随着社交媒体和网络技术的飞速发展，自拍已成为当代社会中一种普遍且流行的自我表达方式。尤其是对于女性来说，自拍在表达自我形象和建立社交联系方面发挥着重要作用。一些社会现象表明，年轻女性倾向于在自拍照片中复刻一些在社交媒体上广泛传播的女性形象模版，这些女性形象模版往往比较符合大众对女性的刻板印象。因此，本文的研究问题为：女性为了迎合大众喜好在社交平台的自拍照片中所呈现出女性性别特征，其传播效果如何？

为研究这一问题，本研究将从视觉传播识别的角度入手，来探究社交媒体的自拍照片中女性性别特征表现与自拍照片传播效果之间的关系。具体而言，本研究选取国内较为知名的图文社交平台——“小红书”作为研究平台。本研究样本为在此平台中以“自拍”为搜索词，按热度排名的前2500篇图文笔记。具体使用Appium自动化技术结合Python编程语言来完成这一抓取工作。在搜集完研究样本后，采用计算机视觉技术和人工编码的方法，对抓取到的样本图片进行图像分析，并完成样本图片的清洗与编码。具体编码项为女性性别特征表现的各个维度，本研究分为了五个维度，分别是：一：自我遮蔽，包括眼神回避和面部遮挡两个指标。二：身体姿势性别化，包括身体裸露程度和头部倾斜两个指标。三：外貌吸引力，包括年龄和颜值两个指标。四：自我抚摸。五：情绪与情绪强度，包括情绪与图片数量两个指标。以上五个维度九个指标，共同组成女性性别特征表现。同时，本研究将传播效果也分为两个维度，分别是传播广度和情绪反馈。其中，传播广度包括点赞、评论和收藏三个指标。情绪反馈则是指用户的评论所反映的情绪，即评论情绪这一指标。评论情绪这一指标具体是通过文本情感分析的方法对用户评论文本进行测量。综合考虑这种多变量对多变量影响的分析，本研究采用了结构方程模型（SEM）来检验其中复杂的关系模式。

研究发现，首先，在小红书平台中，自拍图片中女性性别特征表现，并不会给其内容带来更好的传播效果。具体表现为图片中身体姿势性别化越高，其传播广度越差，也就是点赞、评论和收藏数量越少。自我遮蔽、外貌吸引力程度越高的图片，其收到的社交媒体情绪反馈越消极，主要体现在评论区的用户评论内容中。而图片中自我抚摸、积极的情绪与情绪强度越高，则会对传播广度产生正向的影响。

最后，研究讨论了导致这种结果的深层原因，可以总结为两个方面：一方面人们开始逐渐意识到女性不应该被简单地定义为符合某些刻板印象的形象。小红书平台的用户大多数为女性，性别化姿势符合社会对女性的期望，但同时也可能被视为过度性感或性别刻板印象。因此，这些照片会遭受来自其他女性的反感或批评。另一方面由于社交媒体是一个强化“身体理想化”和“身体不满意两者之间矛盾的渠道，女性用户会将社交媒体上的照片作为自我参照，如果这些自拍照片一味地强调性感和吸引力，会引起其他女性的不满和压力，从而影响这些照片的传播效果。本研究结果可以启发未来研究者进一步探究女性自拍照片的传播机制，并从更全面的角度考虑女性性别特征在社交媒体传播效果中的作用。同时，这一研究结果也为我们反思和拓展提供了机会，对女性形象的认知不应该仅局限于女性外部特征，而是要更加关注女性的多样性和多维度的人格魅力。

With the rapid development of social media and network technology, selfies have become a common and popular way of self-expression in contemporary society. Especially for women, selfies play an important role in expressing self-image and establishing social connections. More and more social phenomena indicate that young women tend to replicate certain female image paradigms that are widely disseminated on social media in their selfies. These female image paradigms often conform to the stereotypical impressions of women held by the public. Therefore, the research question of this article is: Can presenting female gender characteristics in their selfies on social platforms to cater to public preferences really bring better dissemination effects for women's selfies?

To investigate this issue, this study will start from the perspective of visual communication recognition to explore the relationship between the expression of female gender characteristics in selfies on social media and the dissemination effects of selfies. Specifically, this study selected a well-known graphic and text social platform in China, "Xiaohongshu" as the research platform. The research sample consists of the top 2,500 graphic and text notes ranked by popularity on this platform with "selfie" as the search keyword. Appium automation technology combined with the Python programming language was used to complete this data collection process. After collecting the research samples, computer vision technology and manual coding were used to analyze the collected images and complete the cleaning and coding of the sample images. The specific coding items are the various dimensions of female gender characteristic expression, which are divided into five dimensions in this study: 1) self-covering, including eye avoidance and facial coverage; 2) gendered body posture, including body exposure level and head tilt; 3) physical attractiveness, including age and appearance; 4) self-touching; and 5) emotions and emotional intensity, including emotions and the number of images. These five dimensions and nine indicators together form the expression of female gender characteristics. At the same time, this study also divided the dissemination effects into two dimensions: dissemination breadth and emotional feedback.Dissemination breadth includes three indicators: likes, comments, and collections. Emotional feedback refers to the emotions reflected in user comments, specifically, the emotion of comments was measured by text sentiment analysis. To comprehensively consider the analysis of the complex relationship patterns between these multivariate and multivariable factors, this study used the structural equation model (SEM) to test the complex relationship patterns among these variables.

The study found that first, in the Little Red Book platform, the expression of female gender characteristics in self-portrait photos does not bring wider spread. Specifically, the higher the gendered body posture in the picture, the worse its spread, that is, the fewer likes, comments, and collections it receives. Self-shielding and higher levels of attractiveness in appearance lead to more negative social media feedback, primarily reflected in user comments in the comment section. On the other hand, self-touching and higher levels of positive emotions and emotional intensity in the picture have a positive effect on spread, meaning more likes, comments, and collections.

Finally, the study discussed the underlying reasons for these results, which can be summarized into two aspects. On the one hand, this indicates that people are gradually becoming aware that women should not be simply defined by stereotypical images. The majority of users on the Xiaohongshu platform are female, and gendered poses may conform to societal expectations of women but may also be seen as overly sexualized or stereotypical. As a result, these photos may face criticism or backlash from other women. On the other hand, social media is a channel that reinforces the conflict between body idealization and body dissatisfaction. Female users may use social media photos as a self-reference, and if these selfies simply emphasize sexiness and attractiveness, they may trigger dissatisfaction and pressure from other women, which in turn affects the spread of these photos. The results of this study can inspire future researchers to further explore the dissemination mechanism of female selfies and to consider the role of female gender features in the effectiveness of social media dissemination from a more comprehensive perspective. At the same time, these research results also provide us with an opportunity to reflect and expand our understanding of women's image, which should not be limited to external features but should also focus on the diversity and multidimensional attractiveness standards of women.