I thought I'd write up a brief summary of the findings related to romance in
Stern, Stephanie C., Brianne Robbins, Jessica E. Black and Jennifer L. Barnes. "What You Read and What You Believe: Genre Exposure and Beliefs About Relationships." Psychology of Aesthetics, Creativity, and the Arts. Advance online publication. Abstract here.
Stern et al summarise prior research which found that
People who watch more soap operas, which tend to focus heavily on romantic storylines, have been shown to hold beliefs about relationships that detract from one’s relationship satisfaction and longevity (Haferkamp, 1999). Exposure to romantic comedy movies in young adults has also been shown to correlate with greater support of the relationship ideals common to the genre, such as the belief that love conquers all and the idealization of romantic partners (Hefner & Wilson, 2013). Similarly, in a study by Shapiro and Kroeger (1991), participants who endorsed unrealistic beliefs about relationships scored higher on a measure of exposure to romance novels and movie romantic comedies. Specifically, exposure to romance was shown to significantly correlate with greater support of the belief that mindreading and sexual perfectionism should be expected in a relationship.
Taken as a whole, this body of work is consistent with the idea that exposure to Romance fiction, across media, is associated with a variety of unrealistic beliefs about romantic relationships. However, a majority of the studies cited above relied on voluntary self-reported exposure to romantic films, magazines, and/or novels, and the definition of what qualifies as “romance” media varies significantly between experiments. (4)
is focused on the degree to which exposure to seven different literary genres relates to participants’ endorsement of the beliefs that the sexes are inherently different, that all disagreement is destructive, that one’s romantic partner should be able to know one’s thoughts and feelings without being told, that romantic partners cannot change, and that sexual perfection should be expected in relationships (3)
The seven genres were "classics, contemporary literary fiction, romance, fantasy, science fiction, suspense/thriller, horror" (5), later reduced to six because "high collinearity between science fiction and fantasy" (6) led them to combine the two.
For romance, the "association between familiarity with romance authors and the belief that the sexes are difference [sic] was positive: participants who recognized more romance authors scored higher on the sexes are different subscale" (6). In other words, romance readers were more likely than other readers to endorse the idea that men are different from women. However, romance reading was not associated with any of the other "unrealistic" beliefs listed above.
The correlational findings from this study differ from past research on the romance genre in several ways. For example, Fong and colleagues (2015) found no significant correlation between exposure to romantic books and attitudes about sex and gender, whereas the current research found that individuals who read romance novels are more likely to assert that the sexes are inherently different. Conversely, research on other forms of media has shown various correlations between Romance and maladaptive relationship beliefs, such as the belief that mind reading is expected and stronger idealization of relationships as a whole (Shapiro & Kroeger, 1991; Hefner & Wilson, 2013), whereas our study showed no relationship between exposure to the romance genre and these beliefs. This difference may be due, in part, to the present study controlling for exposure to other written fiction genres, including literary classics, which may be particularly important, given that exposure to the classics— or increased knowledge of them—may reflect education more than reading habits per se. (9)
In addition, the results do not show that romance causes its readers to believe that "the sexes are inherently different":
Another limitation that merits consideration is the correlational nature of the results reported here. The current experiment found a relationship between exposure to certain written fiction genres and relationship beliefs, but it does not and cannot determine the nature of that association. It is possible that repeatedly reading specific genres affects our beliefs about relationships, but it is also possible that preexisting beliefs guide genre preferences. (10)
research is needed to explore whether viewing or reading specific genres of fiction affects relationship beliefs and, conversely, whether priming specific relationship beliefs can influence fiction preferences. Finally, future research is needed to explore the role that other variables, such as education, imagination, prior experience with romantic relationships, and transportation may play in the associations found here. (10)