What Kind of Erotic Film Clips Should We Use in Female Sex Research? An Exploratory Study
Terri L. Woodard, MD, Department of Obstetrics and Gynecology, Wayne State University School of Medicine, 3750 Woodward avenue Suite 200D Detroit, MI, USA 48201. Tel: 313-993-4527; Fax: 313-993-4534; E-mail: firstname.lastname@example.org
Introduction. Erotic film clips are used in sex research, including studies of female sexual dysfunction and arousal. However, little is known about which clips optimize female sexual response. Furthermore, their use is not well standardized.
Aims. To identify the types of film clips that are most mentally appealing and physically arousing to women for use in future sexual function and dysfunction studies; to explore the relationship between mental appeal and reported physical arousal; to characterize the content of the films that were found to be the most and least appealing and arousing.
Methods. Twenty-one women viewed 90 segments of erotic film clips. They rated how (i) mentally appealing and (ii) how physically aroused they were by each clip. The data were analyzed by descriptive statistics. The means of the mental and self-reported physical responses were calculated to determine the most and least appealing/arousing film clips. Pearson correlations were calculated to assess the relationship between mental appeal and reported physical arousal.
Main Outcome Measures. Self-reported mental and physical arousal.
Results. Of 90 film clips, 18 were identified as the most mentally appealing and physically arousing while nine were identified as the least mentally appealing and physically arousing. The level of mental appeal positively correlated with the level of perceived physical arousal in both categories (r = 0.61, P < 0.05 and r = 0.62, P < 0.05). The most appealing and physically arousing films tended to exhibit heterosexual behavior with vaginal intercourse. The least appealing and least physically arousing films tended to depict male homosexual behavior, fellatio, and anal intercourse.
Conclusions. Erotic film clips reliably produced a state of self-reported arousal in women. The most appealing and arousing films tended to depict heterosexual vaginal intercourse. Film clips with these attributes should be used in future research of sexual function and response of women. Woodard TL, Collins K, Perez M, Balon R, Tancer ME, Kruger M, Moffat S, and Diamond MP. What kind of erotic film clips should we use in female sex research? An exploratory study. J Sex Med 2008;5:146–154.
Amajor obstacle in sex research has been a lack of standardization in investigative protocols. This is particularly true for the stimuli used in studies of sexual behavior. One of the most common stimuli used in this field is the erotic film clip, which has been employed in studies e.g., of female sexual dysfunction [1,2], sexual arousal [3,4], and sexual orientation in men and women . Although a powerful tool, its effectiveness may not be completely realized because of inappropriate application, lack of standardization, and incomplete knowledge of its use.
Erotic film clips have been shown to elicit higher levels of both subjective and physiologic (genital) arousal than slides , written literature , or audiotape . Previous studies have shown that the more explicit the material, the more physiologic stimulation it causes in both sexes . Because of their efficacy, wide availability, low cost, and ease of use, erotic film clips are considered to be the best stimuli to use in studies of sexual arousal .
In the past, pornography has been a male-dominated industry; thus, previous studies in female sex research have utilized film clips produced by men . However, these may not be appropriate as many of these materials portray degradation and objectification of women, which may make them distasteful or repulsive . There has been considerable interest in gender differences in response to erotic films. Studies have shown that men become more aroused and have a higher degree of correlation between subjective and physiologic arousal than women when watching sexually explicit films [13–16]. This phenomenon has been explained by evolutionary , social [18,19], and sociobiologic  theories which are beyond the scope of this article. Importantly, women's response to pornography may be altered as a result of its negative connotations, societal taboos, religious, social and cultural restrictions, and potential threat to self-image.
Historically, there have been many biases about women's attitudes toward and responses to erotic stimuli. In the 1950s, Kinsey et al. argued that women need greater expression of commitment to maximize arousal to sexually explicit materials . Additionally, women were thought to be less likely to look at pornography  and experience more sexual guilt than men .
However, these notions and findings have been challenged, and it is debatable whether there are significant differences between men and women in response to erotica. A study by Jacobovits showed that men had a greater psychosexual response to emotionally-based erotic material, while women had a greater psychosexual response to nonemotionally-based erotic material, though these findings were never replicated . Contrary to the belief that women needed more romance and affection than men to respond to erotica, Schmidt et al.  and Heiman  did not find a gender difference in the need for romance in erotica. Fisher and Byrne  stated that romantic or affectional emphasis was not a precondition for sexual arousal in women. Another study concluded that gender differences have diminished , while more recently, Quackenbush et al. have reached contrasting conclusions that men and women responded similarly to love and affection in explicit films .
Driven by the assumed irrelevance of traditional erotic films to women, the first erotic films produced by women for women were introduced in 1984. These women-centered films used novel videographic technique and content. The prototype of these films was created by Candida Royalle, a former pornographic film star. She sought to create pornography that was explicit yet possessed integrity, while being nonsexist and life enriching . Her films were considered “hard-core” but maintained conservative perspectives on sexuality; they generally involved older, middle to upper class white characters involved in consensual heterosexual sex. These films rarely depicted anal, group, or lesbian sex, and there were fewer “close-ups.”“Money shots” (i.e., scenes that showed a male ejaculating on a person or object) were eliminated. Most importantly, the viewpoint character was always the woman—someone whom the average woman could identify with . Investigators questioned whether films in this new genre would enhance sexual arousal in women and studies have demonstrated that women have a better response to women-centered erotica [9,11,30]. In a study comparing the responses of women watching male-centered vs. female-centered films, it was demonstrated that women found female erotica more subjectively stimulating than the male-produced erotica; however, there were no differences in physiologic response. Correlations between subjective and physiologic response in these women were nonsignificant. Furthermore, women reported feeling more guilt, shame, and aversion with the male films . The authors concluded that the stimulus content and meaning is what was most important in determining the female sexual arousal, supporting the idea that a particular genre of pornographic film might be more useful in achieving sexual arousal in women. Even so, another study showed that while women reported a more positive response to female-produced videos, there was still some level of disgust .
Few studies have systematically looked at what type of erotic film clips may be most appropriate for the study of female sexual response and arousal. The study by Laan et al. was the first attempt to evaluate subjective and physiologic responses to male- vs. female-centered pornography . Janssen et al. subsequently conducted a study of young men and women who were asked to view both male- and female-selected erotic video clips . They found that gender differences in response were not evident with the female-selected clip, but were significant with the male-selected clip. Furthermore, men and women were more aroused by film clips selected by their own gender. Though limited by exclusive use of young college-aged subjects in the psychology department (introducing volunteer and selection biases) and lack of physiologic assessment, this study illustrated how the way films are selected for sex research can influence the data that are elicited. However, both of these studies are limited by the number of clips used, their inclusion of only heterosexual individuals (many of whom were in the psychology field), and content of clips (only heterosexual, nonfetish content was allowed). Neither study elaborated on the content of the films that were selected.
The goal of our study was to determine what types of erotic film clips were maximally appealing and arousing to a general population of women by measuring levels of mental appeal and self-reported physical arousal in response to a diverse group of film clips. We also wanted to see if we could observe correlations between mental appeal and reported physical arousal.
Description of Clips
The top rental list of an adult video store was obtained. Various erotic films from this list encompassing a broad range of themes were examined. They ranged from R to XXX ratings. Ninety clips were selected from these videos by members of the research team to include in the study. An attempt to include a wide selection of material was made; however, clips from well-known or “mainstream” movies and videos were excluded. These clips represented a variety of themes (see Table 1) which were coded by the investigators, and they varied in their degree of explicitness. Some contained “close-up” shots of genitalia while others did not. Eighty-seven percent (87.8%) involved male and female actors and 12.2% involved a single gender. The film clips were ordered into sets of 30, with each film clip lasting approximately 1 minute with 10 seconds in between each.
Table 1. Characterization of film clips
|Male on female sex position||34||0.38|
|Female on male sex position||20||0.22|
|Female on female sex position||5||0.06|
|Male on male sex position||5||0.06|
|Two females and one male||4||0.04|
|Two males and one female||16||0.18|
|Race other than Caucasian||6||0.07|
After approval was obtained by the Wayne State University Human Investigation Committee, 21 women were recruited via personal contact or flyers to participate in the study. As required by our Institutional Review Board, an inclusion criterion was that the participants had to have viewed pornography in the past. Informed consent was obtained. The mean age of the subjects was 31.2 ± 10.46 years (range: 18–57 years). The majority of the participants were Caucasian (66.7%) and 23.8% were black, 4.8% were Asian and 4.8% did not respond. All of the women had achieved at least a high school level of education, with 9.5% completing high school, 38.1% completing some college, 19% completing a Bachelor's degree, and 33.3% completing a Master's degree or above. The majority of women were single (57.1%); however, 28.6% were married and 14.3% lived as married. With regard to sexual orientation, 76.2% described themselves as heterosexual while 19% were bisexual and 4.8% were unknown. One woman was postmenopausal.
Subjects were instructed to watch the film clips and rate their level of mental appeal and physical arousal on a bidirectional rating scale. They were told to use positive values if they found the clip pleasant and/or if it turned them on and negative values if they found the clip unpleasant and/or it turned them off. The first several subjects (N = 7) rated the films in the privacy of their home while the others rated them in a private room at the Clinical Research Center at Wayne State University to facilitate prompt return of the study digital video disks. They were instructed to watch 30 clips at a time and then pause for 15 minutes prior to viewing the next set of clips. The clips were not shown in the same order to all women to minimize the possibility of habituation.
Description of Rating Scale
The rating scale was a bidirectional scale developed to measure mental appeal and self-reported physical arousal by members of the research team. Each scale ranged from −5 to +5 with 0 being neutral. Positive values indicated that the subject experienced pleasant feelings while watching the clip and that it made them feel physically “turned on.” Negative values indicated that they experienced unpleasant feelings while watching the clip and that it made them feel physically “turned off.”
Main Outcome Measures
The main outcome measures were the level of mental appeal (M) and the level of self-reported physical arousal (P). Secondary outcome measures were correlations between mental appeal and reported physical arousal and identification of clips that were most and least appealing/arousing. For the identification of the clips, the following categories were designated.
- 1Mentally appealing and physically arousing (defined by average score >2 for mental appeal and physical arousal)
- 2More physically arousing but mentally unappealing (defined by M < P with M < 0)
- 3More physically arousing than mentally appealing (defined by P > M with M > 1)
- 4Mentally unappealing and physically unarousing (defined by M and P < 0)
- 5More mentally appealing than physically arousing (defined by M > P)
See Table 2 for a summarization of these categories.
Table 2. Description of video categories
The data were analyzed using descriptive statistics with SPSS 14.0 for Windows (SPSS Inc., Chicago, IL, USA). Frequency tables were generated for each category and the clips in each category were identified. To determine the degree of inter-rater reliability, the coefficient of variation was calculated. Pearson correlations were calculated to determine the relationship between mental appeal and reported physical arousal for each category of clips. Data were expressed as mean ± standard error of mean.
An independent samples t-test was used to assess if there was a difference in ratings between subjects who rated the video clips at home vs. those who rated them in the laboratory. A one-way analysis of variance was used to see if there were any differences in ratings between subjects who rated the video clips in different sequences.
For all 90 film clips, the mean mental appeal score was 0.91 ± 1.34 and the mean reported physical arousal score was 1.56 ± 0.94. These indicate the mean ratings of the 90 videos, not the average scores yielded by the 21 subjects. There was a significant positive correlation between mental appeal and reported physical arousal (r = 0.86, P < 0.05).
There were no significant differences in ratings between subjects who rated the videos at home (N = 7) vs. those who rated them in the laboratory (N = 14) (P > 0.05). Furthermore, there were no significant differences in ratings between subjects who watched the video clips in various sequences (P > 0.05).
Mentally Appealing and Physically Arousing (Defined by Average Score >2 for Mental Appeal and Physical Arousal)
Videos that were reported as being the most mentally appealing and physically arousing had a mean subjective score of 2.54 ± 0.39 and a mean physical score of 2.48 ± 0.24. The coefficients of variation were 0.15 and 0.10, respectively. While collectively for the entire cohort of women studied, there was no significant correlation between mental appeal and reported physical arousal (r = 0.319, P = 0.197); when analyzing ratings on an individual basis, there was a significant positive correlation (r = 0.61, P < 0.05).
Out of the 90 film clips, 18 met these criteria. They were characterized as containing scenes that involved (percentage refers to the number of clips with the characteristic divided by the number of clips in this category) vaginal intercourse (83.33%), male on female sexual positions (77.78%), outdoor settings (55.56%), missionary positions (33.33%), female on male sexual positions (27.78%), rear-entry vaginal intercourse (22.22%), characters with race other than Caucasian (5.56%), and cunnilingus (5.56%).
Mentally Unappealing with More Physical Arousal than Mental Appeal (Defined by M < P with M < 0)
Film clips that were reported as mentally unappealing but more physically arousing had a mean mental appeal score of −0.80 ± 0.77 and a mean physical arousal score of 0.46 ± 0.98. The coefficients of variation were 0.96 and 2.14, respectively. There was a significant positive correlation between mental appeal and reported physical arousal (r = 0.65, P < 0.05). Interestingly, 37.5% of these films were also identified as the least mentally appealing and physically arousing.
Out of the 90 film clips, 24 met the above criteria. They were characterized as having scenes showing fellatio (62.50%), rear-entry anal or vaginal intercourse (41.67%), anal intercourse (37.50%), two males and one female (33.33%), partner masturbation (29.17%), vaginal intercourse (29.17%), male on male sexual positions (25%), group sex (16.67%), cunnilingus (16.67%), self-masturbation (16.67%), female on female sexual positions (8.33%), female on male sexual positions (8.33%), and two females and one male (4.17%).
More Physically Arousing than Mentally Appealing (Defined by P > M and M > 1)
Film clips that were reported as more physically arousing than mentally appealing had a mean mental appeal score of 1.65 ± 0.66 and mean physical arousal score of 2.32 ± 0.52. The coefficients of variation were 0.40 and 0.22, respectively. There was a significant positive correlation between mental appeal and reported physical arousal in this group of clips (r = 0.37, P < 0.05). Out of the 90 films, four met these criteria. They were characterized as having scenes depicting two-person heterosexual behavior (100%) with 50% showing male on female sexual positions and the other 50% showing female on male sexual positions. Half of the clips depicted partner masturbation. Vaginal intercourse was present in 75%. Interestingly, none of these made it to the most mentally/physically appealing tier.
Mentally Unappealing and Physically Unarousing (Defined by M and P < 0)
Film clips that were viewed negatively in terms of mental appeal and physical arousal had a mean mental appeal score of −1.54 ± 0.73 and a mean physical score of −0.66 ± 0.43. The coefficients of variation were 0.47 and 0.65, respectively. There was a significant positive correlation between mental appeal and reported physical arousal (r = 0.62, P < 0.05).
Out of the 90 film clips, nine met these criteria. They were characterized by scenes showing male on male sexual positions (66.67%), fellatio (66.67%), anal intercourse (44.44%), partner masturbation (33.33%), cunnilingus (22.22%), two males and one female (22.22%), vaginal intercourse (11.11%), group sex (11.1%), and self-masturbation (11.1%). Seven of these clips depicted male homosexual behavior. All of these had higher physical scores than mental scores.
More Mentally Appealing than Physically Arousing (Defined by M > P)
Film clips that were reported as being more mentally appealing than physically arousing had a mean mental appeal score of 2.38 ± 0.67 and a mean physical arousal score of 2.00 ± 0.62. The coefficients of variation were 0.28 and 0.31, respectively. There was a significant positive correlation between mental appeal and reported physical arousal scores (r = 0.59, P < 0.05). Interestingly, over half of these clips were also found to be the most mentally and physically arousing (55.55%).
Out of the 90 film clips, 18 met these criteria. They were characterized as having scenes involving male on female sexual positions (66.67%), vaginal intercourse (50.00%), outdoors setting (38.89%), female on male sexual positions (22.22%), characters of non-Caucasian race (16.67%), rear-entry vaginal intercourse (16.67%), cunnilingus (11.11%), partner masturbation (11.11%), missionary position (11.11%), anal intercourse (5.56%), bondage (5.56%), and fellatio (5.56%).
Among this group of video clip raters, the films that were found to be the most mentally appealing and physically arousing all contained heterosexual scenes. They tended to be sexually conservative, depicting male on female vaginal intercourse. This may be explained by the sexual involvement theory, which states that if there is sufficient overlap between the erotic scene and an individual's sexual script and affect, subjective arousal may be increased [31–33]. For example, a woman may be more receptive to, and therefore, aroused by a film that shows an average-appearing male and female couple engaging in vaginal intercourse. Using “realistic” characters with whom she can identify allows her to perceive the situation as “sexy” and results in increased interest in and involvement with the video.
There were films that were found to be more physically arousing than mentally appealing with negative mental appeal ratings. Film clips in this category were generally rated lower than others; however, there was still a significant positive correlation between mental appeal and physical arousal. Films where women felt physically aroused but were subjectively “turned off” were those that contained themes such as group sex, fellatio, and anal intercourse. There are multiple explanations that may account for this finding. Religious, social and cultural values and mores may influence what content an individual woman finds acceptable or appealing. Women may be embarrassed or ashamed to admit that they find certain behaviors mentally appealing. Lastly, women may interpret the behaviors in these clips as being objectifying or degrading to women. In spite of finding these clips mentally unappealing, many women (although not all) still found themselves physically aroused. Perhaps the shock and/or novelty of the clip triggered a sexual response. It is also possible that women may be less selective in their physiologic arousal, being physically aroused by innately recognized sexual features, even when they are out of a “normal” context .
There were some film clips that were identified as being more physically arousing than mentally appealing. Only a few clips were in this category and interestingly, none of them made the most appealing and most physically arousing category. This may suggest and support Laan et al.'s  notion that the stimulus content and meaning may play a larger role in eliciting female sexual response. This also suggests that mental appeal may carry greater weight than reported physical arousal in determining which clips are most appropriate for studies in female sexual function.
There were films that women viewed negatively in both aspects. These films were characterized as having scenes involving male homosexuality, anal intercourse, and fellatio. All of them involved male–male homosexuality or group sex. Again, religious, social and cultural values may play a role, and the absence of male homosexuals from our sample may have limited the appeal of this category of stimuli.
Films that were reported as more mentally than physically arousing were similar to the ones that were considered most mentally and physically arousing in that they all portrayed heterosexual scenes and most of them focused on vaginal intercourse. Over half of these clips were also identified as the most mentally appealing and physically arousing, supporting the theories that more sexually conservative clips tend to optimize arousal and mental appeal and may play a larger role in optimizing female sexual response.
Overall, we were able to demonstrate a significant positive correlation between mental appeal and reported physical arousal, meaning that reported physical arousal increases as mental appeal increases. When the categories were broken down, we were still able to demonstrate this when assessing the data on an individual basis. Additionally, when evaluating inter-rater variability within categories, there was little variation in how women rated the most mentally appealing and physically arousing clips; however, there was more variability in the ratings of clips in other categories. There was a large amount of variation in the ratings of the physically arousing but mentally unappealing category. Interestingly, the content of these clips was similar to those in the mentally unappealing and physically unarousing category. This phenomenon may be explained by the individual differences in how women rated clips containing less conservative sexual behaviors. Our data imply that women agreed more on which clips were the most appropriate, while there was more variation in how women viewed clips that were not as mentally appealing or physically arousing.
It should be noted that these categorical boundaries are generalizations that are relevant to our sample of patients. Obviously, there are women in our sample that have individual preferences that do not conform to the model. It would be interesting to look at these women to see if there are identifiable factors, such as age, sexual orientation, and sexual experiences that influence their responses.
There is clearly a need to identify appropriate stimuli in studies of sexual function. Indeed, the mode of stimulus presentation has important effects on the magnitude of sexual arousal elicited [35,36].
The strengths of our study include a broad range of clips from a large selection of pornographic videos. Most importantly, our study design allowed us to ascertain the level of mental appeal which is distinctly different from asking how mentally or subjectively aroused a subject becomes as a result of viewing an erotic video clip. However, our study has a number of important limitations. First, our study population was small, making it impossible to assess relationships between age, race, education or sexual orientation, and response to erotic films. It is certainly possible that these variables may have contributed to our findings. The heterogeneity of our sample, while appropriate for a pilot study, does not make our findings applicable to all women. Although many women completed the task in the privacy of their own home, some did so within the study environment, which can be inhibitory to sexual response, thus confounding our results. However, we did not find a statistically significant difference in ratings between subjects who rated the clips at home compared with those who rated them in the lab. The level of physical arousal was self-reported and not an objective measure (as is vaginal photoplethysmography, for example ). There may also be limitations inherent to the scale we used. Because it has not yet been validated, it can only provide us with relative measures of reported mental appeal and physical arousal. As with most sex research, there is a potential for selection bias and the “snowball sampling” effect although we made attempts to minimize both. Furthermore, we did not ascertain the sexual function status of the participants by validated questionnaires, which may result in skewed data (i.e., women with sexual arousal disorder may have rated the clips much lower than women with normal sexual function). It is also possible that we exhibited some bias in the selection of film clips based on our own personal preferences. We assumed that the behavioral content of the clips determined how women rated them; however, it is also possible that other variables such as the attractiveness of the characters, the level of the engagement of the actors, and the degree of explicitness (soft-core vs. hard-core) may have influenced their responses. We plan to address these limitations in future studies.
Another important potential limitation to consider is the possibility of habituation. Subjects had to view 90 clips and there is some concern that this could potentially skew ratings of clips viewed toward the end. Meuwissen and Over  found that there was a decrease in genital and subjective arousal with repeated exposure to erotic films or erotic fantasies; however, there were some flaws in measurement and no control for physiologic fatigue. In a subsequent study, Laan and Everaerd found no habituation of genital arousal and anonsignificant decline in subjective arousal with repeated exposure to erotic film . We attempted to minimize the effects of habituation on ratings by counterbalancing the order of presentation to the subjects. In addition, our statistical analysis did not reveal any significant differences in ratings based on the order that the clips were presented in.
Our findings show that it is possible to identify the types of erotic video clips that may optimize sexual arousal in women. The use of appropriate clips provides the opportunity to improve the rigor in studies of female sexual function and dysfunction. This methodology should be considered in sex research studies of all populations, including men, older persons, and people of different sexual orientations.
Although our study suggests a way to improve the methodology of selecting erotic film clips for research in female sexual function, there are still many questions to be answered before it is optimized. Film clips are rich media that consist of multiple variables such as level of explicitness, auditory and visual cues, scene changes, and script, all of which may impact their utility in studies of female sexual function and dysfunction. Future studies should address this, as well as aim to explore the relationship between demographic variables and sexual arousal induced by erotic films in the context of sexual function indices and physiologic data.
Erotic film clips reliably produce a state of mental appeal and self-reported physical arousal in women. The level of mental appeal and physical arousal vary with the content of the clip. It is possible to identify clips that optimize both mental appeal and reported physical arousal. These types of clips should be the standard for future studies of female sexual response.
Conflict of Interest: None declared.