Abstract: A recurring problem with the study of sexual fantasy is that of social desirability bias. Study participants may report fantasies that are consistent with general societal expectations of fantasy content, as opposed to themes characterized by their actual fantasies. The wide availability of erotic material on the Internet, however, facilitates the study of sexual fantasy narratives as they are anonymously expressed and viewed online. By extracting approximately 250,000 text-based erotic fantasies from a user-generated website, we sought to examine “real-world” sexual fantasies, determine the themes that were typical of these narratives, and explore the relationship between themes and story popularity (as assessed by story views per day). A principal components analysis identified 20 themes that commonly occurred across the massive corpus, and a path analysis revealed that these themes played a significant role in predicting the popularity of the sexual fantasy narratives. In particular, the empirically identified themes reflecting familial words (e.g., mother, father) and colloquial sexual words (e.g., cock, fuck) were predictive of story popularity. Other themes identified included those not obviously erotic, such as those consisting of words reflecting domesticity (e.g., towel, shower) and colors (e.g., brown, blue). By analyzing a sexual fantasy corpus of unprecedented size, this study offers unique insight into both the content of sexual fantasies and the popularity of that content.
Keywords: Sexual fantasy Language Meaning extraction method Text analysis
Posts
No comments:
Post a Comment