148. Brown, T. A., Cash, T. F., & Lewis, R. J. ( 1989). Body-image disturbanc es in adolescent female binge-purgers: A brief report of th e results of a national survey in the U.S.A. Journal of Clinical Psychology and Psychiatry, 30, 605-613. Brown, T. A., Cash, T. F., & Mikulka, P. J. (1990). Attitudinal body-image assessment: Factor analysis of the Body Self-Relations Questionnaire. Journal of Personality Assessment 55, 135-144.

PAGE 110

100 Cash, T. F. (1995). Developmental teasing about physical appearance: Retrospective descriptions and relationships with body image. Social Behavior and Personality 23, 123-130. Cash, T. F., & Deagle, E. A. (1997). The nature and extent of body-imag e disturbances in anorexia nervosa and bulimia nervosa: A meta-analysis. International Journal of Eating Disorders 21, 2-19. Cash, T. F., & Flemming, E. C., (2002). Body imag es and social relations. In T. F. Cash & T. Pruzinsky (Eds.), Body image: A handbook of theory, research, and clinical practice (pp. 277-286). New York: Guilford Press. Cash, T. F., Fleming, E. C., Alindogan, J ., Steadman, L., & Whitehead, A. (2002). Beyond body image as a trait: The development and validation of the Body Image States Scale. Eating Disorders: The Journal of Treatment and Prevention 10, 103-113. Cash, T. F., Melnyk, S. E., & Hrabosky, J. I. (2004). The as sessment of body image investment: An extensive revision of the Appearance Schema Inventory. International Journal of Eating Disorders 35, 305-316. Cash, T. F., Winstead, B. A., & Janda, L. H. (1986). Body Image Survey report: The great American shape-up. Psychology Today 24, 30-37. Cattarin, J. and Thompson, J. K. (1994). A th ree-year longitudinal study of body image, eating disturbance, and general psychologi cal functioning in adolescent females. Eating Disorders: Treatment and Prevention, 2, 114-125. Cooper, M. J., Anastasiades, P., & Fairbur n, C. G. (1992). Selec tive processing of eating-, shape-, and weight-related word s in persons with bulimia nervoas. Journal of Abnormal Psychology 101, 352-355. Cooper, M. J., & Fairburn, C. G. (1992). Se lection processing of eating, weight, and shape related words in patterns with eating disorders and dieters. British Journal of Clinical Psychology, 31 363-365. Davidson, E. J., & Wright, P. (2002). Select ive processing of shape and weight-related words in bulimia nervosa: Use of a computerized Stroop test. Eating Behaviors, 3, 261-273. Denniston, C., Roth, D., & Gilroy, F. (1992) Dysphoria and body image among college woman. International Journal of Eating Disorders 12, 449-452.

PAGE 111

101 Eisenberg, M. E., Neumark-Sztainer, D., & St ory M. (2003). Associations of weightbased teasing and emotional well-being among adolescents Archives of Pediatrics and Adol escent Medicine, 157, 733-738. Fabian, L. J., & Thompson, J. K. (1989). Body image and eating disturbance in young females. International Journal of Eating Disorders, 8, 63-74. Garner, D. M. (1997, January/February). The Body Image Survey. Psychology Today 32-84. Garrow, J. S., & Webster J. (1985). Quetelet's index (W/H2) as a measure of fatness. International Journal of Obesity 9, 147-153. Gleason, J.H., Alexander, A. M., & Somers, C. L. (2000). Later adolescents reactions to three types of childhood teasing: Relati ons with self-esteem and body image. Social Behavior and Personality, 28 471-480. Grilo, C. M., Wilfley, D. E., Brownwell, K. D., & Rodin, J. (1994). Teasing, body image, and self-esteem in a clinical sample of obese women. Addictive Behavior, 19, 443-450. Fabian, L. J., & Thompson, J. K. (1989). Body image and eating disturbance in young females. International Journal of Eating Disorders, 8, 63-74. Fairburn, C.G. & Beglin, S.J. (1994). Assessmen t of eating disorders: Interview or selfreport questionnaire? International Journal of Eating Disorders, 16 (4), 363-370. Fairburn, C.G. & Cooper, Z. (1993). The Eati ng Disorder Examination (12th edition). In C. Fairburn & G. Wilson (Eds.), Binge Eating: Nature, assessment, and treatment (pp. 317-360). New York: Guildford Press. Fairburn, C. G., Cooper, P. J., Cooper, M. J ., McKenna, F. P., & Anastasiades, P. (1991). Selective information processing in bulimia nervosa. International Journal of Eating Disorders 10, 415-422. Festinger, L. (1954). A theory of social comparison processes. Human Relations, 7, 117-140. Fuller, R. D., Williamson, D. A., & Anders on, T. W. (1995). Selective information processing of body size and food related stimuli in women who are preoccupied with body size. Advances in Health Care Research 14, 61-66.

PAGE 112

102 Furman, K., & Thompson, J. K. (2002). B ody image, teasing, and mood alterations: An experimental study of exposure to negative verbal commentary. International Journal of Eating Disorders, 32, 449-457. Garner, D. M. (1997, January/February). The Body Image Survey. Psychology Today 32-84. Garner, D. M. (2005). Eating disorder inventory-3: Professional manual. Odessa, FL: Psychological Assessment Resources. Gleason, J.H., Alexander, A. M., & Somers, C. L. (2000). Later adoles cents reactions to three types of childhood teasing: Rela tions with self-esteem and body image. Social Behavior and Personality, 28 471-480. Gleaves, D. H., Williamson, D. A., Eberenz, K. P., Sebastian, S. E., & Barker, S. E. (1995). Testing a multidimensional model of body image construct using structural modeling. Journal of Personality Assessment, 64, 478-493. Heinberg, L. J., & Thompson, J. K. (1992). The e ffects of figure size feedback (positive vs. negative) and target comparison group (particularistic vs. universalistic) on body image disturbance. International Journal of Eating Disorders, 12 441448. Heinberg, L. J., & Thompson, J. K. (1995). Body image and televised images of thinness and attractiveness: A contro lled laboratory investigation. Journal of Social and Clinical Psychology, 14 325. Herbozo, S., & Thompson, J.K. (2006a). Development and validation of the Verbal Commentary on Physical Appearance Scale: Considering both positive and negative commentary. Body Image: An International Journal of Research 3, 335344. Herbozo, S., & Thompson, J.K. (2006b) A ppearance-related commentary, body image, and self-esteem: Does the distress as sociated with the commentary matter? Body Image: An International Journal of Research 3, 255-262. Hintze, J. (2001). NCSS and PASS. Number cruncher statis tical systems. Kaysville, UT. www.ncss.com. Irving, L. M. (1990). Mirror images: Effects of the standard of beauty on the selfand body-esteem of women exhibiting vary ing levels of bulimic symptoms. Journal of Social and Clinical Psychology ,9, 230-242.

PAGE 113

103 Jackman, L. P., Williamson, D. A., Netemeyer, R. G., & Anderson, D. A. (1995). Do weight pre-occupied wome n misinterpret ambiguous stim uli related to body size? Cognitive Therapy and Research 19 341-355. Jackson, T. D., Grilo, C. M., & Masheb, R. M. (2002). Teasing history and eating disorder features: An ageand body ma ss index-matched comparison of bulimia nervosa and binge-eating disorder. Comprehensive Psychiatry, 43 108-113. Kenny, D.A. (2006). Mediation Retrieved March 5, 2006 from http://davidakenny.net/cm/mediate.htm. Kraemer, H.C., Wilson, G.T., Fairburn, C.G., & Agras, W.S. (2002). Mediators and moderators of treatment effects in randomized clinical trials. Archives of General Psychiatry, 59 877-884. Kruglanski, A. W., & Mayseless, O. (1990). Classic and current social comparison research: Expanding the perspective. Psychological Bulletin 108, 195-208. Kulbok, P.A., Carter, K.F., Baldwin, J.H., G ilmartin, M.J., & Kirkwood, B. (1999). The multidimensional health behavior inventory. Journal of Nursing Measurement, 7, 177-195. Lang, P. J. (1984). Cognition in emotion: Concep t and action. In C. Izard, J. Kagan, & R. Zajorc (Eds.), Emotion, cognition, and behavior (pp. 192-226). New York: Cambridge University Press. Long, C. G., Hinton, C., & Gillespie, N. K ., (1994). Selective processing of food and body size words: Application of the Str oop test with obese restrained eaters, anorexics, and normals. International Journal of Eating Disorders 15, 279-283. Lunner, K., Werthem, E., Thomspon, J. K., Paxton, S. J., McDonald, F., & Halvaarson, K. S (2000). A cross-cultural examination of weight-related te asing, body image, and eating disturbance in Swed ish and Australian samples. International Journal of Eating Disorders, 28 430-435. Lyubomirsky, S. & Nolen-Hoeksema, S. ( 1993). Self-perpetua ting properties of dysphoric rumination. Journal of Personality and Social Psychology, 65, 330349. Lyubomirsky, S. & Nolen-Hoeksema, S. (1995). Effects of self-focused rumination on negative thinking and interp ersonal problem solving. Journal of Personality and Social Psychology, 69, 176-190.

PAGE 114

104 Major, B., Testa, M., & Blysma, W. H. (1991). Responses to upward and downward social comparisons: The impact of esteem -relevance and perceived control. In J. Suls & T. A. Ashby (Eds.), Social comparison: Contemporary theory and research (pp. 237-260). Hillsdale, NJ: Erlbaum. Markus, H. (1977). Self-schemata and pr ocessing information about the self. Journal of Personality and Social Psychology 35 63-78. Matz, P. E., Foster, G. D., Faith, M. S ., & Wadden, T. A. (2002). Correlates of body image dissatisfaction among overwei ght women seeking weight loss. Journal of Consulting and Clinical Psychology, 70 1040-1044. Mautner, R. D., Owen, S. V., & Furnham, A. (2000). Cross-cultural explanations of body image disturbance in western cultural samples. International Journal of Eating Disorders, 28 165-172. McKenzie, S. J., Williamson, D. A., & Cubic, B. A. (1993). Stable and reactive body image disturbances in bulimia nervosa. Behavior Therapy, 24, 195-207. McLaren, L., Kuh, D., Hardy, R., & Gauvi n, L. (2004). Positive and negative bodyrelated comments and their relationshi p with body dissatisfaction in middle-aged women. Psychology and Health 19, 261-272. Morrow, J. & Nolen-Hoeksema, S. (1990). Effects of response to depression on the remediation of depressive affect. Journal of Personality and Social Psychology, 58, 519-527. Neumark-Sztainer, D., Faulkner, N., Story, M ., Perry, C., Hannan, P. J., & Mulert, S. (2002). Weight-teasing among adolescents: Co rrelations with weight status and disordered eating behaviors. International Journal of Obesity and Related Metabolic Disorders, 26, 123-131. Noles, S. W., Cash, T. F., & Winstead, B. (1986). Body image, physical attractiveness, and depression. Journal of Consulting and Clinical Psychology, 53, 88-94. Perprina, C., Hemsely, D., Treasure, J., & de Si lva, P (1995). Is the selective information processing of food and body words specifi c to patients with eating disorders. International Journal of Eating Disorders 14, 359-366. Perrin, L. (1995). Interpretation of ambiguous body-r elated stimuli in the eating disorders Unpublished doctoral dissertation, Lo uisiana State University, Baton Rouge.

PAGE 115

105 Powell, M. R., & Hendricks, B. (1999). B ody schema, gender, and other correlates in nonclinical populations, Genetic, Social, and General Psychology Monographs 125, 333-412. Richins, M. L. (1991). Social comparison and the idealized imag es of advertising. Journal of Consumer Research ,18, 71-83. Rieves, L., & Cash, T. F. (1996). Social developmental factors and womens body-image attitudes. Journal of Social Behavior and Personality, 11 63-78. Rosen, James C. (1992). Body-image diso rder: Definition, development, and contribution to eating disorders. In J. H. Crowther, D.L., Tennenbaum, S.E. Hobfoll & M. A. P Stephens (Eds.), The etiology of bulimia: The individual and familial context pp. 155-177. Washington, D.C.: Hemisphere Publishing Corporation. Schwartz, D. J., Phares, V., TantleffDunn, S., & Thompson, J. K. (1999). Body image, psychological functioning, and parental f eedback regarding physical appearance. International Journal of Eating Disorders, 25 339-343. Schotte, D. E., McNally, R. J., & Turner, M. L., (1990). A dichotic listening analysis of body weight concern in bulimia nervosa. International Journal of Eating Disorders, 9 109-113. Sebastian, S. B., Williamson, D. A., & Blouin, D. C. (1996). Memory bias for fatness stimuli in the eating disorders. Cognitive Therapy and Research 20, 275-286. Shroff, H., & Thompson, J. K. (2004). Body image and eating disturbance in India: Media and interpersonal influences International Journal of Eating Disorders 35, 198-203. Stice, E. (2001). Risk factors for eating pathology: Recent advances and future directions. In R. Striegel-Moore & L. Smolak (Eds.), Eating disorders: New direction for future research and practice (pp. 57-73). Washington, DC: American Psychological Association. Stormer, S. M. & Thompson, J. K. (1996). Explanations of Body Image Disturbance: A Test of Maturational Status, Ne gative Verbal Commentary, Social Comparison, and Sociocultural Hypotheses. International Journal of Eating Disorders, 19 193-202. Tantleff-Dunn, S., & Gokee. J. (2002). Interp ersonal influences on body image. In T. F. Cash & T. Pruzinsky (Eds.), Body image: A handbook of theory, research, and clinical practice (pp. 108-116). New York: Guilford Press.

PAGE 116

106 Tantleff-Dunn, S., & Thompson, J. K. (1998) Body image and appearance-related feedback: Recall, judgment, and affective response Journal of Social and Clinical Psychology, 17 319-340. Tantleff-Dunn, S., Thompson, J. K., & Dunn, M. F. (1995). Development and validation of the Feedback on Physical Appearance Scale (FOPAS). Eating Disorders: The Journal of Treatment and Prevention 3, 341-350. Thompson, J. K. (1990). Body image disturbance: Assessment and treatment Elmsford, NY: Pergamon Press. Thompson, J. K. (1991). Body shape preferences: Effects of instructional protocol and level of eating disturbance. International Journal of Eating Disorders, 10, 193198. Thompson, J.K. (1992). Body image: Extent of disturbance, associated features, theoretical models, assessment methodol ogies, intervention strategies, and a proposal for a new DSM-IV diagnostic category Body Image Disorder. In M. Hersen, R. M. Eisler, and P. M. Miller (Eds.), Progress in Behavior Modification (Vol. 28, pp. 3-54). Sycamore, IL: Sycamore Press. Thompson, J.K. (1995). Assessment of body image. In D. Allison (Ed.), Handbook of assessment methods for eating beha viors and weight-related problems (pp. 119148). Thousand Oaks, CA: Sage. Thompson, J.K., & Altabe, M. N. (1991). Psyc hometric properties of the Figure Rating Scale. International Journal of Eating Disorders, 10, 615-619. Thompson, J. K., Coovert, M. D., & Stormer, S. (1999). Body image, social comparison, and eating disturbance: A covariance structure modeling investigation. International Journal of Eating Disorders, 26 43-51. Thompson, J. K., Coovert, M. D., Richards, K. J., Johnson, S., and Cattarin, J. (1995). Development of body image, eating dist urbance, and general psychological functioning in female adolescents: Covariance structure modeling and longitudinal investigations. International Journal of Eating Disorders, 18 221236. Thompson, J. K., Fabian, L. J., Moulton, D. O., Dunn, M. E., & Altabe, M. N. (1991). Development and Validation of the Physical Appearance Related Teasing Scale. Journal of Personality Assessment, 56 (3), 513-521. Thompson, J. K., & Heinberg, L. J, (1993). Preliminary test of two hypotheses of body image disturbance. International Journal of Eating Disorders, 14, 59-63.

PAGE 117

107 Thompson, J. K., Heinberg, L., Altabe, M. N., & Tantleff-Dunn, S. (1999). Exacting Beauty: Theory, Assessment, and Treatment of Body Image Disturbance Washington: American Psychological Association. Thompson, J. K., & Heinberg, L. J, & Tantleff-Dunn, S. (1991). The Physical Appearance Comparison Scale (PACS). The Behavior Therapist 14, 174. Thompson, J. K., & Psaltis, K. (1988). Mult iple aspects and correlates of body figure ratings: A replication and extens ion of Fallon and Rozin (1985). International Journal of Eating Disorders, 7, 813-818. Thompson, J. K., & Stice, E. (2001). Thin-ide al internalization: Mounting evidence for a new risk factor for body-image disturbance and eating pathology. Current Directions in Psychological Science, 10 (5), 181-183. Thompson, J.K., van den Berg, P, Roehrig, M ., Guarda, A.S., & Heinberg, L.J. (2004). The sociocultural attitudes towards appearance questionnaire-3 (SATAQ-3): Development and validation. International Journal of Eating Disorders, 35, 293304. Tiggemann, M., & McGill, B. (2004). The role of social comparison in the effect of magazine advertisements on wo mens mood and body dissatisfaction. Journal of Social and Clinical Psychology 23, 23-44. Tiggemann, M., & Slater, A. (200). Thin ideals in music television: A source of social comparison and body dissatisfaction. International Journal of Eating Disorders 35, 48-58. Wardle, J., Waller, J, & Fox, E. (2002). Age of onset and body dissati sfaction in obesity. Addictive Behaviors, 27, 561-573. Watkins, P. C., Martin, C., Muller, S., & Da y, S. K. (1995). Cognitive biases associated with feelings of fatness: Unhea lthy responses to healthy messages. Advances in Health Care Research 14 67-73. Wiederman, M. W. (2002). Body image and sexual functioning. In T. F. Cash & T. Pruzinsky (Eds.), Body image: A handbook of theory, research, and clinical practice (pp. 287-294). New York: Guilford Press. Williamson, D. A. (1990). Assessment of eating disord ers: Obesity, bulimia, and anorexia. New York: Pergamon. Williamson, D. A. (1996). Body image disturbance in eating disorders: A form of cognitive bias?, Eating Disorders, 4 47-58.

PAGE 118

108 Williamson, D. A., Gleaves, D. H., Watkins, P. C., & Schlundt, D. G. (1993). Validation of a self-ideal body size discrepancy as a measure of body size dissatisfaction. Journal of Psychopathology and Behavioral Assessment 15, 57-68. Williamson, D. A., Muller, S. L., Reas, D. L., & Thaw, J. M. (1999). Cognitive bias in eating disorders: Implications for theory and treatment. Behavior Modification 23, 556-577. Williamson, D. A., Stewart, T. M., White, M. A., & York-Crowe, E. (2004). An information processing perspective on body image. In T. F. Cash & T. Pruzinsky (Eds.), Body image: A handbook of theory, re search, and clinical practice (pp. 47-54). New York: Guilford Press. Wood, J. V. (1989). Theory and research concerning social comparison of personal attributes. Psychological Bulletin 106 231-248. van den Berg, P., Wertheim, E. H., Thom pson, J. K., & Paxton, S. J. (2002). Development of body image, eating dist urbance, and general psychological functioning in adolescent females: A re plication using covariance structure modeling in an Australian sample. International Journal of Eating Disorders 32, 46-51. van Strien, T., Frijters, J.E., Bergers, G.P ., & Defares, P.B. (1986). The Dutch Eating Behavior Questionnaire (DEBQ) for assessment of restrained, emotional, and external eating behavior. International Journal of Eating Disorders, 5 (2), 295315. Vitousek, K. B., & Hollon, S. D. (1990). Th e investigation of sc hematic content and processing in eating disorders. Cognitive Therapy and Research 14, 191-214.

PAGE 119

109 Appendices

PAGE 120

110 Appendix A: Demographic Information Thank you for participating in this study. Please read the directi ons for each group of questions and answer each one to the best of your ability. DEMOGRAPHIC INFORMATION Age: ____________ Height: __________ Weight: __________ Race/Ethnicity: (p lease circle one): Asian-American African-American Caucasian Hispanic Other: Please specify _______________________ Year in School: (p lease circle one) Freshman Sophomore Junior Senior Other: Please specify _______________________

PAGE 121

111 Appendix B: Multidimensional Body -Self Relations QuestionnaireAppearance Evaluation Subscale Using the scale below, please circle the number that best matches your agreement with the following statements. Definitely Disagree Mostly Disagree Neither Agree Nor Disagree Mostly Agree Definitely Agree 1 2 3 4 5 1. My body is sexually appealing. 1 2 3 4 5 2. I like my looks just the way they are. 1 2 3 4 5 3. Most people would consider me good-looking. 1 2 3 4 5 4. I like the way I look without my cl othes on. 1 2 3 4 5 5. I like the way my clothes fit me. 1 2 3 4 5 6. I dislike my physique. 1 2 3 4 5 7. I am physically unattractive. 1 2 3 4 5

PAGE 122

112 Appendix C: Body Image States Scale For each of the items below, check the box beside the one statement that best describes how you feel RIGHT NOW AT THIS VERY MOME NT. Read the items carefully to be sure the statement you choose accurately and ho nestly describes how you feel right now. 1. Right now I feel Extremely dissatisfied with my physical appearance Mostly dissatisfied with my physical appearance Moderately dissatisfied with my physical appearance Slightly dissatisfied with my physical appearance Neither dissatisfied nor satisfied with my physical appearance Slightly satisfied with my physical appearance Moderately satisfied with my physical appearance Mostly satisfied with my physi cal appearance Extremely satisfied with my physical appearance 2. Right now I feel Extremely dissatisfied with my body size and shape Mostly dissatisfied with my body size and shape Moderately dissatisfied with my body size and shape Slightly dissatisfied with my body size and shape Neither dissatisfied nor satisfied with my body size and shape Slightly satisfied with my body size and shape Moderately satisfied with my body size and shape Mostly satisfied with my body size and shape Extremely satisfied with my body size and shape 3. Right now I feel Extremely dissatisfied with m y weight Mostly dissatisfied with my weight Moderately dissatisfied with my weight Slightly dissatisfied with my weight Neither dissatisfied nor satisfied with my weight Slightly satisfied with my weight Moderately satisfied with my weight Mostly satisfied with my weight Extremely satisfied with my weight

PAGE 123

113 Appendix C (Continued) 4. Right now I feel Extremely physically attractive Very physically attractive Moderately physically attractive Slightly physically attractive Neither attractive nor unattractive Slightly physically unattractive Moderately physically unattractive Very physically unattractive Extremely physically unattractive 5. Right now I feel A great deal worse about my looks than I usually feel Much worse about my looks than I usually feel Somewhat worse about my looks than I usually feel Just slightly worse about my looks than I usually feel About the same about my looks as usual Just slightly better about my looks than I usually feel Somewhat better about my looks than I usually feel Much better about my looks than I usually feel A great deal better about my looks than I usually feel 6. Right now I feel that I look A great deal better than the average person looks Much better than the average person looks Somewhat better th an the average person looks Just slightly better than the average person looks About the same as the average person looks Just slightly worse than the average person looks Somewhat worse than the average person looks Much worse than the average person looks A great deal worse than the average person looks

PAGE 124

114 Appendix D: Visual Analog Scales Instructions: Place a mark through the area of the line that matches your current level of feeling for the following emotions: 1. Anxiety None Extreme 2. Depression None Extreme 3. Satisfaction with Overall Appearance None Extreme 4. Anger None Extreme 5. Self-Confidence None Extreme

PAGE 125

115 Appendix E: Appearance Schema Inventory-Revised Short Form The statements below are beliefs that people may or may not have about their physical appearance and the influence of appearance on life. Decide the extent to which you personally disagree or agree with each statement and enter a number from 1 to 5. There are no right or wrong answers. Just be truthful about yo ur personal beliefs. _____ 1. I spend litt le time on my physical appearance. _____ 2. When I see good-looki ng people, I wonder about how my own looks measure up. _____ 3. I try to be as physically attrac tive as I can be. _____ 4. I have never pa id much attention to what I look like. _____ 5. I seldom compare my a ppearance to that of other people I see. _____ 6. I often check my appearance in a mirror just to make sure I look okay. _____ 7. When something makes me feel good or ba d about my looks, I tend to dwell on it. _____ 8. If I like how I look on a gi ven day, its easy to feel happy about other things. _____ 9. If somebody had a negative reaction to what I l ook like, it wouldnt bother me. _____ 10. When it comes to my physical appearance, I have high standards. _____ 11. My physical appearance has had little influence on my life. _____ 12. Dressing well is not a priority for me _____ 13. When I meet people for the first tim e, I wonder what they think about how I look. 1 2 3 4 5 Strongly Disagree Mostly Disagree Neither Agree or Disagree Mostly Agree Strongly Agree

PAGE 126

116 Appendix E (Continued) _____ 14. In my everyday life, lots of thi ngs happen that make me think about what I look like. _____ 15. If I dislike how I look on a given day, its hard to feel happy about other things. _____ 16. I fantasize about what it woul d be like to be better looking than I am. _____ 17. Before going out, I make sure that I look as good as I possibly can. _____ 18. What I look like is an important part of who I am. _____ 19. By controlling my appearance, I can control many of the social and emotional events in my life. _____ 20. My appearance is responsible for much of whats happened to me in my life. 1 2 3 4 5 Strongly Disagree Mostly Disagree Neither Agree or Disagree Mostly Agree Strongly Agree

PAGE 127

117 Appendix F: Sociocultural Attit udes Towards Appearance Scale-3Internalization-General subscale Using the scale below, please write the numbe r that best matches your agreement with the following statements. Definitely Disagree Mostly Disagree Neither Agree Nor Disagree Mostly Agree Definitely Agree 1 2 3 4 5 1. __________ I would like my body to look like the people who are on TV. 2. __________ I compare my body to the bodies of TV and movie stars. 3. __________ I would like my body to look like the models who appear in magazines. 4. __________ I compare my appearance to the appearance of TV and movie stars. 5. __________ I would like my body to l ook like the people who are in movies. 6. __________ I compare my body to the bodies of people who appear in magazines 7. __________ I wish I looked like the models in music videos. 8. __________ I compare my appearan ce to the appearance of people in magazines. 9. __________ I try to look like the people on TV.

PAGE 128

118 Appendix G: Physical A ppearance Comparison Scale Using the scale below, please circle the num ber that best matches your agreement with the following statements. Never Seldom Sometimes Often Always 1 2 3 4 5 1. At parties or other social events, I compare my physical appearance to the physical appearance of others. 1 2 3 4 5 2. The best way for a person to know if they are overweight or underweight is to compare their figure to the figure of others. 1 2 3 4 5 3. At parties or other soci al events, I compare how I Am dressed to how other people are dressed 1 2 3 4 5 4. Comparing your looks to the looks of others is a bad way to determin e if you are attractive or unattractive. 1 2 3 4 5 5. In social situations, I sometimes compare my figure to the figures of other people. 1 2 3 4 5

PAGE 129

119 Appendix H: State Appearance Comparison Scale In the past fifteen minutes to what extent did you 1. Think about your own appearance? No thought A lot of thought about my appearanceabout my appearance 1 2 3 4 5 6 7 2. Compare your overall appearance to that of th e other research partic ipant in the study? No comparison.......A lot of comparison 1 2 3 4 5 6 7 3. Compare your specific body parts to those of the other research pa rticipant in the study? No comparison...A lot of comparison 1 2 3 4 5 6 7

PAGE 130

120 Appendix I: Physical Appear ance-Related Teasing Scale Each question pertains to the time period of when you were growing up. Please respond by circling the appropriate number for the follo wing scale: Never (1), Frequently (5). 1. When you were a child, did you feel th at your peers were Never Frequently staring at because you were overweight? 1 2 3 4 5 2. When you were a child, did you ever f eel like people were Never Frequently making fun of you because of your weight? 1 2 3 4 5 3. Were you ridiculed as a child about being overweight? Never Frequently 1 2 3 4 5 4. When you were a child, did people make jokes about you Never Frequently being too big? 1 2 3 4 5 5. When you were a child, were you laughed at for trying Never Frequently out for sports because you were too heavy? 1 2 3 4 5 6. Did your brother(s) or other male relative s call you Never Frequently names like fatso when they got angry at you? 1 2 3 4 5 7. Did your father ever make jokes that re ferred to your Neve r Frequently weight? 1 2 3 4 5 8. Did other kids call you deroga tory names that related to Ne ver Frequently your size or weight? 1 2 3 4 5 9. Did you ever feel like people were pointing at you Never Frequently because of your size or weight? 1 2 3 4 5 10. Were you the brunt of family jokes because of your Never Frequently weight? 1 2 3 4 5 11. Did people point you out of a crowd because of your Never Frequently weight? 1 2 3 4 5 12. Did you ever hear your classmate snicker when you Never Frequently walked into the classroom alone? 1 2 3 4 5 13. When you were growing up, did people say you Never Frequently dressed funny? 1 2 3 4 5

PAGE 131

121 Appendix I (Continued) 14. Did people say you had funny teeth? Never Frequently 1 2 3 4 5 15. Did kids call you funny looking? Never Frequently 1 2 3 4 5 16. Did other kids tease you about wearing clothes that didnt Never Frequently match or were out of style? 1 2 3 4 5 17. Did other kids ever make jokes about your hair? Never Frequently 1 2 3 4 5 18. When you were a child were you scoffed at for l ooking Never Frequently like a weakling? 1 2 3 4 5

PAGE 132

122 Appendix J: Dutch Eating Behavi or Questionnaire-Restraint Scale Circle the best response to describe your usual behavior : Never Seldom SometimesOften Always 1. Did you eat less than you normally would to lose weight? 1 2 3 4 5 2. Did you try to eat less at meal times than you would like to eat? 1 2 3 4 5 3. How often did you refuse food or drink because you were concerned about your weight? 1 2 3 4 5 4. Did you watch exactly what you ate? 1 2 3 4 5 5. Did you deliberately eat foods that were slimming? 1 2 3 4 5 6. If you ate too much, did you eat less than usual the next day? 1 2 3 4 5 7. Did you deliberately eat less in order not to become heavier? 1 2 3 4 5 8. How often did you try not to eat between meals because you were watching your weight? 1 2 3 4 5 9. How often in the evenings did you try not to eat because you were watching your weight? 1 2 3 4 5 10. Did you take into account your weight in deciding what to eat? 1 2 3 4 5

PAGE 133

123 Appendix K: Modified Dutch Eating Be havior Questionnaire-Restraint Scale Circle the best response to describe the behaviors you intend to engage in : Strongly Disagree Somewhat Disagree Neither Agree Nor Disagree Somewhat Agree Strongly Agree 1. Do you plan to eat less than you normally would to lose weight? 1 2 3 4 5 2. Do you plan to eat less at mealtimes than you would like to eat? 1 2 3 4 5 3. Do you plan to refuse food or drink to lose weight? 1 2 3 4 5 4. Do you plan to watch exactly what you eat? 1 2 3 4 5 5. Do you plan to deliberately eat foods that are slimming? 1 2 3 4 5 6. If you overeat one day, do you plan to eat less than usual the next day? 1 2 3 4 5 7. Do you plan to deliberately eat less in order to not become heavier? 1 2 3 4 5

PAGE 134

124 Appendix L: Eating Disorder Examination-Questionnaire-Bulimia subscale Please circle the response that descri bes your behavior over the past week : No 1 2 3 4 5 6 7 On how many days during the past week ... days days days days days days days days 1. Have you felt fat?...........................................0 1 2 3 4 5 6 7 2. Have you had a definite fear that you might gain weight or become fat?..................0 1 2 3 4 5 6 7 Over the past week ... Not at all Slightly Moderately Extremely 3. Has your weight influenced how you think about (judge) yourself as a person ?................0 1 2 3 4 5 6 4. Has your shape influenced how you think about (judge) yourself as a pers on?................0 1 2 3 4 5 6 5. During the past week have there been times when you felt you have eaten what other people would regard as an unusually large amount of food given the circumstances? YES NO 6. During the times when you ate an unus ually large amount of food, did you experience a loss of control i.e. feel you couldn't stop eating or control what or how much you were eating? YES NO 7. How many times during the past week have you eaten an unusually large amount of food and experienced a loss of control?____________ (please write in number or indicate zero) 8. During the past week have you had other times where you felt you uncontrollably ate a large amount of food, but the amount eaten would not have been considered large by most people? YES NO 9. How many times during the past week have you uncontrollably eaten a large amount of food that others might not consider large?________________ (please write in number or indicate zero) 10. How many times during the past week have you made yourself sick in order to prevent weight gain or counteract the effects of eating?________________ (write in number or indicate zero) 11. How many times during the past week have you used laxatives or diuretics in order to prevent weight gain or counteract the effects of eating?__________ (write in number or indicate zero)

PAGE 135

125 Appendix L (Continued) 12. How many times during the past week have you engaged in excessive exercise specifically for the purpose of counteracting overeating episodes?_______________ (write in number or indicate zero)

PAGE 136

126 Appendix M: Modified Eati ng Disorder Examination-Questionnaire-Bulimia subscale Circle the best response to describe the behaviors you intend to engage in: Strongly Disagree Somewhat Disagree Neither Agree Nor Disagree Somewhat Agree Strongly Agree 1. I plan to make myself sick in order to prevent weight gain or counteract the effects of eating. 1 2 3 4 5 2. I plan to use laxatives or diuretics in order to prevent weight gain or counteract the effects of eating. 1 2 3 4 5 3. I plan to vigorously exercise for an hour or more in order to prevent weight gain or counteract the effects of eating. 1 2 3 4 5 4. I plan to use diet pills in order to prevent weight gain or help me lose weight. 1 2 3 4 5 5. I plan to smoke cigarettes in order to prevent weight gain or help me lose weight. 1 2 3 4 5 6. I plan to skip meals in order to prevent weight gain or help me lose weight. 1 2 3 4 5

PAGE 137

127 Appendix N: Multidimensional Health Be havior Inventory-Ex ercise subscale. Directions: The following statem ents describe a broad range of health-related actions or behaviors that you may or may not do. Read each behavior statement and circle the number following each statement that tells how often you usually do this behavior NEVER RARELY SOMETIMES OFTEN ALWAYS 1. Participate in recreational 1 2 3 4 5 physical activities as walking, biking, dancing or sports regularly at least twice a week. 2. Exercise vigorously for at 1 2 3 4 5 least 20 minutes 3 times a week. 3. Increase your physical 1 2 3 4 5 activity to lose weight. 4. Run, jog, or swim for 1 2 3 4 5 exercise at least 3 times per week.

PAGE 138

128 Appendix O: Modified Multidimensional Health Behavior Inventory-Exercise subscale Directions: The followi ng statements describe a broad range of health-related actions or behaviors that you may or may not do. Read each behavior statement and circle the number following each statement th at tells how often you intend to do this behavior NEVER RARELY SOMETIMES OFTEN ALWAYS 1. Participate in recreational 1 2 3 4 5 physical activities as walking, biking, dancing or sports regularly at least twice a week. 2. Exercise vigorously for at 1 2 3 4 5 least 20 minutes 3 times a week. 3. Increase your physical 1 2 3 4 5 activity to lose weight. 4. Run, jog, or swim for 1 2 3 4 5 exercise at least 3 times per week.

PAGE 139

129 Appendix P: Message Source Rating Form for the Pilot Study As you have been informed, the feedback th at was given to you by the other research participant (e.g., message source) was part of the study. Please read each of the following questions carefully and circle the num ber that best reflect s your response to the question. 1. Overall, how credible (e.g., believable ) was the message source as an actual em ployee of their company? 1 2 3 4 5 Not credible Not very Somewhat Very Extremely at all credible cred ible credible credible 2. Overall, how credible (e.g., believable) wa s the feedback given to you by the message source? 1 2 3 4 5 Not credible Not very Somewhat Very Extremely at all credible cred ible credible credible 3. Overall, how credible (e.g., believable) was the flyer given to you by the message source? 1 2 3 4 5 Not credible Not very Somewhat Very Extremely at all credible cred ible credible credible 4. Overall, how thin was the message source? 1 2 3 4 5 Not thin Not very Somewhat Very Extremely at all thin thin thin thin 5. Overall, how attractive was the message source? 1 2 3 4 5 Not attractive Not very So mewhat Very Extremely at all attractive attr active attractive attractive 6. Overall, how warm was the message source? 1 2 3 4 5 Not warm Not very Somewhat Very Extremely at all warm warm warm warm

PAGE 140

130 Appendix Q: Flyer for Ambiguous App earance-related Feedback Condition of the Pilot Study South Tampa Center for Cosmetic Surgery 1115 W. Platt Street Tampa, FL 33606 Phone: (813) 251-9690 Fax: (813) 251-9691 10% Discount Special for USF students! We have served the South Tampa area for several years and have helped hundreds of people improve their physical appearance. You are invited to come to our center for an initial consultation and have your questions about our se rvices answered. Once the appropriate procedures are determined, you will be provided with detailed information about the procedures and packages that we offe r. You will also be given an estimate for the proposed services and fees at the time of your consultati on. USF students will receive this consultation for free Here are some of the many pr ocedures that we offer: Botox Breast Enlargement Cheek Enlargement or Implants Chin Enlargement or Implant Collagen Eyelid Lift Face Lift Facial Peels Lesion Removal Liposuction and Liposculpture N-Lite Laser Permanent Hair Removal Skin Care Programs Tattoo Removal Treatment for Leg Veins Tummy Tuck To schedule your initial cons ultation, call (813) 251-9690 a nd ask for Maria Gonzalez.

PAGE 141

131 Appendix R: Coupon for Ambiguous App earance-related Feedback Condition of the Pilot Study Student Discount for USF Students! Receive a 10% discount o ff your first cosmetic procedure for being a USF stude nt! Be sure to mention this coupon at yo ur initial consultation. South Tampa Center for Cosmetic Surgery 1115 W. Platt Street Tampa, FL 33606 Phone: (813) 251-9690 Fax: (813) 251-9691

PAGE 142

132 Appendix S: Script for Ambiguous A ppearance-related Feedback Condition of the Pilot Study After the video clip is finished, the confeder ate will turn to the pa rticipant and say, Hi, Im a representative for the South Tampa Cent er for Cosmetic Surgery. We are currently offering a free consultation for USF students. The confederate will then pull out a flyer from her bag. This flyer will list the centers contact information as well as the services being offered by that center. The confeder ate will then note a fe w of the procedures listed by saying, Some of the procedures that we offer include: Botox, breast enlargement, liposuction and permanent hair re moval. You should take advantage of the free consultation. Following the feedback incident, the confederat e will turn to the participant and say Im going to the restroom. If the re searcher returns before I do, can you please tell her where I went? Before leaving, the confederate will also pull out a stack of discount coupons for the South Tampa Center for Cosmetic Surgery. While placing this stack on the table in front of them, she will say Feel free to ta ke a coupon if you want a student discount for our center. The confederate w ill then leave to the restroom. The researcher will reenter the laboratory room while the confederate is in the restroom. The confederate will come back to the laborat ory room shortly after the researcher has returned. The participant and confederate will then be asked to complete a packet of posttest measures.

PAGE 143

133 Appendix T: First Debriefing Form for Ambiguous Appearance-related Feedback Condition of the Pilot Study Debriefing Form Previous research has shown that th e development of body image and eating disturbance is greatly influenced by phys ical appearance-related feedback (Thompson, Heinberg, Altabe, & Tantleff-Dunn, 1999, Tan tleff-Dunn & Gokee, 2003). Most studies in this area have primarily focused on teasi ng and negative appearance -related feedback. It has become evident that such feedback is often associated with body dissatisfaction, maladaptive eating behaviors, and psychol ogical distress (Thompson et al., 1999). However, very little is known about ambi guous appearance-related feedback and its impact on others. Specifically, there are st ill many unanswered questions regarding the processing of ambiguous appearance-related feedback. The purpose of the present study is to examine the influen ce of various factors on how ambiguous appearance-related feedback is processed and leads to ne gative outcomes for only a subset of individuals. It is important that you are aware that decep tion was used in this study. The person who was in the room with you was not another re search participant, but rather an actor working with the research team. Nothing sh e said was in anyway related to how she thought about your appearance. Instead, her fee dback was part of a scripted process to help examine our hypotheses about ambiguous a ppearance-related feedback. Everyone who participated in this study was treated similarly. The findings of this study are likely to provide a better understanding of the manne r in which ambiguous appearance-related feedback may contribute to body im age and eating disturbances.

PAGE 144

134 Appendix U: Second Debriefing Form fo r Ambiguous Appearance-related Feedback Condition of the Pilot Study Your participation in this study on physical appearance-related commentary is greatly appreciated. Sometimes, completing questionn aires about your physical appearance may temporarily result in distressing feelings and/or thoughts. If you experience such negative outcomes for a prolonged period of time after this study or have been experiencing them prior to this study, you may benefit from seeki ng therapy services. Contact the USF Counseling Center for Human Development at 974-2831 or the USF Psychological Services Center at 974-2496 if you are interest ed in learning more about their therapy services for students. If you ha ve any questions about the study or therapy services, feel free to ask one of th e researchers. Suggested Readings: Thompson, J.K., Heinberg, L.J., Altabl e, M., & Tantleff-Dunn, S. (1999). Exacting Beauty: Theory, Assessment, and Treatment of Body Image Disturbance. American Psychological Association: Washington, D.C. Cash, T.F., & Pruzinsky, T. (2002). Body Image: A Handbook of Theory, Research, and Clinical Practice Guilford Press: New York.

PAGE 145

135 Appendix V: Revised Message Source Rating Form for the Current Study As you have been informed, the information that was given to you by the other research participant (e.g., message source) was part of the study. Please read each of the following questions carefully and circle the num ber that best reflect s your response to the question. 1. Overall, how credible (e.g., believable ) was the message source as an actual em ployee of the company that she was supposedly working for? 1 2 3 4 5 Not credible Not very Somewhat Very Extremely at all credible cred ible credible credible 2. Overall, how credible (e.g., believable) was the information given to you by the message source? 1 2 3 4 5 Not credible Not very Somewhat Very Extremely at all credible cred ible credible credible 3. Overall, how credible (e.g., believable) was the flyer given to you by the message source? 1 2 3 4 5 Not credible Not very Somewhat Very Extremely at all credible cred ible credible credible 4. Overall, how thin was the message source? 1 2 3 4 5 Not thin Not very Somewhat Very Extremely at all thin thin thin thin 5. Overall, how attractive was the message source? 1 2 3 4 5 Not attractive Not very So mewhat Very Extremely at all attractive attr active attractive attractive 6. Overall, how warm was the message source? 1 2 3 4 5 Not warm Not very Somewhat Very Extremely at all warm warm warm warm

PAGE 146

136 Appendix W: Flyer for Ambiguous App earance-related Feedback Condition of the Current Study South Tampa Center for Cosmetic Surgery 1115 W. Platt Street Tampa, FL 33606 Phone: (813) 251-9690 Fax: (813) 251-9691 Free Consultation for USF students! We have served the South Tampa area for several years and have helped hundreds of people improve their physical appearance. You are invited to come to our center for an initial consultation and have your questions about our se rvices answered. Once the appropriate procedures for you are determ ined, you will be provided with detailed information about the procedures and packages that we offer. You will also be given an estimate for the recommended procedures and fe es at the time of your consultation. USF students will receive this consultation for free Here are some of the many pr ocedures that we offer: Botox Breast Enlargement Cheek Enlargement or Implants Chin Enlargement or Implant Collagen Eyelid Lift Face Lift Facial Peels Lesion Removal Liposuction and Liposculpture N-Lite Laser Rhinoplasty Tattoo Removal Treatment for Leg Veins Tummy Tuck To schedule your initial cons ultation, call (813) 251-9690 a nd ask for Maria Gonzalez.

PAGE 147

137 Appendix X: Flyer for Ambiguous Nona ppearance-related Feedback Condition of the Current Study South Tampa Center for Academic Enhancement 1115 W. Platt Street Tampa, FL 33606 Phone: (813) 251-9690 Fax: (813) 251-9691 Free Consultation for USF students! We have served the South Tampa area for several years and have helped hundreds of students improve their academic performance. You are invited to come to our center for an initial consultation and have your questions about our services answered. Once the appropriate services for you are determ ined, you will be provided with detailed information about the services and packages th at we offer. You will also be given an estimate for the recommended services and fees at the time of your consultation. USF students will receive this consultation for free! Here are some of the many areas for tu toring that we offer: Anatomy Algebra Biology Calculus Chemistry English French Italian Latin Organic Chemistry Physics Precalculus Spanish Statistics Writing To schedule an appointment, call (813 ) 251-9690 and ask for Maria Gonzalez.

PAGE 148

138 Appendix Y: Coupon for Ambiguous App earance-related Feedback Condition of the Current Study Student Discount for USF Students! Receive a 10% discount o ff your first cosmetic surgery procedure for bein g a USF student! Be sure to mention this coupon at your initial consultation. South Tampa Center for Cosmetic Surgery 1115 W. Platt Street Tampa, FL 33606 Phone: (813) 251-9690 Fax: (813) 251-9691

PAGE 149

139 Appendix Z: Coupon for Ambiguous Nonapp earance-related Feedback Condition of the Current Study Student Discount for USF Students! Receive a 10% discount o ff your first tutoring service for being a USF student! Be sure to mention this coupon at yo ur initial consultation. South Tampa Center for Academic Services 1115 W. Platt Street Tampa, FL 33606 Phone: (813) 251-9690 Fax: (813) 251-9691

PAGE 150

140 Appendix AA: Script for the Ambiguous Appearance-related F eedback Condition of the Current Study After the video clip is finished, the confeder ate will turn to the pa rticipant and say, Hi, Im a representative for the South Tampa Cent er for Cosmetic Surgery. We are currently offering a free consultation for USF students. The confederate will then pull out a flyer from her bag. This flyer will list the cen ters contact information as well as the procedures being offered by that center. The confederate will then note a few of the procedures listed by saying, Some of the proced ures that we offer include: Botox, breast enlargement, liposuction and tummy tuck. You should take advantage of the free consultation being offered. Following the feedback incident, the confederat e will turn to the participant and say Im going to the restroom. If the re searcher returns before I do, can you please tell her where I went? Before leaving, the confederate will also pull out a stack of discount coupons for the South Tampa Center for Cosmetic Surgery. While placing this stack on the table in front of them, she will say Feel free to ta ke more than one coupon for a student discount at our center. The confederate will then leave to the restroom. The researcher will reenter the laboratory room while the confederate is in the restroom. The confederate will come back to the laborat ory room shortly after the researcher has returned. The participant and confederate will then be asked to complete a packet of posttest measures.

PAGE 151

141 Appendix BB: Script for Ambiguous Nona ppearance-related F eedback Condition of the Current Study After the video clip is finished, the confeder ate will turn to the pa rticipant and say, Hi, Im a representative for the South Tampa Ce nter for Academic Enhancement. We are currently offering a free consultation for U SF students. The confederate will then pull out a flyer from her bag. This flyer will list the centers contact information as well as the services being offered by that center. The confederate will then note a few of the services listed by saying, Som e of the areas for tutoring that we offer include: Algebra, Chemistry, Spanish, and Statistics. You shoul d take advantage of the free consultation being offered. Following the feedback incident, the confederat e will turn to the participant and say Im going to the restroom. If the re searcher returns before I do, can you please tell her where I went? Before leaving, the confederate will also pull out a stack of discount coupons for the South Tampa Center for Academic Enha ncement. While placing this stack on the table in front of them, she will say Feel fr ee to take more than one coupon for a student discount at our center. The confeder ate will then leave to the restroom. The researcher will reenter the laboratory room while the confederate is in the restroom. The confederate will come back to the laborat ory room shortly after the researcher has returned. The participant and confederate will then be asked to complete a packet of posttest measures.

PAGE 152

142 Appendix CC: First Debriefing Form fo r Ambiguous Appearance -related Feedback Condition of the Current Study Debriefing Form Previous research has shown that the development of body image and eating disturbance is greatly influenced by phys ical appearance-related feedback (Thompson, Heinberg, Altabe, & Tantleff-Dunn, 1999, Tan tleff-Dunn & Gokee, 2003). Most studies in this area have primarily focused on teasi ng and negative appearance -related feedback. It has become evident that such feedback is often associated with body dissatisfaction, maladaptive eating behaviors, and psychol ogical distress (Thompson et al., 1999). However, very little is known about ambi guous appearance-related feedback and its impact on others. Specifically, there are st ill many unanswered questions regarding the processing of ambiguous appearance-related feedback. The purpose of the present study is to examine the influen ce of various factors on how ambiguous appearance-related feedback is processed and lead s to negative outcomes for only a subset of individuals. It is important that you are aware that deception was used in this study. The person who was in the room with you was not another re search participant, but rather an actor working with the research team. Nothing sh e said was in anyway related to what she thought about your intelligence. Instead, her feedback was part of a scripted process to help examine our hypotheses about ambiguous appearance-related feedback and ambiguous nonappearance-related feedback. Everyone who part icipated in this study was treated similarly. The findings of th is study are likely to provide a better understanding of the manner in which am biguous appearance-related feedback may contribute to body image and eating disturbances.

PAGE 153

143 Appendix DD: First Debriefing Form for Ambiguous Nonappearance-related Feedback Condition of the Current Study Debriefing Form Previous research has shown that the development of body image and eating disturbance is greatly influenced by phys ical appearance-related feedback (Thompson, Heinberg, Altabe, & Tantleff-Dunn, 1999, Tan tleff-Dunn & Gokee, 2003). Most studies in this area have primarily focused on teasi ng and negative appearance -related feedback. It has become evident that such feedback is often associated with body dissatisfaction, maladaptive eating behaviors, and psychol ogical distress (Thompson et al., 1999). However, very little is known about ambi guous appearance-related feedback and its impact on others. Specifically, there are st ill many unanswered questions regarding the processing of ambiguous appearance-related feedback. The purpose of the present study is to examine the influen ce of various factors on how ambiguous appearance-related feedback is processed and lead s to negative outcomes for only a subset of individuals. It is important that you are aware that decep tion was used in this study. The person who was in the room with you was not another re search participant, but rather an actor working with the research team. Nothing sh e said was in anyway related to what she thought about your intelligence. Instead, her feedback was part of a scripted process to help examine our hypotheses about ambiguous appearance-related feedback and ambiguous nonappearance-related feedback. Everyone who part icipated in this study was treated similarly. The findings of th is study are likely to provide a better understanding of the manner in which am biguous appearance-related feedback may contribute to body image a nd eating disturbances.

PAGE 154

144 Appendix EE: Standardized Beta Weights and R2 Values for Moderation Analyses with Trait Appearance Satisfaction Pre-Test Variable Ambiguous Feedback Pre-test Trait Variable Interaction R2 Post-test BISS .83** .03 .13* -.06 .81 Post-test VAS Appear .67** .05 .24** -.09 .65 Post-test VAS Anger .83 .11 .05 .04 .67 Post-test VAS Self-Con .91** .05 .05 -.10* .81 Post-test VAS Anxiety .80** -.02 .01 .04 .63 Post-test VAS Depress .84** -.02 -.11 .15** .74 DEBQ-RS Inten .86** -.09* .03 .05 .72 EDE-Q-B Inten .51** -.20** -.16 .15 .63 MHBI-E Inten .72** -.09 -.10 -.04 .76 Note : Post-test BISS = Post-test Body Image States S cale; Post-test VAS Appe ar = Post-test Visual Analogue Scale Appearance; Post-test VAS Anger = Po st-test Visual Analogue Scale Anger; Post-test VAS Self-Con = Post-test Visual Analogue Scale Se lf-Confidence; Post-test VAS Anxiety = Post-test Visual Analogue Scale Anxiety; PT VAS Depress = Post-test VAS Depression; DEBQ-RS Inten = Dutch Eating Behavior Questionnaire-Restraint Scale Intentions; EDE-Q-B Inten = Eating Disorder Examination-Questionnaire-Bulimia subscale Inten tions; MHBI-E = Multidimensional Health Behavior Inventory-Exercise subscale Intentions. *p<.05 **p<.01

PAGE 155

145 Appendix FF: Standardized Beta Weights and R2 Values for Moderation Analyses with Trait Appearance Comparison Pre-Test Variable Ambiguous Feedback Pre-test Trait Variable Interaction R2 Post-test BISS .91** .03 -.02 .09 .81 Post-test VAS Appear .76** .06 -.11 .07 .63 Post-test VAS Anger .81** .12* -.05 .10 .67 Post-test VAS Self-Con .89** .04 .02 -.03 .80 Post-test VAS Anxiety .78** -.01 .05 -.02 .63 Post-test VAS Depress .81** -.10 .20** -.18** .74 DEBQ-RS Inten .80** -.09* .06 -.00 .72 EDE-Q-B Inten .74** -.10 -.12 .20* .59 MHBI-E Inten .56** -.30** .10 .15 .39 Note : Post-test BISS = Post-test Body Image States S cale; Post-test VAS Appe ar = Post-test Visual Analogue Scale Appearance; Post-test VAS Anger = Po st-test Visual Analogue Scale Anger; Post-test VAS Self-Con = Post-test Visual Analogue Scale Se lf-Confidence; Post-test VAS Anxiety = Post-test Visual Analogue Scale Anxiety; PT VAS Depress = Post-test VAS Depression; DEBQ-RS Inten = Dutch Eating Behavior Questionnaire-Restraint Scale Intentions; EDE-Q-B Inten = Eating Disorder Examination-Questionnaire-Bulimia subscale Inten tions; MHBI-E = Multidimensional Health Behavior Inventory-Exercise subscale Intentions *p<.05 **p<.01

PAGE 156

146 Appendix GG: Standardi zed Beta Weights and R2 Values for Moderation Analyses with Trait Appearance Schematicity Pre-Test Variable Ambiguous Feedback Pre-test Trait Variable Interaction R2 Post-test BISS .75** .06 -.09 .03 .63 Post-test VAS Appear .90** .03 -.01 .05 .81 Post-test VAS Anger .82** .12* -.04 .03 .66 Post-test VAS Self-Con .88** .04 -.01 -.05 .81 Post-test VAS Anxiety .78** -.01 .07 -.04 .63 Post-test VAS Depress .83** -.01 .22** -.19** .75 DEBQ-RS Inten .78** -.08 .09 .04 .73 EDE-Q-B Inten .53** -.20** .12 -.11 .39 MHBI-E Inten .73** -.09 .03 .09 .58 Note : Post-test BISS = Post-test Body Image States S cale; Post-test VAS Appe ar = Post-test Visual Analogue Scale Appearance; Post-test VAS Anger = Po st-test Visual Analogue Scale Anger; Post-test VAS Self-Con = Post-test Visual Analogue Scale Self-Confidence; Post-test VAS Anxiety = Post-test Visual Analogue Scale Anxiety; PT VAS Depress = Post-test VAS Depression; DEBQ-RS Inten = Dutch Eating Behavior Questionnaire-Restraint Scale Intentions; EDE-Q-B Inten = Eating Disorder Examination-Questionnaire-Bulimia subscale Inten tions; MHBI-E = Multidimensional Health Behavior Inventory-Exercise subscale Intentions *p<.05 **p<.01

PAGE 157

147 Appendix HH: Standardiz ed Beta Weights and R2 Values for Moderation Analyses with Trait Thin-Ideal Internalization Pre-Test Variable Ambiguous Feedback Pre-test Trait Variable Interaction R2 Post-test BISS .89** .03 -.01 .01 .81 Post-test VAS Appear .73** .07 -.15 .06 .63 Post-test VAS Anger .83** .12* -.08 .08 .67 Post-test VAS Self-Con .88** .04 -.03 -.03 .81 Post-test VAS Anxiety .78** -.01 .08 -.01 .63 Post-test VAS Depress .82** -.02 .24* -.20* .75 DEBQ-RS Inten .81** -.09* -.00 .07 .72 EDE-Q-B Inten .52** -.21* .16 -.14 .39 MHBI-E Inten .75** -.10 -.09 .16 .58 Note : Post-test BISS = Post-test Body Image States S cale; Post-test VAS Appe ar = Post-test Visual Analogue Scale Appearance; Post-test VAS Anger = Po st-test Visual Analogue Scale Anger; Post-test VAS Self-Con = Post-test Visual Analogue Scale Self-Confidence; Post-test VAS Anxiety = Post-test Visual Analogue Scale Anxiety; PT VAS Depress = Post-test VAS Depression; DEBQ-RS Inten = Dutch Eating Behavior Questionnaire-Restraint Scale Intentions; EDE-Q-B Inten = Eating Disorder Examination-Questionnaire-Bulimia subscale Inten tions; MHBI-E = Multidimensional Health Behavior Inventory-Exercise subscale Intentions *p<.05 **p<.01

PAGE 158

About the Author Sylvia Herbozo received a Bachelor of Arts degree in Psychology from the University of Central Florida in 2000, a nd a Masters of Arts degree in Clinical Psychology from the University of South Fl orida in 2004. She is currently a doctoral candidate in Clinical Psyc hology at the University of South Florida and recently completed a predoctoral internship at the Un iversity of Chicago Hospitals in June 2008. After graduating with her doctoral degree, Sy lvia will start a postdoctoral fellowship at Yale University in September 2008. Her re search focuses on interpersonal factors associated with body image, eating disorders, and obesity. She ha s co-authored several peer-reviewed journal articles and book chapters in this field.