Elsevier

Body Image

Volume 14, June 2015, Pages 54-61
Body Image

Bones, body parts, and sex appeal: An analysis of #thinspiration images on popular social media

https://doi.org/10.1016/j.bodyim.2015.03.006Get rights and content

Highlights

  • We examined thinspiration images, intended to inspire weight loss, on social media.

  • Images tended to be sexually suggestive, with thin, bony, scantily-clad women.

  • Twitter images were characterized by more segmented, bony content.

  • Sexually suggestive images tended to have greater social endorsement.

  • Results support the self-objectification phenomenon and social cognitive theory.

Abstract

The present study extends research on thinspiration images, visual and/or textual images intended to inspire weight loss, from pro-eating disorder websites to popular photo-sharing social media websites. The article reports on a systematic content analysis of thinspiration images (N = 300) on Twitter and Pinterest. Images tended to be sexually suggestive and objectifying with a focus on ultra-thin, bony, scantily-clad women. Results indicated that particular social media channels and labels (i.e., tags) were characterized by more segmented, bony content and greater social endorsement compared to others. In light of theories of media influence, results offer insight into the potentially harmful effects of exposure to sexually suggestive and objectifying content in large online communities on body image, quality of life, and mental health.

Introduction

The proliferation of thin-ideal imagery referred to as thinspiration across social media websites like Pinterest, Tumblr and Twitter has received significant media attention in recent years (Bahadur, 2013, Columbia Broadcasting System New York, 2012, Public Broadcasting Service, 2014, Spiegel, 2013). Thinspiration is thin-ideal media content (i.e., images and/or prose) that intentionally promotes weight loss, often in a manner that encourages or glorifies dangerous behaviors characteristic of eating disorders (Lewis and Arbuthnott, 2012, National Eating Disorders Association, 2013). Such content is frequently accompanied by explicit encouragement or advice on losing weight and staying thin (Lapinski, 2009). The link between exposure to thin-ideal media content and body-related self-discrepancies, body dissatisfaction, and eating disorder symptomatology, particularly for young women consuming Western media, has been demonstrated in numerous studies across a variety of methodologies including experiments (for a meta-analysis, see Groesz, Levine, & Murnen, 2002; e.g., Homan, McHugh, Wells, Watson, & King, 2012) and longitudinal and cross-sectional surveys (e.g., Bell & Dittmar, 2011; for a meta-analysis, see Grabe, Ward, & Hyde, 2008). Although content specifically labeled as thinspiration has traditionally been found on pro-eating disorder (pro-ED) websites that encourage or advocate disordered eating for weight loss and image management (Lapinski, 2009), it is now widely shared and endorsed on popular social media websites (Columbia Broadcasting System New York, 2012).

Social media websites are online communities and social networks in which people can interact with one another and create, share, and exchange information and ideas (Kietzmann, Hermkens, McCarthy, & Silvestre, 2011). These websites are characterized by interactive features that differ from earlier forms of media and have been found to hold different psychological outcomes for users in terms of body image concerns (e.g., Fardouly and Vartanian, 2015, Teodoro and Chang, 2014, Tiggemann and Slater, 2013; for a review, see Perloff, 2014). Given the harmful effects associated with thin-ideal media content (Grabe et al., 2008, Groesz et al., 2002), the ease of access to such images on social media (Columbia Broadcasting System New York, 2012), the degree to which these websites facilitate interaction with a community of like-minded users (Amichai-Hamburger, 2007), and the increasing and widespread use of these channels by a younger, vulnerable audience of preteens and teens (Bahadur, 2013, Duggan and Brenner, 2013), further research into the nature of potentially problematic content on social media websites is warranted. In this study, prior research into thinspiration content was expanded upon by documenting patterns of representation across thin-ideal imagery posted on popular social media. Specifically, the extent to which figures in thinspiration images are objectified, sexualized, and endorsed by other users, and the nature of the other labels or tags (i.e., terms used to label the image and organize it under various categories) connected with these images were examined within the context of social media.

Previous studies examining thinspiration content in the context of explicitly pro-ED websites focused on user network characteristics, terms associated with such content, textual themes, and types of figures depicted within images like celebrities and models (e.g., Borzekowski et al., 2010, Lapinski, 2009, Lewis and Arbuthnott, 2012, Yom-Tov et al., 2012). A majority of pro-ED websites contain media images overtly labeled as thinspiration that feature photos of thin models, celebrities, real people and/or athletes (e.g., Borzekowski et al., 2010, Lapinski, 2009, Norris et al., 2006). These images are often accompanied by poetry, song lyrics, or quotations encouraging disordered eating behaviors (Lapinski, 2009, Norris et al., 2006). An analysis into the photo-sharing behavior of pro-anorexia and pro-recovery (i.e., recovery from an eating disorder) communities on Yahoo's photo-sharing site Flickr revealed that the term thinspiration was among the tags (i.e., labels used to mark and categorize images) with the highest probability of usage by pro-anorexia users (Yom-Tov et al., 2012). Further research has indicated that terms thinspo and thinspiration tended to be associated with the most harmful content, including, graphic images of emaciated models and active encouragement of pro-ED behavior through tips and tricks (Lewis & Arbuthnott, 2012).

Although this research offers insight into where thinspiration content can be found, what it generally consists of, and the terms associated with this material, researchers have generally not attended to the physical characteristics of the figures within the images across various types of websites. As young adults, teenagers, and women have come to rely increasingly on the Internet and social media (Bair et al., 2012, Lenhart et al., 2010), researchers have focused increasingly on the relationship between social media use and body image concerns (e.g., Chrisler et al., 2013, Fardouly and Vartanian, 2015, Tiggemann and Slater, 2013).

Several theories of media influence suggest that thinspiration images on social media may negatively influence the attitudes, beliefs, and behaviors of individuals who view them. These theories also suggest that the likelihood of such effects depends on exposure, and the nature of the effects depends on the particular character of the content. Both objectification and social cognitive theory offer insight into how exposure to and interaction with these images, particularly in a social context such as social media, may result in negative outcomes for viewers.

Objectification refers to the conceptualization of an individual as an object, generally for the use and pleasure of others (Fredrickson & Roberts, 1997). When individuals, principally girls and women, are socialized to internalize an observer's perspective as a primary view of their physical selves, they tend to think of themselves in this way, as objects to be looked at; this phenomenon is known as self-objectification (Fredrickson & Roberts, 1997). This self-objectification occurs, in part, when women view objectifying images or video of other women (Aubrey, 2006). Objectifying images include those that feature a woman in a manner that reduces her to an instrument of other's pleasure or use, such as those that isolate, emphasize, or evaluate specific body parts, particularly with regard to sexual appeal and appearance (Bartky, 1990, Fredrickson and Roberts, 1997).

Self-objectification is associated with a range of negative outcomes. Girls and women who experience self-objectification adopt a third-person perspective in evaluating themselves, leading to preoccupation with self-monitoring and negative self-evaluations (Curry & Choate, 2010). Exposure to objectified images portraying the thin ideal has been shown to increase self-objectification, weight-related appearance anxiety, body dissatisfaction, and disordered eating (Harper and Tiggemann, 2008, Morry and Staska, 2001, Murnen et al., 2003). Additional mental health risks associated with self-objectification include depression, body shame, appearance anxiety and sexual dysfunction (Fredrickson & Roberts, 1997).

The risks associated with exposure to thinspiration content, in particular, may be, if anything, even more pronounced. Thinspiration content serves the express purpose of motivating conformity to a thin appearance ideal. The images are created, posted, shared, and often sought out with the goal of promoting thinness. Their purpose is to motivate and guide self-scrutiny based on appearance. The manner in which they do so is likely to shape, at least in part, the degree to which they induce self-objectification. Inasmuch as they overtly call attention to appearance evaluation, isolate specific body parts, and depict or describe the body as something to be seen or used, they are likely to hold the potential for relatively greater harm. The first research question, then, explores the nature of the visual characteristics of thinspiration content as well as accompanying text that may suggest or guide the meaning likely to be derived from that content.

RQ1: What are the characteristics of thinspiration content on social media?

Within many social media sites that facilitate photo sharing (e.g., Pinterest, Twitter), users have the option to label images with multiple tags (e.g., #proana, #fitspo) that allow images to be classified and more readily searched. Multiple tags allow the image to appear in multiple search results (e.g., a search for “fitness” images would lead to thinspiration content also tagged with the word “fitness”). Understanding what tags tend to co-occur with thinspiration-related tags can offer insight into likely patterns of exposure to this content (Bahadur, 2013). Given the potential for wide exposure, the second research question explores how images are categorized through the use of tags.

RQ2: What additional tags are used to categorize thinspiration content on social media sites?

Social cognitive theory also suggests that exposure to thinspiration content is likely to be harmful. Social cognitive theory proposes that people learn from modeled behaviors and are more likely to imitate such behaviors when they can relate to the model and when the behavior is rewarded socially or otherwise (Bandura, 2009). Idealizing and striving toward a thin ideal is a behavior often modeled in media content; this modeling often takes the form of imagery similar to that found in thinspiration content (for a review, see Perloff, 2014). Learning such a thin ideal ultimately affects beliefs, attitudes, and behavior involving food, eating, and the desirable weight and body shape (Dittmar et al., 2006, Harrison and Cantor, 1997, Levine and Smolak, 2006).

Thinspiration content on social media may provide a particularly influential model owing to its social character. The endorsement of behaviors by others can serve as a form of social reward that further encourages learning from and adopting modeled behaviors (Bandura, 2009); this has been shown to contribute to the internalization of the thin ideal (Borzekowski et al., 2010). Social media websites typically allow users to endorse images through various features, including “repins,” “retweets,” “likes,” or “favorites.” These endorsements are visible, both individually and in aggregate, by other site users. Vulnerable users, then, may be more likely to adopt thin-ideal messages due to repeated exposure to images of successful media figures embodying the thin ideal and the apparent social rewards accruing to those images in the form of endorsement of these images by peers (Borzekowski et al., 2010). Such repeated exposure may result from accessing social media websites and being directly or indirectly exposed to thinspiration content. Given that social endorsement of thinspiration messages coupled with repeated exposure to accompanying imagery may ultimately influence adoption of the thin-ideal, the third research question asks whether certain visual characteristics and/or messages of thinspiration content are particularly endorsed.

RQ3: What type of thinspiration content elicits social endorsement on social media sites?

Of course, social media sites are not all the same. In addition to varying in their specific practices and options, they likely differ in the characteristics of their users, including those who share and view thinspiration content. The demographic characteristics of users tend to differ by age and gender, for example, across a variety of social media websites (Lenhart et al., 2010). Facebook and Pinterest users, for example, are generally older than Twitter and Instagram users (Lenhart et al., 2010). Age differences matter when it comes to body image. Individuals most prone to the negative effects of media images promoting the thin-ideal tended to be those who younger than 19 years old (for a meta-analysis, see Groesz et al., 2002). Inasmuch as user characteristics of social media websites differ outcomes of viewing may also be different. In addition, different audiences may select, promote, and endorse different content. The fourth research question examines potential differences in thinspiration characteristics, additional tags and forms of endorsement across both social media websites.

RQ4: How do characteristics of thinspiration content differ across different social media?

Finally, the tagged terms thinspiration and thinspo (a common abbreviation for thinspiration) were found to be associated with the most harmful content on pro-ED websites and have been used synonymously in past research (Lewis & Arbuthnott, 2012). The last research question looks at whether image characteristics differ across the two tagged terms given the emphasis on the term thinspiration in the news and concern about intentions to limit or censor thinspiration tagged material.

RQ5: How will characteristics of thinspiration content differ between the tags thinspiration and thinspo, if at all?

Section snippets

Sample

The sample was collected through a systematic random sampling of posted images from photo-sharing website Pinterest and social networking website Twitter. These particular social media websites were chosen for their popularity, function, and features. Next to Facebook, Pinterest and Twitter are the two most popular social media websites used by U.S. women (25% of female internet users above the age of 18) and teens (8% internet users ages 12 to 17) (Duggan and Brenner, 2013, Lenhart et al., 2010

Results

The final sample consisted of 300 images from Twitter (n = 150) and Pinterest (n = 150), with half of the images from each website tagged as thinspiration and the other half tagged as thinspo. Both of these search terms yielded significant overlap in their results. An analysis of the images labeled under either tag revealed that thinspiration content typically features a bony, partially clad, and sexually provocative image of a woman with the focal point of the depiction on her pelvis and abdomen.

Discussion

Based on relevant theoretical frameworks, the type of image typical of thinspiration content on social media is the type of image likely to increase self-objectification, promote unhealthy standards of beauty, and encourage the sexualization of women by contributing to an increasingly segmented view of one's self and pressure to appear more sexual in one's demeanor and attire. Although this phenomenon cannot be assumed to strictly pertain to females, the content in question overwhelmingly

References (48)

  • S.L. Bartky

    Femininity and domination: Studies in the phenomenology of oppression

    (1990)
  • B.T. Bell et al.

    Does media type matter? The role of identification in adolescent girls’ media consumption and the impact of different thin-ideal media on body image

    Sex Roles

    (2011)
  • D.L. Borzekowski et al.

    e-ana and e-mia: A content analysis of pro-eating disorder web sites

    American Journal of Public Health

    (2010)
  • J. Cohen

    A coefficient of agreement for nominal scales

    Educational and Psychological Measurement

    (1960)
  • S. Coltrane et al.

    The perpetuation of subtle prejudice: Race and gender imagery in 1990 television advertising

    Sex Roles

    (2000)
  • Columbia Broadcasting System New York. (2012). Thinspiration: Doctors concerned with social media sites like pinterest,...
  • J.R. Curry et al.

    The oversexualization of young adolescent girls: Implications for middle grades educators

    Middle School Journal

    (2010)
  • H. Dittmar et al.

    Does Barbie make girls want to be thin? The effect of experimental exposure to images of dolls on the body image of 5- to 8-year-old girls

    Developmental Psychology

    (2006)
  • Duggan, M., & Brenner, J. (2013). The demographics of social media users. A demographic portrait of users of various...
  • R. Engeln-Maddox

    Cognitive responses to idealized media images of women: The relationship of social comparison and critical processing to body image disturbance in college women

    Journal of Social and Clinical Psychology

    (2005)
  • B.L. Fredrickson et al.

    Objectification theory

    Psychology of Women Quarterly

    (1997)
  • D. Freelon

    Recal: Intercoder reliability calculation as a web service

    International Journal of Internet Science

    (2010)
  • S.J. Gervais et al.

    Seeing women as objects: The sexual body part recognition bias

    European Journal of Social Psychology

    (2012)
  • S. Grabe et al.

    The role of the media in body image concerns among women: A meta-analysis of experimental and correlational studies

    Psychological Bulletin

    (2008)
  • Cited by (167)

    • A structural equation model of intuitive eating with adolescents

      2023, Revue Europeenne de Psychologie Appliquee
    View all citing articles on Scopus
    View full text