Eudaimonic Media in Lived Experience: Retrospective Responses to Eudaimonic vs. Non-Eudaimonic Films

ABSTRACT

The present study examines the self-reported impact of viewing more versus less eudaimonic Hollywood films using a retrospective study design. We investigate the role of three novel constructs in understanding people’s lived experience of eudaimonic narratives. These include two outcomes: acceptance of the human condition, or the perception that a film helped the viewer accept that inevitable challenges in life contribute to a meaningful existence, and viewers’ self-report of the film’s impact on their ability to make sense of life’s difficulties. The third is a mediator: emotional range, or the breadth of emotions experienced during media exposure. Findings indicate that more eudaimonic films can increase viewers’ ability to make sense of difficulties, their acceptance of the human condition, and their motivation to pursue moral goals, relative to less eudaimonic films, thereby extending the Mediated Wisdom of Experience perspective. These effects are mediated by feelings of elevation, poignancy, and emotional range. Additionally, value congruence between participants and the film viewed increases perceptions that the film helped them make sense of difficulties, their acceptance of the human condition, and their motivation to pursue moral goals, irrespective of whether the film was a more or less eudaimonic one.

There has been a growing inter est among media scholars in eudaimonic entertainment media content, which highlight moral virtues and the meaning and purpose in life (Oliver & Bartsch, 2010; Vorderer, 2016; Vorderer & Reinecke, 2015). Increasingly, researchers have examined possible outcomes of exposure to eudaimonic media, such as reduction in death anxiety and willingness to accept delayed rewards (Rieger et al., 2015; Slater et al., 2016, 2018). Studies of the possible effects of consuming eudaimonic media, however, have typically relied on forced exposure experimental designs, use of relatively brief excerpts from longer narratives, use of immediate posttests, and a focus on outcome measures that are informative psychologically, but removed from the outcomes audience members are most likely to seek from eudaimonic media.

We have sought to address these limitations in several ways. First, we chose to study retrospective responses to real-world eudaimonic media exposure, comparing responses to more versus less eudaimonic popular Hollywood films. In this study, research participants were randomly assigned to recall either a more versus a less eudaimonic film from a list of 20 films of each type. This permitted us to study the self-reported impact of actual, volitional exposure on their lived experience, and compare responses between these two types of films. Second, we examined two novel outcome measures that we believe are more likely to reflect the benefits audience members might seek from eudaimonic media experiences, making sense of difficulties and acceptance of the human condition, as well as studying moral motivations (Oliver et al., 2012) as an outcome. Third, we introduced a novel mediator, emotional range. Fourth, we addressed the possibilities that eudaimonic effects are in part an artifact of alignment of narrative content with the viewer’s own values, or that effects may be enhanced when such alignment occurs (see Mastro et al., 2012).

Possible outcomes of eudaimonic film exposure accessible to self-report

Our study is concerned with eudaimonic films as experienced and self-reported by viewers. Therefore, we chose to focus on outcomes that are intuitive and accessible to self-report and likely to be considered by the viewer as desirable outcomes. In addition, we also compared the self-reported effects of the two types of films on hedonic motivations and enjoyment.

Moral motivations

Emotional responses such as elevation, admiration, and gratitude, which are often a response to eudaimonic content, have been associated with motivations to emulate inspiring behaviors (Algoe & Haidt, 2009), also called moral motivations by Oliver et al. (2012). For instance, elevation, which is a positive emotion resulting from seeing acts of moral virtue by exemplars, was found to be associated with motivation to perform prosocial behaviors and wanting to be a better person (Algoe & Haidt, 2009). Similarly, Oliver et al. (2012) found that meaningful films increased feelings of elevation, which was positively associated with moral motivations, whereas pleasurable films were associated with other motivations that are more hedonic in nature, termed hedonic motivations (e.g., to enjoy myself).¹ Given this, we expect that viewing of more eudaimonic films, relative to the comparison films, will be associated with increased moral motivations (H1).

Eudaimonic narratives and making sense of difficulties

We suggest that one possible benefit of narrative exposure is the subjective experience that the narrative helps the audience member make sense of the difficulties they face in their own life. Films that contain many eudaimonic elements often portray characters going through a tragedy (e.g., loss of a loved one) or a challenging life event (e.g., a serious illness), ultimately overcoming the tragedy or challenge, and coming out stronger. Much narrative effects research is based on social cognitive theory, which argues learning can occur vicariously through the observation of models (Bandura, 1986, 2004). Research on such observational learning typically focuses on acquiring specific skills, increasing self-efficacy, and thereby influencing behaviors. We are also particularly interested in how exposure to eudaimonic content can influence more subjective aspects of well-being. Recent research has suggested that eudaimonic media consumption can benefit psychological well-being and vitality (Rieger et al., 2014). We seek to expand upon such benefits to one’s subjective sense of well-being through suggesting a way in which media exposure can instill audience members with psychological resources important for adaptive coping. Specifically, we suggest that the experience of seeing a character encounter and effectively weather life’s difficulties in effect models positive coping with one’s own challenges. One could regard this as akin to increasing self-efficacy (Bandura, 1986) with respect to one’s capacity for managing life difficulties. Indeed, the literature on inspiration (Thrash & Elliot, 2003) by implication suggests that films can inspire the viewer to find meaning and potential for growth in their own struggles and provide resources and ideas for the viewer to cope with challenges.

We believe the construct of making sense of difficulties, then, captures the ability for eudaimonic media content to model successful coping for viewers, so they feel more able to cope with challenges in their own lives. In fact, the self-reported impact of a eudaimonic narrative on making sense of difficulties was found to mediate the effect of eudaimonic narratives on death-related anxiety (Tan, Ott, Silver, & Slater, 2019). In this study, rather than as a mediator, we examine the ability to make sense of difficulties as an outcome. We do so because we believe use of narratives to help people feel able to cope with their life challenges is a fundamentally important potential outcome of eudaimonic narratives. We expect that viewing of more eudaimonic films, relative to the comparison films, will be self-reported as having helped the respondent to make sense of difficulties in life (H2).

Acceptance of the human condition

Previous studies have focused on effects of eudaimonic media content on outcomes such as delay discounting and death acceptance that were argued as being associated with greater maturity, in a phenomenon referred to as Mediated Wisdom of Experience (Slater et al., 2016, 2018). In this study, as explained above, we are interested primarily in subjective benefits of exposure to such content. Accordingly, we propose a variable called acceptance of the human condition. Acceptance of the human condition refers to a recognition that life consists of losses as well as gains, sorrows as well as joys, but that the very human richness of such experiences can make life fulfilling, meaningful, and result in personal growth. We regard such a recognition as a fundamental component of human maturity. The subjective experience of wisdom and maturity gained through the media experience is intrinsically beneficial, and as such may be viewed as a logical extension of the Mediated Wisdom of Experience perspective (Slater et al., 2016, 2018). That is to say, acceptance of the human condition seeks to more directly capture some of the mechanism through which such wisdom is mediated – a greater recognition of the complexities of human experience, and its inherent meaning– which is typically accrued over the course of one’s life, but which also can be occasioned through mediated experience. We view acceptance of the human condition as related to but broader than making sense of difficulties, described above, which more narrowly assesses one’s ability to positively reframe and cope with one’s life challenges as a result of media exposure. Therefore, we predict that eudaimonic films, relative to comparison films, will increase acceptance of the human condition (H3).

As popular films that are typically considered eudaimonic usually also contain many elements that produce hedonic enjoyment, we also ask: Will more eudaimonic films be equally likely to elicit a) hedonic motivations and b) enjoyment as comparison films, or will the less eudaimonic films be more likely to elicit such outcomes? (RQ1)

Mediators of exposure to more eudaimonic films on outcomes

In this study, we examined two previously studied mediators of eudaimonic narrative effects, elevation and poignancy, as well as a novel mediator, emotional range. Elevation has been studied as an emotional response to meaningful media (Oliver et al., 2012). Elevation is specifically concerned with a response to witnessing acts of virtue or moral beauty (Haidt, 2000); while related, such responses can be distinguished from appreciation, which represents a general reflective and introspective response to perceived meaningful and insightful content (Oliver & Bartsch, 2010). Poignancy, meanwhile, focuses on the specific co-occurrence of both happy and sad emotions in response to entertainment content (Slater et al., 2016).

Emotional range

Eudaimonic films may also elicit an especially wide variety of different emotions, as they address a wide range of human experiences, positive and negative. Mixed affect, particularly poignancy or the mixture of happiness and sadness, has been tested as a mediator of eudaimonic narrative effects, though with inconsistent support (Slater et al., 2016, 2018). Many eudaimonic films have story structures that demonstrate characters going through a tragedy or other challenges in life, which may elicit multiple emotions such as happiness, sadness, fear, hope, or feeling uplifted. We propose that poignancy and other pairs of mixed emotions may be special cases of a more general phenomenon: richness of emotional response to a meaningful narrative. Pairing contrasting emotions may capture the complexity of such a rich emotional response less efficiently than assessing the breadth of emotional response itself. Indeed, a classic theory of drama dating back over a millennium argues that the power of dramatic experience for a viewer is largely a function of the range of emotions evoked (see Pollock, 2018). Hence, given the willingness of eudaimonic films to engage with a wider range of human experiences, we propose that more eudaimonic films, relative to comparison films, will evoke correspondingly wider emotional range (H4).

Mediation predictions

Elevation may serve as a potential mediator, since a sense of inspiration or elevation arising from a eudaimonic narrative is likely to increase feeling equipped to face life’s difficulties or accept the human condition. We further sought to find evidence of the mediating role of poignancy, given inconsistent experimental support in earlier studies. The presence of both happy and sad emotions in response to a media experience may convey greater acceptance of both positives and negatives in one’s own life, and facilitate greater sense-making of difficulties. Hence, we predict that the effect of more eudaimonic films on ability to make sense of life’s difficulties will be mediated by elevation (H5a) and poignancy (H5b).

Moral motivations are also likely to be mediated by elevation, given that this emotion is inherently linked to an embracing of higher-order values (Algoe & Haidt, 2009; Oliver et al., 2012). Emotionally complex media experiences often also elicit a mixture of happiness and sadness, or poignancy (Slater et al., 2016), which may similarly work to instill a greater maturity of perspective toward what is ultimately important in life. Therefore, we expect that the effect of more eudaimonic films on pursuit of moral motivations will be mediated by elevation (H6a) and poignancy (H6b).

The presence of poignant mixed affect should lead to a greater sense of the ambivalent and complex nature of life, and that joy and sorrow are often equal partners on the road to meaningful life experiences—that is to say, greater acceptance of the human condition. Elevation is typically a response to viewing admirable responses to trying situations (Oliver et al., 2012); in other words, a response to people rising above challenge and loss to affirm the value of life, which should lead to greater acceptance of the human condition. Given this, we expect that the effect of more eudaimonic films on acceptance of the human condition will be mediated by elevation (H7a) and poignancy (H7b).

Present research on mixed affect is largely limited to a mix of happiness and sadness (e.g., Das et al., 2017; Oliver et al., 2012; Slater et al., 2016); however, related research on emotional flow provides some insights regarding emotional range. Emotional flow in a narrative, exemplified by emotional shifts from positive to negative emotions and vice versa, is posited to increase engagement with the narrative and sustain attention (Alam & So, 2020; Nabi & Green, 2015). In addition, Nabi and Green (2015) suggest that some outcomes of emotional flow include repeated exposure and increased post-narrative message elaboration, whereby the audience member further processes the contents of the narrative. In our view, the experience of multiple emotions, similar to emotional flow, would similarly prompt greater message elaboration and reflection on one’s life experiences, values, and goals. In addition, as emotional range is conceptualized as a broader form of mixed affect, we also expect that emotional range will mediate participants’ ability to make sense of difficulties and their pursuit of moral motivations in ways similar to poignancy described above. As emotional range is more general in scope than poignancy, it is unclear if it will be as effective in facilitating such effects as poignancy. Thus, we ask: Will emotional range mediate the effect of more eudaimonic films on a) participants’ ability to make sense of difficulties, b) acceptance of the human condition, and c) pursuit of moral motivations? (RQ2)

Value match

In this study, we propose that a match in terms of values expressed in the film and the values important to the audience member would be important for our outcomes, based on moral foundations theory (MFT, Haidt & Joseph, 2008) and the model of intuitive morality and exemplars (MIME, Tamborini, 2011). A central tenet of MFT and the MIME is that entertainment media content is more appealing and enjoyable for audiences when it is consistent with the audience’s moral codes (Mastro et al., 2012). In this study, however, instead of the five moral modules laid out in the MFT (see Haidt & Joseph, 2008), we expanded on the values inventory created by Schwartz (Schwartz & Boehnke, 2004). This value inventory was used because we believe that Schwartz’s circumplex model is more comprehensive and more satisfying theoretically as the values are theorized as part of a cohesive and interdependent value structure.

We believe that, while such values are a central part of eudaimonic content, more hedonic media experiences may also feature value-resonant elements. Thus, a match in the values of the viewer and the media content is conceptualized as a variable that may be present in both more and less eudaimonic contexts. We anticipate that a match in values between audience member and film should help to make salient meaningful aspects of lived experience, and may help individuals to better make sense of the difficult and complex situations they face (Klimmt & Rieger, 2021). Focusing on important values can highlight what one finds most important, and may help individuals to reframe difficult circumstances in light of those values (Edwards & Allen, 2008). Further, in the same way that a match in moral codes leads to more entertainment appeal and enjoyment for the audience, we expect that a match in values with the film will heighten the felt sense of what it means to be human as they dramatize those elements of the human condition most personally important to the viewer. Finally, we predict that seeing personally important values portrayed in a film should increase motivations to pursue more moral activities in one’s own life. Being reminded of one’s values is likely to be a strong motivator toward enacting such values in practice (Edwards & Kirven, 2019). Hence, we predict that value match with the film will be associated with greater ability to make sense of difficulties (H8a), pursuit of moral motivations (H8b), and acceptance of the human condition (H8c).

Method

The objective of this study was to better understand people’s lived experience of viewing more versus less eudaimonic films. It was therefore essential to study responses to popular films as they were experienced in the social environment. We employed a retrospective survey design to assess self-reported responses to more versus less eudaimonic films. Respondents were randomly assigned to respond regarding more eudaimonic or less eudaimonic films, permitting us to test hypotheses regarding differences in responses to such films.

Procedure

We created two lists of 20 Hollywood films, one list of more eudaimonic and one list of less eudaimonic films. Participants were randomly assigned to see either the list of the 20 more eudaimonic or the 20 less eudaimonic films, and asked to select all the films they had previously seen out of the list. Out of all the films the participant indicated having previously seen, one film was randomly selected and their responses for the survey were elicited in regard to that specific film. To improve recall, we provided the promotional poster for each film at the selection stage, and then asked participants to take a few moments to recall and describe any scenes from the film they were assigned which they found most memorable.

Stimuli

To create the list of the more eudaimonic films, we reviewed user-generated lists on the movie site IMDB.com that had titles containing at least one of the following words: poignant, moving, meaningful, appreciation, elevating, compassionate, tender, emotional, touching, or inspiring (e.g., “The 30 most moving movies”). These words were chosen as they represent the specific items from previous measures of meaningful affect in response to eudaimonic media (Oliver et al., 2012; Oliver & Raney, 2011), as well as specific terms used to describe responses to eudaimonic media (poignance and appreciation). Our more eudaimonic film list comprised the 20 films that were most frequently mentioned on such lists, and which also met the following criteria: (a) had a current rating on IMDB.com of equal to or greater than 8.0 and a Metascore rating of equal to or greater than 60; (b) were produced after 1985 to maintain a relatively recent and consistent level of production quality, and (c) were not from any genres that present an experience theoretically dissimilar to that typical of eudaimonic media (horror, comedy), or which create a distinctively different narrative experience (foreign language, musical, documentary).

To create our less eudaimonic film list, we matched each of our more eudaimonic films with a less eudaimonic film that met the following conditions: (a) was not mentioned more than once on the user-generated IMDB.com lists containing eudaimonic descriptors, (b) met the three qualifications mentioned above (IMDB and Metascore ratings, produced after 1985, and genre constraints), (c) were produced in a similar year to their matched eudaimonic film (within two years), and (d) had a similar MPAA rating (e.g., G, PG, R). We also sought to avoid films for either list that were part of a series with a broad overarching storyline for which it would be difficult for participants to recall details of just one film by itself (e.g., the Harry Potter series). An example of a more eudaimonic film was The Shawshank Redemption (released in 1994, rated R, IMDB/Meta-score of 9.3/80), matched with the less eudaimonic film, Pulp Fiction (released in 1994, rated R, IMDB/Meta-score of 8.9/94; full list available in Supplementary Table 1, available in supplemental materials on publisher website).

Participants

We contracted with Qualtrics to recruit an online panel of participants that represented the general population aged 18 and over, stratified to be balanced between males and females. Participants in the final sample (n = 1098, 47.72% female) ranged in age from 18 to 94 years old (M = 49.62, SD = 16.25) and predominantly identified as Caucasian/white (78.87%). We screened out potential respondents who indicated that they had not previously seen any of the movies from the condition they were assigned to (n = 293). We excluded respondents who failed the random attention checks requesting that they select a specific response (n = 883) or who had survey completion time that was faster than possible for researchers familiar with the survey (n = 7). All respondents with a response time more than three standard deviations above the mean completion time were removed as outliers. As there were still several very long response times, we also removed all respondents who took longer than 80 minutes to complete the survey, which was just more than 3 standard deviations above the new mean and represented a natural break in the distribution (n = 22 for excessively long responses).

Measures

Value match

In order to assess the extent to which the films aligned with respondents’ values, we asked respondents to select up to three values from a list of 16 that they saw represented in the presented film. The list of values was adapted from the value inventory by Schwartz and Boehnke (2004; e.g., “Achievement and personal success”, “Independence and creativity”, and “Obedience and respect for authority”), and the “values in action” inventory (Peterson, 2006; e.g., “Forgiveness and mercy” and “Love and intimacy”), with a few additional values added to reflect prominent film themes (e.g., “Equality and justice”, “Courage and bravery”, and “Appreciation and gratitude”). After selecting up to three values, participants then rated the personal importance of each value, relative to all the values they hold, from 0 (not important at all) to 10 (the most important value they hold). The personal importance ratings of all three values were summed to form a composite scale of value match ranging from 0 to 30 (M = 23.64, SD = 6.49).

Acceptance of the human condition

Participants were asked to respond to 12 items with the prompt, “Watching [presented film] left me with the feeling that … ” These items were developed by the authors to capture a shift in one’s focus toward a more mature and emotionally meaningful perspective toward lived experience, as suggested by socioemotional selectivity theory (Carstensen et al., 1999) and explored in previous research on Mediated Wisdom of Experience (Slater et al., 2016, 2018). Examples of items include, “Almost every choice in life comes at some sort of cost,” “Both happy and sad experiences give meaning to our life,” and “Part of life is accepting that caring deeply for people means you will suffer losses deeply as well.” CFA results showed that two items did not load well, likely due to being phrased in the reverse, making them more difficult to interpret. A third item also failed to load well, as it seemed to be tapping into individual differences in duration of life events. The three items were removed to improve construct validity, and the 9-item measure was utilized for analyses (Cronbach’s α = .94, M = 7.59, SD = 2.04).

Poignancy

Participants were asked to rate the extent to which they felt “happy” and “sad” while watching the presented film, on a seven-point scale from “Not at all” to “Very much”. We follow the work of Ersner-Hershfield et al. (2008) in computing poignancy as the minimum of a participants’ responses to how “happy” and “sad” they felt while viewing the film (M = 2.69, SD = 1.61).

Elevation

Based on previous work (Slater et al., 2016), elevation was measured with the term “inspired” (“meaningful” was removed as it overlapped with an item for appreciation; see CFA results below) and also with the term “elevated” to improve content validity (Cronbach’s α = .77, M = 4.37, SD = 1.68).

Emotional range

In order to measure emotional range, we asked participants to respond to an affect inventory containing 28 feelings and emotions. This inventory included the items mentioned above for poignancy and elevation, items used in previous research on meaningful affect in response to eudaimonic media (e.g., compassionate, tender, touched; Oliver et al., 2012; Oliver & Raney, 2011), items from the International Positive and Negative Affect Schedule Short Form (I-PANAS–SF, Thompson, 2007), as well as a few additional items to capture a wider range of positive and negative emotional responses (e.g., cheerful, upbeat, depressed, melancholy) as used in previous work on eudaimonic content (Oliver et al., 2012; Slater et al., 2018). This affect inventory was used to assess the full array of emotional responses that were significantly evoked by the presented film. Thus, we simply calculated the number of emotions to which a participant responded above the scale midpoint for each of the 28 emotions assessed. This resulted in a composite score ranging from 0–28 of how many emotions were strongly elicited in response to the presented film (M = 11.02, SD = 6.96).

Enjoyment and appreciation

We followed previous research (Oliver & Bartsch, 2010) in measuring the extent participants evaluated the film in regard to enjoyment (fun, entertaining, a good time) and appreciation (meaningful, moving, thought provoking). All items were measured on scale from 1 (Not at all) to 7 (Very much). CFA results indicated that “entertaining” did not load well onto the factor for enjoyment, and so this item was removed resulting in a two-item measure of enjoyment (Cronbach’s α = .87, M = 4.42, SD = 1.85). Appreciation comprised a reliable three-item scale (Cronbach’s α = .86, M = 5.13, SD = 1.57), and served as the primary manipulation check for our more and less eudaimonic groups.

Moral and hedonic motivations

Our primary outcomes were derived from previous work on meaningful media. We utilized a battery of motivations that tap into both elevating or moral outcomes (e.g., “Be a better person” and “Do good things for other people”) and more hedonic outcomes (e.g., “Enjoy myself” and “Meet new friends”) as a result of viewing the presented film (Oliver et al., 2012). The extent to which the film motivated participants toward each item was assessed on an 11-point scale ranging from 0 (Strongly disagree) to 10 (Strongly agree). The scale of moral motivations was composed of three items (Cronbach’s α = .94, M = 6.49, SD = 2.93), while the scale of hedonic motivations was composed of four items (Cronbach’s α = .89, M = 5.79, SD = 2.75).

Making sense of difficulties

Making sense of difficulties assessed participants’ self-report of the impact of the film on their capacity to cope with challenges they face in life. We used a 12-item measure with items such as “Feel like struggles in life are for a reason”, “Feel equipped to handle the difficult situations life throws at me” and “Understand that, while difficult, overcoming challenges is an essential part of personal growth”. All items were assessed on an 11-point scale ranging from 0 (Strongly disagree) to 10 (Strongly agree) (Cronbach’s α = .98, M = 6.43, SD = 2.62).

Since we assessed participants’ responses to our stimuli films retrospectively, and our pool of films spanned a 22-year period, we also measured time since last viewing and frequency of viewing. We asked, “When did you last see [presented film]?” with responses of “In the past month”, “In the past year”, “2–3 years ago”, “4–6 years ago”, “7–10 years ago”, and “more than 10 years ago”. Responses were recoded to the average of the interval response (e.g., a response of “2–3 years ago” was recoded to 2.5 & a response of “more than 10 years ago” for a film released 20 years prior to the study was recoded to 15 years). 52% of respondents indicated the time since last viewing was 3 years or less (M = 5.30, SD = 5.41). Frequency of viewing was measured by asking participants, “How many times have you seen [presented film]?” with responses from 0–4 or 5+. For the purposes of this study, all responses of 5+ were recoded to 5; 41.8% of respondents indicated they had seen the film only once, while 85.8% indicated they had seen the film 4 times or fewer (M = 2.18, SD = 1.39).

Confirmatory Factor Analysis (CFA)

We conducted a CFA with all the latent mediator and outcome variables using Mplus 8.0 and a maximum likelihood estimation method. This was important as: (1) many of the variables in eudaimonic media studies are conceptually similar, particularly the emotional response variables such as appreciation, enjoyment, and elevation; hence, it was important to test if they were empirically distinguishable constructs here; and (2) this study tested a number of newly created variables and the CFA allowed us to test the factor structure of these variables. To further add confidence that our results were not model-specific, we randomly split the data set into two equal halves (each N = 549). We conducted the initial CFA on the “A” half of the data set, before running the same model on the “B” half of the data set.

The initial measurement model did not have adequate model fit (χ²(836) =2506.90, p < .001, RMSEA = .06, CFI = .93, SRMR = .05). We removed items that were not loading cleanly on a given factor, correlated several error terms between items within the acceptance of the human condition measure and within the making sense of difficulties measure, and removed “meaningful” from elevation as this item also loaded on appreciation, where we believed it fit better. After these changes, we were able to achieve acceptable model fit, (χ²(533) = 1239.48, p < .001, RMSEA = .05, CFI = .96, SRMR = .04). These results were replicated in the “B” half of the data set, (χ²(533) = 1317.60, p < .001, RMSEA = .05, CFI = .96, SRMR = .04). A full list of items and loadings is available in Supplementary Table 2, available in supplemental materials on publisher website.

Results

Manipulation check

Appreciation has been established as a form of gratification in response to eudaimonic media content that is distinct from enjoyment (Hofer, 2013; Oliver & Bartsch, 2010). We therefore expected that participants responding to one of the more eudaimonic films should elicit higher levels of appreciation than those responding to the less eudaimonic comparison films. This served as the primary manipulation validity check that our conditions did in fact contain different levels of eudaimonic content. In order to test this, we utilized an Analysis of Covariance (ANCOVA) model which included film condition as a fixed factor, and both time since last viewing and frequency of viewing as covariates. Results showed that appreciation was significantly higher for participants in our more eudaimonic condition (M = 5.58) than those in the less eudaimonic comparison condition (M = 4.67, p < .001). There was a significant interaction effect between film condition and time since last viewing on appreciation. However, this was a weak interaction (R² change = .01 when including interaction term) that made the difference between conditions slightly larger as time since last viewing increased, but did not change the overall pattern of results (see Table 1 for full results).

Table 1. Results of ANCOVAs for study variables with time since last viewing and number of times seen as covariates

	More Eud. Films		Less Eud. Films		Univariate F-Tests
Variables	M	SE	M	SE	F	DF	$\eta$^2
Appreciation (Manipulation Check)^a,b	5.58	.06	4.67	.06	32.38***	1,1074	.03
Appreciation x Time Since Last Viewing					6.16*	1,1074	.01
Moral Motivations (H1)^a	6.94	.12	6.05	.12	26.45***	1,1059	.02
Making Sense of Difficulties (H2)^a	6.78	.11	6.10	.11	19.05***	1,1067	.02
Hedonic Motivations (RQ1)^a,b	5.86	.11	5.73	.12	.59	1,1065	.00
Enjoyment (RQ1)^a,b	4.21	.07	4.65	.07	.27	1,1075	.00
Enjoyment x Number of Times Seen					4.08*	1,1075	.00
Acceptance of the Human Condition (H3)^a,b	7.87	.08	7.32	.09	20.63***	1,1075	.02
Emotional Range (H4)^a	12.26	.29	9.86	.30	33.46***	1,1076	.03

Note: All interactions between covariates and film condition were tested and removed if non-significant. Significant interactions are shown. Values for partial ${\eta }^2$ are shown.

^aVariables for which time since last viewing was a significant predictor in the model.

^bVariables for which number of times seen was a significant predictor in the model.

*p < .05, **p < .01, ***p < .001.

More eudaimonic films vs. comparison films

We also tested our hypotheses concerning the direct effect of treatment conditions using similar ANCOVA models, including film condition as a fixed factor, and time since last viewing and frequency of viewing as covariates. Our specific hypotheses were that, relative to those in the comparison condition, participants in the more eudaimonic condition would rate films as contributing more to moral motivations (H1), a greater ability to make sense of life difficulties (H2), increased acceptance of the human condition (H3), and a broader range of emotional response (H4). As shown in Table 1, all these variables were significantly higher for the more eudaimonic condition, controlling for time since last viewing and frequency of viewing. Thus, H1–H4 were supported.²

We further asked whether there would be differences between those in the more eudaimonic and comparison conditions in rating of films as contributing to hedonic motivations and enjoyment (RQ1). Additional ANCOVAs revealed no difference between conditions for either hedonic motivations or enjoyment (see Table 1). There was one significant interaction effect between film condition and frequency of viewing on enjoyment. The interaction indicates that frequency of viewing increased reported enjoyment more for the films in the less eudaimonic condition than for those in the more eudaimonic condition.

Mediation analyses

We hypothesized individual mediation paths from recall of more eudaimonic films, through our mediators (elevation, poignancy), to ability to make sense of difficulties (H5a,b), moral motivations (H6a,b), and acceptance of the human condition (H7a,b). We further posed a research question as to whether emotional range would mediate effects on all three outcomes (RQ2). To test these hypothesized indirect effects, we constructed separate mediation models for elevation, poignancy, and emotional range on each of our outcomes: moral motivations, making sense of difficulties, and acceptance of the human condition, with time since last viewing and frequency of viewing as covariates. All tests of indirect effects were constructed using Model 4 of the PROCESS macro for SPSS (Hayes, 2018). As shown in Table 2, elevation, poignancy, and emotional range all significantly mediated effects on ability to make sense of difficulties, with elevation providing the largest point estimate. Significant indirect effects of all three mediators were similarly observed for the outcomes of moral motivations and acceptance of the human condition. Thus, hypotheses 5–7 were supported, and RQ2 was answered in the affirmative.

Table 2. Tests of Indirect Effects and Associated Bootstrapped Confidence Intervals

Outcomes (Mediation paths)
Making Sense of Difficulties	Point Estimate	BootSE	BootLLCI	BootULCI
More Eudaimonic Films → Elevation	.5877	.0891	.4124	.7612
More Eudaimonic Films → Poignancy	.2909	.0488	.2009	.3922
More Eudaimonic Films → Emotional Range	.2936	.0586	.1846	.4156
Moral Motivations	Point Estimate	BootSE	BootLLCI	BootULCI
More Eudaimonic Films → Elevation	.6067	.0910	.4298	.7873
More Eudaimonic Films → Poignancy	.3068	.0538	.2063	.4172
More Eudaimonic Films → Emotional Range	.3193	.0642	.1992	.4505
Acceptance of the Human Condition	Point Estimate	BootSE	BootLLCI	BootULCI
More Eudaimonic Films → Elevation	.3555	.0572	.2486	.4743
More Eudaimonic Films → Poignancy	.1732	.0346	.1097	.2449
More Eudaimonic Films → Emotional Range	.1779	.0374	.1090	.2558

Note: BootLLCI = Bootstrapped lower limit of the 95% confidence interval; BootULCI = Bootstrapped upper limit of the 95% confidence interval. Significant indirect effects are indicated by a bootstrapped confidence interval that does not include zero.

Value match

We hypothesized that a match in values between audience member and film would lead to greater ability to make sense of difficulties (H8a), pursuit of moral motivations (H8b) and acceptance of the human condition (H8c). We tested these effects using hierarchical linear regression models that included our covariates (time since last viewing and frequency of viewing) in the first block and then added value match to determine its unique contribution. The first model showed a significant effect of value match (b = .18, FΔ(1,1065) = 295.93, p < .001, R²Δ = .20), indicating that participants felt better able to make sense of difficulties in their own lives if they rated a film that portrayed values perceived to more closely align with their own. Value match showed a similar positive relationship with our other outcome variables, pursuit of moral motivations (b = .18, FΔ(1,1051) = 217.78, p < .001, R²Δ = .16) and acceptance of the human condition (b = .15, FΔ(1,1066) = 330.99, p < .001, R²Δ = .22). Thus, Hypotheses 8a-c were all supported.

Post-hoc analyses regarding recency of film viewing

Given the retrospective nature of this study, we wanted to assess the possibility that individuals responding to a film that they had not seen in an extended period of time unduly influenced their evaluations. Thus, we restricted the data to include only participants who indicated they had seen the film within the past year or within the past 6 months (N = 302) and reran all analyses with the quarter of the sample who had seen the film more recently. ANCOVA results generally aligned with the full results, with one minor exception.³ While motivation to pursue moral motivations was rated as higher for more eudaimonic films (M = 7.63, SD = 2.62) than less eudaimonic films (M = 7.11, SD = 2.56), this difference was only marginally significant in the restricted model, F(1, 294) = 3.45, p = .06, $\eta$² = .01; by comparison, the F statistic for the full model was over 26. It appears that with more time and reflection, the more eudaimonic films may be relatively more likely to promote these moral motivations in an enduring manner. For the mediation analyses, all of the individual path model results were confirmed for the restricted data set.

Discussion

Through studying retrospective responses to films either relatively high or relatively low in eudaimonic content, we show that such films vary in the range of emotional response they elicit, as well as in the degree that they resonate with viewers’ values. Further, such differences can contribute to outcomes such as helping individuals feel they can make sense of the difficulties in their lives, motivating the pursuit of moral goals, and supporting greater acceptance of the human condition. While narrative effects research traditionally focuses on behavioral outcomes, we believe such effects on subjective well-being are important in their own right, and deserve greater attention; our results suggest the Mediated Wisdom of Experience perspective (Slater et al., 2016, 2018) should be extended to incorporate consideration of these variables.

As predicted, more eudaimonic films elicited higher levels of appreciation, and were associated with greater ability to make sense of difficulties and pursuit of moral motivations, such as wanting to be a better person, doing good for others, or seeking what really matters in life. However, we did not find a difference between conditions in terms of enjoyment or hedonic motivations, which referred to more everyday experiences of pleasure such as enjoying oneself. These findings make sense given the conceptualization of eudaimonic and hedonic elements as representing independent dimensions rather than two ends on a continuum (Vorderer, 2011). Hence, a film that is considered more eudaimonic still can be quite enjoyable, and a more hedonic film (or less eudaimonic film) does not necessarily have to be bereft of meaning.

Participants who recalled more eudaimonic films rated their acceptance of the human condition higher than did those who recalled less eudaimonic films. This suggests that media typically considered eudaimonic can impart a greater felt sense of the joys and sorrows that are an inevitable part of being human. This helps to extend the Mediated Wisdom of Experience perspective (Slater et al., 2016, 2018) by more directly assessing aspects of maturity vicariously imparted through media. While it is impossible to definitively establish causal order given our retrospective design, our findings indicated that typical emotional and cognitive responses to films helped facilitate greater acceptance of the human condition. In other words, experiencing poignancy, elevation, or a broader range of emotions led to a felt sense of the precious, often ambiguous nature of lived experience. This shift in perspective is important as it suggests a corollary shift in goals one might choose to pursue. For example, similar shifts toward emotional meaning typically associated with development across the lifespan have been shown to impact social choices, such as focusing on the depth of close relationships rather than cultivating a wider network (Carstensen et al., 1999). Having an intervention capable of cultivating such a shift in perspective may be useful for a variety of interventions where greater maturity in decision making and goal pursuit are paramount. In tapping into the psychological benefits of broadened perspective concerning the vicissitudes of life, acceptance of the human condition may be important for resilience, coping, or other well-being outcomes, and future research should examine such possibilities.

Similarly, making sense of difficulties as a variable suggests ways that media may be useful as an intervention in specific well-being contexts. Specifically, these results demonstrate potential benefits in regard to subjective psychological well-being. To the extent that media are able to help individuals positively reframe their own challenges in life, they may provide a means of coping with and growing through traumatic experiences (Neimeyer, 2004). While most media effects are strongest directly following exposure, the retrospective design used here suggests that such positive coping skills may well endure over time. It is likely that individuals who prefer such content return to it more frequently, in line with research on emotional flow (Nabi & Green, 2015), and hence such benefits are cumulative in nature. Our results suggest media exposure may not only have the capacity to improve subjective well-being in terms of short-term recovery (Reinecke et al., 2011) but that these effects are recognized by the media user and may be acknowledged even after many years have passed. This points to further questions about why individuals are motivated to select media content that depicts difficult situations in the first place. Future research may be able to use such outcomes to better understand how intentional such media selection choices are, whether as a means for coping or satisfying other intrinsic needs.

Emotional range was another novel construct investigated in this study, meant to capture the broader repertoire of emotional responses to media. It is possible that such a broader range of emotions experienced in response to media content reflects more nuance in how we process and reflect upon the often ambivalent and complicated themes presented. It is not surprising then, that such emotional range would help facilitate greater sense-making of difficulties, acceptance of the often-bittersweet nature of lived experience, and motivate pursuit of more meaningful goals, as seen in our outcomes. It may be that emotional range is a precursor to various kinds of meaningful affect, such as being moved, touched, or inspired. However, trying to establish causal order in this retrospective survey design would be problematic.

Our results also showed that a greater value match with the film was associated with a greater ability to make sense of difficulties, increased moral motivations, and greater acceptance of the human condition among participants. It seems that seeing values that one views as particularly important represented in media can help individuals reframe difficulties around such ideals, as well as motivate them to pursue goals in accordance with such values. It also suggests that the representation of important values helps individuals be more accepting of the complex nature of lived experience, perhaps because such films tend to highlight that such principles are typically achieved through both gains and losses, joys and sorrows. While our less eudaimonic comparison films did not lead directly to these outcomes, the extent to which they were perceived to portray personally important values still facilitated such effects. It is important to emphasize, then, that value resonance is not restricted to more typically meaningful or thought-provoking content. While this study did not differentiate between values, future research may consider what specific outcomes might be expected when there is a match between viewer and content in regard to specific values. It is also worth considering what the implications are for individuals who hold value structures in opposition to those portrayed by the narrative viewed or read.

Limitations

One of the simultaneous strengths and limitations of this study is that it assessed films in a retrospective manner and depended on self-reported outcomes, as opposed to manipulating exposure as part of the experimental procedure. We did attempt to facilitate recall by providing the promotional poster for each film and by asking participants to take a few moments to recall and describe any scenes from the film that they found most memorable. We also controlled for time elapsed since last viewing and frequency of viewing in all analyses. Nonetheless, there was still likely variability in terms of how well individuals may have been able to recall the film and assess their specific responses and evaluations to the content. This suggests that our study presents a relatively conservative test of the outcomes described, and that studying direct exposure may well lead to stronger results. However, to further ensure that time since viewing was not substantively skewing our results, we ran all analyses on a subset of participants who had viewed the film they assessed within the past year. These results largely confirmed the analyses on the full data set. At least with respect to self-reported outcomes, the effects of exposure to entertainment may be relatively durable across time, and not merely a function of short-term exposure. Moreover, some differences between the more eudaimonic and comparison films (e.g., appreciation and impact on moral motivations) in fact tended to slightly increase over time.

Additionally, the broad and diverse list of feature films employed in this study (20 per condition) makes it unlikely that systematic biases would affect the results. Even if some films have taken on new meanings or interpersonal discussions with others regarding a film altered evaluations for some individuals over time, we would expect such effects to be randomly dispersed across the extensive list of films used. The use of popular feature films also makes this study conservative as a test of responses to eudaimonic narratives. It is likely that more literary narratives (either films or novels) may be more strongly eudaimonic than Hollywood films intended to appeal to a mass audience. From an ecological vantagepoint, of course, this is an advantage, as Hollywood movies are a form of eudaimonic narrative more likely to be encountered by the general population.

Another inherent limitation of this retrospective approach is the dependence on self-reports of effects. However, as noted above, it seems to us that the perception that a film impacted one’s capacity to make sense of difficulties in one’s life and inspired a desire for more moral action and behavior, are of social and psychological importance in their own right. Further, we acknowledge that some effects observed in this study were relatively small. While our large sample size contributed to this, we also note that non-significant results were observed for conceptually important variables (such as group differences for hedonic motivations and enjoyment). We also note that use of popular entertainment narratives is a world-wide phenomenon across vast populations; small effects across such large numbers are of substantive as well as theoretical interest.

Finally, it is also important to note that causal order of effects cannot be definitely established given a retrospective approach. We attempted to follow previous research on eudaimonic effects in using emotional responses as mediators of cognitive and motivational responses. In reality, many of these responses may occur so close together as to be virtually simultaneous, or may occur in a somewhat iterative or recursive relationship to each other, and possible more complex causal processes cannot readily be assessed. Thus while causal order is difficult to establish, the associations between variables remains. The fact that many of these associations are evident in the retrospective evaluations utilized here again illustrates the durability of such relationships.

Overall, this study provides novel evidence concerning how elevation, poignancy, and emotional range may mediate the effects of more eudaimonic films. In addition, we assessed how these responses led to self-reported outcomes such as making greater sense of one’s own life difficulties, motivation toward moral behavior, and acceptance of the human condition. In so doing, the study underscores the value that such films have in the lives of audience members, beyond enjoyment alone. We hope that this project will stimulate further research into understanding the nature of eudaimonic media and its beneficial outcomes in the lived experience of viewers and readers, and that the novel variables conceptualized and tested here will serve as useful tools in pursuit of such ends.

Supplementary material

Supplemental data for this article can be accessed on the publisher’s website.

Correction Statement

This article has been republished with minor changes. These changes do not impact the academic content of the article.

Additional information

Notes on contributors

Jared M. Ott

Jared M. Ott is a doctoral student in the School of Communication, Ohio State University. His research focuses on narrative messages with interests in message engagement and processing, persuasive outcomes, and improving measurement.

Naomi Q. P. Tan

Naomi Q. P. Tan (Ph.D., The Ohio State University, 2020) is a postdoctoral fellow at The University of Texas MD Anderson Cancer Center. She is a health communication scholar focusing on shared decision making and the design of educational interventions and health messages that are culturally appropriate in order to address health disparities.

Michael D. Slater

Michael D. Slater (Ph.D. Stanford University, 1988; MPA, New York University, 1982; BA, Columbia University, 1974) is Social and Behavioral Science Distinguished Professor and Director of the School of Communication, The Ohio State University. His research includes theory-building efforts in narrative influences (Temporarily Expanded Boundaries of the Self or TEBOTS and Mediated Wisdom of Experience), persuasion (Extended Elaboration Likelihood Model or E-ELM) and dynamic processes of media selection, media effects, and maintenance of personal and social identity (the Reinforcing Spirals Model), with a particular interest in health outcomes, with over 175 publications and $12 million in funded research.

Notes

¹ Our use of moral and hedonic motivations in this study is following work by Oliver et al. (2012), in which such motivations are used as outcomes of exposure to entertainment content. While some past research has looked at motivations for seeking out specifically meaningful or hedonic media content, our use of moral motivations here is not connected to such choices for entertainment consumption.

² There were main effects for one or both covariates in these models (see Table 1). Thus, how recently and frequently a film had been seen did have a main effect on our outcomes, as one might expect. We also ran analyses without the covariates and results were largely unchanged. As discussed in text, there was little evidence for interactions between these covariates and condition. Detailed results are available from the authors upon request.

³ In the restricted ANCOVA models the covariates for some variables and all interactions with covariates also became non-significant. However, these changes strictly affected control variables, and were not central to any of the hypotheses or research questions in this study.