Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Does Misery Love Company? Exploring the Therapeutic Effects of TV Viewing on Regretted Experiences

Does Misery Love Company? Exploring the Therapeutic Effects of TV Viewing on Regretted Experiences Abstract Testing hypotheses derived from regret and mood management theories, this research explores how regretted experiences impact interest in viewing experience-relevant TV programming and such viewing’s effects on program enjoyment and felt regret. One hundred and forty-four participants, half of whom had been unfaithful in romantic relationships, were asked first to rate their interest in viewing a series of storylines and then to provide their reactions to 1 of 2 versions of a TV program depicting cheating behavior. Largely consistent with hypotheses, results indicated that those who had both cheated and felt regret about their behavior were more likely than others to want to watch experience-related storylines, were no less likely to enjoy watching such programming, and particularly preferred viewing the program version in which the main character rationalized, rather than expressed regret for, her behavior. Both program versions, however, reduced regret equally. A survey of 206 city residents also offered evidence consistent with predictions based on regret theory. Overall, this research speaks to the value of integrating theories of emotion with media theory to enhance the latter’s predictive ability. It is widely accepted that the mass media can be used to meet people’s social and psychological needs, emotional ones in particular (e.g., Rubin, 2002; Zillmann, 1988). However, apart from the research on coping with media-induced fear (Kubey & Peluso, 1990), the literature offers very little insight into how viewers might use media like TV to cope with (vs. escape from) emotions associated with distressing personal situations. Yet, given the multitude of storylines in entertainment programming that deal with such issues, including health threats, violence, and relationship turmoil, it is reasonable to imagine that TV viewing may serve not simply as a potential source of painful reminders of life’s trials and tribulations but also as a potential resource for coping and comfort to help emotional wounds begin to heal. Among some of the most painful and intractable emotional experiences potentially triggered by TV viewing are those related to regret. Marked by active rumination and motivation to understand why certain past events unfolded to lamentable effect (Landman, 1993), regret over past actions (or inactions)—comparing what is to what might have been—can persist or recur for years after the regretted event itself has passed. Although theoretical perspectives, like uses and gratifications or mood management, suggest emotional needs may be met through media exposure, such frameworks, without benefit of understanding the specific needs associated with unique emotional states like regret, may be inadequate to predict how such emotions might impact media consumption and subsequent effects. However, by integrating regret theory with theory of media selection, we may arrive at more precise predictions of how those suffering might interact with the media to alleviate their personal distress and, perhaps in turn, enhance the quality of their lives and social interactions. The purpose of this research, then, is first to explore how regretted experiences influence the desire to view experience-related entertainment TV programming and second to consider how viewing such programming impacts one’s subsequent feelings of regret. In doing so, we hope not only to extend the literature on media and emotion by exploring the role of TV as “therapist” in helping viewers cope with personal problems but also to elaborate current media theory that has yet to fully benefit from the insights available from the theory and research on specific emotional experiences. Media theory and emotional coping Uses and gratifications Extant media theory has a history of recognizing the important role media consumption plays in satisfying emotional needs and adjusting affective states. The uses and gratifications paradigm in particular is frequently referenced as offering a framework through which media selection and use based on these and other needs can be understood. In essence, the uses and gratifications paradigm asserts that individuals are aware of their social, psychological, and biological needs. Based on this awareness, they evaluate various media channels and content, assess functional alternatives, and select the media or interpersonal channel that they believe will best meet their various needs (e.g., Katz, Blumler, & Gurevitch, 1974; Palmgreen, Wenner, & Rosengren, 1985; Rubin, 2002; Rubin & Rubin, 1985). Such needs include those related to diversion (e.g., escapism and arousal), personal relationships (e.g., social utility), personal identity (e.g., reality exploration), and surveillance (e.g., news gathering; McQuail, Blumler, & Brown, 1972; see review by Rubin, 2002). Of particular interest to this investigation are needs associated with “emotional release,” which, though identified early on as a primary function of media consumption (Herzog, 1944), has focused primarily on issues of diversion, entertainment, and arousal (see Finn & Gorr, 1988). Indeed, Finn and Gorr subsume these three motivations (along with information seeking) into a more general category of “mood management,” borrowing the term from Zillmann’s (1988) seminal work examining the selection of media to regulate levels of arousal. Indeed, mood management is the gratification most directly related to this research. Mood management theory In essence, Zillmann’s (1988) mood management theory asserts that people use media to modulate their affective states. More specifically, Zillmann argues that people, driven by hedonistic desires, strive to alter negative moods as well as maintain and prolong positive ones. Consequently, they will arrange their environments to adjust “all conceivable moods” (p. 328), using any type or genre of communication, as well as all specific forms within such genres, available. Moreover, Zillmann, in assuming that mood management processes are based on operant conditioning, argues that people may, but need not, be cognizant of the reasons for their choices. Zillmann goes on to note four message features that might impact message selection based on mood: excitatory potential, absorption potential, semantic affinity, and hedonic valence. For each, the underlying principle is the same. If a message reflects one (or presumably more) feature that might perpetuate the negative state, for example, high absorption potential or mood-related content, the message is likely to be avoided in favor of one that would interrupt the negative state, like a message of an opposite valence or on a topic unrelated to the mood source. Much research supports the predictions of mood management theory (see Oliver, 2003, for a recent review). However, its boundaries have been challenged by paradoxes of media selection, like the enjoyment of horror movies or tearjerkers that are designed to evoke negative effect seemingly in contradiction to mood management theory’s assumption of hedonistic motivation. Zillmann (2000) attempted to subsume such outcomes within the mood management paradigm, arguing that sometimes we postpone immediately hedonistic outcomes in pursuit of other goals that will bring about positive affective states once achieved (i.e., telic hedonism). Yet, it is debatable whether such investigations justifiably fall under the umbrella of mood management theory as they do not employ the key assumption that hedonism is the primary motivation underlying choice, and as Knobloch (2003) notes, nearly all “goal-directed behavior can be construed as maximizing delight or minimizing discomfort at some point along the line” (p. 235). Further, even if we accept the expanded boundaries of mood management theory to include delayed hedonistic gratification, this framework loses its predictive power regarding media selection when particular emotional states are aroused. This becomes evident as we consider Zillmann’s discussion of hedonistic versus information-seeking motives and their implications for media choice. In general, Zillmann suggests that hedonistic motives will guide selection of entertaining media, whereas information needs will guide selection of informative/educational messages. Only under exceptional conditions does he allow that informational needs may temporarily supersede hedonistic ones when selecting entertainment programming. Though he does not specify what those conditions might be, it appears that he does not see emotional states as falling into this category as he argues that extreme emotional states, like anger, would lead to “state-focused cognitive preoccupation” that interrupts hedonistic goals in place of “instrumental coping efforts” (p. 109). Such coping, he asserts, leads one to avoid diversionary media messages altogether. Though he does not say so explicitly, we might infer that Zillmann would suggest that informational sources might be sought to aid efforts to cope with emotional states. Indeed, this position would be supported by studies that demonstrate that TV news is sought to gain information during or after national crises, like the space shuttle disaster (Kubey & Peluso, 1990) and September 11.1 Thus, based on Zillmann’s (1988, 2000) writings, it seems that mood management theory would suggest that an emotion like regret might lead to (a) avoidance of emotion-relevant entertainment programming (due to semantic affinity); (b) avoidance of entertainment programming altogether (due to emotion-focused coping); (c) watching informational/educational programming to aid coping, or, if all else fails; (d) exposure to entertainment programming driven by informational needs. Surely, one of these predictions is bound to be accurate; yet, mood management theory is unable to tell us with reasonable certainty which one is most likely. If we draw from emotion theory, however, we are far better positioned to arrive at precise predictions. Like Zillmann (2000), emotion theorists—Lazarus (1991) in particular—argue that negative emotional experiences are associated with coping needs and efforts. Additional evidence suggests that these emotion-driven coping efforts are likely associated with unique information-seeking goals (Nabi, 1999, 2003). If the emotional states stem from personal and goal-relevant situations, which most emotions likely do (Lazarus, 1991), it stands to reason that entertainment programs that depict stories of life’s trials and tribulations will not serve as sources of diversion, as Zillmann implies, but rather as potential sources of information for those seeking to cope with their distress. In essence, we suggest that even if programming is generally intended to entertain, particular emotional states might draw, not repel, viewers to that programming if they have some expectation that the storylines might offer information useful to their coping needs as generated by their particular emotional states. If the need to regulate these emotional states results in information seeking from entertainment programming, not only do we have a theoretical motivation for proposing the existence of an effect, which Zillmann considered unusual, but there is also an established body of literature from which to draw to predict when this would happen and to what possible result. As our investigation focuses on the emotion of regret, we now turn to this literature. Regret theory Regret, at its foundation, is a cognitively driven affective phenomenon based in having misgivings about decisions or actions that cannot be undone (Landman, 1987). As Sugden (1985) defines it, regret is “the painful sensation of recognising that ‘what is’ compares unfavourably with ‘what might have been.’” Regret can be evoked by past events or decisions, both small (e.g., choosing the wrong line in the grocery story) and large (e.g., cheating on a spouse). Theories of regret arousal (developed primarily from economic models of decision making; see Landman, 1993, for a review) generally suggest it is a function of the value of a chosen option compared with the values of foregone options combined with self-recrimination (e.g., Connolly & Zeelenberg, 2002; Sugden, 1985). That is, one might experience regret not only if the outcome is not what one had hoped for, especially when compared to an alternative course, but also if one feels responsible for the outcome, even if there was nothing one could have done differently. Thus, counterfactual thinking and self-blame are central features marking the experience of regret. At its extreme, regret is associated with cognitive preoccupation that mires one in the past, fruitless and irrational, preventing forward progress (Landman, 1993). However, its benefits include learning from past mistakes, initiating life changes, and making more thoughtful decisions in the hopes of avoiding future regretted experiences (Landman, 1993; Stewart & Vandewater, 1999; Zeelenberg, 1999). The past 10 years have seen a proliferation of studies examining the features of regretted experiences, like the types of decisions people tend to regret (e.g., relationship choices, career or financial decisions, and educational choices; see Gilovich & Medvec, 1994) and the role of action versus inaction (e.g., Gilovich, Medvec, & Kahneman, 1998; Savitsky, Medvec, & Gilovich, 1997; Zeelenberg, van den Bos, van Dijk, & Pieters, 2002). Yet, there is little research on how regret is actually resolved. Boninger, Gleicher, and Strathman (1994) suggest that counterfactual thinking itself serves as a coping technique for regret in that reflection allows for future preparation, and this preparation then serves to distract from the currently felt negative affect. Thus, reflection may not only lead to more functional outcomes in the future but also alleviate negative feelings, and perhaps generate positive ones, in the present. The question most relevant to this research then becomes, how might media content aid counterfactual thinking? Because the research on regret suggests that it is largely an intrapersonal phenomenon, there is no reference in the literature to the role the media might play in either arousing or alleviating feelings of regret. However, to the extent that a regretted experience is portrayed in entertainment TV programming, one could imagine that watching such programming might not simply remind one of the regretted act but those representations might offer insight into that experience that could, in turn, help minimize feelings of regret. So, whereas mood management theory proper suggests that seeing a regretted experience portrayed on TV would exacerbate negative affect and thus lead to avoidance, and the counterhedonistic extension of mood management theory suggests that either avoidance or consumption might occur (but is not clear on which), the regret literature suggests that people experiencing regret would likely be drawn to experience-related entertainment programming for its potential information value and the insight gained from the TV portrayal might, in turn, help to alleviate their regret. Hypotheses Based on the above literature review, mood management theory and regret theory offer somewhat competing predictions about how people with regretted experiences will react to entertainment programming on the topic of their regret (i.e., with semantic affinity).2 Whereas mood management theory suggests people will avoid such programming as it will exacerbate their negative affect, regret theory predicts that the opportunity to ruminate on the regretted experience to hopefully come to terms with their past mistake (i.e., to aid in counterfactual thinking) will be welcome. Thus, those experiencing regret will, in the interest of coping, likely be drawn to programming that reflects the source of their affective disposition. Conversely, for those who have not had the experience to regret, it is reasonable to assume that regular mood management processes will be engaged. That is, viewing preference will be predicted by expectation of the programming’s impact on mood. Those who might associate a behavior with negative affect (e.g., would anticipate regretting an experience, were they to have it) will be less likely to seek out such programming compared to those who have neither had the experience nor associate it with negative affect. As our research focuses on cheating on a romantic partner as the potentially regretted behavior—a common regret among people both young and old—we propose the following interaction hypothesis, based on the logic outlined above: H1: Regret will interact with past cheating behavior such that (a) for those who have cheated in the past, those high in regret will be more interested in viewing experience-related TV storylines than those low in regret, whereas (b) for those who have not cheated, those high in anticipated regret were they to cheat will be less interested in viewing experience-related TV storylines than those low in anticipated regret. If exposed to programming reflecting a regretted experience, like an act of infidelity, regret theory offers insight into how viewers will react. On the one hand, reliving regretted experiences, like reopening a wound, is not pleasant. Yet, as Boninger et al. (1994) suggest, the positive affect associated with the potential for learning effective means of resolving regret may make such programming enjoyable from an informational, though perhaps not an entertaining, standpoint. However, for those without the regretted experience, assuming hedonistic motives are paramount, we predict that those who associate the experience with negative affect (e.g., would anticipate regretting the experience, were they to have it) will enjoy watching such programming less than those who do not associate it with regret. Thus, H2: Regret will interact with past cheating behavior such that (a) for those who have cheated in the past, those high in regret will enjoy watching an experience-related TV storyline more than those low in regret, whereas (b) for those who have not cheated, those high in anticipated regret were they to cheat will enjoy watching an experience-related TV storyline less than those low in anticipated regret. Finally, regret theory suggests that alternative perspectives on one’s past mistakes (e.g., increasing the value of one’s own actions, decreasing the value of the alternative paths, or minimizing self-blame) should help one cope with regret. Thus, the effect of program viewing on both enjoyment and regret over the past behavior should be affected by how the story protagonist reacts to the regretted situation. If the protagonist expresses self-blame and regret, that is, does not offer a model for resolution, viewers who regret their past behavior will be unlikely to receive new insight to aid in their coping. However, if the protagonist does not express self-blame or regret, this portrayal may serve as a model for coping by reducing the negative value of the path chosen and/or reducing self-blame. Consequently, a regretful person may not only enjoy this programming more but will also feel better about his or her own situation. In contrast, those who do not feel regret about their past behavior, and thus have no coping goals, should have no reason to prefer one portrayal over the other. Thus, focusing only on those who have cheated on a relational partner in the past, we suggest that: H3: Regret will interact with character portrayal such that (a) cheaters with lingering regret over their past behavior will enjoy viewing a media character who rationalizes cheating more than a media character who self-blames, whereas (b) cheaters without lingering regret over their past behavior will enjoy both portrayals equally. H4: For those with lingering regret over their past cheating behavior, viewing a media character who rationalizes cheating will reduce cheating-related regret more than viewing a media character who self-blames. Study 1
Method Participants and procedures One hundred and forty-four participants, 54% women, were recruited to participate in a study on TV and relational choices. Of these, the majority were upperclassmen (36% juniors and 44% seniors). Ninety percent had been in a committed romantic relationship in the past, and 36% were currently in such a relationship. Fifty-three percent of respondents reported having cheated on a romantic partner, equally divided among men and women. In groups of 3–12, participants completed a survey in which they were asked about their TV viewing habits, behaviors they believed constituted unfaithfulness, and how they felt (or would feel) if they or their partner had been unfaithful. After several filler scales tapping personality constructs, respondents were given a list of nine hypothetical storylines from a dramatic TV series to rate in terms of how interested they were in watching each. Participants then watched one of two versions of a 7-minute edited segment of a popular reality-based drama in which a young woman cheats on her boyfriend and then either self-blames/expresses regret or rationalizes her behavior. Respondents then provided their emotional reactions to the segment, how enjoyable they found it, their reactions to the protagonist, and their current feelings about any incidents of unfaithfulness in their past. Stimuli To choose a topic for this research, we solicited events regretted by undergraduates (n= 37). The most frequent responses included those related to partying (e.g., drinking too much), relationships (e.g., a bad relationship and indiscretions), and school participation (e.g., not studying enough and missing classes). Of these, a topic frequently mentioned by our sample, frequently portrayed in the media, and generalizable to other age groups is cheating on a romantic partner. Our assumption was that those who cheated on a boyfriend or girlfriend were vulnerable to experiencing regret, and we further assumed that resolution of regret ranged from occurring quickly after the event to not having yet fully occurred. The stimuli for the second half of the study were drawn from the MTV Series (U.S. produced) Real World, in which several episodes across a season were edited into a 7-minute segment using Adobe Premiere Pro 1.5. We chose this particular series not only because it contained the theme we were hoping to display but also because it is a program popular with young people and thus more likely to serve as a source of coping information compared to less popular programming. The first 5 minutes of the stimuli, which were identical for both conditions, introduced the female protagonist, a young woman with a boyfriend back home but a tendency to flirt with other men when drunk. She meets a guy in a bar who she begins to spend time with, though she claims they are just friends. One evening, after drinking a bottle of champagne, she and her friend “make out.” At this point, the two versions of the stimulus diverge. In the “self-blame” condition, the protagonist has a telephone conversation with her boyfriend in which he indicates that if she cheated, he will break up with her. In her subsequent “confessional,” she tearfully questions why she makes such destructive choices in her life. In the “rationalization” condition, the protagonist has a telephone conversation with her new male friend in which she says she anticipates her boyfriend being unhappy with her actions, even though she does not think she did anything wrong. Then, in a conversation with her roommate, she indicates that it is unfair of her boyfriend to limit her behavior, and she concludes that she is “a good girlfriend.” Measures Measures are presented in the order in which they were completed. As this study reports on a subset of the data collected, variables not relevant to the present analyses are omitted. After completing TV viewing and demographic measures, respondents were asked on a scale of 1 (not at all) to 5 (definitely) how much each of six behaviors constituted being unfaithful to a romantic partner and how damaging to a relationship each would be.3 Participants were then asked to indicate how much of several emotions they felt when thinking about being unfaithful to a romantic partner—past, present, or imagined. Of these, four related to regret—regret, remorse, guilt, and shame. The four loaded onto a single factor and formed a reliable index assessing regret (α=.91). Respondents were then asked whether or not they had been unfaithful in a past or present romantic relationship and, if yes, how long ago the incident(s) occurred. Of note, there was minimal association between how long ago the cheating occurred and the degree of regret felt, r(75) =.14, p=.22. Respondents were then asked to assess how they felt when thinking about their partner (past, present, or imagined) being unfaithful to them, using the same emotion items noted above, followed by a question asking whether or not, to the best of their knowledge, a romantic partner had ever been unfaithful to them. After several personality trait measures, the participants responded to the dependent measure related to selective exposure—storyline preference. Nine storylines were presented as having appeared in a popular TV drama. Respondents were asked to rate how interested they were in watching each episode. For women, the protagonist of each storyline was female; for men, the protagonist was male. Of the nine storylines, three related to cheating (presented from the survey for the female participants)—“A young woman jeopardizes her relationship with her boyfriend when she pursues a summer crush, only to regret her decision,”“A stormy relationship comes to a head as a young woman, lonely and away from home, finds comfort with a new guy,” and “A young woman’s suspicions about her boyfriend’s ‘extracurricular activities’ turn out to be true, and she must decide whether to break off the relationship or take him back.” These three were combined to reflect “cheating-related storylines.” The remaining storylines, examples of which include the protagonist hiding an eating disorder (females)/steroid use (males), dealing with a pregnancy scare, pursuing dreams of becoming an actor, and coping with mental illness, were combined to represent “cheating-unrelated storylines.” After viewing the program segment, respondents completed a mood adjective checklist and then indicated how much they enjoyed viewing the segment using four items (enjoyable, entertaining, pleasurable, and captivating) rated on 1 (not at all) to 7 (very much) scales (α=.91) and how much they liked the protagonist using six items (friendly, likable, pleasant, trustworthy, reliable, and honest), again rated on 1 (not at all) to 7 (very much) scales (α=.85). As a manipulation check, we asked how regretful the protagonist seemed about her behavior. As a way of validating the self-reports of cheating experiences, we also asked how much the respondent identified with the situation as experienced by the protagonist as well as her boyfriend. Each of these items was measured with 1 (not at all) to 7 (very much) scales. To test H4, we asked, for those who had cheated in the past, how they currently felt about their behavior. Of the several emotion items included, the four items related to regret were of primary interest (α=.95). Finally, we included several measures to serve as possible control variables. Given the protagonist’s alternative style, we included a homophily scale (McCroskey, Richmond, & Daly, 1975) in case lack of similarity to the character influenced our results (α=.77). We measured perceivedrealism of the TV program using five items adapted from Rubin’s (1981) perceived realism scale (α=.88), and we further measured how frequently respondents had viewed Real World in the average month (0 to 5+ times) and whether they had seen the particular storyline before. Respondents indicated watching slightly more than 2.5 episodes per month (M= 2.69, SD= 1.76), and two-thirds indicated they had seen the particular storyline used for the stimulus (68%). Analysis Given the continuous nature of the regret measure, the hypotheses were tested using regression procedures. However, to more clearly present the interaction results, regret was dichotomized, analyses of covariance were performed, and the group means presented in Tables 1–3. Table 1 Interest in Viewing Cheating-Related Storylines by Past Cheating Behavior and Regret Past Behavior . Regret . Low . High . M . SD . M . SD . Cheated 5.68 (2.43), n= 37 6.44 (2.45), n= 39 Not cheated 6.17 (2.25), n= 37 5.29 (2.10), n= 31 Past Behavior . Regret . Low . High . M . SD . M . SD . Cheated 5.68 (2.43), n= 37 6.44 (2.45), n= 39 Not cheated 6.17 (2.25), n= 37 5.29 (2.10), n= 31 Note: Means are based on an ANCOVA comparable to the regression analysis reported in the text. Regret levels are based on a dichotomous split of regret. Open in new tab Table 1 Interest in Viewing Cheating-Related Storylines by Past Cheating Behavior and Regret Past Behavior . Regret . Low . High . M . SD . M . SD . Cheated 5.68 (2.43), n= 37 6.44 (2.45), n= 39 Not cheated 6.17 (2.25), n= 37 5.29 (2.10), n= 31 Past Behavior . Regret . Low . High . M . SD . M . SD . Cheated 5.68 (2.43), n= 37 6.44 (2.45), n= 39 Not cheated 6.17 (2.25), n= 37 5.29 (2.10), n= 31 Note: Means are based on an ANCOVA comparable to the regression analysis reported in the text. Regret levels are based on a dichotomous split of regret. Open in new tab Table 2 Program Enjoyment by Past Cheating Behavior and Regret Past Behavior . Regret . Low . High . M . SD . M . SD . Cheated 3.89 (1.60), n= 37 3.72 (1.90), n= 24 Not cheated 4.01 (1.62), n= 37 3.45 (1.24), n= 31 Past Behavior . Regret . Low . High . M . SD . M . SD . Cheated 3.89 (1.60), n= 37 3.72 (1.90), n= 24 Not cheated 4.01 (1.62), n= 37 3.45 (1.24), n= 31 Note: Means are based on an ANCOVA comparable to the regression analysis reported in the text. Regret levels are based on a dichotomous split of regret. Open in new tab Table 2 Program Enjoyment by Past Cheating Behavior and Regret Past Behavior . Regret . Low . High . M . SD . M . SD . Cheated 3.89 (1.60), n= 37 3.72 (1.90), n= 24 Not cheated 4.01 (1.62), n= 37 3.45 (1.24), n= 31 Past Behavior . Regret . Low . High . M . SD . M . SD . Cheated 3.89 (1.60), n= 37 3.72 (1.90), n= 24 Not cheated 4.01 (1.62), n= 37 3.45 (1.24), n= 31 Note: Means are based on an ANCOVA comparable to the regression analysis reported in the text. Regret levels are based on a dichotomous split of regret. Open in new tab Table 3 Program Enjoyment by Character Portrayal and Regret (for past cheaters only) Character Portrayal . Regret . Low . High . M . SD . M . SD . Self-blame 3.64 (1.62), n= 14 3.51 (1.91), n= 23 Rationalization 3.70 (1.71), n= 16 4.25 (1.70), n= 23 Character Portrayal . Regret . Low . High . M . SD . M . SD . Self-blame 3.64 (1.62), n= 14 3.51 (1.91), n= 23 Rationalization 3.70 (1.71), n= 16 4.25 (1.70), n= 23 Note: Means are based on an ANCOVA comparable to the regression analysis reported in the text. Regret levels are based on a dichotomous split of regret. Open in new tab Table 3 Program Enjoyment by Character Portrayal and Regret (for past cheaters only) Character Portrayal . Regret . Low . High . M . SD . M . SD . Self-blame 3.64 (1.62), n= 14 3.51 (1.91), n= 23 Rationalization 3.70 (1.71), n= 16 4.25 (1.70), n= 23 Character Portrayal . Regret . Low . High . M . SD . M . SD . Self-blame 3.64 (1.62), n= 14 3.51 (1.91), n= 23 Rationalization 3.70 (1.71), n= 16 4.25 (1.70), n= 23 Note: Means are based on an ANCOVA comparable to the regression analysis reported in the text. Regret levels are based on a dichotomous split of regret. Open in new tab Results Manipulation checks Those who reported having cheated on a relational partner identified with the protagonist’s situation (M= 3.32, SD= 1.80) more than those who had not cheated (M= 2.00, SD= 1.20), F(1, 142) = 26.07, p <.001, η2=.16. However, they did not significantly differ in their identification with her boyfriend (cheated: M= 3.92, SD= 2.19; not cheated: M= 3.38, SD= 2.19), p=.12. Conversely, those who had been cheated on identified with the boyfriend’s situation (M= 4.83, SD= 2.11) more than those who had not been cheated on (M= 3.10, SD= 2.03), F(1, 142) = 22.28, p <.001, η2=.14, but they did not differ in identification with the cheating protagonist (cheated: M= 2.64, SD= 1.71; not cheated: M= 2.72, SD= 1.66), p=.78. These results support the validity of the self-report measure of cheating experiences. Further, the two versions of the stimuli evoked different perceptions of the protagonist’s emotional state. The young woman was perceived as being more regretful in the self-blame version (M= 4.33, SD= 1.89) compared to the rationalization version (M= 2.13, SD= 1.32), F(1, 142) = 65.97, p <.001, η2=.32, suggesting an effective manipulation. Regret and selective exposure Recall that H1a suggested that regretful cheaters would be more interested in watching cheating-related TV storylines than unregretful cheaters. Conversely, H1b suggested that those who have not cheated, but anticipate regretting it if they did, would be less interested in watching cheating-related storylines than those who do not anticipate regretting cheating. The hypotheses were tested with regression analysis, with age and gender entered in Block 1; past cheating behavior and regret (or anticipated regret) for having cheated entered in Block 2; and the interaction between the latter two variables entered in Block 3.4 Interest in viewing the cheating-related storylines served as the dependent variable. Variance explained is noted by change (Δ) in R2. We expected, and found, a significant interaction between cheating experience and regret (β=.33, p=.002, ΔR2=.06). As Table 1 shows, for past cheaters, those high in regret expressed greater interest in cheating-related storylines (M= 6.44) than those low in regret (M= 5.68), whereas for noncheaters, those high in anticipated regret expressed less interest in cheating-related storylines (M= 5.29) compared to those low in anticipated regret (M= 6.17). The main effect for regret (β=−.19, p=.06) and cheating behavior (β=.12, p=.11) did not surpass the conventional threshold of statistical significance, p <.05. Of note, the remaining storylines were combined to serve as the dependent variable in a regression with the same architecture as described above. The main effect and interaction terms were not significant (p s =.08–.84). In sum, H1a and H1b were supported. Regret and program enjoyment H2a suggested that regretful cheaters would enjoy watching a cheating-related TV storyline more than unregretful cheaters. Conversely, H2b suggested that for noncheaters, those high in anticipated regret were they to cheat would enjoy watching a cheating-related storyline less than those low in anticipated regret. To test these hypotheses, a regression was performed in which age, gender, and past experience being cheated on were entered in Block 1; frequency of Real World viewing, having seen the storyline before, and perceived reality of Real World were entered in Block 2; similarity to and liking of the protagonist were entered in Block 3; experimental condition, regret level, and past cheating behavior were entered in Block 4; the interactions between regret and condition as well as regret and cheating were entered in Block 5; and the three-way interaction was entered in Block 6. Given that the three-way interaction explained no variance in enjoyment, it was dropped from the analyses in the interest of parsimony. We expected, and found, a significant interaction between cheating experience and regret (β=.18, p=.05, ΔR2=.02), though the results were not fully in line with predictions. As shown in Table 2, there was no difference in enjoyment between past cheaters high or low in regret (M= 3.72 and M= 3.89, respectively). However, consistent with H2b, noncheaters high in anticipated regret enjoyed the program less (M= 3.45) than their less regretful counterparts (M= 4.01).5 Of note, no significant main effects for experimental condition, regret, or past cheating behavior were revealed (p s =.25–.45). However, there was a significant interaction between regret and story condition (β=.22, p=.02, ΔR2=.02), indicating that enjoyment level depended not only on level of regret/anticipated regret but on character portrayal as well. We explore this further in the context of H3 below. Regretted behavior, character portrayal, enjoyment, and coping H3a and H3b focused on those with a history of cheating, suggesting that regretful cheaters would be more likely to enjoy watching a media character rationalize rather than blame herself (i.e., express regret) for cheating. However, cheaters with little lingering regret would enjoy the two portrayals equally. To test these hypotheses, we examined the interaction between regret over past cheating and character portrayal for cheaters only (n= 76). The architecture of the regression was identical to that mentioned above; however, to conserve power, nonsignificant demographics (age, gender, and been cheated on) were removed. As expected, the regression revealed a significant interaction between character portrayal and regret level (β=.27, p=.04, ΔR2=.03). Consistent with H3a and as shown in Table 3, regretful cheaters enjoyed the rationalization portrayal (M= 4.25) more than the self-blame portrayal (M= 3.51), whereas cheaters with little lingering regret enjoyed the two portrayals equally (rationalization: M= 3.70; self-blame: M= 3.64). The main effect terms for past regret and character portrayal were not significant (p s =.53–.71). Enjoying the rationalization portrayal, however, does not necessarily translate into resolution of regret. To assess H4, that regretful cheaters will experience less regret after viewing the rationalization versus the self-blame portrayal, we first performed a paired t test between pre- and postviewing regret for those who had cheated in the past. Results suggested that the feelings of regret had significantly declined from previewing (M= 5.63, SD= 1.70) to postviewing (M= 3.98, SD= 1.87), t(75) = 7.09, p <.001. However, a regression with initial regret in Block 1 and experimental condition in Block 2 indicated no effect of character portrayal on postviewing regret (β=−.06, p=.57). In other words, both portrayals had equivalent effects on regret reduction. Thus, these data do not support H4. Study 2 H1a and H3a were further explored as part of a larger study in which 206 adults from the community (average age of 43 years, 60% women, 67% Caucasian) were asked to list their three greatest regrets and then, for the first regret listed, to imagine a TV program in which the lead character had (a) made the same choice or (b) made a different choice as they had. For each, respondents were asked to indicate how much they would want to watch such a program on a scale of 0 (not at all) to 10 (very much). An 18-item measure of trait regret was included later in the survey to tap into people’s tendency to feel regret. Of these, exploratory factor analysis identified 10 items that loaded onto a single factor (α=.87). Sample items, assessed on 5-point Likert scales, included “If I had the chance, I would go back and make different decisions in my life,”“I am often bothered by mistakes in my past,” and “I’m happy with the decisions I’ve made in my life” (reverse coded). Controlling for age, education level, gender, race, and proportion of TV viewing that included narrative-based entertainment programming (situation comedies, dramas, and soap operas), partial correlations indicated that respondents higher in trait regret were interested in watching a program in which the character made their same choice, r(184) =.20, p=.007, as well as a program in which the character made a different choice, r(184) =.34, p <.001. We then created a single viewing preference measure by subtracting the similar choice from the different choice score. A positive score on this measure indicates preference for watching a character make a different choice than the respondent had made, and a negative score indicates preference for watching a character make the same choice as the respondent. Correlating this measure with trait regret will indicate whether or not those high in trait regret prefer watching a character make a similar or different choice as they had made in their own lives. Results of the partial correlation (using the same control variables noted above) indicated that those higher in trait regret were significantly more likely to prefer to watch a program in which a character made a different choice than they did, r(184) =.15, p=.04. These data are consistent with the Study 1 findings that not only are those who are also regretful interested in watching TV content about their regretted experience but they are particularly interested in seeing that experience portrayed differently than it had unfolded in their own lives. Discussion Through this research, we sought to assess how regretted personal experiences impact interest in and enjoyment of experience-related entertainment TV programming as well as the effect of viewing such programming on persistent feelings of regret. Results suggested that, consistent with regret theory, regret for past mistakes motivated interest in, rather than avoidance of, experience-related storylines, though this interest did not necessarily translate into greater enjoyment of the programming, generally speaking. Conversely, and consistent with mood management theory, anticipated regret of a mistake yet to be made appeared to motivate both programming avoidance and reduced viewing enjoyment. For regretful cheaters, program viewing appeared to reduce regret (at least temporarily) regardless of how the protagonist reacted to her mistake, though they seemed to enjoy viewing her rationalization over her self-blame. Further, the findings linking regret over past behavior to program interest and preference for media portrayals that offer a different perspective than experienced in life were reinforced by the survey results. As the hypotheses regarding regret were derived from regret theory, though with the counterhedonistic exceptions to mood management theory in mind, this research highlights the value of integrating theory on specific emotions with media theory, like mood management, to arrive at more precise predictions of how emotional states might impact media selection and impact counterhedonistic choices in particular. Although the results were often supportive of predictions, there were two outcomes in this research that deserve closer inspection. First, though overall, the regretful and nonregretful cheaters appeared to enjoy the programming equally, closer analysis revealed that the regretful cheaters in fact enjoyed the rationalization portrayal more than the self-blame portrayal (see Table 3). This finding suggests that both content and context matters. That is, enjoyment is not simply a function of whether or not the issue is portrayed but how it is portrayed. Thus, we must consider more carefully why the two depictions had different effects on enjoyment for regretful cheaters. On the one hand, as we argued earlier, those experiencing regret who see a TV character react to the regretted situation differently than the viewer had in real life may serve as a model for an alternative means of coping. This realization or insight may fulfill the information needs of the viewer, resulting in enjoyment. Alternatively, it may be that watching another’s regretful display compounds one’s own negative affect, thus reducing enjoyment. Were this the case, however, we would expect the less regretful cheaters to respond similarly; yet, they evidenced no difference in enjoyment between the two depictions. Given the heightened enjoyment of the rationalization portrayal by the regretful cheaters, we believe the information-needs explanation is most fitting, though a direct test of this assumption awaits future research. Second, if we grant that the regretful cheaters’ enhanced enjoyment of the rationalization portrayal indeed signaled greater coping, we would have expected that portrayal to reduce regret more so than the self-blame portrayal. Yet, in contrast to H4, no such difference was found. Thus, we might conclude that both portrayals served to aid coping but through different routes, one being more enjoyable than the other. More specifically, perhaps through watching someone else’s regret reaction, regretful cheaters may have realized they were not alone in their transgression and sympathizing with the character’s plight, reduced their self-blame, thus minimizing the associated regret. However, viewing a representation of one’s own transgression, coupled with the affective display of the character and no concrete portrayal of resolution, may not have been a particularly enjoyable experience. Conversely, in watching their plight reenacted with a different solution—no tears and no break- up—regretful cheaters might not only have come to see their behavior as normal and perhaps even understandable but also as something to be considered without self-flagellation. Thus, not only is regret reduced but also enjoyment enhanced. Understanding the underlying cause not only of these differences in enjoyment but also in the mechanisms through which regret is reduced (i.e., reduced self-blame and alteration of the value for the enacted or alternative behaviors) is an important direction for future exploration. Of particular note, though we were able to show that experience-relevant programming reduced levels of regret, we have no way of knowing if this is a temporary or a more long-lived effect. Like an analgesic salve, watching the program may have led to temporary relief of the negative feelings associated with regret, but in all likelihood, without acquisition of unique insight into one’s condition, it is likely the regret will recur. Although we were unable to follow up with our respondents to determine longevity of effects, it is possible that watching the rationalization portrayal, though it had an equal effect on regret in the short run, might have had a more lasting impact, particularly given its enhanced effect on enjoyment. Future research would do well to explore this issue. Despite the insight this research offers into how emotional states might impact selection and reaction to TV programming, this research has several limitations that require discussion. The first relates to study design in that the questions about personal history of, and feelings about, being unfaithful to a romantic partner were asked prior to any of the dependent measures, which raises concerns about demand characteristics. That is, perhaps the participants not only realized the study was related to cheating behavior but also consequently responded in ways desired by the researchers. In response to such concerns, we point out first that it was necessary for us to ask cheating-related questions at the beginning of the study in order to activate feelings of regret, to the extent they still lingered, to examine their effects, in context, on program interest and reactions. Were we to simply rely on those naturally in a state of regret (e.g., those who had cheated only very recently), we would likely have been hard pressed to achieve a sample size suitable for hypothesis testing. More to the point, because our hypotheses are somewhat counterintuitive, even if the respondents knew we were interested in the effect of their past cheating behavior on program selection and viewing, it is unlikely they could have successfully and systematically guessed in what direction we expected the effect to occur. Thus, we believe it is unlikely that the question order biased the findings in favor of our hypotheses. Further, we point out that in Study 2, the program interest items preceded the trait regret items, and the results from those analyses were consistent with those from Study 1. Thus, demand characteristics or question order does not appear to pose a significant threat to the validity of these findings. A second major concern with this research involves the proportion of participants who reported having seen the program storyline previously. If they had indeed followed this storyline before, one might say, they may have responded based not only on what they saw during the study but also on their previous experience with the programming, thus muddying the results. However, we believe the reader can confidently set this worry aside for several reasons. First, the manipulation check clearly showed that participants saw the versions as distinct, regardless of their previous experience. Second, we reran the analyses with only those who had not seen the program before and found results of similar magnitude to those reported with the complete sample (see Footnote 5). If anything, the results suggest that our choosing a popular stimulus program may have created a more conservative test of our hypotheses as we may have lost the advantage of capturing the initial impact of program viewing. Finally, we would like to note that in choosing a program popular with our sample audience, we are more confident in our conclusions about how young people might cope with regretted experience through TV viewing than if we had selected a program less likely to be watched in “the real world.” In sum, these two studies, with different samples, regretted experiences, and assessments of regret, converges to suggest that lingering feelings of regret can motivate consumption of programming relevant to one’s life experience and that such consumption may have a positive impact on coping and, thus, feelings of regret. This research builds on work by Oliver (1993) and others who explore the effects of counterhedonistic media choices (see Oliver, 2003) and simultaneously answers Zillmann’s (2000) call to explore the exceptional circumstances under which informational utility may supersede diversionary goals when selecting entertainment programming. By drawing from regret theory, we were able to arrive at predictions that could not have been made based on mood management theory alone. Thus, future developments in mood management theory specifically, and uses and gratifications generally, would likely benefit from integrating the information-seeking motivations associated with specific emotional states to allow for more predictive models of media selection, content preference, and effects. Understanding the boundaries and persistence of these effects, for regret as well as other intractable emotions, will not only serve to advance media theory, but, as important, will also offer much-needed insight into the potentially beneficial effects of TV portrayals of dramatic and emotional life events. Notes 1 " There is a notable amount of research on coping with fear generated by TV viewing (e.g., Valkenburg, Cantor, & Peeters, 2000), though this approach does not discuss the use of TV as a coping device. 2 " We say this with the understanding that though the counterhedonistic exceptions to mood management theory are acknowledged by Zillmann (2000), the theory is not currently equipped to make counterhedonistic predictions. 3 " Hugging and spending a lot of time with another (opposite sex) person were not considered to be unfaithful acts but holding hands, emotional intimacy, kissing, and physical intimacy were. Further, all but hugging were considered potentially damaging to a relationship. No significant differences between men and women were identified, though women were slightly more likely than men to indicate emotional intimacy as cheating and women saw emotional intimacy as more damaging to a relationship than men. 4 " The regret variable was centered before creating the interaction term with past cheating behavior. In addition, the effect of whether or not one had been cheated on was explored, but as it had no meaningful impact on viewing interest (nor did it significantly interact with negative emotions about that experience), it was omitted from the reported analyses in the interest of parsimony. 5 " Of note, analyses were repeated, separating those who had and those who had not previously seen the Real World storyline that served as the basis for the stimuli. Both groups evidenced comparable effects for regret (seen: β=−.15; not seen: β=−.24), the interaction between regret and past cheating behavior (seen: β=.23; not seen: β=.21), and the interaction between regret and condition (seen: β=.17; not seen: β=.40). The stronger effect for those who had not previously seen the storyline suggests initial viewing may have a more potent effect, though this is clearly speculative given the small sample size on which this observation is based. References Boninger , D. S. , Gleicher , F., & Strathman , A. ( 1994 ). Counterfactual thinking: From what might have been to what may be . Journal of Personality and Social Psychology , 67 , 297 – 307 . Google Scholar Crossref Search ADS WorldCat Connolly , R. , & Zeelenberg , M. ( 2002 ). Regret in decision making . Current Directions in Psychological Science , 11 , 212 – 216 . Google Scholar Crossref Search ADS WorldCat Finn , S. , & Gorr , M. B. ( 1988 ). Social isolation and social support as correlates of television viewing motivation . Communication Research , 15 , 135 – 158 . Google Scholar Crossref Search ADS WorldCat Gilovich , T. , & Medvec , V. H. ( 1994 ). The temporal pattern to the experience of regret . Journal of Personality and Social Psychology , 67 , 357 – 365 . Google Scholar Crossref Search ADS PubMed WorldCat Gilovich , T. , Medvec , V. H ., & Kahneman , D. ( 1998 ). Varieties of regret: A debate and partial resolution . Psychological Review , 105 , 602 – 605 . Google Scholar Crossref Search ADS WorldCat Herzog , H . ( 1944 ). What do we really know about daytime serial listeners? In P. F . Lazarsfeld & F. N . Stanton (Eds.), Radio research 1942–1943 (pp. 3 – 33 ). New York : Duell, Sloan, & Pearce . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Katz , E. , Blumler , J. G., & Gurevitch , M. ( 1974 ). Utilization of mass communication by the individual. In J. G . Blumler & E . Katz (Eds.), The uses of mass communications: Current perspectives on gratifications research (pp. 19 – 32 ). Beverly Hills, CA : Sage . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Knobloch , S . ( 2003 ). Mood adjustment via mass communication . Journal of Communication , 53 , 233 – 250 . Google Scholar Crossref Search ADS WorldCat Kubey , R ., & Peluso , T. ( 1990 ). Emotional response as a cause of interpersonal news diffusion: The case of the space shuttle tragedy . Journal of Broadcasting & Electronic Media , 34 , 69 – 76 . Google Scholar Crossref Search ADS WorldCat Landman , J . ( 1987 ). Regret: A theoretical and conceptual analysis . Journal for the Theory of Social Behaviour , 17 , 135 – 160 . Google Scholar Crossref Search ADS WorldCat Landman , J . ( 1993 ). Regret: The persistence of the possible . New York : Oxford . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Lazarus , R. S . ( 1991 ). Emotion and adaptation . New York : Oxford . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC McCroskey , J. C. , Richmond , V. P., & Daly , J. A. ( 1975 ). The development of a measure of perceived homophily in interpersonal communication . Human Communication Research , 1 , 323 – 332 . Google Scholar Crossref Search ADS WorldCat McQuail , D. , Blumler , J. G., & Brown , J. R. ( 1972 ). The television audience: A revised perspective. In D. McQuail (Ed.), Sociology of mass communication (pp. 135 – 165 ). Middlesex, U.K.: Penguin . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Nabi , R. L . ( 1999 ). A cognitive-functional model for the effects of discrete negative emotions on information processing, attitude change, and recall . Communication Theory , 9 , 292 – 320 . Google Scholar Crossref Search ADS WorldCat Nabi , R. L . ( 2003 ). The framing effects of emotion: Can discrete emotions influence information recall and policy preference? Communication Research , 30 , 224 – 247 . Google Scholar Crossref Search ADS WorldCat Oliver , M. B . ( 1993 ). Exploring the paradox of the enjoyment of sad films . Human Communication Research , 19 , 315 – 342 . Google Scholar Crossref Search ADS WorldCat Oliver , M. B . ( 2003 ). Mood management and selective exposure. In J. Bryant, D. Roskos-Ewoldsen, & J. Cantor (Eds.), Communication and emotion: Essays in honor of Dolf Zillmann (pp. 85 – 106 ). Mahwah, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Palmgreen , P. , Wenner , L. A., & Rosengren , K. E. ( 1985 ). Uses and gratifications research: The past ten years. In K. E . Rosengren, L. A . Wenner, & P. Palmgreen (Eds.), Media gratifications research: Current perspectives (pp. 11 – 37 ). Beverly Hills, CA : Sage . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Rubin , A. M . ( 1981 ). An examination of television viewing motivations . Communication Research , 8 , 141 – 165 . Google Scholar Crossref Search ADS WorldCat Rubin , A. M . ( 2002 ). The uses-and-gratifications perspective of media effects. In J. Bryant & D. Zillmann (Eds.), Media effects: Advances in theory and research (pp. 525 – 548 ). Hillsdale, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Rubin , A. M. , & Rubin , R. B. ( 1985 ). Interface of personal and mediated communication: A research agenda . Critical Studies in Mass Communication , 2 , 36 – 53 . Google Scholar Crossref Search ADS WorldCat Savitsky , K. , Medvec , V. H., & Gilovich , T. ( 1997 ). Remembering and regretting: The Zeigarnik effect and the cognitive availability of regrettable actions and inactions . Personality and Social Psychology Bulletin , 23 , 248 – 257 . Google Scholar Crossref Search ADS WorldCat Stewart , A. J. , & Vandewater , E. A. ( 1999 ). “If I had it to do over again …”: Midlife review, midcourse corrections, and women’s well-being in midlife . Journal of Personality and Social Psychology , 76 , 270 – 283 . Google Scholar Crossref Search ADS PubMed WorldCat Sugden , R . ( 1985 ). Regret, recrimination, and rationality . Theory and Decision , 19 , 77 – 99 . Google Scholar Crossref Search ADS WorldCat Valkenburg , P. M. , Cantor , J., & Peeters , A. L. ( 2000 ). Fright reactions to television . Communication Research , 27 , 82 – 100 . Google Scholar Crossref Search ADS WorldCat Zeelenberg , M . ( 1999 ). The use of crying over spilled milk: A note on the rationality and functionality of regret . Philosophical Psychology , 12 , 325 – 340 . Google Scholar Crossref Search ADS WorldCat Zeelenberg , M. , Van Den Bos , K., Van Dijk , E., & Pieters , R. ( 2002 ). The inaction effect in the psychology of regret . Journal of Personality and Social Psychology , 82 , 314 – 327 . Google Scholar Crossref Search ADS PubMed WorldCat Zillmann , D . ( 1988 ). Mood management: Using entertainment to full advantage. In L. Donohew & H. E . Sypher (Eds.), Communication, social cognition, and affect (pp. 147 – 171 ). Hillsdale, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Zillmann , D . ( 2000 ). Mood management in the context of selective exposure theory. In M. Roloff (Ed.), Communication yearbook , 23 (pp. 103 – 123 ). Thousand Oaks, CA : Sage . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC © 2006 International Communication Association http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Communication Oxford University Press

Does Misery Love Company? Exploring the Therapeutic Effects of TV Viewing on Regretted Experiences

Download PDF
18 pages

Loading next page...
 
/lp/oxford-university-press/does-misery-love-company-exploring-the-therapeutic-effects-of-tv-E4FWYCwBqM

References (41)

Publisher
Oxford University Press
Copyright
© 2006 International Communication Association
ISSN
0021-9916
eISSN
1460-2466
DOI
10.1111/j.1460-2466.2006.00315.x
Publisher site
See Article on Publisher Site

Abstract

Abstract Testing hypotheses derived from regret and mood management theories, this research explores how regretted experiences impact interest in viewing experience-relevant TV programming and such viewing’s effects on program enjoyment and felt regret. One hundred and forty-four participants, half of whom had been unfaithful in romantic relationships, were asked first to rate their interest in viewing a series of storylines and then to provide their reactions to 1 of 2 versions of a TV program depicting cheating behavior. Largely consistent with hypotheses, results indicated that those who had both cheated and felt regret about their behavior were more likely than others to want to watch experience-related storylines, were no less likely to enjoy watching such programming, and particularly preferred viewing the program version in which the main character rationalized, rather than expressed regret for, her behavior. Both program versions, however, reduced regret equally. A survey of 206 city residents also offered evidence consistent with predictions based on regret theory. Overall, this research speaks to the value of integrating theories of emotion with media theory to enhance the latter’s predictive ability. It is widely accepted that the mass media can be used to meet people’s social and psychological needs, emotional ones in particular (e.g., Rubin, 2002; Zillmann, 1988). However, apart from the research on coping with media-induced fear (Kubey & Peluso, 1990), the literature offers very little insight into how viewers might use media like TV to cope with (vs. escape from) emotions associated with distressing personal situations. Yet, given the multitude of storylines in entertainment programming that deal with such issues, including health threats, violence, and relationship turmoil, it is reasonable to imagine that TV viewing may serve not simply as a potential source of painful reminders of life’s trials and tribulations but also as a potential resource for coping and comfort to help emotional wounds begin to heal. Among some of the most painful and intractable emotional experiences potentially triggered by TV viewing are those related to regret. Marked by active rumination and motivation to understand why certain past events unfolded to lamentable effect (Landman, 1993), regret over past actions (or inactions)—comparing what is to what might have been—can persist or recur for years after the regretted event itself has passed. Although theoretical perspectives, like uses and gratifications or mood management, suggest emotional needs may be met through media exposure, such frameworks, without benefit of understanding the specific needs associated with unique emotional states like regret, may be inadequate to predict how such emotions might impact media consumption and subsequent effects. However, by integrating regret theory with theory of media selection, we may arrive at more precise predictions of how those suffering might interact with the media to alleviate their personal distress and, perhaps in turn, enhance the quality of their lives and social interactions. The purpose of this research, then, is first to explore how regretted experiences influence the desire to view experience-related entertainment TV programming and second to consider how viewing such programming impacts one’s subsequent feelings of regret. In doing so, we hope not only to extend the literature on media and emotion by exploring the role of TV as “therapist” in helping viewers cope with personal problems but also to elaborate current media theory that has yet to fully benefit from the insights available from the theory and research on specific emotional experiences. Media theory and emotional coping Uses and gratifications Extant media theory has a history of recognizing the important role media consumption plays in satisfying emotional needs and adjusting affective states. The uses and gratifications paradigm in particular is frequently referenced as offering a framework through which media selection and use based on these and other needs can be understood. In essence, the uses and gratifications paradigm asserts that individuals are aware of their social, psychological, and biological needs. Based on this awareness, they evaluate various media channels and content, assess functional alternatives, and select the media or interpersonal channel that they believe will best meet their various needs (e.g., Katz, Blumler, & Gurevitch, 1974; Palmgreen, Wenner, & Rosengren, 1985; Rubin, 2002; Rubin & Rubin, 1985). Such needs include those related to diversion (e.g., escapism and arousal), personal relationships (e.g., social utility), personal identity (e.g., reality exploration), and surveillance (e.g., news gathering; McQuail, Blumler, & Brown, 1972; see review by Rubin, 2002). Of particular interest to this investigation are needs associated with “emotional release,” which, though identified early on as a primary function of media consumption (Herzog, 1944), has focused primarily on issues of diversion, entertainment, and arousal (see Finn & Gorr, 1988). Indeed, Finn and Gorr subsume these three motivations (along with information seeking) into a more general category of “mood management,” borrowing the term from Zillmann’s (1988) seminal work examining the selection of media to regulate levels of arousal. Indeed, mood management is the gratification most directly related to this research. Mood management theory In essence, Zillmann’s (1988) mood management theory asserts that people use media to modulate their affective states. More specifically, Zillmann argues that people, driven by hedonistic desires, strive to alter negative moods as well as maintain and prolong positive ones. Consequently, they will arrange their environments to adjust “all conceivable moods” (p. 328), using any type or genre of communication, as well as all specific forms within such genres, available. Moreover, Zillmann, in assuming that mood management processes are based on operant conditioning, argues that people may, but need not, be cognizant of the reasons for their choices. Zillmann goes on to note four message features that might impact message selection based on mood: excitatory potential, absorption potential, semantic affinity, and hedonic valence. For each, the underlying principle is the same. If a message reflects one (or presumably more) feature that might perpetuate the negative state, for example, high absorption potential or mood-related content, the message is likely to be avoided in favor of one that would interrupt the negative state, like a message of an opposite valence or on a topic unrelated to the mood source. Much research supports the predictions of mood management theory (see Oliver, 2003, for a recent review). However, its boundaries have been challenged by paradoxes of media selection, like the enjoyment of horror movies or tearjerkers that are designed to evoke negative effect seemingly in contradiction to mood management theory’s assumption of hedonistic motivation. Zillmann (2000) attempted to subsume such outcomes within the mood management paradigm, arguing that sometimes we postpone immediately hedonistic outcomes in pursuit of other goals that will bring about positive affective states once achieved (i.e., telic hedonism). Yet, it is debatable whether such investigations justifiably fall under the umbrella of mood management theory as they do not employ the key assumption that hedonism is the primary motivation underlying choice, and as Knobloch (2003) notes, nearly all “goal-directed behavior can be construed as maximizing delight or minimizing discomfort at some point along the line” (p. 235). Further, even if we accept the expanded boundaries of mood management theory to include delayed hedonistic gratification, this framework loses its predictive power regarding media selection when particular emotional states are aroused. This becomes evident as we consider Zillmann’s discussion of hedonistic versus information-seeking motives and their implications for media choice. In general, Zillmann suggests that hedonistic motives will guide selection of entertaining media, whereas information needs will guide selection of informative/educational messages. Only under exceptional conditions does he allow that informational needs may temporarily supersede hedonistic ones when selecting entertainment programming. Though he does not specify what those conditions might be, it appears that he does not see emotional states as falling into this category as he argues that extreme emotional states, like anger, would lead to “state-focused cognitive preoccupation” that interrupts hedonistic goals in place of “instrumental coping efforts” (p. 109). Such coping, he asserts, leads one to avoid diversionary media messages altogether. Though he does not say so explicitly, we might infer that Zillmann would suggest that informational sources might be sought to aid efforts to cope with emotional states. Indeed, this position would be supported by studies that demonstrate that TV news is sought to gain information during or after national crises, like the space shuttle disaster (Kubey & Peluso, 1990) and September 11.1 Thus, based on Zillmann’s (1988, 2000) writings, it seems that mood management theory would suggest that an emotion like regret might lead to (a) avoidance of emotion-relevant entertainment programming (due to semantic affinity); (b) avoidance of entertainment programming altogether (due to emotion-focused coping); (c) watching informational/educational programming to aid coping, or, if all else fails; (d) exposure to entertainment programming driven by informational needs. Surely, one of these predictions is bound to be accurate; yet, mood management theory is unable to tell us with reasonable certainty which one is most likely. If we draw from emotion theory, however, we are far better positioned to arrive at precise predictions. Like Zillmann (2000), emotion theorists—Lazarus (1991) in particular—argue that negative emotional experiences are associated with coping needs and efforts. Additional evidence suggests that these emotion-driven coping efforts are likely associated with unique information-seeking goals (Nabi, 1999, 2003). If the emotional states stem from personal and goal-relevant situations, which most emotions likely do (Lazarus, 1991), it stands to reason that entertainment programs that depict stories of life’s trials and tribulations will not serve as sources of diversion, as Zillmann implies, but rather as potential sources of information for those seeking to cope with their distress. In essence, we suggest that even if programming is generally intended to entertain, particular emotional states might draw, not repel, viewers to that programming if they have some expectation that the storylines might offer information useful to their coping needs as generated by their particular emotional states. If the need to regulate these emotional states results in information seeking from entertainment programming, not only do we have a theoretical motivation for proposing the existence of an effect, which Zillmann considered unusual, but there is also an established body of literature from which to draw to predict when this would happen and to what possible result. As our investigation focuses on the emotion of regret, we now turn to this literature. Regret theory Regret, at its foundation, is a cognitively driven affective phenomenon based in having misgivings about decisions or actions that cannot be undone (Landman, 1987). As Sugden (1985) defines it, regret is “the painful sensation of recognising that ‘what is’ compares unfavourably with ‘what might have been.’” Regret can be evoked by past events or decisions, both small (e.g., choosing the wrong line in the grocery story) and large (e.g., cheating on a spouse). Theories of regret arousal (developed primarily from economic models of decision making; see Landman, 1993, for a review) generally suggest it is a function of the value of a chosen option compared with the values of foregone options combined with self-recrimination (e.g., Connolly & Zeelenberg, 2002; Sugden, 1985). That is, one might experience regret not only if the outcome is not what one had hoped for, especially when compared to an alternative course, but also if one feels responsible for the outcome, even if there was nothing one could have done differently. Thus, counterfactual thinking and self-blame are central features marking the experience of regret. At its extreme, regret is associated with cognitive preoccupation that mires one in the past, fruitless and irrational, preventing forward progress (Landman, 1993). However, its benefits include learning from past mistakes, initiating life changes, and making more thoughtful decisions in the hopes of avoiding future regretted experiences (Landman, 1993; Stewart & Vandewater, 1999; Zeelenberg, 1999). The past 10 years have seen a proliferation of studies examining the features of regretted experiences, like the types of decisions people tend to regret (e.g., relationship choices, career or financial decisions, and educational choices; see Gilovich & Medvec, 1994) and the role of action versus inaction (e.g., Gilovich, Medvec, & Kahneman, 1998; Savitsky, Medvec, & Gilovich, 1997; Zeelenberg, van den Bos, van Dijk, & Pieters, 2002). Yet, there is little research on how regret is actually resolved. Boninger, Gleicher, and Strathman (1994) suggest that counterfactual thinking itself serves as a coping technique for regret in that reflection allows for future preparation, and this preparation then serves to distract from the currently felt negative affect. Thus, reflection may not only lead to more functional outcomes in the future but also alleviate negative feelings, and perhaps generate positive ones, in the present. The question most relevant to this research then becomes, how might media content aid counterfactual thinking? Because the research on regret suggests that it is largely an intrapersonal phenomenon, there is no reference in the literature to the role the media might play in either arousing or alleviating feelings of regret. However, to the extent that a regretted experience is portrayed in entertainment TV programming, one could imagine that watching such programming might not simply remind one of the regretted act but those representations might offer insight into that experience that could, in turn, help minimize feelings of regret. So, whereas mood management theory proper suggests that seeing a regretted experience portrayed on TV would exacerbate negative affect and thus lead to avoidance, and the counterhedonistic extension of mood management theory suggests that either avoidance or consumption might occur (but is not clear on which), the regret literature suggests that people experiencing regret would likely be drawn to experience-related entertainment programming for its potential information value and the insight gained from the TV portrayal might, in turn, help to alleviate their regret. Hypotheses Based on the above literature review, mood management theory and regret theory offer somewhat competing predictions about how people with regretted experiences will react to entertainment programming on the topic of their regret (i.e., with semantic affinity).2 Whereas mood management theory suggests people will avoid such programming as it will exacerbate their negative affect, regret theory predicts that the opportunity to ruminate on the regretted experience to hopefully come to terms with their past mistake (i.e., to aid in counterfactual thinking) will be welcome. Thus, those experiencing regret will, in the interest of coping, likely be drawn to programming that reflects the source of their affective disposition. Conversely, for those who have not had the experience to regret, it is reasonable to assume that regular mood management processes will be engaged. That is, viewing preference will be predicted by expectation of the programming’s impact on mood. Those who might associate a behavior with negative affect (e.g., would anticipate regretting an experience, were they to have it) will be less likely to seek out such programming compared to those who have neither had the experience nor associate it with negative affect. As our research focuses on cheating on a romantic partner as the potentially regretted behavior—a common regret among people both young and old—we propose the following interaction hypothesis, based on the logic outlined above: H1: Regret will interact with past cheating behavior such that (a) for those who have cheated in the past, those high in regret will be more interested in viewing experience-related TV storylines than those low in regret, whereas (b) for those who have not cheated, those high in anticipated regret were they to cheat will be less interested in viewing experience-related TV storylines than those low in anticipated regret. If exposed to programming reflecting a regretted experience, like an act of infidelity, regret theory offers insight into how viewers will react. On the one hand, reliving regretted experiences, like reopening a wound, is not pleasant. Yet, as Boninger et al. (1994) suggest, the positive affect associated with the potential for learning effective means of resolving regret may make such programming enjoyable from an informational, though perhaps not an entertaining, standpoint. However, for those without the regretted experience, assuming hedonistic motives are paramount, we predict that those who associate the experience with negative affect (e.g., would anticipate regretting the experience, were they to have it) will enjoy watching such programming less than those who do not associate it with regret. Thus, H2: Regret will interact with past cheating behavior such that (a) for those who have cheated in the past, those high in regret will enjoy watching an experience-related TV storyline more than those low in regret, whereas (b) for those who have not cheated, those high in anticipated regret were they to cheat will enjoy watching an experience-related TV storyline less than those low in anticipated regret. Finally, regret theory suggests that alternative perspectives on one’s past mistakes (e.g., increasing the value of one’s own actions, decreasing the value of the alternative paths, or minimizing self-blame) should help one cope with regret. Thus, the effect of program viewing on both enjoyment and regret over the past behavior should be affected by how the story protagonist reacts to the regretted situation. If the protagonist expresses self-blame and regret, that is, does not offer a model for resolution, viewers who regret their past behavior will be unlikely to receive new insight to aid in their coping. However, if the protagonist does not express self-blame or regret, this portrayal may serve as a model for coping by reducing the negative value of the path chosen and/or reducing self-blame. Consequently, a regretful person may not only enjoy this programming more but will also feel better about his or her own situation. In contrast, those who do not feel regret about their past behavior, and thus have no coping goals, should have no reason to prefer one portrayal over the other. Thus, focusing only on those who have cheated on a relational partner in the past, we suggest that: H3: Regret will interact with character portrayal such that (a) cheaters with lingering regret over their past behavior will enjoy viewing a media character who rationalizes cheating more than a media character who self-blames, whereas (b) cheaters without lingering regret over their past behavior will enjoy both portrayals equally. H4: For those with lingering regret over their past cheating behavior, viewing a media character who rationalizes cheating will reduce cheating-related regret more than viewing a media character who self-blames. Study 1
Method Participants and procedures One hundred and forty-four participants, 54% women, were recruited to participate in a study on TV and relational choices. Of these, the majority were upperclassmen (36% juniors and 44% seniors). Ninety percent had been in a committed romantic relationship in the past, and 36% were currently in such a relationship. Fifty-three percent of respondents reported having cheated on a romantic partner, equally divided among men and women. In groups of 3–12, participants completed a survey in which they were asked about their TV viewing habits, behaviors they believed constituted unfaithfulness, and how they felt (or would feel) if they or their partner had been unfaithful. After several filler scales tapping personality constructs, respondents were given a list of nine hypothetical storylines from a dramatic TV series to rate in terms of how interested they were in watching each. Participants then watched one of two versions of a 7-minute edited segment of a popular reality-based drama in which a young woman cheats on her boyfriend and then either self-blames/expresses regret or rationalizes her behavior. Respondents then provided their emotional reactions to the segment, how enjoyable they found it, their reactions to the protagonist, and their current feelings about any incidents of unfaithfulness in their past. Stimuli To choose a topic for this research, we solicited events regretted by undergraduates (n= 37). The most frequent responses included those related to partying (e.g., drinking too much), relationships (e.g., a bad relationship and indiscretions), and school participation (e.g., not studying enough and missing classes). Of these, a topic frequently mentioned by our sample, frequently portrayed in the media, and generalizable to other age groups is cheating on a romantic partner. Our assumption was that those who cheated on a boyfriend or girlfriend were vulnerable to experiencing regret, and we further assumed that resolution of regret ranged from occurring quickly after the event to not having yet fully occurred. The stimuli for the second half of the study were drawn from the MTV Series (U.S. produced) Real World, in which several episodes across a season were edited into a 7-minute segment using Adobe Premiere Pro 1.5. We chose this particular series not only because it contained the theme we were hoping to display but also because it is a program popular with young people and thus more likely to serve as a source of coping information compared to less popular programming. The first 5 minutes of the stimuli, which were identical for both conditions, introduced the female protagonist, a young woman with a boyfriend back home but a tendency to flirt with other men when drunk. She meets a guy in a bar who she begins to spend time with, though she claims they are just friends. One evening, after drinking a bottle of champagne, she and her friend “make out.” At this point, the two versions of the stimulus diverge. In the “self-blame” condition, the protagonist has a telephone conversation with her boyfriend in which he indicates that if she cheated, he will break up with her. In her subsequent “confessional,” she tearfully questions why she makes such destructive choices in her life. In the “rationalization” condition, the protagonist has a telephone conversation with her new male friend in which she says she anticipates her boyfriend being unhappy with her actions, even though she does not think she did anything wrong. Then, in a conversation with her roommate, she indicates that it is unfair of her boyfriend to limit her behavior, and she concludes that she is “a good girlfriend.” Measures Measures are presented in the order in which they were completed. As this study reports on a subset of the data collected, variables not relevant to the present analyses are omitted. After completing TV viewing and demographic measures, respondents were asked on a scale of 1 (not at all) to 5 (definitely) how much each of six behaviors constituted being unfaithful to a romantic partner and how damaging to a relationship each would be.3 Participants were then asked to indicate how much of several emotions they felt when thinking about being unfaithful to a romantic partner—past, present, or imagined. Of these, four related to regret—regret, remorse, guilt, and shame. The four loaded onto a single factor and formed a reliable index assessing regret (α=.91). Respondents were then asked whether or not they had been unfaithful in a past or present romantic relationship and, if yes, how long ago the incident(s) occurred. Of note, there was minimal association between how long ago the cheating occurred and the degree of regret felt, r(75) =.14, p=.22. Respondents were then asked to assess how they felt when thinking about their partner (past, present, or imagined) being unfaithful to them, using the same emotion items noted above, followed by a question asking whether or not, to the best of their knowledge, a romantic partner had ever been unfaithful to them. After several personality trait measures, the participants responded to the dependent measure related to selective exposure—storyline preference. Nine storylines were presented as having appeared in a popular TV drama. Respondents were asked to rate how interested they were in watching each episode. For women, the protagonist of each storyline was female; for men, the protagonist was male. Of the nine storylines, three related to cheating (presented from the survey for the female participants)—“A young woman jeopardizes her relationship with her boyfriend when she pursues a summer crush, only to regret her decision,”“A stormy relationship comes to a head as a young woman, lonely and away from home, finds comfort with a new guy,” and “A young woman’s suspicions about her boyfriend’s ‘extracurricular activities’ turn out to be true, and she must decide whether to break off the relationship or take him back.” These three were combined to reflect “cheating-related storylines.” The remaining storylines, examples of which include the protagonist hiding an eating disorder (females)/steroid use (males), dealing with a pregnancy scare, pursuing dreams of becoming an actor, and coping with mental illness, were combined to represent “cheating-unrelated storylines.” After viewing the program segment, respondents completed a mood adjective checklist and then indicated how much they enjoyed viewing the segment using four items (enjoyable, entertaining, pleasurable, and captivating) rated on 1 (not at all) to 7 (very much) scales (α=.91) and how much they liked the protagonist using six items (friendly, likable, pleasant, trustworthy, reliable, and honest), again rated on 1 (not at all) to 7 (very much) scales (α=.85). As a manipulation check, we asked how regretful the protagonist seemed about her behavior. As a way of validating the self-reports of cheating experiences, we also asked how much the respondent identified with the situation as experienced by the protagonist as well as her boyfriend. Each of these items was measured with 1 (not at all) to 7 (very much) scales. To test H4, we asked, for those who had cheated in the past, how they currently felt about their behavior. Of the several emotion items included, the four items related to regret were of primary interest (α=.95). Finally, we included several measures to serve as possible control variables. Given the protagonist’s alternative style, we included a homophily scale (McCroskey, Richmond, & Daly, 1975) in case lack of similarity to the character influenced our results (α=.77). We measured perceivedrealism of the TV program using five items adapted from Rubin’s (1981) perceived realism scale (α=.88), and we further measured how frequently respondents had viewed Real World in the average month (0 to 5+ times) and whether they had seen the particular storyline before. Respondents indicated watching slightly more than 2.5 episodes per month (M= 2.69, SD= 1.76), and two-thirds indicated they had seen the particular storyline used for the stimulus (68%). Analysis Given the continuous nature of the regret measure, the hypotheses were tested using regression procedures. However, to more clearly present the interaction results, regret was dichotomized, analyses of covariance were performed, and the group means presented in Tables 1–3. Table 1 Interest in Viewing Cheating-Related Storylines by Past Cheating Behavior and Regret Past Behavior . Regret . Low . High . M . SD . M . SD . Cheated 5.68 (2.43), n= 37 6.44 (2.45), n= 39 Not cheated 6.17 (2.25), n= 37 5.29 (2.10), n= 31 Past Behavior . Regret . Low . High . M . SD . M . SD . Cheated 5.68 (2.43), n= 37 6.44 (2.45), n= 39 Not cheated 6.17 (2.25), n= 37 5.29 (2.10), n= 31 Note: Means are based on an ANCOVA comparable to the regression analysis reported in the text. Regret levels are based on a dichotomous split of regret. Open in new tab Table 1 Interest in Viewing Cheating-Related Storylines by Past Cheating Behavior and Regret Past Behavior . Regret . Low . High . M . SD . M . SD . Cheated 5.68 (2.43), n= 37 6.44 (2.45), n= 39 Not cheated 6.17 (2.25), n= 37 5.29 (2.10), n= 31 Past Behavior . Regret . Low . High . M . SD . M . SD . Cheated 5.68 (2.43), n= 37 6.44 (2.45), n= 39 Not cheated 6.17 (2.25), n= 37 5.29 (2.10), n= 31 Note: Means are based on an ANCOVA comparable to the regression analysis reported in the text. Regret levels are based on a dichotomous split of regret. Open in new tab Table 2 Program Enjoyment by Past Cheating Behavior and Regret Past Behavior . Regret . Low . High . M . SD . M . SD . Cheated 3.89 (1.60), n= 37 3.72 (1.90), n= 24 Not cheated 4.01 (1.62), n= 37 3.45 (1.24), n= 31 Past Behavior . Regret . Low . High . M . SD . M . SD . Cheated 3.89 (1.60), n= 37 3.72 (1.90), n= 24 Not cheated 4.01 (1.62), n= 37 3.45 (1.24), n= 31 Note: Means are based on an ANCOVA comparable to the regression analysis reported in the text. Regret levels are based on a dichotomous split of regret. Open in new tab Table 2 Program Enjoyment by Past Cheating Behavior and Regret Past Behavior . Regret . Low . High . M . SD . M . SD . Cheated 3.89 (1.60), n= 37 3.72 (1.90), n= 24 Not cheated 4.01 (1.62), n= 37 3.45 (1.24), n= 31 Past Behavior . Regret . Low . High . M . SD . M . SD . Cheated 3.89 (1.60), n= 37 3.72 (1.90), n= 24 Not cheated 4.01 (1.62), n= 37 3.45 (1.24), n= 31 Note: Means are based on an ANCOVA comparable to the regression analysis reported in the text. Regret levels are based on a dichotomous split of regret. Open in new tab Table 3 Program Enjoyment by Character Portrayal and Regret (for past cheaters only) Character Portrayal . Regret . Low . High . M . SD . M . SD . Self-blame 3.64 (1.62), n= 14 3.51 (1.91), n= 23 Rationalization 3.70 (1.71), n= 16 4.25 (1.70), n= 23 Character Portrayal . Regret . Low . High . M . SD . M . SD . Self-blame 3.64 (1.62), n= 14 3.51 (1.91), n= 23 Rationalization 3.70 (1.71), n= 16 4.25 (1.70), n= 23 Note: Means are based on an ANCOVA comparable to the regression analysis reported in the text. Regret levels are based on a dichotomous split of regret. Open in new tab Table 3 Program Enjoyment by Character Portrayal and Regret (for past cheaters only) Character Portrayal . Regret . Low . High . M . SD . M . SD . Self-blame 3.64 (1.62), n= 14 3.51 (1.91), n= 23 Rationalization 3.70 (1.71), n= 16 4.25 (1.70), n= 23 Character Portrayal . Regret . Low . High . M . SD . M . SD . Self-blame 3.64 (1.62), n= 14 3.51 (1.91), n= 23 Rationalization 3.70 (1.71), n= 16 4.25 (1.70), n= 23 Note: Means are based on an ANCOVA comparable to the regression analysis reported in the text. Regret levels are based on a dichotomous split of regret. Open in new tab Results Manipulation checks Those who reported having cheated on a relational partner identified with the protagonist’s situation (M= 3.32, SD= 1.80) more than those who had not cheated (M= 2.00, SD= 1.20), F(1, 142) = 26.07, p <.001, η2=.16. However, they did not significantly differ in their identification with her boyfriend (cheated: M= 3.92, SD= 2.19; not cheated: M= 3.38, SD= 2.19), p=.12. Conversely, those who had been cheated on identified with the boyfriend’s situation (M= 4.83, SD= 2.11) more than those who had not been cheated on (M= 3.10, SD= 2.03), F(1, 142) = 22.28, p <.001, η2=.14, but they did not differ in identification with the cheating protagonist (cheated: M= 2.64, SD= 1.71; not cheated: M= 2.72, SD= 1.66), p=.78. These results support the validity of the self-report measure of cheating experiences. Further, the two versions of the stimuli evoked different perceptions of the protagonist’s emotional state. The young woman was perceived as being more regretful in the self-blame version (M= 4.33, SD= 1.89) compared to the rationalization version (M= 2.13, SD= 1.32), F(1, 142) = 65.97, p <.001, η2=.32, suggesting an effective manipulation. Regret and selective exposure Recall that H1a suggested that regretful cheaters would be more interested in watching cheating-related TV storylines than unregretful cheaters. Conversely, H1b suggested that those who have not cheated, but anticipate regretting it if they did, would be less interested in watching cheating-related storylines than those who do not anticipate regretting cheating. The hypotheses were tested with regression analysis, with age and gender entered in Block 1; past cheating behavior and regret (or anticipated regret) for having cheated entered in Block 2; and the interaction between the latter two variables entered in Block 3.4 Interest in viewing the cheating-related storylines served as the dependent variable. Variance explained is noted by change (Δ) in R2. We expected, and found, a significant interaction between cheating experience and regret (β=.33, p=.002, ΔR2=.06). As Table 1 shows, for past cheaters, those high in regret expressed greater interest in cheating-related storylines (M= 6.44) than those low in regret (M= 5.68), whereas for noncheaters, those high in anticipated regret expressed less interest in cheating-related storylines (M= 5.29) compared to those low in anticipated regret (M= 6.17). The main effect for regret (β=−.19, p=.06) and cheating behavior (β=.12, p=.11) did not surpass the conventional threshold of statistical significance, p <.05. Of note, the remaining storylines were combined to serve as the dependent variable in a regression with the same architecture as described above. The main effect and interaction terms were not significant (p s =.08–.84). In sum, H1a and H1b were supported. Regret and program enjoyment H2a suggested that regretful cheaters would enjoy watching a cheating-related TV storyline more than unregretful cheaters. Conversely, H2b suggested that for noncheaters, those high in anticipated regret were they to cheat would enjoy watching a cheating-related storyline less than those low in anticipated regret. To test these hypotheses, a regression was performed in which age, gender, and past experience being cheated on were entered in Block 1; frequency of Real World viewing, having seen the storyline before, and perceived reality of Real World were entered in Block 2; similarity to and liking of the protagonist were entered in Block 3; experimental condition, regret level, and past cheating behavior were entered in Block 4; the interactions between regret and condition as well as regret and cheating were entered in Block 5; and the three-way interaction was entered in Block 6. Given that the three-way interaction explained no variance in enjoyment, it was dropped from the analyses in the interest of parsimony. We expected, and found, a significant interaction between cheating experience and regret (β=.18, p=.05, ΔR2=.02), though the results were not fully in line with predictions. As shown in Table 2, there was no difference in enjoyment between past cheaters high or low in regret (M= 3.72 and M= 3.89, respectively). However, consistent with H2b, noncheaters high in anticipated regret enjoyed the program less (M= 3.45) than their less regretful counterparts (M= 4.01).5 Of note, no significant main effects for experimental condition, regret, or past cheating behavior were revealed (p s =.25–.45). However, there was a significant interaction between regret and story condition (β=.22, p=.02, ΔR2=.02), indicating that enjoyment level depended not only on level of regret/anticipated regret but on character portrayal as well. We explore this further in the context of H3 below. Regretted behavior, character portrayal, enjoyment, and coping H3a and H3b focused on those with a history of cheating, suggesting that regretful cheaters would be more likely to enjoy watching a media character rationalize rather than blame herself (i.e., express regret) for cheating. However, cheaters with little lingering regret would enjoy the two portrayals equally. To test these hypotheses, we examined the interaction between regret over past cheating and character portrayal for cheaters only (n= 76). The architecture of the regression was identical to that mentioned above; however, to conserve power, nonsignificant demographics (age, gender, and been cheated on) were removed. As expected, the regression revealed a significant interaction between character portrayal and regret level (β=.27, p=.04, ΔR2=.03). Consistent with H3a and as shown in Table 3, regretful cheaters enjoyed the rationalization portrayal (M= 4.25) more than the self-blame portrayal (M= 3.51), whereas cheaters with little lingering regret enjoyed the two portrayals equally (rationalization: M= 3.70; self-blame: M= 3.64). The main effect terms for past regret and character portrayal were not significant (p s =.53–.71). Enjoying the rationalization portrayal, however, does not necessarily translate into resolution of regret. To assess H4, that regretful cheaters will experience less regret after viewing the rationalization versus the self-blame portrayal, we first performed a paired t test between pre- and postviewing regret for those who had cheated in the past. Results suggested that the feelings of regret had significantly declined from previewing (M= 5.63, SD= 1.70) to postviewing (M= 3.98, SD= 1.87), t(75) = 7.09, p <.001. However, a regression with initial regret in Block 1 and experimental condition in Block 2 indicated no effect of character portrayal on postviewing regret (β=−.06, p=.57). In other words, both portrayals had equivalent effects on regret reduction. Thus, these data do not support H4. Study 2 H1a and H3a were further explored as part of a larger study in which 206 adults from the community (average age of 43 years, 60% women, 67% Caucasian) were asked to list their three greatest regrets and then, for the first regret listed, to imagine a TV program in which the lead character had (a) made the same choice or (b) made a different choice as they had. For each, respondents were asked to indicate how much they would want to watch such a program on a scale of 0 (not at all) to 10 (very much). An 18-item measure of trait regret was included later in the survey to tap into people’s tendency to feel regret. Of these, exploratory factor analysis identified 10 items that loaded onto a single factor (α=.87). Sample items, assessed on 5-point Likert scales, included “If I had the chance, I would go back and make different decisions in my life,”“I am often bothered by mistakes in my past,” and “I’m happy with the decisions I’ve made in my life” (reverse coded). Controlling for age, education level, gender, race, and proportion of TV viewing that included narrative-based entertainment programming (situation comedies, dramas, and soap operas), partial correlations indicated that respondents higher in trait regret were interested in watching a program in which the character made their same choice, r(184) =.20, p=.007, as well as a program in which the character made a different choice, r(184) =.34, p <.001. We then created a single viewing preference measure by subtracting the similar choice from the different choice score. A positive score on this measure indicates preference for watching a character make a different choice than the respondent had made, and a negative score indicates preference for watching a character make the same choice as the respondent. Correlating this measure with trait regret will indicate whether or not those high in trait regret prefer watching a character make a similar or different choice as they had made in their own lives. Results of the partial correlation (using the same control variables noted above) indicated that those higher in trait regret were significantly more likely to prefer to watch a program in which a character made a different choice than they did, r(184) =.15, p=.04. These data are consistent with the Study 1 findings that not only are those who are also regretful interested in watching TV content about their regretted experience but they are particularly interested in seeing that experience portrayed differently than it had unfolded in their own lives. Discussion Through this research, we sought to assess how regretted personal experiences impact interest in and enjoyment of experience-related entertainment TV programming as well as the effect of viewing such programming on persistent feelings of regret. Results suggested that, consistent with regret theory, regret for past mistakes motivated interest in, rather than avoidance of, experience-related storylines, though this interest did not necessarily translate into greater enjoyment of the programming, generally speaking. Conversely, and consistent with mood management theory, anticipated regret of a mistake yet to be made appeared to motivate both programming avoidance and reduced viewing enjoyment. For regretful cheaters, program viewing appeared to reduce regret (at least temporarily) regardless of how the protagonist reacted to her mistake, though they seemed to enjoy viewing her rationalization over her self-blame. Further, the findings linking regret over past behavior to program interest and preference for media portrayals that offer a different perspective than experienced in life were reinforced by the survey results. As the hypotheses regarding regret were derived from regret theory, though with the counterhedonistic exceptions to mood management theory in mind, this research highlights the value of integrating theory on specific emotions with media theory, like mood management, to arrive at more precise predictions of how emotional states might impact media selection and impact counterhedonistic choices in particular. Although the results were often supportive of predictions, there were two outcomes in this research that deserve closer inspection. First, though overall, the regretful and nonregretful cheaters appeared to enjoy the programming equally, closer analysis revealed that the regretful cheaters in fact enjoyed the rationalization portrayal more than the self-blame portrayal (see Table 3). This finding suggests that both content and context matters. That is, enjoyment is not simply a function of whether or not the issue is portrayed but how it is portrayed. Thus, we must consider more carefully why the two depictions had different effects on enjoyment for regretful cheaters. On the one hand, as we argued earlier, those experiencing regret who see a TV character react to the regretted situation differently than the viewer had in real life may serve as a model for an alternative means of coping. This realization or insight may fulfill the information needs of the viewer, resulting in enjoyment. Alternatively, it may be that watching another’s regretful display compounds one’s own negative affect, thus reducing enjoyment. Were this the case, however, we would expect the less regretful cheaters to respond similarly; yet, they evidenced no difference in enjoyment between the two depictions. Given the heightened enjoyment of the rationalization portrayal by the regretful cheaters, we believe the information-needs explanation is most fitting, though a direct test of this assumption awaits future research. Second, if we grant that the regretful cheaters’ enhanced enjoyment of the rationalization portrayal indeed signaled greater coping, we would have expected that portrayal to reduce regret more so than the self-blame portrayal. Yet, in contrast to H4, no such difference was found. Thus, we might conclude that both portrayals served to aid coping but through different routes, one being more enjoyable than the other. More specifically, perhaps through watching someone else’s regret reaction, regretful cheaters may have realized they were not alone in their transgression and sympathizing with the character’s plight, reduced their self-blame, thus minimizing the associated regret. However, viewing a representation of one’s own transgression, coupled with the affective display of the character and no concrete portrayal of resolution, may not have been a particularly enjoyable experience. Conversely, in watching their plight reenacted with a different solution—no tears and no break- up—regretful cheaters might not only have come to see their behavior as normal and perhaps even understandable but also as something to be considered without self-flagellation. Thus, not only is regret reduced but also enjoyment enhanced. Understanding the underlying cause not only of these differences in enjoyment but also in the mechanisms through which regret is reduced (i.e., reduced self-blame and alteration of the value for the enacted or alternative behaviors) is an important direction for future exploration. Of particular note, though we were able to show that experience-relevant programming reduced levels of regret, we have no way of knowing if this is a temporary or a more long-lived effect. Like an analgesic salve, watching the program may have led to temporary relief of the negative feelings associated with regret, but in all likelihood, without acquisition of unique insight into one’s condition, it is likely the regret will recur. Although we were unable to follow up with our respondents to determine longevity of effects, it is possible that watching the rationalization portrayal, though it had an equal effect on regret in the short run, might have had a more lasting impact, particularly given its enhanced effect on enjoyment. Future research would do well to explore this issue. Despite the insight this research offers into how emotional states might impact selection and reaction to TV programming, this research has several limitations that require discussion. The first relates to study design in that the questions about personal history of, and feelings about, being unfaithful to a romantic partner were asked prior to any of the dependent measures, which raises concerns about demand characteristics. That is, perhaps the participants not only realized the study was related to cheating behavior but also consequently responded in ways desired by the researchers. In response to such concerns, we point out first that it was necessary for us to ask cheating-related questions at the beginning of the study in order to activate feelings of regret, to the extent they still lingered, to examine their effects, in context, on program interest and reactions. Were we to simply rely on those naturally in a state of regret (e.g., those who had cheated only very recently), we would likely have been hard pressed to achieve a sample size suitable for hypothesis testing. More to the point, because our hypotheses are somewhat counterintuitive, even if the respondents knew we were interested in the effect of their past cheating behavior on program selection and viewing, it is unlikely they could have successfully and systematically guessed in what direction we expected the effect to occur. Thus, we believe it is unlikely that the question order biased the findings in favor of our hypotheses. Further, we point out that in Study 2, the program interest items preceded the trait regret items, and the results from those analyses were consistent with those from Study 1. Thus, demand characteristics or question order does not appear to pose a significant threat to the validity of these findings. A second major concern with this research involves the proportion of participants who reported having seen the program storyline previously. If they had indeed followed this storyline before, one might say, they may have responded based not only on what they saw during the study but also on their previous experience with the programming, thus muddying the results. However, we believe the reader can confidently set this worry aside for several reasons. First, the manipulation check clearly showed that participants saw the versions as distinct, regardless of their previous experience. Second, we reran the analyses with only those who had not seen the program before and found results of similar magnitude to those reported with the complete sample (see Footnote 5). If anything, the results suggest that our choosing a popular stimulus program may have created a more conservative test of our hypotheses as we may have lost the advantage of capturing the initial impact of program viewing. Finally, we would like to note that in choosing a program popular with our sample audience, we are more confident in our conclusions about how young people might cope with regretted experience through TV viewing than if we had selected a program less likely to be watched in “the real world.” In sum, these two studies, with different samples, regretted experiences, and assessments of regret, converges to suggest that lingering feelings of regret can motivate consumption of programming relevant to one’s life experience and that such consumption may have a positive impact on coping and, thus, feelings of regret. This research builds on work by Oliver (1993) and others who explore the effects of counterhedonistic media choices (see Oliver, 2003) and simultaneously answers Zillmann’s (2000) call to explore the exceptional circumstances under which informational utility may supersede diversionary goals when selecting entertainment programming. By drawing from regret theory, we were able to arrive at predictions that could not have been made based on mood management theory alone. Thus, future developments in mood management theory specifically, and uses and gratifications generally, would likely benefit from integrating the information-seeking motivations associated with specific emotional states to allow for more predictive models of media selection, content preference, and effects. Understanding the boundaries and persistence of these effects, for regret as well as other intractable emotions, will not only serve to advance media theory, but, as important, will also offer much-needed insight into the potentially beneficial effects of TV portrayals of dramatic and emotional life events. Notes 1 " There is a notable amount of research on coping with fear generated by TV viewing (e.g., Valkenburg, Cantor, & Peeters, 2000), though this approach does not discuss the use of TV as a coping device. 2 " We say this with the understanding that though the counterhedonistic exceptions to mood management theory are acknowledged by Zillmann (2000), the theory is not currently equipped to make counterhedonistic predictions. 3 " Hugging and spending a lot of time with another (opposite sex) person were not considered to be unfaithful acts but holding hands, emotional intimacy, kissing, and physical intimacy were. Further, all but hugging were considered potentially damaging to a relationship. No significant differences between men and women were identified, though women were slightly more likely than men to indicate emotional intimacy as cheating and women saw emotional intimacy as more damaging to a relationship than men. 4 " The regret variable was centered before creating the interaction term with past cheating behavior. In addition, the effect of whether or not one had been cheated on was explored, but as it had no meaningful impact on viewing interest (nor did it significantly interact with negative emotions about that experience), it was omitted from the reported analyses in the interest of parsimony. 5 " Of note, analyses were repeated, separating those who had and those who had not previously seen the Real World storyline that served as the basis for the stimuli. Both groups evidenced comparable effects for regret (seen: β=−.15; not seen: β=−.24), the interaction between regret and past cheating behavior (seen: β=.23; not seen: β=.21), and the interaction between regret and condition (seen: β=.17; not seen: β=.40). The stronger effect for those who had not previously seen the storyline suggests initial viewing may have a more potent effect, though this is clearly speculative given the small sample size on which this observation is based. References Boninger , D. S. , Gleicher , F., & Strathman , A. ( 1994 ). Counterfactual thinking: From what might have been to what may be . Journal of Personality and Social Psychology , 67 , 297 – 307 . Google Scholar Crossref Search ADS WorldCat Connolly , R. , & Zeelenberg , M. ( 2002 ). Regret in decision making . Current Directions in Psychological Science , 11 , 212 – 216 . Google Scholar Crossref Search ADS WorldCat Finn , S. , & Gorr , M. B. ( 1988 ). Social isolation and social support as correlates of television viewing motivation . Communication Research , 15 , 135 – 158 . Google Scholar Crossref Search ADS WorldCat Gilovich , T. , & Medvec , V. H. ( 1994 ). The temporal pattern to the experience of regret . Journal of Personality and Social Psychology , 67 , 357 – 365 . Google Scholar Crossref Search ADS PubMed WorldCat Gilovich , T. , Medvec , V. H ., & Kahneman , D. ( 1998 ). Varieties of regret: A debate and partial resolution . Psychological Review , 105 , 602 – 605 . Google Scholar Crossref Search ADS WorldCat Herzog , H . ( 1944 ). What do we really know about daytime serial listeners? In P. F . Lazarsfeld & F. N . Stanton (Eds.), Radio research 1942–1943 (pp. 3 – 33 ). New York : Duell, Sloan, & Pearce . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Katz , E. , Blumler , J. G., & Gurevitch , M. ( 1974 ). Utilization of mass communication by the individual. In J. G . Blumler & E . Katz (Eds.), The uses of mass communications: Current perspectives on gratifications research (pp. 19 – 32 ). Beverly Hills, CA : Sage . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Knobloch , S . ( 2003 ). Mood adjustment via mass communication . Journal of Communication , 53 , 233 – 250 . Google Scholar Crossref Search ADS WorldCat Kubey , R ., & Peluso , T. ( 1990 ). Emotional response as a cause of interpersonal news diffusion: The case of the space shuttle tragedy . Journal of Broadcasting & Electronic Media , 34 , 69 – 76 . Google Scholar Crossref Search ADS WorldCat Landman , J . ( 1987 ). Regret: A theoretical and conceptual analysis . Journal for the Theory of Social Behaviour , 17 , 135 – 160 . Google Scholar Crossref Search ADS WorldCat Landman , J . ( 1993 ). Regret: The persistence of the possible . New York : Oxford . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Lazarus , R. S . ( 1991 ). Emotion and adaptation . New York : Oxford . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC McCroskey , J. C. , Richmond , V. P., & Daly , J. A. ( 1975 ). The development of a measure of perceived homophily in interpersonal communication . Human Communication Research , 1 , 323 – 332 . Google Scholar Crossref Search ADS WorldCat McQuail , D. , Blumler , J. G., & Brown , J. R. ( 1972 ). The television audience: A revised perspective. In D. McQuail (Ed.), Sociology of mass communication (pp. 135 – 165 ). Middlesex, U.K.: Penguin . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Nabi , R. L . ( 1999 ). A cognitive-functional model for the effects of discrete negative emotions on information processing, attitude change, and recall . Communication Theory , 9 , 292 – 320 . Google Scholar Crossref Search ADS WorldCat Nabi , R. L . ( 2003 ). The framing effects of emotion: Can discrete emotions influence information recall and policy preference? Communication Research , 30 , 224 – 247 . Google Scholar Crossref Search ADS WorldCat Oliver , M. B . ( 1993 ). Exploring the paradox of the enjoyment of sad films . Human Communication Research , 19 , 315 – 342 . Google Scholar Crossref Search ADS WorldCat Oliver , M. B . ( 2003 ). Mood management and selective exposure. In J. Bryant, D. Roskos-Ewoldsen, & J. Cantor (Eds.), Communication and emotion: Essays in honor of Dolf Zillmann (pp. 85 – 106 ). Mahwah, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Palmgreen , P. , Wenner , L. A., & Rosengren , K. E. ( 1985 ). Uses and gratifications research: The past ten years. In K. E . Rosengren, L. A . Wenner, & P. Palmgreen (Eds.), Media gratifications research: Current perspectives (pp. 11 – 37 ). Beverly Hills, CA : Sage . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Rubin , A. M . ( 1981 ). An examination of television viewing motivations . Communication Research , 8 , 141 – 165 . Google Scholar Crossref Search ADS WorldCat Rubin , A. M . ( 2002 ). The uses-and-gratifications perspective of media effects. In J. Bryant & D. Zillmann (Eds.), Media effects: Advances in theory and research (pp. 525 – 548 ). Hillsdale, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Rubin , A. M. , & Rubin , R. B. ( 1985 ). Interface of personal and mediated communication: A research agenda . Critical Studies in Mass Communication , 2 , 36 – 53 . Google Scholar Crossref Search ADS WorldCat Savitsky , K. , Medvec , V. H., & Gilovich , T. ( 1997 ). Remembering and regretting: The Zeigarnik effect and the cognitive availability of regrettable actions and inactions . Personality and Social Psychology Bulletin , 23 , 248 – 257 . Google Scholar Crossref Search ADS WorldCat Stewart , A. J. , & Vandewater , E. A. ( 1999 ). “If I had it to do over again …”: Midlife review, midcourse corrections, and women’s well-being in midlife . Journal of Personality and Social Psychology , 76 , 270 – 283 . Google Scholar Crossref Search ADS PubMed WorldCat Sugden , R . ( 1985 ). Regret, recrimination, and rationality . Theory and Decision , 19 , 77 – 99 . Google Scholar Crossref Search ADS WorldCat Valkenburg , P. M. , Cantor , J., & Peeters , A. L. ( 2000 ). Fright reactions to television . Communication Research , 27 , 82 – 100 . Google Scholar Crossref Search ADS WorldCat Zeelenberg , M . ( 1999 ). The use of crying over spilled milk: A note on the rationality and functionality of regret . Philosophical Psychology , 12 , 325 – 340 . Google Scholar Crossref Search ADS WorldCat Zeelenberg , M. , Van Den Bos , K., Van Dijk , E., & Pieters , R. ( 2002 ). The inaction effect in the psychology of regret . Journal of Personality and Social Psychology , 82 , 314 – 327 . Google Scholar Crossref Search ADS PubMed WorldCat Zillmann , D . ( 1988 ). Mood management: Using entertainment to full advantage. In L. Donohew & H. E . Sypher (Eds.), Communication, social cognition, and affect (pp. 147 – 171 ). Hillsdale, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Zillmann , D . ( 2000 ). Mood management in the context of selective exposure theory. In M. Roloff (Ed.), Communication yearbook , 23 (pp. 103 – 123 ). Thousand Oaks, CA : Sage . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC © 2006 International Communication Association

Journal

Journal of Communication – Oxford University Press

Published: Dec 1, 2006

There are no references for this article.