Trait Empathy, Narrative Engagement, and Emotional Reactivity in
Response to Film: A Mediation Analysis
Research Master Thesis Graduate School of Communication
Name: Mariana Orozco Fernandez Student #: 10392807
Supervisor: Dr. Helen Vossen Date: June 27, 2014
Abstract
Emotions are an integral part of our experience with media, and can be influenced by individual factors, among them empathy. This study aimed to analyze the mediating role of narrative engagement in the relationship between individual trait empathy and emotional reactivity while viewing fiction films, as well as their relationship to enjoyment. Emotional reactivity was operationalized as the amount and intensity of facial expressions while viewing and was measured using Noldus FaceReader software. Trait empathy was considered to be a multidimensional concept, encompassing both an affective and a cognitive factor. Participants (n=94) watched fragments of fiction films, edited to elicit the emotions of sadness and happiness. No direct relationship was found between either dimension of empathy and emotional reactivity. Results show that affective empathy is positively related to narrative presence, a dimension of engagement, while narrative focus, another dimension of engagement, is related to emotional reactivity. Affective empathy has an indirect effect on enjoyment through narrative engagement. Narrative engagement appears to be a complex construct that affects parts of the viewing experience in different ways.
Keywords: affective empathy, cognitive empathy, narrative engagement, emotional reactivity, facial expressions, enjoyment
Trait Empathy, Narrative Engagement, and Emotional Reactivity in Response to Film; A Mediation Analysis
Fictional stories across many genres have the ability to touch our emotions (Wirth & Schramm, 2005). The experience of emotions while viewing media is one of the key determinants of why we seek out, or avoid, certain types of content, and whether or not we are entertained. But why do we experience these emotions? Zillman (1994) proposed that empathy is fundamental to explain emotional involvement with drama, and many scholars have cited it as a key determinant of emotional responses to media (Hoffner, 2009). In order to like a character, viewers have to care about the character and share his or her emotions. If they remain indifferent to the character’s fate, there might be no entertainment. In essence, an essential component of emotional experience with media is considered to be empathy (Vorderer, Klimmt, & Rittefeld, 2004). The study of empathy in relation to media has recently gained attention, particularly because of findings that suggest that fictional stories can increase our ability to empathize (Mar, Oatley, & Peterson, 2009; Bal & Veltkamp, 2013; Comer & Castano, 2013). Understanding the processes through which that may occur could have considerable repercussions on the discussion of the beneficial aspects of literature and film.
Emotions are essential for our relationships with others, and are guided by empathy. Different levels of trait empathy among individuals could account for the degree of an emotional reaction, among other individual differences that influence the response to media. It is possible that, regardless of level of empathy, when a viewer is not fully engaged and focused on a story it is difficult to have an emotional reaction. However, little research has focused on how the extent that a person is engaged with a narrative affects the emotional reception to media. This leads to the following research question:
RQ: How is trait empathy related to emotional reactivity while viewing media, and how
does narrative engagement mediate this relationship?
The aim of this study is to investigate how trait empathy is related to emotional reactivity while viewing media, and whether people higher in trait empathy also exhibit more emotional reactivity. Empathy is considered to be a multidimensional concept, encompassing both affective and cognitive dimensions.Narrative engagement, the extent to which a viewer is immersed in the narrative, is considered to be a response state that mediates the relationship between trait empathy and emotional reactivity. Furthermore, given that emotional reactivity has been consistently connected to enjoyment, the effects of empathy and narrative engagement on enjoyment are also analyzed.
Emotional Reactivity
The experience of emotions is a key component of reception to media and entertainment. Emotional reactivity, sometimes referred to as emotional expressiveness, encompasses the intensity of the experience and display of an emotion in response to a stimuli (Roberts & Strayer, 1996). As Tan (2013) explains in his review of the psychological literature on empathy and film, films activate the same neuronal circuits as real life, more specifically, the Perception Action Mechanism, which perceives the actions of others as well as our own. In our daily life, we imagine future outcomes or remember past ones, which also produces emotions. These emotions have been compared to those elicited by fictional narratives (Busselle & Bilandzic, 2008). Fictionality is not a problem when experiencing emotions because audiences are prepared and motivated to accept this new reality. Films might even be more likely to produce emotional
responses than real events because, while viewing, social rules are suspended and viewers are free from the burden of regulating their emotions (Tan, 2013).
Understanding the emotions elicited by film can give insight into other media responses while viewing, which are at the core of entertainment, namely enjoyment and appreciation. In the comprehensive theory of enjoyment, emotions are considered to be a key component and a visible manifestation of this pleasant state, encompassing physiological, affective, and cognitive factors (Vorderer et al., 2004). Enjoyment is most often associated with feelings of happiness and joy (Oliver, 2008). But even negatively valenced emotions induced in films are often associated with enjoyment (Hoffner & Levine, 2005). As Oliver (1993) found, feelings of sadness are interpreted by many viewers as pleasurable. The one exception is when the content portrayed is very violent (Hoffner & Levine, 2005). Feeling pleasure from negatively valenced emotions in this line of research is explained by meta-emotions, the appraisal of negatively valenced emotions as worthwhile and suitable for the situation (Vorderer et.al, 2004). Drama will be appraised as positive if respondents are emotionally touched (Zillmann, 1994).
In contrast to enjoyment, the reason for viewing might be to experience appreciation and meaningfulness through emotions. While theories like Mood Management and Disposition Theory have a hedonic view of entertainment, Wirth, Hoffer and Schramm (2012) posit that the entertainment experience is not only pleasurable, but can also lead to more profound feelings. This research approach considers there to be both hedonic (pleasure, fun, suspense) and eudaimonic (insight, meaning) motivations to view media. The experience of sadness and distress while watching tragedies and dramas is thus not an undesirable activity. More emotional reactivity, regardless of its valence, would then lead to a higher appreciation of a film.
Emotional reactivity is not only determined by the content of the narrative, but also by dispositional factors. In this study, trait empathy is considered to have an effect on the presence and intensity of an emotional reaction to a visual narrative.
Trait Empathy and Media
What is empathy? While definitions and measurements vary across the literature and across different scientific fields, it is now widely accepted that empathy exists on two dimensions, cognitive and affective, and is differentiated from sympathy (Nathanson, 2003; Keen, 2006; Tan, 2013). Affective empathy refers to the activation and experience of the emotional state of someone else; cognitive empathy to understanding the emotions of another person; and a related emotion, sympathy, entails having concern for the distress of another (Clark, 2010). It is also important to distinguish between trait empathy and state empathy. State empathy is considered to be the experience of empathy while viewing (Shen, 2010). In contrast, the present study approaches empathy as a dispositional characteristic.
Dispositional characteristics can predispose our receptiveness to media. According to the Differential Susceptibility to Media Effects Model (DSMM), media effects are both conditional and indirect (Valkenburg & Peter, 2013). In the model, media-effects theories are synthesized into four propositions regarding media and non-media variables. In the first proposition, the authors posit that media effects are conditional on three types of variables: dispositional, developmental, and social. Developmental susceptibility refers to media use across the lifespan. Social susceptibility refers to contextual factors that influence media use. Of interest to this study is dispositional susceptibility, which refers to personal dimensions that predispose how we react to and select media, including personality, temperament, values, attitudes, etc., that are relatively stable.
Trait Empathy can be considered as one of these dispositional, individual, relatively stable traits. Eysenck & Eysenck (1978) considered it a fundamental personality trait. In communication science, Zillmann (1994) proposed that people possess empathic dispositions, partly reflexive, partly learned, which they bring to the viewing experience. It has been posited that this ability and readiness to empathize influences the degree to which viewers like and empathize with a character (Vorderer et al., 2004). And there is evidence that trait empathy can affect the viewing experience outside of the experience of emotions. For example, people with higher trait empathy evaluate content more highly when they perceive that it is low in fictionality (Argo, Zhu, & Dahl, 2008), pointing to individual differences dictated by trait empathy.
Trait Empathy and Emotional Reactivity. Trait empathy has been associated to emotional reactivity across the psychological and communication science literature. Recent evidence has shown that individuals with higher empathy have enhanced perception of emotional facial expressions, particularly that of fear (Johnson, 2012). Furthermore, high-empathy subjects have a higher degree of mimicking behavior, such as postures, mannerisms, and facial expressions (Sonny-Borgstrom, 2002). They react more quickly to facial expressions with mimicry (unconscious imitation), while those low in empathy do not differ. They are more sensitive to the feedback from the facial reactions, which will induce a similar emotion in the receiver, as manifested in a higher experience of the respective emotions (Dimberg & Thunberg, 2012).
In the communication field, there is evidence that trait empathy affects the experience of emotions. Film has the potential to allow viewers to not build protective layers of reasoning that can inhibit empathy in the real world (Keen, 2006). Across the literature, high and low empathizers differ on several emotional responses. Davis, Hull, Young, & Warren (1987) found
that the trait dimensions of empathic concern and perspective taking, part of the Interpersonal Reactivity Index (IRI), had effects on different affective responses to dramatic film stimuli measured with a Mood Adjective Checklist. This study also pointed to the importance of evaluating the different dimensions of empathy, since the empathic concern subscale had strong effects on negative affective reactions, and perspective-taking had strong effects on positive affective reactions. Various dimensions of empathy have been shown to influence emotional reactions to graphic horror scenes (Tamborini, Stiff, & Heidel, 1990). De Wied, Zillmann, and Ordman (1994) found that high empathizers had stronger physical reactions and emotions to a sad film, and also enjoyed the film more. De Wied, Hoffman, and Roskos (1997) describe that participants with high empathic sensitivity were more likely to experience suspense and distress when watching suspenseful films. Roberts & Strayer (1996) observed a relationship between emotional expressiveness and empathy in children. And Argo et al. (2008) found that high empathizers elicited more emotional responses to a fictional story.
As previously discussed, emotional reactivity can be considered a manifestation of enjoyment, which has also been associated with empathy (Vorderer et al., 2004). Enjoyment has been associated to dimensions of the IRI relating to cognitive empathy, specifically the fantasy empathy subscale (Harris et al., 2000; Harris et al., 2009; Oliver, 1993). In one study, perspective taking and empathic concern were positively related to sympathy for a victim, which led to higher enjoyment of the clip that included retribution for the villain (Raney, 2004). Trait empathy also enhances responsiveness of negative emotions, but that does not mean enjoyment decreases, and can even increase enjoyment of danger and excitement (Hoffner, 2009). Oliver (1993) also found this relationship between affective empathy and enjoyment of negatively valenced emotions. Viewers high in affective components of empathy are more likely to experience fear
and distress, but this only reduces enjoyment when the material contains graphic violence (Hoffner & Levine, 2005) or when characters suffer or are severely injured (Hoffner, 2009).
In this study, trait empathy is measured with two subscales: affective empathy and cognitive empathy. The cognitive and affective dimensions of empathy have been shown to be distinct processes (Shamay-Tsoory, Aharon-Peretz, & Perry, 2009). The two dimensions are related differently to measures such as physical aggression and indirect aggression (Yeo, And, Loh, & Fu, 2011). However, they are also closely interconnected. For example, both subscales are related to prosocial behavior (McDonald & Messinger, 2011). Tan (2013) considers cognitive empathy to be an enabling factor for affective empathy, while also being fuelled by it. Similarly, some argue that shared affect without shared cognition is only mimicry or emotional contagion, while shared cognition without affect is more an indication of sympathy (Shen, 2010).
Since previous scales have not been very clear on the distinction between dimensions, the question arises of whether affective or cognitive empathy will both have the same effects on emotional reactivity. Since affective empathy is more related to the actual experience of emotions, it is likely that it will induce more emotional reactivity. However, this emotional reactivity might not be independent from cognitive empathy. It is also possible that each dimension can affect the experience of emotions separately. Therefore, the hypothesis is twofold:
H1a: Participants with higher affective trait empathy will show greater emotional reactivity than those low in affective trait empathy.
H1b: Participants with higher cognitive trait empathy will show greater emotional reactivity than those low in cognitive trait empathy.
Furthermore, since emotional reactivity is considered to be an integral part of enjoyment, it is expected that a measure of enjoyment will be related to emotional reactivity. Therefore:
H2: Emotional reactivity will be positively related to enjoyment.
However, trait empathy does not guarantee an emotional experience. As Vignemont and Singer (2006) point out “if we were to consciously feel what others feel all the time, we would be in permanent emotional turmoil, leaving no room for our own emotions.” While there is research regarding the processes that might moderate empathy while exposed to media (Nathanson, 2003; Roberts & Strayer, 1996; Vignemont & Singer, 2006), not much research has focused on the processes that mediate the relationship between empathy and emotional reactivity. It stands to argue that people of a similar empathic level do not react emotionally in the same way to all media stimuli. Therefore, in this study, narrative engagement is considered to be a media response state that acts as the mechanism that allows trait empathy to materialize into an emotional experience in the world of fiction.
Narrative engagement
As previously mentioned, according to the DSMM, media effects are not only conditional, but also indirect. The second proposition of the model states that there are three types of media response states that mediate the relationship between media use and media effects. These response states emerge from media use and can be cognitive, emotional, and excitative. Cognitive response states entail the extent to which viewers attend to and invest cognitive effort in content, while emotional response states refer to affectively valenced reactions, and excitative to physiological arousal. The higher a response state, the longer and more evident a subsequent effect is expected to be (Valkenburg & Peter, 2013).
In this study, narrative engagement is hypothesized as a cognitive response state that mediates the relationship between trait empathy and emotional reactivity. Narrative engagement is known by many names, including absorption, immersion, flow, and, most common in the communication literature, transportation. Green and Brock (2000), whose transportation scale is typically used to study the process with written narratives, defined transportation as “a convergent process, where all mental systems and capacities become focused on events occurring in the narrative.” Film is also considered a proper medium to experience the process, given its rich visual imagery and the way it focuses the viewer's attention, able to demand equally complicated mental processes as novels (Mar & Oatley, 2008). While viewing a film, through storytelling and film techniques, viewers become engaged with the story (Busselle & Bilandzik, 2009).
Narrative engagement can be affected by individual factors. For example, Green (2004) demonstrated that those with personal experience relating to the themes of a story had higher transportation, and prior knowledge and perceived realism also influence the experience. Busselle and Bilandzic (2008) similarly posit that if external realism and narrative realism (story coherence) are perceived to be inconsistent, then engagement might be disrupted, but they present evidence that fictionality does not affect narrative processing.
Narrative engagement and trait empathy. There is not much research analyzing the relationship between trait empathy and narrative engagement. Green, Brock, and Kaufman (2004) speculated that transportation could both draw upon individual's tendency towards empathy as well as develop it. Hall and Bracken (2011) found a positive relationship between the Fantasy subscale of trait empathy and transportation. However, the Fantasy subscale includes several items that are similar to those of transportation, namely, those that inquire whether the viewer
was “involved” or “caught up” in the story. Argo et al. (2008) found an interaction effect between empathy and fictionality that led to higher transportation. However, empathy here was operationalized as gender. It is also possible that the dimensions of empathy could come into play, affecting the dimensions of engagement in different ways. Therefore, further clarification of the relationship between the two concepts is necessary.
Narrative Engagement and Emotional Reactivity. In communication science, not much empirical research has focused on the direct relationship between narrative engagement and emotions. In some cases, emotions are treated as consequences of narrative engagement, while in others, a component of this process (Green & Brock, 2000; Busselle & Bilandzic, 2009). Furthermore, there is some overlap regarding emotional reactivity and state empathy in the literature. However, there is a distinction between affective reactions and those that are actually empathy. Situations that produce affect in an observer that matches the affect of the observed should be construed as affect-inducing situations, not empathy-inducing situations (Nathanson, 2003). In this study, the aim is not to distinguish between self-oriented or other-oriented emotions, but rather, the overall emotional response. Of course, as it is done in the context of trait empathy, it is expected that a participant will have an emotion matching the valence of that of the character.
Using a mixing console analogy, the DSMM posits that cognitive and emotional response states are represented by sliders that can coexist, yet be at different levels. While the authors concur that emotions and cognitions are not fully distinct forces and typically interact, they endorse studying these responses as separate entities (Valkenburg & Peter, 2013). Therefore, more narrative engagement is proposed to increase emotional reactivity.
Flowing from this, cognitive aspects of narrative engagement have been related to emotional reactivity in several ways. From a neurological perspective, there is evidence that attention, a key component of engagement, modulates empathic brain responses to pain (Hein & Singer, 2008). Cognitive appraisals have been found to be antecedents of emotions (Smith & Lazarus, 1993). Tan (2013) suggests that film-induced emotions without cognitive appraisal might be little more than mimicry. Furthermore, appraisal of a character’s actions affects our dispositions towards them, which leads to emotions, for example, hope or fear for certain outcomes (Zillman, 1994). Transportation has also been associated with more positive evaluations of characters (Busselle & Bilandzic, 2008) potentially leading to an increased emotional experience.
Emotional reactivity has been found to be a main indicator of enjoyment, and enjoyment has been consistently related to transportation in written narratives (Green et al., 2004). The sense of being transported has been considered as a prerequisite for enjoyment (Vorderer et al., 2004), and becoming engaged as a pleasurable experience in and of itself (Busselle & Bilandzik, 2009). Oliver (1993) measured fictional involvement, the tendency to become engrossed in the experiences of other. This measure was associated with higher scores on the Sad-Film scale, developed to reflect the enjoyment of sad films. And finally, Green et al. (2004) suggest that we find pleasure in narrative engagement because of the ability of narrative to allow us to “travel to the dark side” and explore the boundaries of emotions, including unpleasant ones, such as fear, sadness and rage, while still being objectively safe. This implies that transportation is necessary for the experience of emotions.
There are not many studies that have analyzed the possible role of narrative engagement as a mediator between empathy and other media effects. Argo et al. (2008) found transportation
to be the mediator between empathy, operationalized as gender, and positive evaluations of fiction pieces. In one study, Johnson (2012) did find that individuals who were more transported into the story also reported higher compassion and sympathy, which points to the influence of narrative engagement on enhanced emotions. And in contrast to other studies, Hall and Bracken (2011) did not find a direct relationship between either dimension of trait empathy and enjoyment. Instead, they found that three dimensions of the IRI (perspective-taking, fantasy, and empathic concern) were significantly correlated with transportation, and that transportation was positively correlated with enjoyment, which points to emotional reactivity, providing initial evidence that transportation mediates this relationship. Therefore, the hypothesis is (see figure 1):
H3: The relationship between trait empathy and emotional reactivity is mediated by narrative engagement.
Figure 1. Conceptual model of the proposed mediation relationship
Furthermore, given that empathy, emotional reactivity and enjoyment have been found to be related across the literature, and narrative engagement has been consistently found to lead to enjoyment, the same mediation model could hold with enjoyment as the dependent variable. Therefore, the hypothesis is:
H4: The relationship between trait empathy and enjoyment is mediated by narrative engagement.
Method Sample
All participants were female, between the ages of 18 and 27 (M =20.9, SD =1.75). The sample was limited to women, as they have been shown to have stronger emotional reactions to media, and more experience of sadness, fear and distress (Oliver, 1993; Oliver, Sargent, & Weaver, 1998; de Wied at. al., 1997), as well as generally scoring higher on measures of empathy (Oliver et al., 1998; Mestre, Samper, Frias, & Tur, 2009). All participants were Dutch students recruited at the University of Amsterdam, and they received either credit points or ten euros for their participation. Seven participants were excluded because of technical problems such as FaceReader failure (e.g. because the participant wore glasses) or participant error (e.g. failing to activate full screen and leaning in too close to the camera). No participants denied the request to be filmed. The final sample consisted of 94 participants.
Stimuli
Pilot Study. Four films were chosen from a large selection, edited to last around 10 minutes each, and included in a pretest. They consisted of four dramas, selected to elicit the emotions of sadness and happiness. Ten participants, not part of the final sample, were included in the pilot study. In an online questionnaire, they were able to watch the video clips at home, and then answered questions regarding whether the storyline was clear, if the length was appropriate, whether the film influenced their emotions, and how much they enjoyed it. They were also asked in an open ended question to describe the emotions they felt while watching. Participants submitted extensive reports detailing their reaction to several parts of each clip. The two films of each genre were compared on the measures, as well as on written descriptions of emotions. Two clips from the pilot study were finally selected for the experiment. The selection consisted of the
clips that elicited stronger emotional responses, and higher scores on whether the storyline was clear and easy to follow, and judged to have most appropriate length. The clips were edited out of feature films, and their duration ranged between 10:00 and 11:00 minutes each.
For this study, two dramas, 50/50 and October Sky, were analyzed. Edited to elicit sadness, 50/50 tells the story of a young man who is diagnosed with cancer, and at the end of the clip goes into surgery, leaving the outcome of the operation unclear. October Sky, edited to elicit happiness, recounts the story of a young man in the 1960’s who desires to build a rocket. After a series of mishaps, he eventually succeeds, winning a science competition. The films were shown in English, their original language, with Dutch subtitles. These two films were selected for two reasons. First, because they elicit two of the main emotions that FaceReader can easily identify, sadness and happiness. While viewers can feel several emotions in response to the same stimuli, the films were edited with the purpose of isolating the emotion of interest as much as possible. Second, while it is not possible to know whether an emotion was directed towards the self or towards the characters, these two emotions can more accurately represent a match in emotional valence because of the genre.
Procedure
The design of the study is quasi-experimental. Participants first read instructions and then signed a consent form where they agreed to being filmed. Participants were instructed not to touch their face during the experiment, as this can interfere with FaceReader analysis. Then, they were escorted to a computer cubicle where they were left on their own. First, they filled out the trait empathy scale in an online questionnaire, and then they watched the film clips in randomized order. After each clip, questions were asked regarding narrative engagement and enjoyment.
Participants were filmed using a webcam, they were aware that they were being recorded, and had the possibility to stop watching a certain fragment at any time.
Measures
Emotional Reactivity. Measuring emotions is usually done by self-report, such as the Mood Adjective Checklist (Davis et al., 1987). Due to non-immediate reporting and retrospective bias, this can have important limitations (Lewinski, Fransen, & Tan, 2014), including social desirability, as well as a subjective interpretation of the intensity of the emotion. A more objective measure can avoid these problems. For these reasons, the Noldus FaceReader 5.1 was the selected measurement tool. The FaceReader is a software program that automatically analyzes facial expressions by identifying a face, and then using a 3-layer neural network, which is the product of 10,000 manually coded images. It has been used before to analyze reactions to taste experience (Danner, Sidorkina, Joechl, & Deurrschmid, 2013), Web page complexity (Goldberg, 2012), and the role of amusement in persuasiveness (Lewinski et al., 2013). It is an unobtrusive tool, especially compared to other physiological options. It is possible to code large amounts of data, with an estimated 89% accuracy rate (Uyl & Kuilenburg, 2005). It can also minimize the impact of social desirability, since it captures physical reactions than cannot be controlled very easily (Lewinski et al., 2014).
After the experiment, the videos were edited and then analyzed by FaceReader. Continuous Calibration, a feature of the software, was selected for standardization. The output from the FaceReader consists of a frame-by-frame score on each of the six basic emotions identified by Ekman (1972), Happy, Sad, Disgusted, Surprised, Angry, and Scared, as well as neutral status. Figure 2 shows a graphical representation of the output of the FaceReader. The sum of all emotions should add to 1.0, adjusting for FaceReader modelling failure.
Figure 2. Example of graphic output of FaceReader
In both films, only the emotion that was associated with genre was used for the study, sadness for 50/50, and happiness for October Sky. This approach was adopted in order to focus more on the experience of empathy, which would entail a matching valence response, and to not muddle interpretation. Emotional reactivity for these emotions was operationalized and calculated in two ways. First, similar to the approach followed by Lewinski et al. (2013) the global mean of the top 10% peak values for each emotion was used. To compute this value, the frequency of each score is calculated into a percentile distribution, and all values below the 90th percentile are removed. This is done in order to analyze the most prominent expressions. Secondly, an overall intensity score was calculated. This was done to differentiate between shorter, yet more intense expressions, from longer, less prominent ones. For this calculation, the values in the top ten percent of each emotion were distributed into four groups, and then assigned the values of 3, 2, 1, and 0, respectively and then multiplied by 100. In this way, higher emotion scores are given more weight than lower scores.
Trait empathy. Trait empathy was measured using the Children and Adolescent Scale of Empathy and Sympathy (CAMES) (Vossen, Piotrowski, & Valkenburg, in preparation). The scale is meant to capture a stable level of sympathy and trait empathy, which can differ among the affective and cognitive dimensions. The two subscales of empathy were taken as separate variables. All answers ranged from Never to Always on a 5-point Likert Scale. Cognitive empathy is also composed of four items related to understanding another’s emotions (e.g. “I can often understand how people are feeling even before they tell me”). The scale was reliable (α =.75) Affective empathy is composed of items related to experiencing another’s emotion (e.g. “When my friend is sad, I become sad too”). Scale reliability was adequate (α = .55).
Narrative Engagement. Narrative Engagement was measured with Busselle and Bilandzik’s (2009) scale of Narrative Engagement. The scale focuses on visual stimuli, taking into account imagery and sensory stimulation. Three dimensions were included, for a total of 8 items that captured narrative understanding (e.g. “I had a hard time recognizing the thread of the story”), attentional focus (e.g. “I found my mind wandering while the program was on”), and narrative presence (i.e. “During the program, my body was in the room, but my mind was inside the world created by the story”). The answers were rated on a 5-point Likert scale from Completely Disagree (1) to Completely Agree (5).
An Exploratory Factor Analysis (EFA) was done to analyze the structure of the narrative engagement scale. The analysis showed narrative presence as a distinct factor in both films, however, there was not a clear distinction between narrative understanding and attentional focus. For October Sky, a two-factor solution emerged, explaining 74.3% of the variance. The first factor, narrative presence, explained 54.8% of the variance, while the second factor, the 5 items of narrative understanding and attentional focus combined, explained 19.5% of the variance. In
50/50, a three component solution emerged, explaining 75% of the variance. The first factor, narrative presence, explained 40% of the variance. There was also overlap between the items of narrative understanding and attentional focus, with items of narrative understanding with a high loading on the attentional focus component. Therefore, they were considered to be one factor, with all factor loadings above .5, explaining 35.1% of the variance. Subsequently, for further analyses, a separate subscale was created under the name narrative focus. This was done by combining the items of the two subscales and dividing by five. This narrative focus scale was reliable in 50/50 (α = .81) and October Sky (α =.88). Narrative presence was also reliable in both films in both 50/50 (α = .77) and October Sky (α =.89).
Enjoyment. Two items were created to measure the experience of enjoyment (“I enjoyed the fragment”; “I would like to see the entire film”). The answers were rated on a 5-point Likert scale from Completely Disagree (1) to Completely Agree (5). The two items were positively correlated for 50/50 (r=.280, p = .005) and for October Sky (r=.714, p<.001).
Analysis Procedure
Since each respondent was exposed to two different films, parallel investigations were done for all hypotheses. A preliminary correlation analysis using Pearson’s coefficient was done to examine whether the different scales of trait empathy, narrative engagement, and emotional reactivity were related in the expected direction, and to investigate hypothesis 2, the relationship between enjoyment and emotional reactivity. To investigate hypothesis 1a and 1b, whether cognitive and affective empathy predict emotional reactivity, multiple regression analysis was done. To investigate hypothesis 3, the mediated relationship between empathy, engagement, and emotional reactivity, and hypothesis 4, the same mediated relationship with enjoyment as dependent variable, the PROCESS macro of Andrew F. Hayes was used, which is designed for
mediation analyses. This macro works by using bootstraps for the indirect effect, and outputs a 95% confidence interval. If zero is included in the interval, then the indirect effect is not significant (Hayes, 2013). For hypothesis 3, all analyses were done twice, first with the value of emotional reactivity operationalized as length, and then as intensity.
Results Descriptive Statistics
For 50/50, sadness was the main outcome, while for October Sky, happiness was the main outcome. Both films elicited both emotions, so to check whether 50/50 indeed elicited more sadness and October Sky more happiness, paired-sample T-tests were done using the measure of mean score of the top 10%. Participants showed significantly more sadness during 50/50, t(93)=4.01, p < .001, 95% CI[.04, .13], and more happiness during October Sky, t(93)= -4.36, p < .001, 95% CI [-.16, -.06]. All other emotions were barely present, and did not differ between films. The means and standard deviations for all variables can be seen in Table 1.
Correlations
Pearson’s correlations between the variables of importance for the study were calculated separately for each film, and are shown in Table 2.
50/50. Affective empathy was only positively correlated to narrative presence in 50/50 and not to narrative focus. Cognitive empathy was not correlated to the dimensions of narrative engagement. Narrative focus was positively correlated to sadness, both in length and intensity. Enjoyment was positively correlated to narrative focus, but not to narrative presence.
October Sky. For October Sky, affective empathy was also positively correlated with narrative presence, but not narrative focus, and there were no effects of cognitive empathy on the
dimensions of narrative engagement. Neither dimension of engagement was related to happiness. Enjoyment was positively correlated with both dimensions of narrative engagement.
Hypothesis 1a and 1b
Hypothesis 1a and 1b stated that the two dimensions of empathy would have a direct effect on emotional reactivity. Using multiple regression, affective empathy and cognitive empathy were introduced as independent variables, with mean sadness or mean happiness as the dependent variable, depending on the film. The regression model with the two predictors and sadness as outcome was not significant, F(2, 91) = .77, p = .47. Neither was the regression model with happiness as outcome F(2, 91) = .14, p =.87. No significant differences were observed when introducing sad intensity or happiness intensity as outcomes. As shown in Table 3, the regression analyses demonstrate no evidence that empathy predicts emotional reactivity for either film. H1a and H1b were not supported.
Hypothesis 2
According to hypothesis 2, emotional reactivity was expected to be positively related to enjoyment. Correlation analyses found that enjoyment was positively correlated with the 10% average measure of happiness for the film October Sky (r=.286, p < .001), however, not with sadness for 50/50 (r=.08, p =.44) Hypothesis 2 was partially supported.
Hypothesis 3
Hypothesis number 3 stated that the relationship between trait empathy and emotional reactivity would be mediated by narrative engagement. As can be observed from the correlations, there do appear to be certain relationships among different subscales of empathy and narrative engagement, and different emotions. Based on these correlation analyses, a direct effect is not present, therefore hypothesis 3 is not supported, but an indirect effect could still be observed.
50/50. Four analyses were done using Andrew F. Hayes’ PROCESS macro for mediation
analysis. First, cognitive empathy was introduced as independent variable, narrative presence and narrative focus as mediators, and mean sadness, and later sadness intensity, as outcomes. Secondly, the same procedure was followed but with affective empathy as independent variable. As expected from the correlations, no direct effect of empathy on emotional reactivity was observed. Indirect effects of cognitive empathy on mean sadness were not statistically significant through either narrative presence (95% BCI: [-.05, .002]) or narrative focus (95% BCI: [-.03, .03]); neither were indirect effects of affective empathy on mean sadness either for narrative presence (95% BCI: [-.08, .002]) or narrative focus (95% BCI: [-.02, .04]). No significantly different results were observed when sadness intensity was introduced in the analysis as outcome measure.
October Sky. The same procedure was done, but with mean happiness and happiness
intense as outcome variables. Again, there was no direct effect of empathy on emotional reactivity. Indirect effects of cognitive empathy on mean happiness were not statistically significant through either narrative presence (95% BCI .01, .05]) or narrative focus (95% BCI [-.06, .01], neither were indirect effects of affective empathy on happy mean for narrative presence (95% BCI [-.08, .01]) or narrative focus (95% BCI [-.01, .06]). Also, when introducing happy intense as outcome measure there were no significantly different results.
Hypothesis 4
Hypothesis 4 expected that narrative engagement would mediate the relationship between empathy and enjoyment.
50/50. In separate analyses using the PROCESS macro, affective empathy and cognitive
mediators, and enjoyment as dependent variable. There were no direct effects of empathy on enjoyment. The indirect effects of cognitive empathy were not significant for either narrative presence (95% BCI [-.01, .16]) or narrative focus (95% BCI [-.06, .09]). There were also no significant indirect effects of affective empathy for either narrative presence (95% BCI [-.01, .22]) or narrative focus (95% BCI [-.05, .11]).
October Sky. The same procedure was done as with 50/50. There were no direct effects of
empathy on enjoyment. The indirect effects of cognitive empathy on enjoyment were not significant for narrative presence (95% BCI [-.21, .09]) or narrative focus (95% BCI [-.28, .06]). However, the indirect effect of affective empathy on enjoyment was significant through narrative presence (95% BCI [.03, .41]) though not narrative focus (95% BCI [-.11, .32]). Hypothesis 4 was partially supported.
Discussion
The broader research question asked how trait empathy is related to emotional reactivity while viewing media, and how narrative engagement mediates this relationship. This study has found no clear evidence that this mediation relationship exists. No direct relationship between empathy and emotional reactivity was found. Affective empathy was positively related to narrative presence, and narrative focus was positively related to sadness. Narrative engagement appears to be a complex construct, with different subscales being influenced by and affecting variables differently.
Based on previous studies (Davis et al., 1987; Tamborini, et al., 1990; de Wied et al., 1994; de Wied, et al., 1997) it was expected that people higher in trait affective and cognitive empathy show more emotional reactivity when watching a movie (H1a and H1b). The results show no relation between either factor of trait empathy and emotional reactivity as measured with
the FaceReader. While the emotion of sadness was present during 50/50, and happiness during October Sky, this does not seem to be in response to the participants’ level of trait empathy. Of course, it could be that trait empathy does not lead to emotional reactions in all cases. Most of the research on this relationship has been done on fear reactions to scary and horror films, so perhaps it is essential here to consider a film’s genre and the emotions it elicits. As Shen (2010) points out, there has been a bias towards studying empathy with negatively valenced emotions. No previous study had explicitly looked at the effects of trait empathy in feeling happiness while watching the success of a character. It could be that we are faster to have an emotional response when witnessing others in negative circumstances than in positive circumstances. However, there was also no effect of trait empathy on the negatively-valenced emotion of sadness. Other processes might also influence this relationship, such as mood regulation. Tan (2013) suggested that while watching, people have less burden to regulate their emotions, however, the artificial context might have given way to regulation, rather than avoid it. Previous experience with what is portrayed in the film could also have an influence (Green, 2004). For example, participants who had personal experience with cancer might have reacted more emotionally, regardless of their dispositional level of empathy. And finally, this could be due to the way the emotions were measured. Other studies have used varying scales of empathy and several subjective and objective measures of emotional reactivity. Subjective measures are usually administered post-viewing, which could lead to information about emotions not outwardly expressed, but also socially desirable answers. Measures of physiological arousal have been used with the emotion of fear, but not with other emotions. These different ways of measuring emotions could account for the lack of a relationship when the indicator is facial expressions.
Narrative engagement was hypothesized to mediate the relationship between empathy and emotional reactivity (H3). While this was not the case, the study did find initial evidence that narrative engagement is related to trait empathy and to emotional reactivity. With regard to the effects of trait empathy on narrative engagement, research on this relationship is scarce. In this study, affective empathy was found to be positively correlated to narrative presence in both films, and not narrative focus. Those higher in affective empathy appear to more easily accept a fictional world, though this may not necessarily lead to increased emotions. It is interesting to note that only the affective dimension of empathy had a positive relationship with narrative presence. This lends support to the distinction between cognitive and affective empathy, and the importance of taking into account their distinct influence on media effects. Research should further investigate the nature of narrative presence, and whether it is more an affective or cognitive response state. And while it is surprising that cognitive empathy was not related to narrative focus, this could perhaps be explained by the fact that cognitive empathy is more related to understanding a character’s emotion, not the storyline as a whole. This could entail quite different mental processes.
In terms of the relationship between narrative engagement and emotions, narrative focus was only related to sadness, and not happiness. This could be explained by the psychological concept of negativity bias, the innate predisposition to give more weight to negative entities (Rozin, 2001), which affects how much attention we invest in them. This was previously supported by Lang, Newhagen and Reeves (1996), who found that televised news stories with negative events (e.g. war, death, accidents) increased attention. Perhaps the valence of the emotions that a genre elicits moderate how much we attend to the stimuli, which in turn increases the emotional response. It should be noted that the happy film was edited in a way that did not
include much struggle for the main character, in an attempt to isolate the emotion of happy. However, in most dramatic films, the feelings of happiness occur after the main character overcomes difficulties. This would then also have the effect of increasing attention.
It was also expected that enjoyment would be related to emotional reactivity (H2). Results show that enjoyment was positively related to happiness, but not to sadness, questioning the theory that people enjoy feelings of all emotional valences. This is contradictory to previous research that posits that we evaluate negative emotions as appropriate, leading to enjoyment and/or appreciation. One possible explanation for 50/50 could be that the film had no satisfying ending; the participant does not know if the protagonist lives or dies. However, the participants overall enjoyed the sad film more than the happy film. This puts into doubt the validity of the data for the emotion of sadness. Also, the fact that sadness was present in both clips lends suspicion to its experience during the sad film. In any case, the experience of sadness might not be enough to explain viewer’s attraction to sad movies.
Finally, it was expected that narrative engagement would also mediate the relationship between empathy and enjoyment (H4). This was expected because emotional reactivity is considered to be an integral part of enjoyment. However, empathy had no direct effect on enjoyment. This is similar to the recent findings of Hall & Bracken (2011), who also suggest that enjoyment will vary depending on genre. Furthermore, as Vorderer et al., (2004) suggest in their model of enjoyment, the evaluation of enjoyment is not simple, and depends on a series of user prerequisites and motives. For example, it might depend on how appropriate a reaction to a film was judged to be (Oliver, 1993), or if the film had a satisfying resolution (Hoffner & Levine, 2005), which might not have been the case for these films. However, enjoyment was positively correlated with narrative focus for both films, and with narrative presence for October Sky. This
is consistent with previous research that suggests that narrative engagement leads to a pleasurable experience, regardless of the genre or the emotions it elicits. However, here again the different dimensions of engagement come into play, as this only seems to hold for narrative focus. Furthermore, for October Sky, there was an indirect effect of affective empathy on enjoyment through narrative presence. This shows that narrative engagement in some form is necessary for enjoyment, though it might depend on the type of genre and emotion.
The question remains, what does emotional reactivity mean? It could be an other-oriented response to a character, namely state empathy. But it could also be a response to a dramatic scene that creates the emotion (Zillman, 1994), or a self-oriented response as a scene can trigger a viewer’s memory of their own personal experience. An affective response with low narrative engagement could point to a more superficial process, such as mimicry or emotional contagion (Coplan, 2006). Future research could investigate these different ways of interpreting an emotional reaction to film stimuli. This could be done, for example, by comparing a scale of state empathy with one of items related to self-focused emotional reactions, and by giving participants questionnaires at various points during a film.
Limitations
Like any experimental setting, the study had issues of ecological validity. Participants were in an artificial environment, and they were aware that they were being filmed. This might have interfered with how they consciously, or unconsciously, controlled their emotions and facial expressions. The developers of the FaceReader point out that people who watch something on a computer monitor might show less facial expression than when they are in the company of someone else (Loijens & Krip, 2013). And the fact that participants were aware that they should
not touch their face could have interfered with narrative engagement, though this does not seem to have occurred as, overall, the scores for engagement were high.
While the films were randomized in order to control for order effects, this could have had a subsequent influence due to residual emotions. A participant who felt a lot of sadness might have found it hard to smile during the happy film, or one who watched a comedy first might have still had a facial expression in the form of a smile while watching a sad movie. It might be necessary to show each film on separate instances.
There are also limitations regarding the research tool. The FaceReader appears to be an appropriate tool to measure reactions related to happiness, but perhaps more thorough calibration is necessary for the emotion of sad. False positive have been reported in other FaceReader studies (Terzis, Morides, & Economides, 2010), mainly false answers of anger, when participants were simply showing focus. The same could have happened in this case with sadness, which could perhaps be also interpreted as boredom. In this study, continuous calibration was used, one of the functionalities of the software. Doing more careful calibrations is possible, but it requires much more time and effort, as participants have to act out each emotion in turn. It might be useful for researchers to manually analyze any unexpected or surprising feedback from the FaceReader to ensure validity. Zaman and Shrimpton (2006) concluded that the data of a researcher and the FaceReader are complementary, and observations by the researcher might be necessary. Of course, depending on the amount of data, this would nullify the ease of using FaceReader. In general, the FaceReader can be a beneficial software program. Technical limitations can be easily solved, but the main problem concerns interpretation. Depending on the emotion of interest and the goals of the project, it could prove to be a valuable tool.
The study contributes to the increasing amount of research exploring the relationship between trait empathy, narrative engagement, and emotions. At the end, what we get from this study is that there is not always a direct effect between empathy and emotions. This may depend on the type of emotions elicited by a particular genre, and possible influence of other moderators. The study does suggest that narrative engagement in some form is affected by empathy, and it can also affect how much emotion and enjoyment we experience while viewing. The role of narrative engagement in entertainment should not be underestimated. It could be worthwhile to further analyze this relationship in future research, taking into account different types of genres and emotions. Further research should also distinguish between the different roles of the dimensions of narrative engagement, as they can have varying influence on media effects.
References
Bal, P. M., & Veltkamp, M. (2013). How does fiction reading influence empathy? An experimental investigation on the role of emotional transportation. PloS One, 8(1), e55341.
Busselle, R., & Bilandzic, H. (2008). Fictionality and perceived realism in experiencing stories: A model of narrative comprehension and engagement. Communication Theory, 18(2), 255– 280.
Busselle, R., & Bilandzic, H. (2009). Measuring narrative engagement. Media Psychology, 12(4), 321–347.
Clark, A. J. (2010). Empathy and sympathy : Therapeutic distinctions in counseling. Journal of Mental Health Counseling. 32(2), 95–101.
Coplan, A. (2006). Catching characters’ emotions: Emotional contagion responses to narrative fiction film. Film Studies, 8, 26–39.
Danner, L., Sidorkina, L., Joechl, M., & Duerrschmid, K. (2014). Make a face! Implicit and explicit measurement of facial expressions elicited by orange juices using face reading technology. Food Quality and Preference, 32, 167–172.
Davis, M. H., Hull, J. G., Young, R. D., & Warren, G. G. (1987). Emotional reactions to dramatic film stimuli: the influence of cognitive and emotional empathy. Journal of Personality and Social Psychology, 52(1), 126–33. Retrieved from
http://www.ncbi.nlm.nih.gov/pubmed/3820067
De Wied, M., Hoffman, K., & Roskos-Ewoldsen, D. R. (1997). Forewarning of graphic portrayal of violence and the experience of suspenseful Drama. Cognition & Emotion, 11(4), 481–494. De Wied, M., Zillmann, D., & Ordman, V. (1994). The role of empathic distress in the enjoyment
of cinematic tragedy. Poetics, 23, 91–106.
Dimberg, U., & Thunberg, M. (2012). Empathy, emotional contagion, and rapid facial reactions to angry and happy facial expressions. PsyCh Journal, 1(2), 118–127.
Ekman, P. (1972). Universals and cultural differences in facial expressions of emotion. In J. Cole (Ed.), Nebraska Symposium on Motivation, 1971 (pp. 207–282). Lincoln: University of Nebraska Press.
Eysenck, S., & Eysneck H. (1978). Impulsiveness and venturesomeness: Their position in a dimensional system of personality description. Psychological Reports, 43, 1247-1255. Goldberg, J. (2012). Relating perceived Web page complexity to emotional valence and eye
movement metrics. Proceedings of the Human Factors and Ergonomics Society Annual Meeting, 56(1), 501–505.
Green, M., Brock, T., & Kaufman, G. (2004). Understanding media enjoyment : The role of transportation into narrative worlds. Communication Theory, 4, 311–327.
Green, M. (2004). Transportation into narrative worlds : The role of prior knowledge and perceived realism. Discourse Processes, 38(2), 37–41.
Green, M., & Brock, T. (2000). The role of transportation in the persuasiveness of public narratives. Journal of Personality and Social Psychology, 79(5), 701–721.
Hall, A. E., & Bracken, C. C. (2011). “I really liked that movie”: Testing the relationship between trait empathy, transportation, perceived realism, and movie enjoyment. Journal of Media Psychology: Theories, Methods, and Applications, 23(2), 90–99.
Harris, R. J., Hoekstra, S. J., Scott, C. L., Sanborn, F. W., Karafa, J. A., & Brandenburg, J. D. (2000). Young men’s and women’s different autobiographical memories of the experience of seeing frightening movies on a date. Media Psychology, 2, 245–268.
Harris, R. J., Hoekstra, S. J., Scott, C. L., Sanborn, F. W., Dodds, L. A., & Brandenburg, J. D. (2009). Autobiographical memories for seeing romantic movies on a date: Romance is not just for women. Media Psychology, 6, 37–41.
Hayes, A. F. 2013. Introduction to Mediation, Moderation and Conditional Process Analysis. New York, New York: Guilford Press
Hoekstra, S. J., Harris, R. J.,& Helmick, A. L. (1999). Autobiographical memories about the experience of seeing frightening movies in childhood. Media Psychology, 1, 117–140. Hein, G., & Singer, T. (2008). I feel how you feel but not always: the empathic brain and its
modulation. Current Opinion in Neurobiology, 18(2), 153–8.
Hoffner, C. (2009). Affective responses and exposure to frightening films: The role of empathy and different types of content. Communication Research Reports, 26(4), 285–296.
Hoffner, C., & Levine, K. (2005). Enjoyment of mediated fright and violence: A meta-analysis. Media Psychology, 7(2), 37–41.
Johnson, D. R. (2012). Transportation into a story increases empathy, prosocial behavior, and perceptual bias toward fearful expressions. Personality and Individual Differences, 52(2), 150–155.
Keen, S. (2006). A Theory of Narrative Empathy. Narrative, 14(3), 207–236.
Kidd, D. C., & Castano, E. (2013). Reading literary fiction improves theory of mind. Science, 342, 377–80.
Lang, A., Newhagen, J., & Reeves, B. (2014). Attention, capacity, and memory. Journal of Broadcasting & Electronic Media, 40, 460–477.
Lewinski, P. Fransen, M., Tan, E. (2014). Predicting advertising effectiveness by facial expressions in response to amusing persuasive stimuli. Journal of Neuroscience, Psychology, and Economics, 7(1), 1-14.
Mar, R. a., & Oatley, K. (2008). The function of fiction is the abstraction and simulation of social experience. Perspectives on Psychological Science, 3(3), 173–192.
Mar, R. a., Oatley, K., & Peterson, J. B. (2009). Exploring the link between reading fiction and empathy: Ruling out individual differences and examining outcomes. Communications, 34(4).
McDonald, N., & Messinger, D. (2011). The development of empathy: How, when, and why. In A. Acerbi, J. A.Lombo, & J. J.Sanguineti (Eds), Free will, Emotions, and Moral Actions: Philosophy and Neuroscience in Dialogue. IF-Press.
Mestre, M. V., Samper, P., Frías, M. D., & Tur, A. M. (2009). Are women more empathetic than men? A longitudinal study in adolescence. The Spanish Journal of Psychology, 12(1), 76– 83.
Nathanson, A. I. (2003). Rethinking empathy. In J. Bryant, D. Roskos-Ewoldsen, & J. Cantor (Eds.), Communication and emotion: Essays in honor of Dolf Zillmann (pp. 107-130). Mahwah, NJ: Lawrence Erlbaum Associates.
Oliver, M. (1993). Exploring the paradox of the enjoyment of sad films. Human Communication Research, 19(3).
Oliver, M. (2008). Tender affective states as predictors of entertainment preference. Journal of Communication, 58(1), 40–61.
Oliver, M., Sargent, S. L., & Waver, J.B. (1998). The impact of sex and gender role self-perception on affective reactions to different types of film. Sex Roles, 38(112).
Raney, A. (2004). Expanding disposition theory: Reconsidering character liking, moral evaluations, and enjoyment. Communication Theory, 14(4), 348–369.
Roberts, W., Strayer, J., Url, S., Roberts, W., & Strayer, J. (2013). Empathy, emotional expressiveness, and prosocial behavior. Child Development, 67(2), 449–470.
Rozin, P. & Royzman, E. (2001). Negativity, negativity dominance, and contagion. Personality and Social Psychology Review, 5(4), 296-320.
Shamay-Tsoory, S. G., Aharon-Peretz, J., & Perry, D. (2009). Two systems for empathy: a double dissociation between emotional and cognitive empathy in inferior frontal gyrus versus ventromedial prefrontal lesions. Brain : A Journal of Neurology, 132(Pt 3), 617–27. Shen, L. (2010). On a scale of state empathy during message processing. Western Journal of
Communication, 74(5), 504–524.
Singer, T. (2006). The neuronal basis and ontogeny of empathy and mind reading: review of literature and implications for future research. Neuroscience and Biobehavioral Reviews, 30(6), 855–63.
Smith, C.A., & Lazarus R.S. (1993). Appraisal components, core relational themes, and the emotions. Cognition and Emotion, 7, 233-269.
Sonnby-borgström, M. (2002). Automatic mimicry reactions as related to differences. Scandinavian Journal of Psychology, 43, 433–443.
Tamborini, R., Stiff, J., & Heidel, C. (1990). Reacting to graphic horror: A model of empathy and emotional behavior. Communication Research, 17(5), 616–640.
Tan, E. (1994). Film-induced affect as a witness emotion. Poetics, 23, 7–32.
Tan, E. (2013). The empathic animal meets the inquisitive animal in the cinema: Notes on a
psychocinematics of mind reading. In Shimamura, A.P. (Ed.), Psychocinematics: Exploring Cognition at the Movies. (pp.337- 352). New York, NY: Oxford University Press.
Valkenburg, P. M., & Peter, J. (2013). The differential susceptibility to media effects model. Journal of Communication, 63(2), 221–243.
Vignemont, F., & Singer, T. (2006). The empathic brain: how, when and why? Trends in Cognitive Sciences, 10(10), 435–41.
Vorderer, P., Klimmt, C., & Ritterfeld, U. (2004). Enjoyment: At the heart of media entertainment. Communication Theory, 4, 388–408.
Vossen, H.G.M., Piotrowski, J.T., Valkenburg, P.M. (2014, May) Development of the Child and Adolescent Measure of Empathy and Sympathy (CAMES). Poster presented at the Empathy Symposium, Utrecht, The Netherlands.
Wirth, W., & Schramm, H. (2005). Media and emotions. A Quarterly Review of Communication Research, 24(3).
Wirth, W., Hofer, M., & Schramm, H. (2012). Beyond pleasure: Exploring the eudaimonic entertainment experience. Human Communication Research, 38(4), 406–428.
Yeo, L. S., Ang, R. P., Loh, S., Fu, K. J., & Karre, J. K. (2011). The role of affective and cognitive empathy in physical, verbal, and indirect aggression of a Singaporean sample of boys. The Journal of Psychology, 145(4), 313–330.
Appendix Table 1.
Means and Standard Deviations
M SD α Cognitive empathy 3.67 .49 .75 Affective empathy 3.23 .48 .55 50/50 October Sky M SD α M SD α NP 3.45 .83 .77 2.88 .95 .89 NF 4.24 .61 .81 3.65 .89 .88 E 3.72 .76 3.22 .98 Sad mean .60 .23 - - Sad intensity 140.49 56.49 - - Happy mean - - .45 .32 Happy intensity - - 160.15 100.72
Notes. NP = narrative presence; NF = narrative focus; E = enjoyment
Table 2.
Correlations between empathy dimensions, narrative engagement dimensions, emotional reactivity depending on film (sad/happy), and enjoyment
Cognitive empathy Affective empathy Narrative presence Narrative focus Enjoyment Happy mean Happy intense Cognitive empathy - .30** -.06 -.10 -.15 -.02 -.05 Affective empathy .30** - .23* .09 .09 -.05 .03 Narrative presence .11 .27** - .46** .52*** -.02 .10 Narrative focus .19 .03 .22* - .55*** .11 .13 Enjoyment -.15 -.07 .17 .23* - .29** .15 Sad mean .02 .13 -.07 .25* .08 - - Sad intensity .02 .17 .14 .24* .00 - - *p<.05, **p<.01, ***p<.001
Table 3.
Unstandardized regression coefficients of empathy predicting emotional reactivity
50/50 October Sky
Sad mean Sad intensity Happy mean Happy intensity
b SE b SE b SE b SE Cognitive empathy -.01 .22 -3.22 12.27 -.001 .07 -12.92 22.14 Affective empathy .07 .05 20.74 12.73 -.04 .07 10.66 22.98 *p<.05, **p<.01, ***p<.001
