1. Introduction

In a digital age where technologies increasingly mediate everyday life, social media has transitioned from a passive information consumption platform to an active site of social communication and self-identity. Visual production-oriented platforms like Xiaohongshu, through their body-centered content production mechanisms and recommendation algorithms, have also redefined the ways and meanings of female bodily presentation. The widespread dissemination of aesthetic templates (e.g., “pale, youthful, and slim” ideal), body comparison photos with heavy retouching, and quantified beauty standards have made women’s bodies as images to be gazed at, evaluated, and disciplined. These have given rise to new eating dilemmas, either avoiding excess food intake (e.g., extreme dieting and calorie counting) or excessive emotional eating followed by compensatory behaviors (e.g., purging and excessive exercise). Both are driven by two typical social phenomena. The first is the distortion of Xiaohongshu so-called “self-disciplined eating” content—under the guise of “health,” many shared weight-loss meals show extreme cases of daily calorie intake below 800 kcal. The second is the mental health crisis of content creators. According to a 2024 survey by The Paper, 35% of food bloggers surveyed reported eating disorder symptoms to be “effortlessly slim” while attempting to maintain an “effortlessly slim” image [1]. All of these phenomena reveal the bodily cost behind the prosperity of the platform. When eating behaviors become a performance and food choices become tokens for attracting online traffic, does eating dilemma also contain the seeds of self-empowerment?

Previous studies have already shown the connection between social media and women’s body anxiety. Yufan Liu proposes the concept of “bodily enclosure,” stating that capital and technology jointly enmesh women’s bodies into a discourse of perfection that generates negative body images [2]. Empirical research by Cohen and Blaszczynski show that images viewed on social media are more likely to induce appearance comparisons than images viewed on traditional media, and that such comparison is directly associated with increased risks of eating disorders [3]. Subsequent research by Dahlgren et al. on Instagram and TikTok also shows a significant positive association between the degree of visual media use and the prevalence of pathological eating disorders among adolescent girls [4]. In sum, these studies have exposed the central paradox that while social media provides spaces for self-expression for women, it also intensifies their bodily alienation through disciplining mechanisms of the “beauty industry”.

However, there are still many gaps in the literature. First, platform-specific research is lacking. In addition to Herrick and colleagues’ study of TikTok’s EDrecovery hashtag, because the “planting-grass” mechanism of Xiaohongshu is significantly different from that of other platforms in terms of community atmosphere and content format, the impact of these mechanisms on eating behaviors has not been studied [5]. Second, analyses of identity dynamics are missing. Existing scholarship has not discussed how eating-disorder bloggers navigate the identity dual bind of being both “patients” and “content producers,” nor how content production affects their self-identity. Third, the exploration of empowerment pathways is limited. Branley and Covey only found that there is “resistance” content existing on social media but did not analyze how the bloggers themselves produce resistance in their everyday content creation [6] . Most of the existing scholarship focused on the general influence of social media on ordinary women, ignoring the particularity of bloggers as producers whose eating behaviors are affected not only by platform discipline but also by the identity conflict of “illness status” and “social expectations.” In addition, most studies are confined to unidirectional arguments of “negative effects,” ignoring the dynamic process by which bloggers achieve self-empowerment through content creation. Facing the above-mentioned situations, what eating dilemmas do female eating-disorder bloggers experience on Xiaohongshu present? How does the eating disorder dilemma of bloggers achieve the balance between “illness status” and “social expectations” in content creation? Does the eating dilemma itself contain the pathways to self-empowerment? By exploring the eating dilemmas of female bloggers on Xiaohongshu, this study aims to enrich the related theoretical corpus of body studies in the digital age and provide a new perspective to explore the dialectical relationship between self-objectification and self-empowerment.

This study is motivated by two theoretical origins. One is Foucault’s theory of discipline, and the other is feminist body studies. Foucault argued that in modern society, power enters the body through norms and standards, and social media is a digital form of disciplinary power. The algorithm influences cognition by always pushing “ideal body” content; likes and quantitative feedback influence behavior, and food control becomes self-discipline internalized by users. Simultaneously, feminist scholar Yang Xizhen’s analysis of the “male gaze” shows that it has evolved into a more harmful “data gaze” on Xiaohongshu. That is, bloggers’ dietary choices must continuously meet the demands of followers and brands [7].

This study makes two innovations in terms of research. One is in research perspective. This study treats the “eating dilemma” as a struggle space between discipline and resistance rather than a purely psychological issue. The other is in research method. This study uses in-depth interviews and content analysis in conjunction to explore bloggers’ micro-strategies in recipe design, food styling, and video editing.

The main part of this article is organized into five parts. Part One introduces the research. Part Two is literature review. This part systematically reviews the theoretical origins of social media and body studies as well as eating-disorder research to clarify the theoretical starting point of this study. Part Three explains the research methodology. This part explains the criteria of selecting research subjects, the method of data collection, and the analytical framework. Part Four explores the expression of eating dilemmas of female bloggers on Xiaohongshu. This part uses representative cases to explore the characteristics of eating dilemmas of female Xiaohongshu bloggers in food control, emotional feeling, and social performance. This part analyzes the generative mechanism of eating dilemmas by exploring how the platform, commercial logic, and socio-cultural context shape bloggers’ eating behaviors in turn. At the same time, this part explores bloggers’ agency in digital platforms. This part explains the strategies bloggers adopt through content creation to resist discipline, and explores the effectiveness of their resistance. Part Five concludes with the main findings of this study. This part explores a dynamic relational model of self-objectification and self-empowerment. This part also explores concrete recommendations for platform governance, blogger support, and public education, and explains the limitations of this study and explores the research direction.

2. Literature review

In terms of the impact of social media on women’s bodies and eating behavior, Yufan Liu put forward the concept of “bodily confinement,” believing that young women’s bodily representations on social media are confined by technology and platform rules, and then give rise to negative body images under the drive of capital in the “beauty industry” [2]. Cohen and Blaszczynski found that images on social media are more likely to trigger appearance comparisons than those in traditional media, and appearance comparison is positively correlated with body dissatisfaction and eating disorder risk [3]. Dahlgren et al. also found that social media, especially Instagram and TikTok, have a negative effect on girls’ body image, and eating disorder pathology is closely related to internalization of appearance ideals [4]. As for Xiaohongshu and women’s objectification of their bodies, Suhong Song et al. put forward the concept of “social monetization,” which refers to women monetizing their bodies as social resources for exchange on social media [8]. Yiran Wang discussed how female influencers on Xiaohongshu participate in material and spiritual communication by producing content to build their ideal self [9]. Lonergan et al. found that investment in and modification of photos are associated with eating disorder risk [10]. However, none of these studies have discussed how eating disorder influencers balance their illness status and social expectations in identity construction, nor have they discussed how content creation affects their identity construction. In terms of self-empowerment, studies on women with eating disorders have found alternative narratives. Branley and Covey found that in addition to “pro-anorexia” content, “recovery support” content also exists [6]. Herrick et al. found that content tagged with EDrecovery on TikTok can assist in building recovery communities [5]. Li and Tang found that the motivations of Chinese patient-bloggers in sharing their experiences are related to their needs for autonomy, connectedness, and other intrinsic factors [11]. However, none of these studies have focused on Xiaohongshu, nor have they discussed the multiple possibilities of “self-objectification” versus “self-empowerment” that eating disorder influencers face in their identity construction. Furthermore, other scholars such as Fangfang Wu and Sihang Feng have discussed the images of fitness and lifestyle influencers in detail, but they did not discuss eating disorder influencers [12,13]. The research on beauty influencers on Xiaohongshu by Liyuan Liu provides valuable references for bodily presentation, but it did not discuss the issue of eating disorders [14].

In summary, previous studies have found the influences of social media on women’s body image and eating behaviors as well as general mechanisms of influencer identity construction. However, they have neglected the influence of Xiaohongshu on female influencers with eating disorders, failed to find the differences in their identity construction and the issues of self-objectification and empowerment such construction reflects, and have not sufficiently explored their agency and mechanisms of identity building in content creation. This study attempts to explore eating dilemmas that female influencers on Xiaohongshu face in contemporary social media environment in depth and tries to find out whether such eating dilemmas will lead to self-objectification or self-empowerment in turn.

3. Research methods

To explore the issues of self-objectification and empowerment reflected in platform content and find new perspectives for body studies in digital age, this study uses a netnographic method. Data were collected through non-participant observation and analyzed based on thematic analysis.

3.1. Data collection

Xiaohongshu was selected as the main data source, and purposive sampling was adopted. First, female influencers with more than 10,000 followers were selected as basic sample accounts, and only when the followers and topics were satisfied, contents tags containing “eating disorder,” “binge eating,” “anorexia,” and “body anxiety” were chosen. Then, representative cases were added into the sample library, including “Shaonvshenpopo” (12.7k followers, and single post received up to 14.9k likes), “Xiaoxilai” (3.2k followers and posts were all about binge-eating self-help), “Qianlixunye” (519.8k followers and posts were about her recovery process), “Y bushi xiaoy” (10.7k followers and posts were about eating with low appetite). Combined together, these accounts formed a dynamic sample library of influencers of different follower scales, eating disorders, and content styles. To ensure the timeliness of collected data, data were collected from January 2022 to July 2025, covering nearly 3 years of content production and dissemination on Xiaohongshu. On one hand, this time span covered the long-term features of influencer production, including the narrative cycle of binge-eating records, psychological reflection, and recovery check-ins. On the other hand, this time span covered the short-term features of influencer production, including the influence of platform algorithms and users’ interactions. In addition, the researcher also participated in eating disorder–related communities and tracked the influencers mentioned in Table 2, for example, when “Shaonvshenpopo” had curatorial and public-speaking events, the researcher attended and recorded the logic of her public-oriented contents. When attending “Xiaoxilai”’s community, the researcher found that most interactions with her followers were on “confessional suffering” and “seeking encouragement.” From these, a cycle of content production was found, which could be described as “problem disclosure–emotional release–solution offering–recovery feedback.”.

3.2. Data analysis

The content of sampled influencers were categorized thematically into three dimensions: food control, emotional association, and social performance. Food Control: “Y bushi xiaoy” recorded detailed quantified information of food intake, such as “500 calories” or “liquid fasting,” to highlight the subject’s control over her body. “Qianlixunye” documented her experience of alternating between “fasting–eating–purging”. Emotional Association: “Xiaoxilai” recorded her binge episodes and described the foods that triggered bingeing, such as “chocolate cheesecake bagels,” which revealed how emptiness in emotions was filled by compulsive bingeing. Social Performance: “Shaonvshenpopo” posted a series of body-ratio installations that visually expressed her resistance against aesthetic oppression. The caption “thin enough” broke through the logic of objectification. In contrast, some influencers reinforced the beauty quantification set by the platform by presenting pictures of their bodies along with height and weight information. Additional information helped to frame influencers’ descriptions of their experiences. These included the documentary They Are “Thin” Enough and its audience feedback (danmu). In addition, the feedback in the topic binge-eating self-help showed a change in communication patterns from “confessing pain” to “seeking encouragement.” All of these information together formed a complete cycle of “binge records – psychological reflection – recovery check-ins.”

4. Objectification and empowerment of female influencers on digital platforms

In terms of the type of eating disorder–related posts made by female influencers on Xiaohongshu, they can be roughly classified into three categories: experience sharing, everyday recording, and public interest/science popularization. Most of them also exhibit moderate commercial integration. Her eating behaviors can be summarized into two types: more control and compensatory loss of control. Extreme dieting refers to daily food intake of <800 kcal. “500 kcal eating logs” refers to careful tracking of every consumed calorie. Liquid fasting means eating nothing but water or beverages. Compensatory behaviors include emotional binge eating (mechanical eating, eating hated foods), self-induced vomiting, taking laxatives, and excessive exercise. In addition, numerous influencers reported various physiological and psychological comorbidities, such as amenorrhea, bipolar disorder, and anxiety disorders. The dilemma comes from the conflict between two identities: patients and influencers. As patients, they record their pains in real time—such as fearing that the register clerk in a convenience store will see her binging. As influencers, they need to bear in mind the expectations of their followers while adhering to platform rules. Some of them feel ashamed of their lack of self-control and consequently reinforce their self-denial and further deteriorate their eating behaviors. Further, the research finds that the dilemma faced by Xiaohongshu influencers are shaped by three intertwined forces: platform algorithms, commercial logics, and socio-cultural discipline. Platform Algorithms: Through the visual orientation and algorithmic recommendation, Xiaohongshu forces “white-young-thin” aesthetic templates and highly interactive body-quantification content. It guides influencers to sell themselves by “low calorie diet” and “thin body weight”, which aggravates the food-control behaviors of influencers. The traffic concentrates on “body anxiety” related keywords, putting influencers in a “paradox of being exposed to be seen”: in order to be noticed, they have to show their suffering. For example, the everyday-recording influencer “Y bushi xiaoy” reduced her daily calorie intake from 600 kcal to 400 kcal to attract audience attention, and presented her “0.5 kg every day weight loss” in a visible way, which gained high interaction. The bodily involution caused by algorithm recommendation is one of the typical mechanisms that aggravates eating disorder dilemma. Commercial Logics: In the trend of social commodification, users’ bodies become the traffic and commerce intermediaries. Some influencers expose their physiques in order to gain brand collaborations. Maintaining the “never-gain-weight” image provokes eating disorder further. Our research finding resonates with the 2024 report of Pengpai News: Eating disorder symptoms in food-related influencers account for 35%. Socio-Cultural Discipline: The prevailing perfectionist “white-young-thin” aesthetic is internalized by women as a personal obligation, covering up the more profound origin of their low self-acceptance and failure to gain recognition for their true selves. Influencers are forced to engage in extreme eating to meet social recognition, but they get stuck more deeply. For example, as a public-interest influencer, “Shaonvshenpopo” publicly revealed the psychological trauma of post-purging, but was then stigmatized and attacked as “you just want to draw attention”. The stigma forces some influencers to pretend that they are not as severe as they are, putting them in a more distressed state. However, some influencers resist the above constraints through authentic narration, aesthetic reconstruction, and community-based mutual support. For example, “Qianlixunye” shared her experience from purging to acceptance, encouraging followers to face their disorders. “Shaonvshenpopo” used her artistic installations to criticize the thinness and guided her followers to escape from hating to recognizing themselves. In the comments section, followers also shared their experiences and encouragements, forming relatively coherent communities promoting recovery. For example, the relatively coherent comment interactions of “Xiaoxilai” changed from confessional suffering to mutual encouragement, reflecting the emotional support value of communities.

5. Conclusion

Based on the behavioral characteristics and mechanism of dilemmas posed to female eating disorder bloggers on digital platforms, this study uses the empirical analysis of Xiaohongshu platform and takes a mixed approach combining subject analysis and quantitative methods. By systematically analyzing content production, eating behaviors and logic of platform operation, this study promotes the following three central arguments: the dialectical unity of objectification and empowerment, the alienation of illness narrative under commercial logic and the complicity of algorithm and stigma for aggravating above-mentioned dilemma. The study can be concluded in the following three aspects. One is that the content produced by Xiaohongshu female eating-disorder bloggers can be categorized into three types of experience sharing, daily recording and public-interest science popularization and is closely related to commercial need. Two is that the eating behaviors of them are characterized by two features of over-control and compensatory loss of control, and the high degree of physiological and psychological comorbidities also exists. Three is that the above-mentioned dilemma is caused by the combined effect of platform algorithm, commercial capital and social stigma, which forms a vicious cycle of behavioral compulsion, cognitive distortion and lack of help. In addition, this study also has some limitations. This study is based on a single platform, that is, Xiaohongshu and it is based on cross-sectional data. There is no multi-platform comparison or longitudinal follow-up. In the future, the research can be extended to compare the differences across multiple platforms, carry out long-term follow-up and explore more comprehensive intervention ways through optimizing the algorithm, regulating commercial ethics and reducing stigma.