Abstract
Although there is growing evidence that receiving positive emotional support (e.g., empathy) facilitates improved mental health outcomes among intimate partner violence (IPV) victims, there has been minimal exploration of factors that might undermine the likelihood of such supportive responses. The current study addressed this issue by examining whether exposure to sexualized music videos would affect IPV victim-directed empathic responding of third-party respondents. In a three-condition design, 243 female Fijian university students viewed sexualized, nonsexualized, or neutral music videos. They then read about a male-to-female IPV incident involving a university student victim who focused heavily on academic success and rated aspiration-related culpability and empathic responding for the victim. Relative to those who viewed neutral and nonsexualized videos, those who viewed the sexualized video reported less victim-directed empathy. Moreover, the impact of video type on empathy was mediated by aspiration-related culpability (i.e., the perception that the victim studied too much). The present research examined, in an understudied, patriarchal population (Fijian women) with an extremely high rate of IPV, how exposure to sexualized music videos can contribute to both greater blame and greater desensitization to the suffering of an IPV victim. The importance of studying third-party responders (bystanders) is that they may represent a fundamental resource for the victim, or by contrast, if they fail to respond empathically, they would be unsupportive to a victim. This provides some directions for facilitating social controls and decreasing social tolerance for harmful patriarchal beliefs and gender-based violence in the Pacific Region of the world.
Intimate partner violence (IPV) can be defined as the physical, sexual, or psychological harm perpetrated by a current or former partner. In the developing country Fiji, 64% of women who have ever been in an intimate relationship experienced physical violence and 58% of ever-partnered women experienced emotional violence (i.e., psychological abuse) from a husband or intimate partner in their lifetime (Swami, 2016). By comparison, 25% of ever-partnered women in America will experience such violence in their lifetime (Smith et al., 2015). However, there has been limited research focused on third-party reactions (i.e., those who are aware of the incident but not directly involved) to IPV in Fiji. Importantly, there is clear evidence that third-party emotional support tends to buffer IPV victims from the deleterious mental health outcomes associated with their violent experiences (Fortin et al., 2012; Lee et al., 2007; Žukauskienė et al., 2019). More broadly, previous research has shown that the simple recognition that others empathize with one’s suffering has been shown to actually reduce the subjective pain appraisal from that suffering (Fauchon et al., 2017; Sambo et al., 2010).
Given the favorable mental health consequences associated with third-party emotional support toward IPV victims, it is striking that there has been minimal exploration of factors that might undermine the likelihood of such supportive responses. The current study addressed this lacuna in the literature by examining whether music video exposure (i.e., sexualized, neutral, or nonsexualized music videos) would affect Fijian university women’s emotional reactions (i.e., empathy) toward a female IPV victim. Moreover, the victim was described as a university student with strong academic aspirations (i.e., planning to be an engineer) and spending a significant amount of time studying. Importantly, recent findings indicate that girls and women have been socialized to prioritize fostering relationships with males and give less focus to academic and intellectual pursuits (Nelson & Brown, 2019; Stone et al., 2015). As a consequence of this gender-biased socialization, for instance, women’s underrepresentation in Science, Technology, Engineering, and Mathematics (STEM) fields may reflect larger social resistance to women entering male-dominated fields because of rigid stereotypes that link men (vs. women) more strongly with STEM competence (Nosek et al., 2002).Further, exposure to sexualized media is thought to play an important role in generally reinforcing such perspectives among young women (Collins, 2011; Ward et al., 2005, 2020). Consequently, we further investigated whether one underlying mechanism for measuring the effect of music video type on empathic responding was aspiration-related culpability (e.g., she studied too much and neglected her relationship).
Although there are a number of denotations of empathy (see Johnson & Lecci, 2019, for greater discussion), in the current examination we will adopt the “empathic concern” perspective (see Batson et al., 2002). A focus on intragroup empathic responding among women is relevant because of the evidence that women IPV victims are most likely to seek emotional support from other women (see Sylaska & Edwards, 2013 for a review). These findings are not surprising given that individuals tend to show a high degree of empathy for in-group members who are facing adversity across a variety of situations (Batson & Ahmad, 2009; Cikara et al., 2011; Mathur et al., 2010). A high degree of intragroup empathy may be due to a strong sense of identification (referred to as“feelings of oneness”) that is critical to evoking empathic concern for others (Cialdini et al., 1997; Maner et al., 2002) and is experienced for in-group members (Smith & Henry, 1996; Tropp & Wright, 2001). In contrast, people often have little empathy for out-group members who are in need (see Avenanti et al., 2010; Gutsell & Inzlicht, 2010). However, there has been minimal empirical explorations of factors that might moderate intragroup empathic responding for those inneed. An examination of the limits of IPV-related intragroup empathic responding among university women seems important given the exceptionally high prevalence of dating violence on college campuses (Sabina & Straus, 2008; Smith et al., 2003) with the concomitant serious physical and mental health consequences for victims (Black et al., 2011; Kilpatrick et al., 2003).
Third-party Influences on Intimate Partner Violence: The Role of Sexualized Media
According to literature on the third-party perceptions of, and reactions to, a dangerous situation—that is, the so-called bystander effect (Latané& Darley, 1970)—bystanders are meant as individuals who are not directly involved in the emergency situation and do not necessarily know the victim of maltreatment. Nevertheless, by their very presence, bystanders might represent a fundamental resource for the victim, or by contrast have the potential to do nothing,and being unsupportive in response to a victim (Banyard, 2011). This is why social psychologists have been widely investigating factors that may lead bystanders to intervene (or not) in favor of the victim herself (for a recent review, see Pagliaro et al. [2020]): their intervention, in fact, “operates like a feedback loop, where the consequences of taking action have repercussions for future action” (Banyard et al., 2019, p. 1) and have the potential to change culture and group norms, fostering actionism among third parties.
Importantly, Gracia and Herrero (2007) contend that third-party strong beliefs regarding the problematic nature of the perpetrator’s action tend to create a climate of social intolerance toward IPV victims which could lead to improved social control of IPV. The necessity for such controls is amplified in patriarchal societies such as Fiji. Specifically, the Global Gender Gap Index is a measure of women’s relative standing compared to men in four key areas: health, education, economy, and politics. In addition to being in the lowest third of ranked nations, Fiji’s ranking on this index has been falling (i.e., from 108th out of 145 countries in 2010 to 121st in 2015), which demonstrates worsening gender inequality (World Economic Forum, 2010, 2015).
An earlier study found that the amount of time adolescents watch music videos exceeds the amount of time they spend watching TV or with any other medium (Roberts & Foehr, 2004). Importantly, the women in these types of media tend to be depicted in a sexually submissive manner (Conrad et al., 2009) while wearing provocative clothing (King et al., 2006; Seidman, 1992; Smith, 2005). Relevant for our study, the portrayal of women and girls by western media in the last decades has become increasingly sexualized (e.g., American Psychological Association, 2007; Zurbriggen & Roberts, 2013). Female sexualization, which is visible in all fields (from advertising, to politics, to academic and working settings; see Pagliaro et al.(2020) for a greater discussion), reinforces a rape culture, according to which victims of IPV should be blamed instead of supported for what had happened. Given that exposure to sexualized music videos may reinforce negative gender stereotypes, the damaging impact of this form of media may be amplified for women in patriarchal societies such as Fiji.
Not surprisingly, there is evidence that exposure to sexualized music videos can influence third-party negative responses toward women. For example, Kistler and Lee (2009) found that men who viewed highly sexual hip-hop music videos expressed greater rape myth acceptance, objectification of women, and stereotypical gender attitudes relative to those who viewed hip-hop videos with less sexual content. In one of the few explorations of the impact of sexualized videos on dating violence in women relative to men, Johnson and colleagues (Johnson et al., 1995) found that adolescent girls who viewed sexualized videos were more likely to accept teen dating violence than those who did not view such videos. Conversely, adolescent males were not influenced by exposure to such videos.
Media and Desensitization
Although there is an expectation that most humans have innately negative physiological and psychological reactions to observed violence, the central tenet of desensitization theory (see Anderson & Bushman, 2002; Anderson et al., 2010; Bartholow et al., 2006) is that repeated exposure to violence in the media can lead to an attitude of numbness toward such violence. More specifically, Rule and Ferguson (1986) suggest that desensitization is the attenuation of distressing reactivity to observed violence after some degree of exposure to violence either in real-life or in the media. From a similar perspective, Vossen et al.(2017) contended that desensitization can be understood as a decreased physiological, emotional, and cognitive response to real-world violence and is thought to be an adaptive process to help individuals dealing with distress associated with the confrontation of violence.
For the present study, and consistent with the perspectives of a number of scholars in this area of research (Carnagey et al., 2007; Dexter et al., 1997; Funket al.,2003), desensitization will be operationalized as reduced empathy for the pain and suffering of another individual. In research directly relevant to the current study, Weiner (1980, 1986) proposed that exposure to the needs of another person stimulates a search for the causes of the person’s struggles by an observer. A central component of this process involves whether the person is perceived to be culpable for their own suffering and this has been referred to as the locus of responsibility. This locus of responsibility attribution, in turn, influences the emotional reaction to the situation. A high culpability attribution tends to lead observers to blame the person-in-need for his or her situation and significantly reduce the likelihood of positive emotional responses, such as empathy.
The victim in the present research is engaged in certain empowering behaviors (i.e., studying a lot, hoping to become a female engineer). There is clear evidence that such goals are in direct contradiction to societal pressures for young women to be relationship focused and suppress agentic and/or intellectual pursuits while prioritizing their romantic relationships (see Starr & Ferguson, 2012; Stone et al., 2015). We expected that exposure to sexualized media would prime these types of relationship-focused stereotypical and misogynist beliefs, which would lead to greater perceived blame (i.e., she should not be engaging in such behaviors and focus on her relationship) of a victim with intellectual goals. This increased perceived “aspiration-related” culpability should, in turn, reduce empathic responding (i.e., desensitize) for her suffering.
The Present Study
Female participants were exposed to 30 minutes of various sexualized, neutral, or nonsexualized music videos. 1 Subsequently, they read a passage depicting an incident of male-to-female IPV. Afterward, participants reported their perceptions of victim culpability (i.e., aspiration-related culpability) and empathic responding for the victim.
On the basis of the rationale described above, we predicted an effect of the video type for both aspiration-related culpability and empathic responding. We expected that participants in the sexualized condition, compared to those in the nonsexualized and neutral conditions, should report (a) greater perceptions of aspiration-related culpability (i.e., the victim neglected her relationship for academic pursuits) (Hp1) and (b) less reported empathic respondingfor the victim (Hp2). We further expected mediational effects of perceived culpability, whereby exposure to sexualized media would increase aspiration-related culpability which would then facilitate reduced empathic responding (i.e., desensitization) for the victim’s suffering (Hp3).
Method
Participants
Two hundred and forty-three female students from the University of the South Pacific (Suva, Fiji) participated in the study. Their age ranged from 18 to 37 (M = 21.16, SD = 3.23). They were compensated for their time with $10. They were recruited from student societies on the campus using blanket email requests to society members.
Procedure
This study was fully reviewed and approved by the university institutional review board. The University of the South Pacific does not have a dedicated pool of participants (i.e., subject pool) to conduct experimental research. Consequently, for psychology research, participants are recruited through student organizations via email requests. For the current study, recruitment emails provided the time and location for the session, and participants gathered in a large auditorium to view the videos. Once the requisite number of participants arrived (a minimum of 75 per condition), the session was closed. Participants were told that psychologists were very interested in their views on a number of media, social, and life issues. Specifically, they were informed,“First, you will be viewing and rating video materials. Then, we will be asking you a number of questions regarding the videos and your ‘Life and Societal Impressions.’” In the sexualized video condition (N = 81), participants viewed eight videos which were chosen because they had included some or all of the following: (a) sexual themes and lyrics, (b) scantily dressed women, and (c) women who danced provocatively around men. Examples of the songs included Nikki Minaj “Anaconda,” Rita Ora ft. Chris Brown—“Body on Me,” and Jennifer Lopez ft. Iggy Azalea—“Booty.” In the nonsexualized video condition (N = 76), participants viewed eight videos that did not involve any of the sexual themes or images cited above in the sexual video condition. Examples of the songs included Adele—“Rolling in the Deep,” Wiz Khalifa ft. Charlie Puth—“See You Again,” and Taylor Swift and Ed Sheeran—“Everything Has Changed.” Finally, in the neutral condition (N = 86), participants viewed a neutral video focused on ocean life. The decision to use music videos is based on the fact that they are familiar, engaging, and brief, and can also convey sexualized content, thereby making them appropriate to represent the larger concept of sexualized media consumption. (Note: Assignment to condition was random in the sense that there were no systematic differences between the three scheduled sessions, and participants attended whichever of the three sessions was most convenient for their schedule, but they were not aware of which condition would be assigned to their session.)
In a previous study that also used the current data set with a different research aim and hypotheses (Johnson et al., 2020), the video conditions were validated by 20 independent raters (females from the same university as the participants), demonstrating that the sexualized videos had significantly greater sexual imagery, sexual lyrics, and female sexual objectification relative to the nonsexualized video condition. Further, the results also demonstrated that both types of videos were perceived to have virtually no violent content. The neutral condition was not included in these ratings because it involved ocean life, making the questions inapplicable.
Consistent with the instructions, participants rated the videos on the extent they enjoyed them and whether they would purchase them. They were then given a passage to read entitled “Life and Society Impressions” with a subheading of “Relationship Issues.” The passage involved Dawn and Robert who were third year college students who had been dating for a year. Dawn was described as an engineering major, having good grades, and spending a lot of time studying. Dawn told Robert that “being the first female engineer in a major engineering company” was her goal. Robert informed Dawn that he was a bit lonely and she should be a “good girlfriend” and spend more time with him. The next section followed:
One evening, Dawn arrived home late. Before she opened the door completely, Robert grabbed her by the arm and pulled her into the house. “Sometimes I just don’t understand the things that you do!” shouted Robert. “You are just never here for me! I am so stressed right now.” She tried to explain but before she could utter two words, his hand whipped across her face sending a crack of skin contacting skin echoing across the house. She attempted to explain again, but this time hails of punches reached her face and her abdomen. As she fell to the ground in a crouching position, he kicked her in the stomach and shouted, “Look what you made me do! This is all your fault!
Participants responded to three items (which were averaged) that assessed their perceived aspiration-related culpability (i.e., women should not prioritize career over relationships, Spearman Brown = .70) on a 5-point Likert scale (1= strongly disagree, 5 = strongly agree). The specific items were: (a) “Dawn did not behave like a good girlfriend by studying so much;” (b) “Dawn should stop spending so much time studyingand focus more on her relationship;” and (c) “Dawn deserved what happened to her because of her actions.” Next, participants completed five items measuring their victim-directed empathic responding (α = .79). Specifically, they rated their experience of sympathy, compassion, warmth, soft-heartedness, and being moved toward Dawn on a 1 (not very much) to 7 (extremely) Likert-type scale (see Batson, 1991; Batson et al., 2002). Participants were then debriefed, paid, and released.
Results
Power Analysis
With an expected effect size of η2 = .05 based on the available literature (e.g., Aubrey et al., 2011; Kahlor & Morrison, 2007) and a sample size of 243 participants, the power for the main effect of video exposure with three levels equals .90. The literature suggests R2 values of approximately .10 for direct and indirect effects in the hypothesized mediation analyses, which would yield a power of > .99 for the achieved sample size.
Perceived Aspiration-related Culpability
For all of the following analyses, any differences in the degrees of freedom are due to instances of missing values. Consistent with Hp1, a one-way analysis of variance (ANOVA) showed a significant effect of video type on relationship-neglect culpability, F (2, 240) = 11.39, p< .000, η2 = .09. Post-hoc analyses showed that those in the sexualized video condition (M= 2.28, SD= .94) reported stronger perceived culpability than those in the nonsexualized condition (M= 1.66, SD= .73, p< .001) and those in the neutral condition (M= 1.84, SD= .82, p= .001). Perceived culpability between those in the nonsexualized and neutral conditions was not significantly different (p= .17).
Victim-directed Empathy
Consistent with Hp2, a one-way ANOVA showed a reliable effect of video type on victim-directed empathy, F (2, 236) = 3.53, p = .031, η2 = .029. Post-hoc analyses of the means showed that those in the sexualized video condition (M= 5.25, SD= 1.07) reported less empathy than those in the nonsexualized condition (M= 5.62, SD= .88, p= .023) and those in the neutral condition (M= 5.61,SD = 1.00, p= .022). Empathic responding between those in the nonsexualized and neutral conditions was not significantly different (p= .964).
Mediation Effects
A PROCESS Macro Model 4 (Hayes, 2013) was used to determine whether the impact of video type on victim-directed empathic responding was mediated by perceived culpability. We created two coded vectors to represent the three video type conditions. The first vector represented the effect of sexualized video relative to neutral video condition, while the second vector represented the effect of sexualized video relative to the nonsexualized video condition. Using this procedure, we estimated (relative to the neutral and nonsexualized conditions) the direct and indirect effects of video type.
Consistent with Hp3 (see Figure 1), a bootstrapping procedure with 5,000 resamples showed that the indirect effect of the experimental condition on victim-directed empathic responding was mediated by aspiration-related culpability (B = –.03, 95% CI [–.0818, –.0090]). The 5,000 bootstraps results revealed full mediation.

Note. Dashed lines represent non-significant effects. The −.62 coefficient from the sexualized vs non-sexualized vector reflects greater aspiration-related culpability. Direct effects of video type on victim-directed empathy are shown in parentheses. *p< .05; **p< .01.
General Discussion
The present research examined, in an understudied patriarchal population with an extremely high rate of IPV (Swami, 2016;WHO, 2013), one factor (i.e., exposure to sexualized music videos) that can contribute to both greater blame and greater desensitization to the suffering of an IPV victim. Relative to those in the neutral and nonsexualized video conditions, sexualized video condition participants reported greater perceived culpability and reduced empathic support for the IPV victim. Most importantly, a mediational analysis revealed that the impact of video type on empathic responding was mediated by perceivedaspiration-related culpability (i.e., women should not prioritize career over relationships). This represents one potential epidemiological pathway for the alarmingly high rates of IPV in this region of the world.
The current study also focused on the perceptions of third-party players. This has practical relevance because a recent report indicated that a significant majority of IPV victims chose to confide in a third party instead of legal authorities. Thus, the author of the report contended that much greater empirical attention should be given to the role of third-party responses in reducing such violence (Parker, 2015). This is also in line with research showing the fundamental role of bystanders in reducing and preventing violence against women, in particular (but not only) on college campuses (Banyard, 2008, 2011). In other words, engaged bystanders may contribute to changing misogynistic culture and group norms, while disengaged ones may reinforce a masculine culture where maltreatment is tolerated and even condoned. Consequently, third parties (i.e., bystanders) can represent a valuable source of support for the victims of maltreatment, even when they do not know the victim.
In terms of the contribution to the IPV literature, the bulk of extant research on reactions to IPV victims has focused on the deleterious consequences of sexuality-related actions of the actual victim such as infidelity (Baldry & Pagliaro, 2014; Johnson et al., 2019; Pagliaro et al., 2021; Spaccatiniet al., 2019; Baldry et al., 2015; Pacilli et al., 2017). The current examination demonstrated that stimuli not related to the IPV victim’s own actions (i.e., sexualized music videos) has direct implications for victim-directed perceived culpability and empathic responding in Fiji.
Sexualized Music Videos and Intragroup Empathy
Previous research has demonstrated that emotional support buffers IPV victims from the deleterious mental health outcomes associated with their violent experiences (Fortin et al., 2012; Lee et al., 2007; Žukauskienė et al., 2019). Moreover, there is evidence that women IPV victims are most likely to receive such support from other women (Sylaska & Edwards, 2013). These findings are consistent with the extensive evidence that individuals show high degrees of empathy for ingroup members (Cikara et al., 2011; Mathur, et al., 2010; Xu et al., 2009). Consequently, the current examination of factors that might influence such intragroup empathic responding for women IPV victims has both practical and theoretical implications. One of the central findings for the present study was that sexualized music exposure reduced women’s intragroup empathic responding toward a female IPV victim. The pattern of results demonstrates that, even in cases of extreme violent acts, intragroup empathy is malleable (reduced by sexualized media exposure). It is noteworthy that perceived culpability ratings in the sexual music video were not very high (an average of 2.33 on a 5-point scale). However, this minimal degree of blame attribution was sufficient to lead to a reduced empathic responding (i.e., desensitization) toward the victim’s suffering. Further, given the nature of the scenario and the brutality of the assault, any degree of victim culpability attribution and desensitizationis extremely problematic.
Media and Desensitization
Both cross-sectional and longitudinal studies have shown that exposure to violent media tends to facilitate desensitization toward the violent treatment of others (e.g., Carnagey et al., 2007; Dexter et al., 1997; Fraser et al., 2012; Krahé & Möller, 2010). Another contribution of the current investigation involves demonstrating that nonviolent media exposure (i.e., sexualized music videos) can also facilitate such desensitization. This particular finding extends the media desensitization literature by revealing that, in addition to the more direct and clearly violent media forms, exposure to the less direct, more subtle, and commonly available forms of media can also desensitize individuals to the suffering of others. It would be interesting for future research to explore what types of media might increase sensitivity to the struggles of IPV victims. For example, there is growing evidence that exposure to prosocial songs enhance the likelihood of prosocial responding (Greitemeyer, 2009a, 2009b). It is possible that these types of stimuli might also increase third-party empathic responding to suffering others.
Although a full understanding of the underlying mechanisms associated with the current finding is unknown, it is possible that exposure to sexualized music videos may have primed certain biased and stereotypical constructs (e.g., female objectification and misogynous thoughts). Because those constructs would be less compatible with the victim’s empowering activities (i.e., intellectual pursuits, see Starr & Ferguson, 2012; Stone et al., 2015), participants would more easily assign culpability (i.e., she should not be engaging in such behaviors) to the IPV victim which, in turn, would desensitize (i.e., reduce empathic responding) third parties to her suffering. Importantly, this possible mechanism would be consistent with the perspective of Weiner and colleagues who argue that exposure to the needs of others stimulates a search for the causes of the person’s struggles (see Weiner, 1980, 1986). Because the victim expressed strong career-focused (vs relationship-focused) goals, this would lead observers to blame the victim for her situation and significantly reduce the likelihood of empathic responding.
Limitations and Conclusion
Several limitations are noted that might undermine the generalizability of the findings. First, the IPV information was conveyed to the participants through a written passage rather than a more realistic method such as video depictions, thereby potentially limiting realism and generalizability, though the general public could conceivably learn of such incidents by reading similar summaries in the newspaper or online forums. Second,the participants in the current study were limited to university women in the Fijian population. Thus, the findings may not generalize to males, to other age groups,or to those from western or more developed countries (though as noted, much of the research in other parts of the world yields similar patterns). Third, as with any social psychological research, demand effects and socially desirable responding may have contributed to the findings. However, the mediational analysis findings likely reduce the concerns for demand bias as they provide strong evidence for complex inter-variable interactions that should be less responsive to more straightforward demand effects. Of course, we recognize that tested mediation is not a substitute for demonstrating temporally ordered causality. Fourth, the obtained effects, while statistically significant, were nevertheless modest. This is to be expected given the many other factors that can influence empathic responses/attitudes to IPV, and the brief video exposure would be dwarfed by the lifetime of media exposure all participants have experienced. Fifth, other forms of media exposure could be investigated to determine if the effects are similar, though presumably any engaging media regardless of the specific source or platform would be expected to yield similar results. Finally, the current study does not speak to the duration of the experimental effect, as the outcome measures were administered shortly after the video exposure. Thus, future research could evaluate the course and durability of the effect over longer periods of time.
With respect to diversity, it is important to acknowledge that although there is clear evidence of greater media access for non-WEIRD samples (Poushter et al., 2018), the current research is among the first to document the potential harmful gender-based violence consequences of such enhanced access to sexualized music videos. The fact that the sample comes from a typically under-studied part of the world, further highlights the potential contributions of the findings. Indeed, given the degree of pervasive patriarchy in Fijian society (i.e., 121 out of 145 countries on the Global Gender Gap Index; World Economic Forum, 2015), exposure to these types of stimuli could be especially damaging for young women in this region of the world. Moreover, there is also evidence that the simple recognition that others empathize with one’s suffering can actually reduce the subjective pain appraisal from that suffering (Fauchon et al., 2017; Sambo et al., 2010). The present findings suggest that third-party exposure to sexualized music video may, in fact, reduce the likelihood that an IPV victim in Fiji would receive such healing emotional support. At any rate, it is hoped that the current findings might contribute to the efforts to both facilitate social controls and decrease social tolerance for all forms of harmful patriarchal beliefs and gender-based violence in the Pacific Region of the world.
Footnotes
Declaration of Conflicting Interests
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
