Encouraging Replotting to Promote Persuasion: How Imagining Alternative Plotlines Influences Message Processing and Intentions

Abstract

This set of studies investigated replotting as a mechanism of narrative persuasion. Replotting involves both the cognitive act of imagining alternative plot lines to avoid an undesirable story outcome and an accompanying emotion such as anger, anxiety, or sadness. Both studies utilized a 2 (story outcome: death vs. survivor) × 2 (efficacy appeal: present vs. absent) message experiment design. Study 1 (N = 1,207) tested a non-narrative efficacy appeal appended to the story and assessed replotting anger. Study 2 (N = 716) tested an efficacy appeal embedded within the narrative and assessed replotting anger, anxiety, and sadness. Death narratives generated greater replotting sadness across efficacy conditions and greater replotting anger and anxiety when a narrative efficacy appeal was not included in the story. Replotting was negatively related to counterarguing and positively related to message elaboration. Replotting influenced behavioral intention either via counterarguing or message elaboration dependent on the efficacy condition.

Keywords

replotting anger message elaboration counterarguing efficacy narratives

Persuasion research has provided strong support for the effectiveness of narratives in promoting attitude and behavior change (Braddock & Dillard, 2016; F. Shen et al., 2015), particularly in the cancer context (Hinyard & Kreuter, 2007). Narratives are an effective persuasive tool in part because they can produce deeper emotional involvement in the message (Zebregs et al., 2015). Emotions are theorized to motivate behavior change alongside and through a variety of mechanisms, such as message elaboration and transportation (Nabi & Green, 2015; Turner et al., 2020). One under-explored mechanism of narrative persuasion that captures both cognitive and emotional processes is replotting.

Replotting is the act of generating alternative plotlines to change unwanted story outcomes (Gerrig, 1993). Replotting can occur during narrative consumption, particularly when audiences already know how a story will end, or it can occur after exposure. Audience members can engage in replotting internally or via interaction with the narrative (e.g., talking out loud to characters while watching a film) and other individuals, such as in conversation or on fanfiction forums (Gerrig & Bezdek, 2013). There is also an affective component to replotting. Specifically, replotting involves a typically negative emotional response that drives and is intertwined with the generation of alternative plot lines (Gerrig, 1993; Green & Brock, 2000).

The current research focuses primarily on replotting anger, meaning the cognitive act of replotting intertwined with anger. Anger as a discrete emotion has received ample attention in the broader persuasion literature (Walter et al., 2019). Theories such as the cognitive functional model (CFM; Nabi, 1999) and the anger activism model (AAM; Turner, 2007) propose conditions under which anger is an effective mechanism through which messages influence audiences. A key condition supported by empirical evidence is that anger is most persuasive when it is message-relevant, although incidental anger can be influential as well (Lu, 2020; Walter et al., 2019). For narrative messaging, anger related to replotting could be a particularly effective type of message-relevant anger (Gerrig, 1993).

The current research addresses theoretical questions regarding replotting, emotion, and persuasion. Across two studies situated in the context of melanoma prevention, the effect of two message features (story outcome and efficacy) on replotting is assessed, and the influence of replotting on behavioral intentions via message elaboration and counterarguing is investigated. Study 1 focuses on replotting anger, grounding hypotheses in the AAM (Turner, 2007) and the CFM (Nabi, 1999). Study 2 expands on Study 1 by adding an examination of replotting anxiety and replotting sadness. Below, we define replotting, describe message features that we propose will trigger replotting, and present theoretical rationale for the predicted relationships between replotting and both message elaboration and counterarguing.

Narrative Replotting

Replotting includes both the cognitive act of identifying possible pathways through which an unwanted story outcome could be prevented or changed and an accompanying emotional reaction (Gerrig, 1993; Green & Brock, 2000). Gerrig (1989) provides examples of replotting, describing how “I may find myself hoping that – this time – Anna Karenina will stay away from the train station or that – just this once – Darth Vader will be slain in his duel with Luke Skywalker” (p. 278). Audiences may engage in participatory responses when replotting (Gerrig & Bezdek, 2013), such as telling the main character: “Don’t wait to go to the doctor.” These responses are evidence of the audience’s deep involvement in the narrative and represent the influence audience members would like to have if they were participants in the narrative (Gerrig, 1993; Gerrig & Bezdek, 2013). We propose that replotting anger is an important mechanism of narrative persuasion.

Anger is an approach emotion that motivates audiences to take action against anger-inducing stimuli (Lazarus, 1991; Turner, 2007). Anger occurs in response to an offense or threat to either the self or an important other (Nabi, 1999). The AAM (Turner, 2007) organizes audiences into four groups including activists who have strong feelings of anger and efficacy, empowered who have low feelings of anger and a strong sense of efficacy, angry who have strong feelings of anger and a low sense of efficacy, and disinterested who have low feelings of anger and efficacy. Persuasive messaging will be most effective on activists, then the empowered. Angry and disinterested audiences will either be unaffected or reactant to persuasive messaging.

The AAM also proposes that anger must be directed at the threat presented in the message, not at the message or its source, in order to be persuasive (Turner, 2007). The CFM (Nabi, 1999), which explicates how different cognitive appraisals elicit discrete emotions in response to stimuli, echoes these assertions. Although incidental emotions, meaning emotions unrelated to the message or threat, can be persuasive, incidental anger is less likely to result in desired attitude and behavior change (Lu, 2020; Nan, 2017). Of greater concern is anger directed at the message itself or at the message source (e.g., I’m angry that this message is trying to manipulate my emotions). This type of anger is a component of reactance and is negatively related to persuasion (Dillard & Shen, 2005; Ratcliff, 2021). Ideally for persuasion, the audience’s anger will be related to the threat itself, referred to as message-relevant anger.

Replotting anger involves feeling angry that an undesirable story outcome could have been avoided but was not. This is a unique type of message-relevant anger because it is directed specifically at an avoidable unwanted outcome and is intertwined with the cognitive act of creating alternative plot lines. Replotting anger is likely a beneficial audience response for narrative persuasion both because anger is an approach emotion (Nabi, 1999) and because replotting directs audience cognitions toward an efficacious response to the threat presented in the narrative. This should result in an activist audience, experiencing high levels of anger and efficacy (Turner, 2007). Replotting involves identifying a course of action that would prevent an unwanted outcome. In the context of cancer narratives, for example, this course of action should align with recommended prevention behaviors, prompting the audience to engage in these behaviors themselves. Anger has been found to motivate preventative behaviors and intentions in past studies (e.g., Oh et al., 2020), and a recent meta-analysis found that message-relevant anger is positively related to behavioral intentions (Walter et al., 2019). Therefore, an important theoretical and practical question for persuasion scholars is which message features trigger replotting. One likely feature is a death story outcome.

Death Narratives

Death narratives, or stories in which the main character ultimately dies, will likely generate greater replotting than narratives of survivorship. Replotting is triggered by undesirable story outcomes (Gerrig, 1993). In most cases, death of the main character will be considered undesirable. According to the just-world hypothesis (Lerner & Miller, 1978), people generally believe that individuals should be rewarded for good or moral behavior and punished for bad or immoral behavior. Similarly, moral sanction theory posits that audiences enjoy media in which characters that they deem to be moral are rewarded and immoral characters receive commensurate punishment (Zillmann, 2000). Stigma associated with some behaviors may lead to blame and the evaluation that character death is justified (e.g., heavy smoking leading to lung cancer; Else-Quest & Jackson, 2014). In many contexts, however, death would not be considered a justifiable consequence of failing to take a preventative action. For example, dying from melanoma is unlikely to be considered a just consequence of failing to regularly use sunscreen. Even if death makes logical sense given a lack of prevention, audiences still desire liked or otherwise moral characters to have positive endings (Janicke & Raney, 2018). Therefore, it is expected that a death narrative, particularly when the death outcome is foreshadowed, will generate greater replotting than a survivor narrative will. Audiences will have both greater motivation to engage in the cognitive act of replotting and a stronger accompanying emotional reaction when a severe negative ending, specifically death, occurs in the narrative.

H1: Death narratives will generate greater replotting than survivor narratives will.

Message Elaboration

Replotting likely influences behavioral intentions by increasing message elaboration beyond consideration of alternative plotlines. Message elaboration refers to the degree of audience engagement in issue-relevant thinking about a message (Niederdeppe et al., 2012; Petty & Cacioppo, 1986). Greater message elaboration enhances persuasion (Briñol & Petty, 2012). The cognitive act of replotting should lead audiences to elaborate on the issues within the narrative and how these issues relate to their own lives. The inclusion of counterfactuals in narratives, meaning the characters themselves replot within the narrative, is hypothesized to increase persuasion by motivating audiences to apply the replotting to their own lives (Shaffer et al., 2018). Further, Tal-Or et al. (2004) suggest that replotting leads to long-term persuasive effects because replotting prompts greater elaboration.

In addition to the effects of the cognitive aspect of replotting, its emotional component should lead to greater elaboration as well. Anger, specifically, is theorized to motivate deeper information processing (Nabi, 2002). Past research has demonstrated mixed effects of anger on message elaboration (Moons & Mackie, 2007; Shields et al., 2016). Yet, inconsistent findings could be the result of anger being incidental or reactant in some cases and message-relevant in others (Walter et al., 2019). The AAM and the CFM propose that message-relevant anger should increase message elaboration (Nabi, 1999; Turner, 2007). Walter et al. (2021) found that anger increased message elaboration when hope was high. Further, Turner et al. (2020) found that participants reading an op-ed piece designed to induce high message-relevant anger (along with high efficacy) experienced greater positive-relevant thoughts (i.e., message elaboration that was relevant to and in favor of the message), leading to increased intentions to protest. This suggests that message-relevant anger should increase message elaboration, leading to persuasion. In-line with these findings and anger theorizing, it is expected that replotting will lead to greater message elaboration and therefore increased behavioral intentions.

H2: Replotting will be positively associated with message elaboration.

H3: Message elaboration will be positively associated with behavioral intention.

Counterarguing

Replotting likely diminishes counterarguing against the underlying claims of the story (e.g., sunscreen protects against melanoma). Counterarguing is the act of producing arguments against the claims of the message or arguments in support of an opposing view (Bilandzic & Busselle, 2013). Although counterarguing can enhance persuasion for those high in need for cognition (Walter & Cohen, 2019), counterarguing is generally considered to be counter-productive for persuasion (Cialdini et al., 1981; Dillard & Shen, 2005; Wegener et al., 2004).

Although replotting may seem like a form of counterarguing, they are two distinct types of message engagement. Replotting involves wishing the events of a narrative had occurred differently, not arguing against the veracity of the events or claims of the narrative. For example, audiences engaging in replotting may wish a character had gone to a doctor sooner about a suspicious mole so that the character would not have died from melanoma. This is a distinct cognitive act from counterarguing, which could involve disagreeing with the premise that delaying treatment for a suspicious mole could result in death or disagreeing that a person would have delayed in those circumstances. Counterarguing implies disagreement with the truth value of events or features of a narrative (Bilandzic & Busselle, 2013), but replotting does not. Both replotting and counterarguing are tied to anger. Reactance is composed of counterarguing paired with anger at having one’s freedom threatened (Dillard & Shen, 2005; Ratcliff, 2021) whereas replotting can include anger at an avoidable, negative story outcome.

Replotting should direct cognitive effort away from counterarguing, focusing the audience’s attentions on the ways an undesirable outcome could have been prevented within the narrative world (Bilandzic & Busselle, 2013). Further, replotting within the established narrative world involves a tacit acceptance of that world and its rules (e.g., tanning can cause melanoma or medical delay is common). Acceptance of the narrative world helps form or reinforce mental models about the underlying story topic in-line with the message position, weakening counterarguments (Miller-Day & Hecht, 2013).

H4: Replotting will be negatively associated with counterarguing.

H5: Counterarguing will be negatively associated with behavioral intention.

Efficacy

Efficacy appeals include both an effective means of protecting against a threat and an indication that the audience has the ability to engage in that protective action (Bandura, 1982; Witte, 1992). Efficacy appeals should encourage replotting and enhance its persuasive effects. Like counterfactuals, efficacy appeals identify avenues through which main characters, and consequently the audience, could have protected themselves from the threat, encouraging replotting (Shaffer et al., 2018). Death narratives specifically should benefit from the inclusion of an efficacy appeal. The presence of a death outcome, particularly if the main character engaged in some protective actions, may diminish audiences’ perceived response efficacy. This could be especially problematic because a recent meta-analysis found that positive response efficacy information significantly improved the persuasiveness of fear appeals (Bigsby & Albarracín, 2022). An efficacy appeal would be necessary to counteract the potentially negative effect of a death outcome on response efficacy. Therefore, the combination of a death narrative and an efficacy appeal should lead to greater replotting.

Efficacy appeals should also reinforce the persuasiveness of replotting by arguing for the effectiveness and ease of the protective action, subsequently increasing message elaboration and diminishing counterarguing. Regarding the emotion component of replotting, anger is theorized to be most effective when an efficacy appeal is present (Turner, 2007). A recent meta-analysis concluded that high anger/high efficacy messages affect attitudes and behaviors more than high anger/low efficacy messages (Walter et al., 2019). Therefore, it is expected that the hypothesized associations between replotting and death narratives, message elaboration, and counterarguing will be conditional upon the presence of an efficacy appeal.

H6: Death narratives will generate greater replotting than survivor narratives will only when an efficacy appeal is included, with the efficacy appeal serving as a contingent, divergent positive moderator (see Holbert & Park, 2020).

H7: Replotting will be positively associated with message elaboration only when an efficacy appeal is included.

H8: Replotting will be negatively associated with counterarguing only when an efficacy appeal is included.

Study 1

Methods

Participants and procedures

Participants were enrolled via Qualtrics Panels into a web-based experiment study. Following collection of demographic data, participants were randomized into one of 24 conditions in a 2 (story outcome: death vs. survivor) × 2 (efficacy appeal: present vs. absent) × 6 (story: Alex, Don, Erin, Fern, Marla, and Ray) between-participants study design. Participants read the melanoma narrative corresponding to their study condition and then answered measures about their reading experience, such as replotting, counterarguing, and message elaboration, and about their sun safe behavior intentions.

Power was calculated using G*Power (Faul et al., 2009). Past research in this area (Jensen et al., 2017; Lillie, Pokharel, et al., 2021) has identified small main effects. Consistently detecting small effects requires relatively large sample sizes guided by a priori power analysis (Cohen, 1992). To achieve .80 power to detect small effect sizes (f = .10), a priori power analysis identified a target sample size of 787 for main effects and 1,289 for the full 2 × 2 × 6 model. A total of 1,300 participants completed the study. Data were removed if participants spent less than 5 seconds on the narrative and/or were unable to answer open-ended questions about the narrative (including what happened to the main character at the end of the story). Data from 93 participants were removed. The final sample (N = 1,207) provided sufficient power for the main effects of story outcome and efficacy appeal and was just short of the target for the full 2 × 2 × 6 model.

Participants were on average 45.03 years old (age range: 18–91), 79.2% identified as Caucasian (Black or African American: 9.4%; Asian or Asian American: 3.2%; American Indian or Alaskan Native: 2.0%; Other: 3.2%), and 11.4% identified as Hispanic or Latina/o. Participants were evenly split between those with high school education or less (50.3%) and those with more than high school education (49.7%) and between females (50.7%) and males (49.3%).

Stimuli

Six different melanoma narratives were utilized in the current study to represent a range of experiences, including variation in main characters’ age, sex, and race, to produce within-study replication (see Supplemental Appendix). Within-study replication is important because it allows for assessment of effects across variations within a message category, rather than relying on one message which may differ from other messages in its category in meaningful ways (Jackson & Brashers, 1994). All six text-based narratives were based on stories featured by the Skin Cancer Foundation, the American Academy of Dermatology, and the Melanoma Research Foundation and followed the same basic structure of character introduction, mole discovery, and outcome. Each story had two versions: (1) the main character dies (death narrative) or (2) the main character survives (survivor narrative), manipulated in the final paragraph of the narrative. Further, half of the conditions included an informational efficacy appeal following the narrative. The appeal emphasized the effectiveness and ease of performing sun safe behaviors and was worded such that it did not emphasize death or survivorship to avoid confounding the story outcome. The first author wrote the efficacy message based on information provided by the Skin Cancer Foundation, the American Academy of Dermatology, and the Melanoma Research Foundation.

Measures

Replotting

Replotting was measured using a single item created for the current study. On a scale ranging from 1 (strongly disagree) to 5 (strongly agree), participants indicated their agreement with the statement “I felt angry about how what happened to the main character could have been avoided” (M = 3.56, SD = 1.38). The item was embedded in a list of seven other items about aspects of the story the participants could have been angry about which served as distractors, such as “what happened to the main character,” “having to read this type of story,” and “the main character’s actions/choices.” These other items were not included in the replotting measure.

Message elaboration

Message elaboration was measured using Reynolds’ (1997) message elaboration measure based on recommendations by L. Shen and Seung (2018). Participants responded to 12 items on a five-point scale ranging from 1 (strongly disagree) to 5 (strongly agree) about their experience while reading the message. Sample items include: “attempting to analyze the issues in the message” and “reflecting on the implications of the arguments.” Half of the items are reverse-coded. To achieve acceptable fit for the measurement model (detailed below) the reverse-coded items were removed from analysis (for the six-item measure: α = .81, M = 3.41, SD = 0.77).

Counterarguing

Counterarguing was measured using four items developed by Nabi et al. (2007) and one item developed by Krakow et al. (2018). Sample items include “I was looking for flaws in the message” and “I wanted to correct one or more points in the message.” Participants responded to the items on a scale ranging from 1 (strongly disagree) to 7 (strongly agree) (α = .74, M = 2.25, SD = 0.74).

Sun safe behavior intention

Sun safe behavior (SSB) intention was measured with eight items on a seven-point scale ranging from 1 (extremely unlikely) to 7 (extremely likely). Participants indicated how likely they were to perform a range of sun safe behaviors such as “use sunscreen,” “wear protective clothing (long pants and sleeves),” and “check your body for strange moles” (α = .90, M = 5.26, SD = 1.48). For means and standard deviations by study condition, see Table 1.

Table 1.

Study 1 Means and Standard Deviations by Study Condition.

Variable	Death narratives	Survivor narratives	Death narratives		Survivor narratives
Variable	Death narratives	Survivor narratives	High efficacy	Low efficacy	High efficacy	Low efficacy
Replotting anger	4.05 (1.13)	3.08 (1.44)	4.06 (1.12)	4.04 (1.14)	3.05 (1.45)	3.10 (1.43)
Counterarguing	2.33 (0.74)	2.17 (0.73)	2.29 (0.74)	2.37 (0.73)	2.15 (0.73)	2.18 (0.73)
Elaboration	3.44 (0.77)	3.38 (0.78)	3.50 (0.75)	3.39 (0.76)	3.39 (0.78)	3.36 (0.78)
SSB^a intention	5.22 (1.51)	5.29 (1.44)	5.27 (1.46)	5.18 (1.55)	5.38 (1.37)	5.20 (1.52)

Note. Means with standard deviations in parentheses. Bolded pairs represent significant main effects of death/survivor or interaction effects with efficacy condition at p < .001.

Sun safe behavior.

Manipulation check

For the manipulation check of the story outcome condition (death vs. survivor), participants responded to two items on a scale ranging from 1 (strongly disagree) to 5 (strongly agree): “This story was about death” and “The main character died of cancer.” Confirming that the manipulation was successful, there was a main effect of story outcome on both the first item, F(1, 1183) = 612.36, p < .001, and the second item, F(1, 1183) = 2,702.33, p < .001. Participants who received death narratives were more likely to endorse the statement that the story was about death (M = 3.36, SE = 0.05) compared to those who received survivor narratives (M = 1.67, SE = 0.05, r = .58).¹ Participants who received death narratives were more likely to endorse the statement that the main character died of cancer (M = 4.53, SE = 0.04) compared to those who received survivor narratives (M = 1.42, SE = 0.04, r = .83).

For the manipulation check of the efficacy condition, participants responded to an eight-item measure capturing self and response efficacy related to sun safe behaviors (α = .86, M = 5.25, SD = 1.14). Example items include: “Using sunscreen is easy for me” and “Sunscreen is effective at preventing skin cancer.” Confirming that the manipulation was successful, there was a main effect of efficacy condition, F(1, 1183) = 203.70, p = .045. Participants who received the efficacy message reported greater efficacy (M = 5.32, SE = 0.05) compared to those who did not receive the efficacy message (M = 5.19, SE = 0.05, r = .05).

Results

The hypothesized model was tested with structural equation modeling (SEM) in SPSS Amos version 27. Analyses utilized Anderson and Gerbing’s (1988) two-step process for SEM. First, a measurement model with all latent variables was tested, with the latent variables covaried with one another. Second, a structural model including the hypothesized associations was tested. The two-step process allows for differentiating whether model fit issues arise from measurement, indicated by a measurement model that does not achieve acceptable fit, or the hypothesized associations not fitting the data. Model fit is determined using the comparative fit index (CFI) and the root mean square error of approximation (RMSEA). Model fit is considered acceptable if CFI ≥ .90, RMSEA ≤ .08, and χ²/df ≤ 3.00, but it is preferable for CFI ≥ .95 and RMSEA ≤ .05 (Hu & Bentler, 1999; Schermelleh-Engel et al., 2003).

To test differences in associations based on efficacy condition, multi-group SEM was used (Kline, 2016). A model constraining measurement weights and intercepts and structural weights, means, covariances, and residuals to be the same was compared to an unconstrained model. A chi square statistic was used to calculate if there is a significant difference between the models (Kline, 2016). If there is a difference, then the associations between the variables differ based on efficacy condition. Prior to model testing, bivariate correlations were calculated to examine zero order relationships between study variables (Table 2).

Table 2.

Study 1 Correlation Matrix.

	Variable	1	2	3	4	5
1	Story outcome	—
2	Efficacy condition	−.02	—
3	Replotting anger	.35±	−.01	—
4	Counterarguing	.11±	−.04	−.06	—
5	Elaboration	.04	.04	.27±	−.18±	—
6	SSB^a intention	−.02	.05	.18±	−.15±	.35±

Note. For story outcome: survivor = 0, death = 1.

Sun safe behavior.

p ≤ .001.

The measurement model achieved acceptable fit after removing the reverse-coded message elaboration items (see Figure 1). Multi-group SEM identified a significant difference between the constrained and unconstrained structural models, χ² = 66.11, df = 47, p = .034. Therefore, the efficacy and no-efficacy conditions were significantly different. Regarding H1–5, death narratives produced greater replotting compared to survivor narratives, supporting H1. Replotting was positively related to message elaboration, supporting H2, and negatively related to counterarguing, supporting H4. Message elaboration was positively related to SSB intention, supporting H3, but counterarguing was unrelated to SSB intention, rejecting H5. Although the same pathways were significant across the efficacy conditions (see Figure 1), rejecting H6–8, there are two key differences. First, the association between counterarguing and sun safe behavior (SSB) intention approached significance in the no-efficacy condition (p = .066) but not in the efficacy condition (p = .689). Second, the effect of replotting on counterarguing appears to be higher in the no-efficacy condition (standardized regression weight = −.13) compared to the efficacy condition (standardized regression weight = −.08).

Figure 1.

Study 1 replotting anger structural model.

Study 2

Study 1 had three key limitations. First, replotting was measured using a single item. Although single-item measures can be effective for some constructs, replotting is more nuanced. Specifically, measurement should account for both the cognitive act of replotting and the accompanying emotion. Second, Study 1 focused only on anger. Replotting is likely linked to other emotions, such as anxiety or sadness. Thirdly, the efficacy manipulation was in a non-narrative format rather than embedded in the story, and its effect on perceived efficacy was just barely significant (p = .045). Therefore, a second study was conducted using a multi-item measure of replotting with anger, anxiety, and sadness and a narrative efficacy appeal.

Methods

Participants and procedures

Participants were enrolled via Qualtrics Panels into a web-based experiment study. Participants were randomized into one of eight conditions in a 2 (story outcome: death vs. survivor) × 2 (efficacy appeal: present vs. absent) × 2 (story: Alex or Fern) between-participants study design. Power was calculated using G*Power (Faul et al., 2009). To achieve .80 power to detect small effect sizes (f = .10), a priori power analysis identified a target sample size of 787 for the full model. A total of 803 participants completed the study. Data were removed if participants spent less than 5 seconds on the narrative and/or were unable to answer close-ended yes/no questions about the narrative including “In the story you just read, did one of the characters die from melanoma” and “In the story you just read, did one of the characters decide to start doing sun safe behaviors (like wearing sunscreen or doing skin self-exams).” Data from 87 participants were removed, resulting in a total sample of 716 participants. The final sample was just short of the target for the full 2 × 2 × 2 model.

Participants were on average 58.54 years old (age range: 18–93), 89.1% identified as Caucasian (Black or African American: 7.4%; Asian or Asian American: 2.5%; American Indian or Alaskan Native: 1.4%; Other: 1.5%), and 4.2% identified as Hispanic or Latina/o. Participants were evenly split between those with high school education or less (46.6%) and those with more than high school education (53.4%) and between females (52.0%) and males (48.0%).

Stimuli

A key goal of Study 2 was to strengthen the efficacy appeal by integrating it into the narrative, removing a potential confound presented in Study 1 with the non-narrative format. The efficacy appeal for Study 2 included a secondary character who decides to focus on sun safe behavior as a result of the main character’s melanoma diagnosis:

After hearing about [main character’s] diagnosis, [friend] became more careful about protecting [his/her] skin from melanoma. It was easy to do – wearing sunscreen and protective clothing when [s/he] was outside, avoiding tanning, and checking [her/his] skin for suspicious moles. [Friend] just hadn’t realized how important it was before for preventing cancer.

Prior to launching Study 2, a pilot study was conducted using a separate Qualtrics Panels sample (N = 888) to test if the narrative efficacy appeal worked for all six stories used in Study 1. Success of the efficacy appeal was assessed using the same manipulation check efficacy measure from Study 1. The efficacy manipulation was successful for three of the six narratives. Specifically, the narrative efficacy appeal produced greater perceived efficacy for the Alex, F(1, 864) = 7.70, p = .005, Fern, F(1, 864) = 12.74, p < .001, and Erin, F(1, 864) = 5.43, p = .02, narratives. Given concerns about the relatability of the Erin narrative across participants (Erin frequently used tanning beds), the authors utilized the Alex and Fern narratives for Study 2 (see Supplemental Appendix).

Measures

Sun safe behavior intention, message elaboration, and counterarguing

The same measures from Study 1 were used to measure sun safe behavior (SSB) intention, (α = .89, M = 5.38, SD = 1.29), message elaboration (α = .84, M = 3.46, SD = 0.79; reverse-coded items were removed), and counterarguing (α = .68, M = 2.38, SD = 0.71).

Replotting

A new measure of replotting was developed for Study 2. Replotting was modeled as a second-order latent variable comprised of a cognitive and an affective component. Before answering replotting items, participants were given a brief definition of replotting to ensure they understood what was being asked of them. Specifically, they were informed that replotting “means imagining different ways a story could have gone, including things characters could have done differently to change story events.” The cognitive component of replotting was assessed with four items on a scale from 1 (strongly disagree) to 5 (strongly agree), including “I imagined how the plot could have been different,” “I thought about things the characters could have done differently,” “I replotted the story as it was unfolding, in ways that could have resulted in a more pleasant outcome,” and “I imagined how unpleasant events in the story could have been avoided” (α = .93, M = 3.44, SD = 1.19). Participants were then asked: “To what extent did you feel each of the following emotions when replotting?” and received the instruction that “If you did not replot the story, indicate that you did not really feel these emotions.” Participants responded on a scale from 1 (none of this emotion) to 7 (a great deal of this emotion) to items capturing anger (irritated, angry, annoyed, aggravated, outraged, and infuriated; α = .96, M = 2.77, SD = 1.86), anxiety (tense, nervous, and anxious; α = .95, M = 2.85, SD = 1.90), and sadness (sad, dreary, and dismal; α = .83, M = 2.81, SD = 1.70). Emotion items were derived from Dillard and Shen (2007) and Lillie, Jensen, et al. (2021).

Previous sun safe behavior

Study 2 added a measure of previous sun safe behavior (SSB) as a control. Four SSB items were embedded in the Good Health Practices scale (Hampson et al., 2019). On a scale from 1 (not at all like me) to 5 (very much like me) participants indicated if the following behaviors were typical of them: “I avoid peak UV exposure from 10 AM to 4 PM,” “I use sunscreen regularly,” “I wear protective clothing (long pants and sleeves) when outside,” and “I check my body regularly for strange moles” (α = .73, M = 2.84, SD = 1.06). For means and standard deviations by study condition, see Tables 3 and 4.

Table 3.

Study 2 Means and Standard Deviations by Study Condition.

Variable	Death narratives	Survivor narratives	Death narratives		Survivor narratives
Variable	Death narratives	Survivor narratives	High efficacy	Low efficacy	High efficacy	Low efficacy
Cognitive replotting	3.67 (1.19)	3.25 (1.16)	3.55 (1.18)	3.77 (1.20)	3.18 (1.21)	3.31 (1.11)
Anger	3.06 (1.89)	2.52 (1.79)	2.76 (1.77)	3.31 (1.95)	2.47 (1.81)	2.57 (1.78)
Anxiety	2.99 (1.94)	2.73 (1.85)	2.77 (1.92)	3.17 (1.95)	2.61 (1.81)	2.83 (1.88)
Sadness	3.34 (1.71)	2.35 (1.56)	3.26 (1.71)	3.41 (1.70)	2.28 (1.50)	2.42 (1.60)
Counterarguing	2.41 (0.76)	2.35 (0.66)	2.22 (0.70)	2.57 (0.78)	2.24 (0.63)	2.44 (0.67)
Elaboration	3.49 (0.79)	3.44 (0.78)	3.40 (0.80)	3.56 (0.79)	3.43 (0.76)	3.45 (0.80)
SSB^a intention	5.44 (1.30)	5.33 (1.29)	5.51 (1.24)	5.38 (1.36)	5.42 (1.22)	5.26 (1.34)

Note. Means with standard deviations in parentheses. Bolded pairs represent significant main effects of death/survivor or interaction effects with efficacy condition at p < .001.

Sun safe behavior.

Table 4.

Study 2 Means and Standard Deviations by Story.

Variable	Fern	Alex	Fern		Alex
Variable	Fern	Alex	Death narrative	Survivor narrative	Death narrative	Survivor narrative
Cognitive replotting	3.46 (1.21)	3.42 (1.17)	3.62 (1.25)	3.33 (1.17)	3.72 (1.14)	3.17 (1.14)
Anger	3.27 (1.98)	2.26 (1.57)	3.46 (1.95)	3.10 (2.00)	2.65 (1.74)	1.94 (1.34)
Anxiety	3.01 (1.97)	2.69 (1.81)	3.07 (1.97)	2.96 (1.98)	2.91 (1.92)	2.51 (1.69)
Sadness	2.90 (1.71)	2.71 (1.68)	3.33 (1.69)	2.53 (1.65)	3.36 (1.73)	2.17 (1.44)
Counterarguing	2.40 (0.71)	2.36 (0.70)	2.46 (0.78)	2.35 (0.65)	2.37 (0.74)	2.35 (0.67)
Elaboration	3.48 (0.83)	3.44 (0.74)	3.52 (0.85)	3.46 (0.81)	3.46 (0.73)	3.42 (0.75)
SSB^a intention	5.33 (1.37)	5.44 (1.22)	5.31 (1.39)	5.34 (1.35)	5.58 (1.20)	5.32 (1.22)

Note. Means with standard deviations in parentheses. Bolded pairs represent significant main effects of Fern/Alex or interaction effects with story outcome condition at p < .05.

Sun safe behavior.

Manipulation check

For Study 2, success of the narrative efficacy message was assessed using the same manipulation check measure from Study 1 and the pilot (α = .83, M = 5.28, SD = 1.00). Confirming that the manipulation was successful, there was a main effect of efficacy condition, F(1, 712) = 4.02, p = .045. Participants who received the efficacy message reported greater efficacy (M = 5.37, SE = 0.06) compared to those who did not receive the efficacy message (M = 5.21, SE = 0.05, r = .07).

Results

In-line with Study 1, the hypothesized model was tested with SEM in SPSS Amos version 27, using Anderson and Gerbing’s (1988) two-step process for SEM and multi-group SEM. Unlike Study 1, prior sun safe behavior was included in the model as a control variable. Separate models were tested for the three emotions, including anger, anxiety, and sadness to determine if effects differed for the three emotions. Prior to model testing, bivariate correlations were calculated to examine zero order relationships between study variables (Table 5).

Table 5.

Study 2 Correlation Matrix.

	Variable	1	2	3	4	5	6	7	8	9
1	Story outcome	—
2	Efficacy condition	−.01	—
3	Cognitive replotting	.17±	−.07*	—
4	Anger	.14±	−.08*	.48±	—
5	Anxiety	.07	−.08*	.44±	.67±	—
6	Sadness	.29±	−.05	.46±	.66±	.72±	—
7	Counterarguing	.04	−.19±	.11**	.14±	.13±	.15±	—
8	Elaboration	.03	−.05	.55±	.43±	.49±	.46±	−.03	—
9	Previous SSB^a	.02	.01	.21±	.25±	.24±	.20±	−.02	.32±	—
10	SSB^a intention	.04	.06	.27±	.25±	.24±	.28±	−.12**	.37±	.62±

Note. For story outcome: survivor = 0, death = 1.

Sun safe behavior.

p ≤ .05. **p ≤ .01. ±p ≤ .001.

The measurement models achieved acceptable fit (see Figures 2 –4). Multi-group SEM identified a significant difference between the constrained and unconstrained structural models for anger, χ² = 136.47, df = 79, p < .001, anxiety, χ² = 120.76, df = 73, p < .001, and sadness, χ² = 120.14, df = 73, p < .001. Therefore, the efficacy and no-efficacy conditions were significantly different. All three structural models achieved acceptable fit (see Figures 2 –4). Regarding H1–5, death narratives produced greater replotting anger and replotting anxiety compared to survivor narratives in the no-efficacy condition only, and greater replotting sadness for both efficacy conditions, partially supporting H1. Replotting was positively related to message elaboration, supporting H2, and negatively related to counterarguing, supporting H4. Message elaboration was positively related to SSB intention in the no-efficacy condition only, partially supporting H3. Counterarguing was negatively related to SSB intention in the efficacy condition only, partially supporting H5. Three key differences emerged between the efficacy conditions. First, for the anger and anxiety models, death narratives only had a significant effect on replotting in the no-efficacy condition. Second, message elaboration was only related to SSB intention in the no-efficacy condition. Third, counterarguing was only related to SSB intention in the efficacy condition. Therefore, H6–8 were rejected.

Figure 2.

Study 2 replotting anger structural model.

Figure 3.

Study 2 replotting anxiety structural model.

Figure 4.

Study 2 replotting sadness structural model.

Discussion

The current research explicated the concept of replotting, including its cognitive and affective components. Findings suggest that audiences engage in replotting when confronted with an undesirable story outcome, such as death. Replotting influenced behavioral intention through greater message elaboration and reduced counterarguing. However, death narratives only produced greater replotting anger and anxiety when story characters did not engage in replotting. Findings have important implications for the inclusion of efficacy appeals in narratives.

Death Narratives and Replotting

Death narratives triggered greater replotting anger and anxiety compared to survivor narratives when a narrative efficacy appeal was not included in the story and greater replotting sadness regardless of efficacy condition. Replotting involves imagining a different outcome for a narrative by finding ways the outcome could have been changed (Gerrig, 1993). It makes sense that audiences would have a stronger motivation to replot for more negative story outcomes, such as character death. The death of the main character was foreshadowed, potentially triggering replotting during story consumption. Past research has found that death narratives are more persuasive when they include foreshadowing (e.g., Lillie, Pokharel, et al., 2021). It is possible that replotting is one of the mechanisms that makes foreshadowed death narratives more effective. Perceived realism could play a role as well. The narratives were based on actual cancer stories. More realistic narratives could result in a greater urge to replot, making perceived realism an important moderator for future research to examine.

Whether death narratives produced greater replotting anger and anxiety was dependent upon the narrative efficacy appeal. The narrative efficacy appeal included a friend deciding that sun safe behaviors were important in response to the main character’s melanoma diagnosis. That this hindered the replotting process is interesting because narratives are effective in part because they model desired behaviors (Bandura, 1977; Miller-Day & Hecht, 2013). Transitional characters that alter their behavior to be in-line with the message position are particularly powerful (Sabido, 2004). Yet, having a transitional character advocate for sun safe behavior diminished the death narratives’ persuasive power. A key element of replotting is the cognitive effort audiences put toward imagining alternative outcomes. The narrative efficacy appeal essentially performed the replotting for the audience, potentially impeding the replotting process by doing it for the audience rather than letting them do it themselves (Tal-Or et al., 2004).

It is interesting that death narratives produced greater replotting sadness with or without the narrative efficacy appeal and that sadness was subsequently persuasive. Sadness has received less attention in the literature compared with anger and anxiety. Previous research has connected death narratives with sadness, but sadness is often detrimental to persuasion (e.g., Banerjee & Greene, 2012). Sadness is expected to be debilitating rather than motivating. However, when sadness was related to the cognitive effort of replotting, it increased message elaboration. Future research should confirm these results in other contexts as well as test the effects of positive emotions related to replotting, such as hope.

Narrative Efficacy Statements

The current research raises questions about the creation and inclusion of efficacy appeals in narrative messaging. Study 1 had utilized a non-narrative efficacy appeal after the story. However, this may have caused a disconnect between the narrative and the efficacy appeal, diminishing its effectiveness. Indeed, although the non-narrative efficacy appeal significantly affected perceived efficacy, the effect was small (r = .05).

When piloting a narrative efficacy appeal, the manipulation only had a significant effect on perceived efficacy for the Erin, Fern, and Alex narratives. The no-efficacy conditions for these stories produced noticeably lower perceived efficacy (4.86–5.15) than both the efficacy and no-efficacy conditions for the Don and Ray stories (5.31–5.54). This suggests that either the Don and Ray stories already included elements that bolstered efficacy or that the Erin, Fern, and Alex stories included elements that diminished efficacy. Both the Erin and Fern stories included delay. Erin waited to go to the doctor because she did not have medical insurance, and Fern’s first doctor did not take her concerns seriously. It is possible that the inclusion of delay caused by external forces may diminish feelings of efficacy. Therefore, having a character advocate for the effectiveness of prevention and early detection behaviors was necessary.

Practically, health communicators must weigh the trade-offs of including or not including narrative efficacy appeals. Story elements like medical delay could diminish audience efficacy. This may be less of a concern when the audience already has a high level of perceived efficacy. For example, in Study 2 the average perceived efficacy for the narratives without an efficacy appeal was still above the midpoint (M = 5.21, SD = 1.04). Importantly, inclusion of an efficacy appeal meant that death narratives did not trigger greater replotting anger or anxiety, a process that was subsequently related to reduced counterarguing and greater message elaboration and SSB intention. Therefore, in a context in which efficacy is already high, inclusion of a narrative efficacy appeal may not be an effective messaging strategy.

Message Elaboration and Counterarguing

The current study tested the relationship between replotting and both message elaboration and counterarguing. Findings validate the theoretical claim that message-relevant anger increases message elaboration, leading to persuasion (Nabi, 1999; Turner, 2007). Findings also demonstrate that message-relevant anger leads to diminished counterarguing. Interestingly, replotting was positively related to message elaboration and negatively related to counterarguing regardless of which replotting emotion (anger, anxiety, or sadness) was examined. This finding runs counter to previous research that has found differing effects for these discrete emotions (e.g., Durkin et al., 2018; Nabi, 2002), suggesting that emotion functions differently when tied to the cognitive act of replotting. However, it is important to note that there were some differences in the magnitude of the relationships. For example, in the no-efficacy conditions, replotting anger (estimate = −.09, p = .037) and replotting sadness (estimate = −.09, p = .047) had only marginally significant relationships with counterarguing while the association between replotting anxiety and counterarguing was stronger (estimate = −.14, p < .001).

Replotting may be effective because of the cognitive effort involved in replotting, encouraging message elaboration and diminishing counterarguing. Additionally, individuals who engage in replotting the story outcome have become engrossed in the story and perhaps feel empathy or a parasocial connection to the character who died (Green & Brock, 2000)—elements of narrative engagement that reduce counterarguing (Ratcliff & Sun, 2020). Such narrative engagement is theorized to reduce counterarguing by diminishing the extent to which the audience perceives the persuasive subtext or processes the message critically (Green & Brock, 2000; Moyer-Gusé, 2008; Slater & Rouner, 2002). Future research should examine replotting in the context of parasocial relationships, which are both deeper and occur over a longer term. More intense forms of replotting, such as creating fanfiction, could be particularly powerful.

Interestingly, message elaboration was only related to behavioral intention when there was not a narrative efficacy appeal. Conversely, counterarguing was only related to behavioral intention when there was a narrative efficacy appeal. The narrative efficacy appeal essentially replotted the story for the audience via a secondary character. When the story replots for the audience, persuasion may be dependent on whether the audience argues against that in-story replot. When audiences have to do the replotting themselves, persuasion may depend on the extent to which they elaborate on how the replot relates to their lives.

Limitations

Although the current study provides useful insight into messaging and emotion theory, it is not without limitations. Measures were taken after message exposure. In order to determine when replotting occurs during message consumption, researchers should utilize data collection methods that allow for real-time emotion measurement during message exposure (Nabi, 2015). Additionally, the studies tested the effects of replotting in a single context, melanoma prevention. Replotting and efficacy both may function differently in contexts where preventative measures are less widely known. Researchers could assess participant knowledge of preventative behaviors prior to message exposure (Nabi et al., 2008). Future research should formally validate the replotting measure, with a focus on demonstrating how it is distinct from anger, anxiety, and sadness more broadly. This research could also test whether the effects of efficacy appeals differ for replotting emotions compared to general discrete emotions. Further, it would be interesting to test the longitudinal effects of replotting on behavior.

Conclusion

This set of studies examined the persuasive effects of replotting in the context of melanoma. Death narratives generated greater replotting anger and anxiety when a narrative efficacy appeal was not included and greater replotting sadness across efficacy conditions. Replotting motivated audiences to accept the behavioral recommendations of the message by increasing message elaboration and inhibiting counterarguing. Whether message elaboration or counterarguing ultimately influenced intentions was dependent upon the narrative efficacy appeal. Persuasion scholars must weigh the value of boosting efficacy against encouraging replotting anger and anxiety when deciding whether to use narrative efficacy appeals.

Supplemental Material

sj-docx-1-crx-10.1177_00936502221114308 – Supplemental material for Encouraging Replotting to Promote Persuasion: How Imagining Alternative Plotlines Influences Message Processing and Intentions

Supplemental material, sj-docx-1-crx-10.1177_00936502221114308 for Encouraging Replotting to Promote Persuasion: How Imagining Alternative Plotlines Influences Message Processing and Intentions by Helen M. Lillie, Chelsea L. Ratcliff, Manusheela Pokharel and Jakob D. Jensen in Communication Research

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This manuscript was written with support from NIH grant 1DP2EB022360-01 (PI: J. Jensen) and 3P30CA042014-29S7 (PI: J. Jensen).

ORCID iDs

Helen Lillie

Jakob D. Jensen

Supplemental Material

Supplemental material for this article is available online.

Notes

Author Biographies

Helen Lillie (PhD, Purdue University) is an Assistant Professor in the Department of Communication Studies at the University of Iowa.

Chelsea L. Ratcliff (PhD, University of Utah) is an Assistant Professor in the Department of Communication Studies at the University of Georgia.

Manusheela Pokharel (PhD, University of Utah) is an Assistant Professor in the Department of Communication Studies at Texas State University.

Jakob D. Jensen (PhD, University of Illinois) is a Professor in the Department of Communication at the University of Utah.

References

Anderson

J. C.

Gerbing

D. W.

(1988). Structural equation modeling in practice: A review and recommended two-step approach. Psychological Bulletin, 103(3), 411–423. https://doi.org/10.1037/0033-2909.103.3.411

Bandura

(1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84(2), 191–215. https://doi.org/10.1037/0033-295X.84.2.191

Bandura

(1982). Self-efficacy mechanism in human agency. American Psychologist, 37(2), 122–147. https://doi.org/10.1037/0003-066X.37.2.122

Banerjee

S. C.

Greene

(2012). Role of transportation in the persuasion process: Cognitive and affective responses to antidrug narratives. Journal of Health Communication, 17(5), 564–581. https://doi.org/10.1080/10810730.2011.635779

Bigsby

Albarracín

(2022). Self-and response efficacy information in fear appeals: A meta-analysis. Journal of Communication, 72(2), 241–263. https://doi.org/10.1093/joc/jqab048

Bilandzic

Busselle

(2013). Narrative persuasion. In Dillard

J. P.

Shen

(Eds.), The SAGE handbook of persuasion: Developments in theory and practice (pp. 200–219). SAGE.

Braddock

Dillard

J. P.

(2016). Meta-analytic evidence for the persuasive effect of narratives on beliefs, attitudes, intentions, and behaviors. Communication Monographs, 83(4), 446–467. https://doi.org/10.1080/03637751.2015.1128555

Briñol

Petty

R. E.

(2012). The history of attitudes and persuasion research. In Kruglanski

A. W.

Stroebe

(Eds.), Handbook of the history of social psychology (pp. 285–320). Psychology Press.

Cialdini

R. B.

Petty

R. E.

Cacioppo

J. T.

(1981). Attitude and attitude change. Annual Review of Psychology, 32(1), 357–404. https://doi.org/10.1146/annurev.ps.32.020181.002041

10.

Cohen

(1992). A power primer. Psychological Bulletin, 112(1), 155–159. https://doi.org/10.1037/0033-2909.112.1.155

11.

Dillard

J. P.

Shen

(2005). On the nature of reactance and its role in persuasive health communication. Communication Monographs, 72(2), 144–168. https://doi.org/10.1080/03637750500111815

12.

Dillard

J. P.

Shen

(2007). Self-report of discrete emotions. In Reynolds

R. A.

Woods

Baker

J. D.

(Eds.), Handbook of research on electronic surveys and measurements (pp. 330–333). Idea Group Reference.

13.

Durkin

Bayly

Brennan

Biener

Wakefield

(2018). Fear, sadness and hope: Which emotions maximize impact of anti-tobacco mass media advertisements among lower and higher SES groups?. Journal of Health Communication, 23(5), 445–461. https://doi.org/10.1080/10810730.2018.1463320

14.

Else-Quest

N. M.

Jackson

T. L.

(2014). Cancer stigma. In Corrigan

P. W.

(Ed.), The stigma of disease and disability: Understanding causes and overcoming injustices (pp. 165–181). American Psychological Association

15.

Faul

Erdfelder

Buchner

Lang

A. G.

(2009). Statistical power analyses using G*Power 3.1: Tests for correlation and regression analyses. Behavior Research Methods, 41(4), 1149–1160. https://doi.org/10.3758/BRM.41.4.1149

16.

Gerrig

R. J.

(1989). Suspense in the absence of uncertainty. Journal of Memory and Language, 28(6), 633–648. https://doi.org/10.1016/0749-596X(89)90001-6

17.

Gerrig

R. J.

(1993). Experiencing narrative worlds: On the psychological activities of reading. Yale University Press.

18.

Gerrig

R. J.

Bezdek

M. A.

(2013). The role of participation in aesthetic illusion. In Wolf

Bernhart

Mahler

(Eds.), Immersion and distance: Aesthetic illusion in literature and other media (pp. 89–112). Rodopi.

19.

Green

M. C.

Brock

T. C.

(2000). The role of transportation in the persuasiveness of public narratives. Journal of Personality and Social Psychology, 79(5), 701–721. https://doi.org/10.1037/0022-3514.79.5.701

20.

Hampson

S. E.

Edmonds

G. W.

Goldberg

L. R.

(2019). The health behavior checklist: Factor structure in community samples and validity of a revised good health practices scale. Journal of Health Psychology, 24(8), 1103–1109. https://doi.org/10.1177/1359105316687629

21.

Hinyard

L. J.

Kreuter

M. W.

(2007). Using narrative communication as a tool for health behavior change: A conceptual, theoretical, and empirical overview. Health Education & Behavior, 34(5), 777–792. https://doi.org/10.1177/1090198106291963

22.

Holbert

R. L.

Park

(2020). Conceptualizing, organizing, and positing moderation in communication research. Communication Theory, 30(3), 227–246. https://doi.org/10.1093/ct/qtz006

23.

L. T.

Bentler

P. M.

(1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal, 6(1), 1–55. https://doi.org/10.1080/10705519909540118

24.

Jackson

Brashers

D. E.

(1994). Random factors in ANOVA. SAGE.

25.

Janicke

S. H.

Raney

A. A.

(2018). Modeling the antihero narrative enjoyment process. Psychology of Popular Media Culture, 7(4), 533–546. https://doi.org/10.1037/ppm0000152

26.

Jensen

J. D.

Yale

R. N.

Krakow

John

K. K.

King

A. J.

(2017). Theorizing foreshadowed death narratives: Examining the impact of character death on narrative processing and skin self-exam intentions. Journal of Health Communication, 22(1), 84–93. https://doi.org/10.1080/10810730.2016.1252816

27.

Kline

R. B.

(2016). Principles and practice of structural equation modeling (4th ed.). The Guilford Press.

28.

Krakow

M. M.

Yale

R. N.

Jensen

J. D.

Carcioppolo

Ratcliff

C. L.

(2018). Comparing mediational pathways for narrative-and argument-based messages: Believability, counterarguing, and emotional reaction. Human Communication Research, 44(3), 299–321. https://doi.org/10.1093/hcr/hqy002

29.

Lazarus

R. S.

(1991). Emotion and adaptation. Oxford University Press.

30.

Lerner

M. J.

Miller

D. T.

(1978). Just world research and the attribution process: Looking back and ahead. Psychological Bulletin, 85(5), 1030–1051. https://doi.org/10.1037/0033-2909.85.5.1030

31.

Lillie

H. M.

Jensen

J. D.

Pokharel

Upshaw

S. J.

(2021). Death narratives, negative emotion, and counterarguing: Testing fear, anger, and sadness as mechanisms of effect. Journal of Health Communication, 26(8), 586–595. https://doi.org/10.1080/10810730.2021.1981495

32.

Lillie

H. M.

Pokharel

John

K. K.

Christy

K. R.

Upshaw

Giorgi

E. A.

Jensen

J. D.

(2021). Does it matter if a story character lives or dies? A message experiment comparing survivor and death narratives. Psychology & Health, 37(4), 419–439. https://doi.org/10.1080/08870446.2021.1873337

33.

(2020). Exploring the role of incidental and integral compassion and anger in health communication about pollution. Health Communication, 35(6), 675–684. https://doi.org/10.1080/10410236.2019.1582313

34.

Miller-Day

Hecht

M. L.

(2013). Narrative means to preventative ends: A narrative engagement framework for designing prevention interventions. Health Communication, 28(7), 657–670. https://doi.org/10.1080/10410236.2012.762861

35.

Moons

W. G.

Mackie

D. M.

(2007). Thinking straight while seeing red: The influence of anger on information processing. Personality and Social Psychology Bulletin, 33(5), 706–720. https://doi.org/10.1177/0146167206298566

36.

Moyer-Gusé

(2008). Toward a theory of entertainment persuasion: Explaining the persuasive effects of entertainment-education messages. Communication Theory, 18(3), 407–425. https://doi.org/10.1111/j.1468-2885.2008.00328.x

37.

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(3), 292–320 https://doi.org/10.1111/j.1468-2885.1999.tb00172.x

38.

Nabi

R. L.

(2002). Anger, fear, uncertainty, and attitudes: A test of the cognitive-functional model. Communication Monographs, 69(3), 204–216. https://doi.org/10.1080/03637750216541

39.

Nabi

R. L.

(2015). Emotional flow in persuasive health messages. Health Communication, 30(2), 114–124. https://doi.org/10.1080/10410236.2014.974129

40.

Nabi

R. L.

Green

M. C.

(2015). The role of a narrative’s emotional flow in promoting persuasive outcomes. Media Psychology, 18(2), 137–162. https://doi.org/10.1080/15213269.2014.912585

41.

Nabi

R. L.

Moyer-Gusé

Byrne

(2007). All joking aside: A serious investigation into the persuasive effect of funny social issue messages. Communication Monographs, 74(1), 29–54. https://doi.org/10.1080/03637750701196896

42.

Nabi

R. L.

Roskos-Ewoldsen

Dillman Carpentier

(2008). Subjective knowledge and fear appeal effectiveness: Implications for message design. Health Communication, 23(2), 191–201. https://doi.org/10.1080/10410230701808327

43.

Nan

(2017). Influence of incidental discrete emotions on health risk perception and persuasion. Health Communication, 32(6), 721–729. https://doi.org/10.1080/10410236.2016.1168004

44.

Niederdeppe

Kim

H. K.

Lundell

Fazili

Frazier

(2012). Beyond counterarguing: Simple elaboration, complex integration, and counterelaboration in response to variations in narrative focus and sidedness. Journal of Communication, 62(5), 758–777. https://doi.org/10.1111/j.1460-2466.2012.01671.x

45.

S. H.

Lee

S. Y.

Han

(2020). The effects of social media use on preventive behaviors during infectious disease outbreaks: The mediating role of self-relevant emotions and public risk perception. Health Communication, 36(8), 972–981. https://doi.org/10.1080/10410236.2020.1724639

46.

Petty

R. E.

Cacioppo

J. T.

(1986). Communication and persuasion: Central and peripheral routes to attitude change. Springer-Verlag.

47.

Ratcliff

C. L.

(2021). Characterizing reactance in communication research: A review of conceptual and operational approaches. Communication Research, 48(7), 1033–1058. https://doi.org/10.1177/0093650219872126

48.

Ratcliff

C. L.

Sun

(2020). Overcoming resistance through narratives: Findings from a meta-analytic review. Human Communication Research, 46(4), 412–443. https://doi.org/10.1093/hcr/hqz017

49.

Reynolds

R. A.

(1997). A validation test of a message elaboration measure. Communication Research Reports, 14(3), 269–278. https://doi.org/10.1080/08824099709388670

50.

Sabido

(2004). The origins of entertainment-education. In Singhal

Cody

M. J.

Rogers

E. M.

Sabido

(Eds.), Entertainment-education and social change: History, research, and practice (pp. 61–74). Lawrence Erlbaum Associates

51.

Schermelleh-Engel

Moosbrugger

Müller

(2003). Evaluating the fit of structural equation models: Tests of significance and descriptive goodness-of-fit measures. Methods of Psychological Research, 8(2), 23–74.

52.

Shaffer

V. A.

Focella

E. S.

Hathaway

Scherer

L. D.

Zikmund-Fisher

B. J.

(2018). On the usefulness of narratives: An interdisciplinary review and theoretical model. Annals of Behavioral Medicine, 52(5), 429–442. https://doi.org/10.1093/abm/kax008

53.

Shen

Sheer

V. C.

(2015). Impact of narratives on persuasion in health communication: A meta-analysis. Journal of Advertising, 44(2), 105–113. https://doi.org/10.1080/00913367.2015.1018467

54.

Shen

Seung

S. Y.

(2018). On measures of message elaboration in narrative communication. Communication Quarterly, 66(1), 79–95. https://doi.org/10.1080/01463373.2017.1334682

55.

Shields

G. S.

Moons

W. G.

Tewell

C. A.

Yonelinas

A. P.

(2016). The effect of negative affect on cognition: Anxiety, not anger, impairs executive function. Emotion, 16(6), 792–797. https://doi.org/10.1037/emo0000151

56.

Slater

M. D.

Rouner

(2002). Entertainment-education and elaboration likelihood: Understanding the processing of narrative persuasion. Communication Theory, 12(2), 173–191. https://doi.org/10.1111/j.1468-2885.2002.tb00265.x

57.

Tal-Or

Boninger

D. S.

Poran

Gleicher

(2004). Counterfactual thinking as a mechanism in narrative persuasion. Human Communication Research, 30(3), 301–328. https://doi.org/10.1111/j.1468-2958.2004.tb00734.x

58.

Turner

M. M.

(2007). Using emotion in risk communication: The anger activism model. Public Relations Review, 33(2), 114–119. https://doi.org/10.1016/j.pubrev.2006.11.013

59.

Turner

M. M.

Richards

A. S.

Bessarabova

Magid

(2020). The effects of anger appeals on systematic processing and intentions: The moderating role of efficacy. Communication Reports, 33(1), 14–26. https://doi.org/10.1080/08934215.2019.1682175

60.

Walter

Cohen

(2019). When less is more and more is less: The paradoxical metacognitive effects of counterarguing. Communication Monographs, 86(3), 377–397. https://doi.org/10.1080/03637751.2019.1580378

61.

Walter

Demetriades

S. Z.

Nabi

R. L.

(2021). Seeing red through rose-colored glasses: Subjective hope as a moderator of the persuasive influence of anger. Journal of Communication, 71(1), 79–103. https://doi.org/10.1093/joc/jqaa037

62.

Walter

Tukachinsky

Pelled

Nabi

(2019). Meta-analysis of anger and persuasion: An empirical integration of four models. Journal of Communication, 69(1), 73–93. https://doi.org/10.1093/joc/jqy054

63.

Wegener

D. T.

Petty

R. E.

Smoak

N. D.

Fabrigar

L. R.

(2004). Multiple routes to resisting attitude change. In Knowles

E. S.

Linn

J. A.

(Eds.), Resistance and persuasion (pp. 13–38). Lawrence Erlbaum Associates.

64.

Witte

(1992). Putting the fear back into fear appeals: The extended parallel process model. Communication Monographs, 59(4), 329–349. https://doi.org/10.1080/03637759209376276

65.

Zebregs

van den Putte

Neijens

de Graaf

(2015). The differential impact of statistical and narrative evidence on beliefs, attitude, and intention: A meta-analysis. Health Communication, 30(3), 282–289. https://doi.org/10.1080/10410236.2013.842528

66.

Zillmann

(2000). Basal morality in drama appreciation. In Bondebjerg

(Ed.), Moving images, culture, and the mind (pp. 53–63). University of Luton Press.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.02 MB