Over the past decades the number of women diagnosed with an eating disorder has risen substantially (Silverstein et al., 1986). Garner, Garfinkel, Schwartz, & Thompson, (1980) demonstrated that from 1960 to 1978 the size of Miss America and Playboy centerfolds has become increasingly thinner. In contrast, the actual population of American female is getting heavier (Garner et al., 1980). A more recent study found this to be true for pageant contestants form 1979 to 1988, but did not observe the same results for centerfolds (Wiseman, Gray, Mosimann, & Ahrens, 1992). This discontinuity of shape and weight among women in the media and the population in general may to lead to unhealthy eating patterns (Richins, 1991), and researchers have been encouraged to further investigate the role of societal influences in the development of eating disorders (Levine et al., 1996; Stice, 1998; Striegle-Moore, 1993). It has been suggested that the messages that females receive from the media, family, and peers, help to perpetuate and validate the female thin-ideal (Irving, 1990; Levine et al., 1994, Silverstein et al., 1986). Social reinforcement of the thin-ideal leads to internalization of this ideal (Levine, 1994; Stice & Agras, 1998; Stice et al., 1994) thus leading to the development of body dissatisfaction, negative affect and dieting which are precursors to the manifestation of an eating disorder (Cattarin & Thompson, 1994; Killen et al.1996; Stice & Agras 1998).
In this paper, I first discuss the role of the media in the perpetuation and validation of the thin ideal. I also provide an empirical overview of the role of family and peers in the perpetuation of the thin ideal and the development of eating disorders. Also an overview of three factors that have been implicated in the etiology of eating disorders are included in this paper. The three factors that are discussed are negative affect, body dissatisfaction, and dieting. I then present a study that seeks to explain the causal role of social reinforcement in the form of indirect peer influences in the subsequent manifestation of body dissatisfaction and negative affect.
A content analysis of women's magazines from 1901 to 1981 revealed that the women portrayed in the media have become increasingly non-curvaceous; evidence that may help explain the rise in the number of eating disorders in recent decades (Silverstein et al., 1986). Garner et al. (1980) performed a content analysis of women's magazines from 1959 to 1978 and discovered that the number of dieting articles increased substantially. The same result was found in an analysis of the same women's magazines extended to 1988. Silverstein et al. (1986) also provided documentation that, in television and magazines, the standard for bodily attractiveness is thinner for women than men, and that women are exposed to more emphasis on the importance of dieting than men. Women also have substantially higher rates of eating disorders than men. Taken together this evidence supports that the media may play a strong role in reinforcing the thin ideal for women and that this may manifest itself in disturbed eating patterns.
There is evidence that internalization of the thin ideal plays a role in the impact of media exposure on body dissatisfaction. Stice et al. (1994) found that the relationship between media exposure and eating pathologies were mediated by internalization of the thin ideal. However, the direct relationship between media exposure and thin-ideal internalization was not significant. This implies that internalization of the thin ideal is more directly related to the development of an eating disorder than just media exposure. The fact that there was not a direct relationship between media exposure to the thin ideal and thin ideal internalization implies that there are other mechanisms by which the thin ideal is internalized. Therefore the initial correlation between media exposure and eating pathologies proposed by Stice (1994) may exist through the mechanism of the internalization of the thin ideal, however the internalization of the thin ideal it not entirely due to media exposure.
There is a substantial body of experimental literature that links the mass media to body dissatisfaction and bulimic symptoms. Using a posttest only design, Irving (1990) found that university females were less satisfied with their weight and exhibited lower self-esteem after exposure to pictures of thin models, than were those exposed to average models. Subjects who were shown pictures of oversized models were the most satisfied with their bodies as compared to the groups who saw thin or average models. Results also indicated that poor body image was related to varying levels of bulimic symptoms.
A study designed by Cash, Cash, & Butters (1983) also demonstrated that participants shown pictures of attractive, thin women rated themselves as less attractive after exposure. An interesting part of the results was that participants in the “attractive” condition which contained pictures of attractive peers reported themselves as less attractive than did participants in the condition in which the attractive women in the photographs were presented as professional models. This finding suggests that peer comparison maybe more important than professional comparison. The implications for these results will be expounded upon in the following sections.
Another experimental study split participants into groups based on high or low scores of levels of body disturbance (negative body image) and sociocultural attitudes regarding appearance (Heinberg & Thompson,1995). Participants scoring high on body dissatisfaction were less satisfied with their body after viewing appearance related commercials as compared with those who scored lower on body disturbance. Also, participants with high levels of body disturbance and/or high levels of acceptance of sociocultural attitudes regarding appearance became more depressed after viewing the appearance related tape. These results show that if a person has internalized the thin-ideal, exposure to sources validating this ideal can increase body dissatisfaction and negative affect.
Stice and Shaw (1994) conducted a prospective study that manipulated exposure to the thin ideal through magazine images. Subjects were shown pictures of either extremely thin models, average weight models, or pictures that did not contain people. The results indicated that feelings of depression, unhappiness, shame, guilt, stress, and body dissatisfaction increased after viewing images of the extremely thin models and levels of confidence decreased in this same condition. However, after exposure to the thin ideal, women's affirmation of the thin-ideal stereotype did not change. Results may reflect that the opinions about the thin ideal is a fairly stable trait that is not subject to change in by an experimental manipulation (Stice & Shaw, 1994). On this view, if a person has internalized the thin ideal, all interactions that take place with society regarding shape and weight may be interpreted with this ideal in mind. A discontinuity between a person's actual weight and the reinforcing messages of the increasingly thinner ideal may result in a dissatisfaction with that person's own appearance which in turn can lead to a search for remedies (i.e. extreme dieting measures).
Interactions with Family and Peers
Literature on peer and family influences on the development of eating disorders has also yielded significant results. However, as I will discuss in this section, there has been very little experimental research conducted to conclude that a causal relationship exists between family and peers exerting pressure to be thin, and the subsequent occurrence of body dissatisfaction and eating disorders.
The family functions as an important socializing agent, and thus serves as a messenger of societal values to be thin. An evaluation of 141 high school dieting females found that subjects' perception of parental pressure to lose weight was predictive of dieting behaviors (Stong & Huon, 1998). Schwartz et al. (1999) found a significant correlation between parents negative feedback frequency regarding appearance and daughter's body dissatisfaction. Psychological distress was also related to negative feedback frequency. Pike and Rodin (1991) demonstrated that mothers of daughters with eating disorders exhibited more eating disturbances themselves and also wanted their daughters to lose more weight in contrast to mothers of daughters without an eating disorder. This finding suggests that the family's attitudes towards eating are passed on to the child. Whether intended or not, daughters are aware of the drive for thinness that some mothers have for their daughters and for themselves.
Teasing is a type of negative feedback delivered from family or peers that also affects body dissatisfaction, negative affect, and the development eating disturbances (Cattarin & Thompson, 1994; Stormer & Thompson, 1996). Studies have found that, after accounting for general negative affect, teasing regarding weight and size significantly predicted body dissatisfaction and eating disturbances (Stormer & Thompson, 1996; Thompson & Heinberg, 1993). The frequency of teasing has been found to be correlated with body dissatisfaction, and the impact of this teasing is also correlated with negative affect and eating dysfunction (Fabian & Thompson, 1989). Cattarin and Thompson (1994) conducted a longitudinal study and found that teasing predicted time 2 levels of body image disturbance and account for 6% of the overall variance in appearance dissatisfaction.
Social pressure to be thin appears to have a more negative impact on an individual when the message comes from certain important others, including family and peers, compared to more tangential members of society. Heinberg and Thompson (1992) indicated that subjects who received feedback regarding their weight in reference to a particularistic group (the average student) experienced more body dissatisfaction than did subjects who received feedback in reference to a universal group (the average USA citizen). This finding implicates peers as one of the groups that is important to an individual when searching for acceptance in the social environment. Heinberg (1993) found that the importance of others as comparison targets affects body dissatisfaction and eating dysfunction more than did the frequencies of social comparison. People who experience the drive to compare oneself to others, typically others in the peer group, are at risk for developing a negative body image if they receive negative feedback from a particularistic group, typically peers or family.
A study done by Crandall (1988) demonstrated that peers influence an individual's search for normative or accepted ways to behave. His results indicated that peers have an impact on each other in regards to binge eating. In a sample of sorority members, a girls binge eating could be predicted by her friends' amount of binge eating and this correlation increased as the friendship grew stronger. This finding can be explained by the fact that sororities are made up of a group of peers and peers appear to be a more important group for social comparison (Cash, Cash, & Butters, 1983; Heinberg & Thompson, 1992). The members of the sorority may serve as models and reinforcers of what is considered attractive and what can be done to be attractive. They may also transmit the message that it is socially acceptable to engage in unhealthy eating patterns.
A study by Mitchell (1986) found that forty-five percent of the bulimics in his sample reported that pressure by a peer to lose weight occurred before the onset of binging and purging. The disorder may first manifest itself in voluntary dieting, but pressure from family and friends to lose weight often coincides with this dieting. A major limitation of this research is that the evidence for these relationships is correlational, thus preventing causal statements from being made.
Steiger et al. (1999) used a naturalistic procedure to examine the relationship between social interactions and subsequent binge eating. The subject pool consisted of actively bulimic women, formerly bulimic women, and women who have never been diagnosed with an eating disorder. The results indicate that actively bulimic women perceived their interaction with others to be more negative, had poorer moods, and were more self-critical than the perceived interactions and moods reported by the other two groups. The results also suggest that both actively and formerly bulimic women appear to be hypersensitive to negativity in social interactions when compared to women who have never had an eating disorder. The major flaw of this study is its inability to differentiate between the subjects' perceptions of the interaction and the actual content of the interaction. It could be that the subjects with eating disorders actually did elicit more negative interactions from their environment. It is also possible that these subjects just perceived more negativity from these interactions, although, the actual content of the interaction may be construed as neutral. Without an experimental design, it is impossible to know what is actually occurring in these interactions, and thus, why eating disorder subjects and non eating-disordered subjects are different in these respects.
Bulimics hypersensitivity to negative feedback was demonstrated by an experimental study performed by Craighead et al. (1996). In one condition, negative social feedback was given following incorrect choices that were made while navigating through a computerized maze. In a second condition, negative social feedback was given in the form of appearance related content. The results demonstrate that bulimic subjects learn the maze faster when exposed to negative rather than positive feedback compared to individuals with past depression, individuals with no past depression and no history of an eating disorder. These results suggest that negative social feedback may be more salient and more threatening to individuals with body image concerns, thus encouraging them to avoid this kind of aversion through bulimic behaviors (Craighead et al., 1996).
It is interesting to note that although this negative information may be perceived as more threatening, bulimics also seem to pay more attention and are more interested in the very feedback that is hurting their body esteem (Joiner, 1999). On this view, simple exposure to negative comments regarding appearance will be salient to bulimics and perhaps to those at risk of developing an eating disorder. The salience of these comments may make them aware of their unhappiness with their body more frequently, which may promote the development of an eating disorder. As mentioned above, negative verbal commentary has been correlated with body dissatisfaction and eating pathology. Comments that perpetuate the thin-ideal without directly targeting an individual may also serve this function. A few correlational studies are consistent with this, but an experimental study that can provide insight into the temporal relationship between these factors has not yet been done.
While the Craighead et al. (1996) study is experimental in type and gives insight into the impact of negative social interaction on bulimic individuals, this study did not manipulate social pressure to be thin. Because Craighead et al.(1996) did find a hypersensitivity to negative feedback about task performance in bulimic individuals, and Joiner (1999) found that bulimics pay more attention to negative feedback , one would suspect that a negative social experience that involved negativity about shape and weight would also be salient to these individuals. One way to address this question is to examine whether females in a normal college sample demonstrate a sensitivity to discourse about dissatisfaction about weight and shape. Demonstrating this point, would provide temporal evidence for the relationship of social pressure to be thin and subsequent increases in body dissatisfaction and in negative affect. Both body dissatisfaction and negative affect seem to play a role in the development of eating disorders.
The Role of Negative Affect
Many studies have revealed that negative affect may be an important factor in the etiology of disordered eating. It has been proposed that binging and purging is a method used by individuals to regulate negative affect (Heatherton & Baumeister, 1991; Stice & Shaw, 1994). The negative affect that is experienced by a person with bulimia is thought to be induced by a discontinuity between perceived self and societal standards (Heatherton & Baumeister, 1991). A person may binge to reduce negative affect (Polivy et al., 1994). Polivy et al (1994) found that food did make restrained eaters feel better i.e. happier, but food did not relieve their anxiety. Dieters may still feel depressed, anxious or guilty after the binge (Johnson & Larson, 1982), and then purge to relieve the emotions caused by the binge (Johnson & Larson, 1982; Schupack-Neuberg & Nemeroff, 1993). A longitudinal study (Stice, 1998) found initial negative affect to be positively correlated with bulimic symptoms, however, this relationship did not hold over the course of the disorder, a result which supports previous findings (Keel et al., 1997; Leon et al. 1995; Streigel-Moore et al., 1989). Leon et al. (1995) found that neither negative emotionality nor depression were predictive of eating disturbance at one, two or three year follow up of a sample of 852 girls and 815 boys, grades 7-10.
Despite the lack of evidence for a causal progression from negative affect to bulimic pathology, there are many studies that do confirm a relationship between these two variables (Killen et al, 1996; Leon, Fulkerson, Perry, & Cudeck, 1993; Stice & Shaw, 1994; Shisslak, Prazda, & Crago, 1990; Stice, Nemeroff, & Shaw, 1996). Stice and Shaw (1994) conducted a prospective study on the effect of media exposure on the risk factors for bulimia and found that components of negative affect such as depression, stress, guilt, shame, and lack of confidence were all factors that correlated with an increase in bulimic symptoms. Exposure to the thin ideal increased these aspects of negative affect. This finding demonstrates that negative affect is closely linked with eating disorders and that circumstances that reinforce the thin ideal may be the mechanism through which negative affect is triggered.
The Role of Body Dissatisfaction
There is also evidence that body dissatisfaction is positively correlated with eating disorders (Stice et al., 1994; Stice & Shaw,1994). Stice and Agras (1997) found that binge eating and compensatory behaviors (e.g. vomiting and laxative abuse) could be predicted by increased body dissatisfaction and that cessation of compensatory behaviors was predicted by decreases in body dissatisfaction. Garfinkel et al. (1992) demonstrated that individuals with bulimia nervosa had higher levels of body dissatisfaction than normal controls, and higher levels of body dissatisfaction were correlated with heightened bulimic pathology. It has been demonstrated in experimental studies that exposure to images of the thin-ideal is a mechanism through which body dissatisfaction is increased (Irving, 1990; Stice & Shaw, 1994). Longitudinal (Attie & Brooks-Gunn, 1989) as well as experimental evidence has indicated that body dissatisfaction can predict eating pathology. Heatherton et al. (1997) conducted a 10-year longitudinal study that revealed that although levels of eating pathology declined over this period, body dissatisfaction remained a problem for both men and women. Taken together, this evidence implies that exposure to the thin-ideal leads to body dissatisfaction, and body dissatisfaction triggers disordered eating and compensatory behaviors.
The Role of Dieting
Dieting has also been implicated in the development of eating disorders. Stice et al (1998) suggested that body mass, pressure to be thin, body dissatisfaction and binge eating were predictive of increased dieting over a 9-month period. Byley et al. (2000) examined the relationship between body dissatisfaction, dieting, and family relations in a longitudinal study. They found that girls' perception of the family's perception of her weight and the mother's actual perception of their daughter's weight predicted the girls' dieting behaviors over the course of a year. This provides evidence for the family's role in the development of eating disorders through the mechanism of dieting.
Stice and Agras (1998) indicated that dieting was the most influential predictor of the onset of binge eating and the onset and cessation of compensatory behaviors. Leon et al. (1995) also reported results indicating the predictive power of dieting on future bulimic pathology.
The Present Study
The present study seeks to further explore the relationship between social pressure to be thin and subsequent body dissatisfaction and negative affect. The social pressure in question in this study is the pressure that is indirectly exerted on individuals from their peers. Indirect peer pressure refers to everyday conversations in which peers relate body dissatisfaction and weight loss strategies to each other, which in turn are overheard by other peers. In an effort to increase the amount of literature on the causal relationship between these variables, the present study will employ an experimental design. A pretest and posttest measure of body dissatisfaction and negative affect that is interrupted by the experimental manipulation of social pressure, will lend a causal description of the results. More specifically after a participant has taken the pretest, the participant will be left in a room to overhear a conversation between peers about dieting and other weight loss strategies, as well as the importance of being thin. After the experimental conversation, the posttest will be given to assess levels of negative affect and body dissatisfaction.
It is predicted that those participants that are exposed to this conversation about dieting and exercise will experience an increase in negative affect and an increase in body dissatisfaction compared to their individual scores on the pretest and compared to a control group who heard a neutral conversation.
Because there are many other factors that may influence an individual's response to the manipulation, other questionnaires that measure the relevant variables will be included after the posttest. These variables include degree of social conformity (Haimes & Katz, 1989; Mason & Chaney,1996; Twamley & Davis, 1999), amount of social support (Grisset & Norvell, 1992; Sohlberg, Norring, & Rosmark, 1993; Tiller et al, 1997), and intentions to diet ( Stice & Argas, 1998). All of these variables have been implicated in the degree that a person will negatively respond to social reinforcement of the thin ideal.