Master’s programme in Communication Science Graduate School of Communication
Master Thesis
Is Facereader Applicable in Studying Political Satire Processing?
Exploring the Relationships Among Political Preference, Openness to
Experience, Expression of Happiness, and Attitudes Toward the Satirized
Object.
Student: Yuan Xi
Student number: 11181559 Supervisor: Marieke Fransen
Abstract
Using facial expression as a pioneer to study political satire, this research explores the relationship between one’s political preference and expression of happiness, as well as
subsequent attitude toward the satirized object. A new individual characteristic, openness to experience, is also proposed as a moderator. Based on the disposition theory of humor, this study hypothesizes that viewers will show more expressions of happiness toward a satirical message when its intended meaning is congruent with their political preferences, rather than when it is incongruent (H1). It further speculates when viewers are more open to experiences,
they are more likely to appreciate political satire that is contrary to their current ideology (H2).
This study also explores whether one’s emotional response toward political satire, as indicated by expressions of happiness, can influence an attitude toward the satirized object. Eighty-four female participants were recruited for the study where their faces were recorded while viewing a video about the Me Too movement. These recordings were analyzed by the FaceReader software provided by the Noldus company. Though there are no significant effects, the hypothesized directions of H1 and H2 were correct. This research demonstrates
that the two constructs (i.e., self-report humor appreciation and the expressions of happiness) may theoretically differ. It also suggests a possibility to reconsider the underlying mechanism of humor processing. Other theoretical and practical implications are also discussed.
Keywords: Political Satire, Facial Expression, Facereader, Openness to Experience, Attitudes, Persuasive Effect
Humor is ubiquitous and studied in many political contexts. Research has
demonstrated that media contents of political humor can influence the audiences’ knowledge (Warner, Hawthorne, & Hawthorne, 2015), attitude (Polk, Young, & Holbert, 2009), and engagement (Bore & Reid, 2014). Among the various types of political humor, political satire, a particular form of humor that often attacks a certain object (Colletta, 2009) such as a
politician, has become a popular research focus (e.g., Colletta, 2009; Higgie, 2017; Hmielowski, Holbert, & Lee, 2011). The rising influence of this genre is apparent in late-night talk shows and news satire TV programs, such as the Daily Show, The Colbert Report, or Saturday Night Live. Drawing their comedies from recent news stories and relevant figures, these TV programs have attracted attention from many scholars
(Baumgartner & Morris, 2006; Baym, 2005; Lamarre, Landreville, & Beam, 2009). With an audience dominated by young people, political satire is considered a tool for engaging the youth in civic events, which closes the age gap in political attentiveness between younger and older citizens (Hmielowski et al., 2011). It also makes a potentially significant democratic contribution by offering perspectives that differ from other serious media contents (Holbert, 2013).
In the line of political satire literature, there are many scholars investigating how people process this kind of humorous message and its consequence of the influence on the attitude toward the satirized object. Boukes, Boomgaarden, Moorman, & de Vreese (2015) elaborated upon the mechanism. They identified two different mediators that evoke and inhibit counterarguing of the criticism put forth in satire following the construct of perceived funniness. And no overall persuasion effect was revealed.
This study argues that the role of the emotional response to humor could be further explored in humor process studies. Instead of asking participants to report their evaluation of humor (i.e., perceive it as funny or not), as previous researchers have done, a more direct measure could be introduced for responding to humor. One could speculate that the
self-report as an after-exposure measurement could be inaccurate to measure if one enjoys the humor or not. Viewers may laugh at a humorous plot because of its amusing nature, but later
report it as not funny, since the intended meaning is not congruent with their ideology. To further explore the humor processing mechanism, this study will introduce FaceReader technology to detect viewers’ facial expressions while they are exposed to political satire. Different from the self-report measure, facial expression reading is a measurement that can reflect the viewers’ real-time spontaneous reactions. This paper will thus investigate whether the facial expression can be adopted as a construct in the study of political satire processing and possibly predict attitude toward the satirized object. The self-report measurement will also be included to explore if these two constructs are related or different.
Also, previous studies based on the disposition theory of humor have found that viewers appreciate the humor more when its intended meaning is congruent with their
political preference (Becker, 2014; Boukes et al., 2015). Viewers may find the humor funnier when they agree with its underlying meaning. Except for predisposition, there are other individual characteristics that can influence one’s appreciation of humor. Peifer & Holbert (2016) have called for attention to the role of openness to experience in political humor processing. They argued that the liberals are more likely to appreciate humor that is opposed to their established beliefs since they are more open-minded, curious, and comfortable with ambiguity as compared to conservatives (Peifer & Holbert, 2016). However, no study in political satire, to date, has focused on this effect. Thus, this paper will further explore the role of political preference, especially considering its interaction with openness to experience in political satire processing, and will assess whether these two variables influence the viewers’ facial expressions when watching political satire.
In sum, this paper will focus on whether political preference will influence the viewers’ facial expressions while watching political satire and their subsequent attitude changes, and whether openness to experience moderates this process. This paper addresses the need to further explore the complex mechanism beneath the humor process, especially for the political satire genre. It will also contribute to the existing literature on humor processing and persuasive effectiveness of political humor.
Humor in the Political Context
Previous research has suggested that politicians find humor to be a useful tool for engaging and influencing their audience (Meyer, 2000). They believe that using humor in public communication could widen their reach and make their image seem more personable, which could help them create a more affable public image (Higgie, 2017). According to Speck (1991), humor is a particular genre of a play, a dramatic work intended for
performance by its actors. Unlike other types of plays, humor can carry the viewers away from the normal life; it takes meaning from, but also gives meaning to the normal world. In humor, the world of play and the normal world overlap, which makes the appreciation of humor both playful and serious, involved and detached. A humorous plot often starts with a manipulation that generates tension or arousal, followed by a mechanism that allows one to reduce tension and enjoy the arousal. Viewers would experience a resolution as well as a joyful feeling that allows them to switch to the implied meaning. In other words, processing humor could lead to pleasurable sensations or even laughter, along with inspiration from some implied meanings.
Amid different genres of humor, satire is one of the most studied types in political communication research. According to Colletta (2009), satire distinguishes itself from other types of humor by its attacking nature. It is defined as “a form that holds up human vices and follies to ridicule and score, and an attack on or criticism of any stupidity or vice in the form of scathing humor” (Colletta, 2009, p. 859). Satire is different from the general comic; in comedy, the comedians often ridicule themselves while satire often derides a certain object. Satire uses laughter as a weapon to attack ideas, behaviors, institutions, or individuals.
Two main streams of political satire exist. One is focused on issue-oriented satire and is more often seen in Europe (Matthes & Rauchfleisch, 2013), and the other is
person-oriented satire, which is more prevalent in the United States (US; Haigh & Heresco, 2010). In both types of political satire, there is always an object that is satirized in the
humorous plot. The object can be either (a) a political issue, such as a particular policy, or (b) a person, such as a political candidate. Such political satire often intends to demean the
satirized objects. With prior exposure to other media content, viewers are likely to have existing knowledge about the satirized topic and may even hold opinions toward them. Such preexisting opinions constitute the viewer’s preferences toward the topic and as such, viewers may either agree or disagree with the intended meaning of the satire. Researchers have
revealed that preexisting opinions influence the appreciation of the humor; preferences regarding a political topic are likely to influence the appreciation of the political humor. Political Preference and Humor Appreciation
Political preference effects humor appreciation. This relies on the assumption that the consumption of political satire is a participatory act in which the satirical message is
deconstructed by the viewer’s existing knowledge, often acquired through other media resources, and constructed differently by the viewer’s understandings. Individual
characteristics also play a significant role in such processes (Nabi, Moyer-Gusé, & Byrne, 2007). When viewers process political satire, their pre-existing opinions toward the satirized object will influence their perception of the humor. This assumption is based on the
disposition theory of humor. First developed in the 1970s, the theory posits that
humor appreciation varies inversely with the favorableness of the disposition toward the agent or entity being disparaged, and varies directly with the favorableness of the disposition toward the agent or entity disparaging it. Appreciation should be maximal when our friends humiliate our enemies, and minimal when our enemies manage to get the upper hand over our friends (Zillmann & Cantor, 1976, pp. 100–101). Disposition theory of humor presumes that one’s preexisting opinion toward the humorous subject is a salient factor influencing humor appreciation. Meyer (2000) further illustrated that when people process humor, they only laugh when the joke is not perceived as a threat to one’s self-image. Humor appreciation, thus, depends on one’s preferences toward the humor subject. Research has revealed consistent findings for the relationship between political preference and humor appreciation. Becker (2014) studied the 2008 American presidential campaign, where Obama (i.e., the Democrat candidate) and Romney (i.e., the Republican candidate) competed against each other. Results of the study suggested that both
liking Obama and disliking Romney were integral to the appreciation of the various stimuli (Becker, 2014). Those who liked Obama were more likely to appreciate humor attacking the Republican party, and those who liked Romney were more likely to appreciate humor directed toward the Democratic party (Becker, 2014). Similar findings are present for those who dislike the politician and their appreciation in the party-directed humor. When viewers’ candidate and party preferences are congruent with the intended meaning of the humor, people will appreciate the humor more than those who experience incongruency. Instead of focusing on politicians, Boukes et al. (2015) focused on policies; they used the Dutch
government’s plan for reducing funding for the public broadcasting organization as the topic of interest. The results for this study echo claims that viewers find political satire funnier when the intended meaning is congruent with their policy preference than when it is incongruent (Boukes et al., 2015).
Conclusively, it is presumable that when viewers watch a political satire, they will find the humor funnier if they do not like the satirized object, based on the disposition theory of humor. They may enjoy this humor because they find their “enemy” is demeaned.
However, if they like the satirized object, they may find the satire less funny since they feel uncomfortable seeing their “friends” being attacked.
Measurement of Humor Appreciation
Self-reported humor appreciation and facial expressions are two different constructs that can reflect one’s emotional reaction to a humorous plot. As an automatic measure of emotional response, facial expressions should be introduced to political humor processing studies.
Though it has not been clearly defined, self-reported humor appreciation is often treated as the extent to which a humorous plot is found to be funny, which is regarded as an emotional response to humor. In most political studies, the viewers’ emotional reactions are measured by this construct. An example of this application is seen in Becker's (2014) study that provides four items (i.e., funny, humorous, amusing, and playful) using Likert scales immediately following exposure to a humorous stimulus. Peifer (2018), alternatively,
proposed a “feeling of mirth” construct, which can be regarded as an indicator of humor appreciation as well. In his study, he applied semantic differential scales with four items (i.e., not funny vs. funny, not amusing vs. amusing, not entertaining vs. entertaining, and not humorous vs. humorous).
According to Poels and Dewitte (2006), emotion measurement can be classified into two main categories—self-report and autonomic measurements. The self-report measurement includes verbal, visual, and moment-to-moment measurements and the autonomic
measurements include heart rate, skin conductance, and facial expressions. Self-reports have two main shortcomings concerning validity (Lewinski, Fransen, & Tan, 2014). First, they often fail to capture low-order emotions (i.e., ones resulting from low-complexity automatic processes such as pleasure and arousal). To accurately gauge emotions with self-reports, the emotions must be strong enough for the viewers to be aware of their change in state and be able to report it later. Self-reports can also elicit insincere answers for reasons such as social desirability. On the other hand, autonomic measures capture bodily reactions that are often beyond the person’s conscious control. Though these can be costly and require a
sophisticated research set-up, they have shown good predictive power in the study of persuasive effectiveness (Lewinski et al., 2014).
Facial expressions of emotions are semi-universal sequences of facial muscle contractions that are linked with a person’s emotional state. More simply, emotions cause facial expressions (e.g., “I feel funny, so I smile”). Capturing facial expressions is one of the automatic measures that can reflect affective states and possibly predict associated behavior and attitude change. In advertising studies, facial expressions are regarded as a valid indicator of one’s emotional state, and its predictive power has been proven in persuasive outcomes including attitude toward the ads, product, or brands, as well as purchase intention (Lewinski et al., 2014).
Political satire as a form of humor could elicit happiness in the viewers; they may laugh at the political satire when they find it funny and amusing. The viewers’ facial expressions of happiness could serve as an indicator of how much they appreciate the
humorous plot. It is speculated that facial expressions could obtain more direct and genuine reactions to the political satire when considering low-ordered emotions and potential social desirability biases. When viewers fill in self-reported measures of humor appreciation, they may reevaluate the humorous plot by filtering through their experiences with and knowledge about the topic or humor narrative. It is possible that the viewers would rate the political satire as funny since they believe the plot is humorous, but not actually laugh at it since the intended meaning is incongruous with their ideology. It is also possible that the viewers do laugh at the political satire, but rate it as not funny since they realized they do not agree with the intended meaning after viewing. Thus, it is believed that the expression of happiness is distinct from the self-reported humor appreciation.
Applying the disposition theory of humor, it is presumable that viewers will laugh more at the political satire when its intended meaning is congruent with their political preference than when it is incongruent. As such, the first hypothesis is proposed:
H1: Viewers show more expressions of happiness toward a satirical message when its
intended meaning is congruent with their political preferences than when it is incongruent with their preferences.
Moreover, facial expression reading is a new measure for the reaction to political satire; it has not yet been validated in this research field. Facial expressions of happiness and self-report humor appreciation may share some similarity, though, since both can indicate the pleasure that viewers experience when viewing political satire. However, it is still unclear whether the real-time measure (i.e., facial expression reading) will lead to the same outcomes as the post-exposure measure (i.e., self-reports). Thus, it is interesting to explore the
relationship between the viewers’ facial expressions and their self-report measures, which is addressed in the first research question:
RQ1: Does the expression of happiness during the exposure to the satire correlate with self-reported humor appreciation following the exposure?
Additional individual characteristics can influence one’s reactions to political humor. One possible characteristic is the openness to experience, a basic dimension of personality. This characteristic distinguishes between people who prefer to seek novelty and variety from those who prefer familiarity and routine. People who score highly on this personality
dimension tend to be broad-minded, intellectually curious, imaginative, unconventional, and aesthetically sensitive, while people who score lower are often characterized as close-minded, lacking curiosity, and preferring familiar people, places, and things (Brandt, Chambers, Crawford, Wetherell, & Reyna, 2015; McCrae & Costa, 1997).
Regarding their cognitive styles, people who score highly for openness to experience are more likely to seek out new information and experiences that challenge their status quo, whereas people who score lower on openness to experience are more likely to stick to their positions (Brandt et al., 2015). To understand such differences in cognitive processing styles, it is necessary to explain a narrower concept, which is Webster and Kruglanski's (1994) epistemic need for cognitive closure (NFC). Need for closure (NFC) refers to a tendency to obtain closure and reach a conclusion, without specifying its content, as long as it solves ambiguity (Onraet, Hiel, Roets, & Cornelis, 2011). Openness to experience and NFC are highly related to each other; they share content at the level of cognition, which is a preference for novelty and variety (Onraet et al., 2011; Perry & Sibley, 2013). Openness to experience is a more stable trait and assumed to be a broader concept, while NFC is conceptualized as a motivated psychological state. People with higher levels of openness tend to have a lower NFC, so they are more comfortable with ambiguity and more likely to accept ideas that contradict their values.
Disposition theory of humor posits that people only appreciate humor when the intended meaning is congruent with their current ideologies. By introducing openness to experience, it is presumable that the more open individuals are to experiences, the more they will appreciate humor that is incongruent with their current ideas. Combining these ideas with the previous discussion on political satire and expressions of happiness, the second hypothesis is presented:
H2: Openness to experience moderates the relationship between political preference
and expressions of happiness, such that the more open viewers are to experiences, the smaller the differences in expressions of happiness will be between the congruent and incongruent messages.
Political Satire and its Persuasive Effects
Scholars have previously attempted to study the ultimate goal of satire, its persuasive effectiveness (Boukes et al., 2015); however, the findings are quite mixed. Whether humor can influence attitude change is also a popular research focus in advertising research. In a review by Weinberger and Gulas (1992), the following four conclusions are proposed
regarding the persuasion effect of humor in advertising research. Firstly, humorous ads attract more attention compared to non-humorous ads. Secondly, humor neither harms nor helps comprehension of the message. Thirdly, strong evidence has shown that humor can increase source liking. Finally, humor is believed to be persuasive for feeling-oriented products (e.g., clothes and perfumes), or non-durable consumer goods with low-involvement (e.g., toilet paper).
Recent studies have investigated this topic further by examining its underlying cognitive process. In political communication, findings have not fully supported the effectiveness of humor use in attitude change. Nabi et al. (2007) identified several contradicting processes that both evoke and inhibit the persuasive effect through
counterarguing. On the one hand, finding the content humorous could increase liking of the source and the depth of processing the humorous messages, which in turn decreases
counterarguing toward the messages. The less the viewers counterargue with the message, the more likely they are to be influenced by the intended meaning of the message, and the more they value the quality of the message’s argument; subsequently, their attitudes are more likely to be influenced by the humorous message. On the other hand, when viewers regard the message as humorous, they may consider that it barely contains any content relevant to serious judgement, which in turn increases their counterarguing and decreases the persuasion effect. Based on the study by Nabi et al. (2007), Boukes et al. (2015), who studied political
satire, also revealed the following two opposing processes. Their findings showed that if the viewers considered the message humorous, they would counterargue with it since they regard it as frivolous. Simultaneously, the perceived funniness absorbed the viewers in message processing, which in turn would decrease their counterarguing. Negatively related to attitude change, the level of counterarguing can subsequently predict the persuasive outcomes, which is also marginal.
Conclusively, the exposure to political humor does not guarantee effects on attitude change. Whether or not the viewers find the humorous message to be funny is a crucial construct in the humor processing studies (Boukes et al., 2015; Nabi et al., 2007). Since perceived funniness further evokes two contradicting effects on counterarguing, no overall persuasive effect of humor appreciation has been detected. In other words, those who exposure to the political satire may either appreciate the humor or not. If they find it funny, two opposing effects evoke, and no persuasive effect can be found. If they do not think it as funny, there is also no overall persuasive effect. Moreover, based on disposition theory of humor, Boukes et al. (2015) revealed that perceived funniness could be predicted by political preference, which is also the attitude about the satirized topic. Political preference, perceived funniness, and attitudes towards the satirized objects are thus three consistent constructs in the proposed underlying mechanism of political satire processing (Boukes et al., 2015). In their story, it is possible that perceived funniness as the self-reported measure of humor appreciation may not precisely indicate one’s emotional response to political humor. The self-report measures of emotion may involve some cognitive evaluation since it is reported by the viewers after exposure to the humorous plot. That could be the reason why the viewers’ evaluations of their emotions are consistent with their attitudes toward the humor’s topic. As previously discussed, facial expressions might be different from self-reported humor
appreciation measures, as they reflect the viewer’s real-time bodily reaction to political satire. In this research design, the expression of happiness is adopted to reevaluate political satire processing. It is interesting to know if facial expression would be a useful construct in the
mechanism especially in predicting attitude change. This query is formulated as two research questions:
RQ2: Is the expression of happiness during exposure to the satire related to attitudes toward the satirized subject?
RQ3: Does the expression of happiness mediate the relationship between political preference and attitudes toward the satirized subjects?
Based on the proposed research questions and hypotheses, the conceptual model in Figure 1 is presented.
Figure 1: Conceptual Model.
Methods Research Design
There were two independent variables in the research design. The first was political preference, which was treated as a dichotomous variable (congruent preference vs.
incongruent preference). The other was the openness to experience moderator, which was treated as a continuous variable.
Research Material
A controversial topic was selected for the political satire material. A pre-test was conducted to choose from three topics—the Me Too movement, animal rights, and climate change. Each topic was framed as a statement with a strong tone and clear position (e.g., “Climate change is definitely real and bad. Everyone should take actions to build a low carbon society. Otherwise, nobody can be exempt from the terrible consequences.”).
Participants were asked to rate these statements according to their opinions with five seven-point semantic differentials, which were bad–good, unfavorable–favorable,
unacceptable–acceptable, foolish–wise, and wrong–right (Holbert, Hmielowski, Jain, Lather, & Morey, 2011). The questionnaire was distributed online among peer students (N = 41, nmen = 15, nwomen = 26). Gender was included since the Me Too movement topic is sensitive to gender influence. Means of the five semantic differentials were computed for each statement. A factorial repeated measures ANOVA was conducted, with the mean of each scale of the topic as the three within-subject variables, and gender (men vs. women) as the
between-subject variable. The descriptive statistics suggest that the Me Too movement topic scored the highest for standard deviation among the three options (M = 4.40, SD = 1.89); within this topic, women (M = 4.37, SD = 2.14) scored higher than men (M = 4.45, SD = 1.42) for standard deviation. The animal rights (M = 3.74, SD = 1.43) and global warming (M = 5.17, SD = 1.60) topics scored lower for standard deviation. Since the standard deviation describes the distribution of the dataset, the higher standard deviation value indicated a wider spread of data. The results thus suggested that the Me Too movement was the most
controversial among the three topics, and the opinions of women are even more polarized than for men.
Thus, the Me Too movement was selected as the research topic. The research material was a one-minute humorous video about the Me Too movement. It depicted a male actor’s interactions with women before and after the Me Too movement. This satirized the way that men are threatened by the movement, and that they are not willing to show their appreciation and affection toward women anymore because of it.
Participants
Considering the topic of the research material is sensitive to gender, only women were included for the full-scale study (N = 83). Participants were recruited through
convenience sampling at a university campus and had an average age of 21.87 (SD = 3.17) years.
The researcher approached potential participants in the university campus and asked if they would be willing to partake in a study about the research topic. Participants were told that a video would play during the study and while watching the video, their face would be recorded to make sure that they were paying attention. After participants gave their
permission, they began the online study through Qualtrics on the laptop provided by the researcher.
A brief introduction to the research was provided at the start of the study and the participants had to provide their informed consent to continue. Participants then filled in the question regarding their political preference, and they were guided on how to record their face during the viewing. Once recording, participants watched the humorous video.
Additional questions were presented after they successfully uploaded their face record files. Measures
Independent variable—Political preference. Similar to previous studies (Boukes et al., 2015; Nabi et al., 2007), political preference was measured with a belief-based statement specifically invented for the research on the selected topic. Participants were asked to
indicate their opinion on whether they think the Me Too movement is overcorrected
(explained as a tendency making too much of a correction). Participants were forced to take sides on the topic by choosing among four options (i.e., absolutely not; no, not really; yes, a little bit; and absolutely yes). This was then treated as a dichotomous variable; those who agree that the Me Too movement is overcorrected (i.e., yes, a little bit or absolutely yes) were assigned to the congruent political preference toward the political satire group (n = 38). Those who posit that the Me Too movement is not overcorrected (i.e., no, not really and absolutely not) were assigned to the incongruent political preference toward the political satire group (n = 45).
Moderator—Openness to experience. Openness to experience is measured by the Neo-FFI (Costa & McCrae, 1992), the well-established scales for the Big Five personality traits. The measurement has 60 five-point items in total, but this study only included the 12 items for openness to experience. The scale for measuring this particular trait was reliable (α
= .78). After recoding the revised items, the mean scores of the 12 items were indexed to create a new variable that served as the final measurement (M = 3.51, SD = 0.52)
Mediator—Expression of happiness. Expression of happiness was measured using FaceReader software version 6.1 from the Noldus company. FaceReader is an automated facial coding software; it is validated as a reliable tool to detect the six basic emotions (i.e., happiness, sadness, anger, surprise, fear, and disgust). Happiness exhibits the most accurate predictive power, with the best recognition score among the six emotions (Lewinski & Butler, 2015). In the study by Lewinski et al. (2014), they found that results of facial expression analyses by FacerReader served as a major indicator of the persuasion effect for amusing advertising, which validates using such a measurement of facial expression in persuasive effectiveness studies.
Participants were guided to record their faces when watching the research material. Each participant generated one video of their facial expressions during their viewing. These videos were then imported into Facereader. Facial expressions of those participants were analyzed by the software (M = .19, SD = .18).
Dependent variable—Attitudes toward the satirized subject. Attitude toward the Me Too movement was measured with six semantic differentials (Holbert et al., 2011), which were negative–positive, bad–good, unfavorable –favorable, unacceptable–acceptable,
foolish–wise, and wrong– right (α = .95). The mean scores of all the items were indexed to create a new variable that served as the final measurement for this construct (M = 5.79, SD = 0.95).
Exploring variable—Self-reported humor appreciation. Humor appreciation was measured with four seven-point items, which were not funny/funny, not amusing/amusing, not entertaining/entertaining, and not humorous/humorous (Nabi et al., 2007). With good reliability (α = .96), the means of the four items were indexed to create a new variable that served as the final measurement for this construct (M = 4.40, SD = 1.67).
Randomization check—Age. Age was included in the study for the randomization check. Participants filled in their age in numbers in an open-ended question.
Results
There were 84 total cases in the original dataset. However, one problematic case where the participant put her hand on her face was removed, as it was too difficult to analyze her facial expressions with FaceReader. Thus, 83 cases were used in the final analyses. Randomization
An independent-sample t-test was conducted to compare age across the two groups of participants (i.e., congruent political preference or incongruent political preference). Given that the Levene’s test for homogeneity of variances was not violated, F(1, 81) < 0.01, p = .994, the result assuming homogeneous variances was calculated. There was no significant difference in age between the congruent political preference group (M = 22.03, SD = 2.79) and the incongruent group (M = 21.73, SD = 3.49), t(81) = -0.42, p = .677. Thus, the two groups did not differ with each other concerning age, and the randomization was successful. Hypothesis Testing
H1 posited that viewers would show more expressions of happiness toward a satirical
message when its intended meaning is congruent with their political preferences than when it is incongruent with their preferences. To test this, an independent samples t-test was
conducted that compared expressions of happiness between the congruent political preference group and the incongruent group. The Levene’s test for homogeneity of variances was not violated, F(1, 81) = 2.50, p = .118, and so the result assuming homogeneous variances was calculated. No significant differences were found for expressions of happiness between the congruent political preference group (M = 0.22, SD = 0.21) and the incongruent group (M = 0.18, SD = 0.16), t(81) = -0.97, p = .337. Thus, H1 is rejected. Viewers who regarded the
MeToo movement as an overcorrected initiative do not show more expressions of happiness than those who think it not to the humorous video that satirized the movement. It seems that viewers do not show more expressions of happiness while watching a satirical message, regardless of the congruency with their political preferences.
RQ1 asked whether expressions of happiness during exposure to the satirical message correlated with self-reported humor appreciation following the exposure. To answer this, expressions of happiness and self-reported humor appreciation were standardized into two
new variables based on their z-scores. A Pearson’s correlation coefficient was then conducted to assess the relationship between the new variables and a moderate correlation was found (r = .31, N = 83, p = .004). This suggests that the two constructs may share some similarities, yet still differ from each other. Expressions of happiness and self-reported humor
appreciation are two relevant constructs, but still different. In other words, viewers who express a certain amount of happiness may not necessarily report experiencing a similar level of humor appreciation while watching a humorous video.
An exploratory analysis was further conducted to see if the levels of self-reported humor appreciation were different between the congruent and incongruent political preference groups. Given that the Levene’s test for homogeneity of variances was not violated, F(1, 81) = 0.74, p = .392, the result assuming homogeneous variances was
calculated. There was no significant difference in self-reported humor appreciation between the congruent political preference group (M = 4.71, SD = 1.59) and the incongruent group (M = 4.14, SD = 1.71), t(81) = -1.57, p = .121. Viewers who regarded the MeToo movement as an overcorrected initiative did not report the humorous video to be funnier than those who thought it not. Thus, viewers did not report the satirical video as significantly funnier when its intended meaning was congruent with their political preferences than when it was incongruent with their preferences.
H2 stated that openness to experience would moderate the relationship between
political preference and expressions of happiness. In other words, the more open viewers are to experiences, the smaller the differences in expressions of happiness would be between the congruent and incongruent message. To test this hypothesis, an analysis of the moderation effect of openness to experience on the relationship between political preference and expression of happiness was conducted using the Process macro (Hayes, 2012) model 1. Political preference did not have a direct effect on expressions of happiness, b = -0.11, t(79) = -0.38, p = .702. Openness to experience did not have a direct effect on expressions of happiness, either, b = 0.63, t(79) = 1.24, p = .219. Further, the effect of political preferences on the expression of happiness was not moderated by the openness to experience, R2
F(1, 79) = 0.22, p = . 639. Probing the interaction effect, the direction of the results suggest that the more participants are open to experiences, the more likely that their congruent political preference would result in more expressions of happiness, though this was not significant.
RQ2 asked whether the expression of happiness during exposure to the satire was related to attitudes toward the satirized subject. RQ3 further addressed whether expressions of happiness mediate the relationship between political preference and attitudes toward the satirized subjects. To answer these two research questions, an analysis of the mediation effect of expressions of happiness between the political preference and attitudes toward the satirized subject was conducted using the Process macro model 4. Political preference had a direct effect on attitudes toward the satirized subject, b = -0.74, t(80) = -3.82, p < .001. However, political preference did not have an effect on the expression of happiness, b = 0.04, t(81) = 0.97, p = .337. Expression of happiness also did not have an effect on attitudes toward the satirized subject, b = -0.71, t(80) = -1.34, p = .138. Viewers who believed the MeToo
movement to be overcorrected reported more negative attitudes towards the movement. Their expression of happiness cannot predict their attitude towards the MeToo movement. Thus, expression of happiness does not relate to the attitudes toward the satirized subject when viewers watch political satire (RQ2). The expression of happiness does not mediate the relationship between political preference and attitudes toward the satirized subject either (RQ3).
An explorative analysis of self-reported humor appreciation was also conducted. The Process macro model 7 was used with political preference as the independent variable, openness to experience as the moderator, self-reported humor appreciation as the mediator, and attitudes toward the satirized subject as the dependent variable. Political preference did not have an effect on humor appreciation, b = -0.30, t(80) = -0.11, p = .901. Further, there was no interaction effect between political preference and openness to experience on the self-reported humor appreciation, b = 0.28, t(79) = 0.37, p = .712. Self-reported humor
t(80) = -2.78, p < .05. Similarly, political preference does have a direct effect on attitudes toward the satirized subject, b = -0.69, t(80) = -3.61, p < .001. These results suggest that self-reported humor appreciation can predict attitudes toward the satirized subject. Essentially, when participants report the video as funnier, they show less positive attitudes toward the Me Too movement.
Discussion
This study investigated the effect of political preference on attitudes toward satirized subjects, employing openness to experience as a moderator and expressions of happiness as a mediator. No effects proposed in the conceptual model (see Figure 1) were found to be significant.
Due to the limited research resources and effortful data collection, this research lacks power, and the limited sample size causes problems for the study. In the dataset, most
participants did not indicate strong inclinations in political preference. Especially considering they were forced to take sides in the questionnaire, it is speculated that many may not hold strong opinions on the Me Too movement, or simply have a neutral attitude toward it. With a larger sample size, this problem could be solved by intentionally selecting participants with polarized opinions for the analyses (i.e., those who chose "absolutely yes” and “absolutely not” in this study). A further limitation of this study, the participants were recruited using convenience sampling and so most were students at a university. This group can be
homogeneous in many ways. The level of openness to experience, for example, did not differ much across the sample (M = 3.51, SD = 0.52), which would make it difficult to test the relevant hypothesis. Also, the experimental settings were not uniform for participants.
Participants were approached when they were doing schoolwork (e.g., a quieter environment) or eating in the university canteen (e.g., a noisier environment) which could have influenced results. Finally, the amount of expressions of happiness was relatively small for all
participants, which makes it difficult to generate significant findings for relevant hypotheses and research questions.
This research does have those inevitable limitations. However, it still generates some interesting implications.
Firstly, the statistical analyses suggested that political preference did not predict viewers’ expressions of happiness when exposed to political satire. However, the mean value of expressions of happiness for participants who have congruent political preferences (M = 0.22, SD = 0.21) was larger than those who have incongruent political preferences (M = 0.18, SD = .16), which is consistent with the direction of the hypothesis. Thus, the responses are possibly consistent with the disposition theory of humor; people may only laugh more at the humor when its implied meaning is congruent with their current ideology. Moreover, it is speculated that expression of happiness can be adopted as an indicator of the viewers’ response toward the political satire. Expressions of happiness could also be a valid indicator of whether the participants appreciate the humor. It may also be true that facial expressions can, in turn, predict one’s disposition toward the topic of political humor.
Secondly, though no moderation effect of openness to experience was detected for the relationship between political preferences and expressions of happiness, the interaction results did suggest that when participants are more open to experiences, the moderation effect is more likely to be significant. This is consistent with the theoretical argument that people who are more open to experiences are more likely to appreciate humor when its intended meaning is incongruent with their current ideology. This could be true for humor appreciation and other types of messages. People who report higher levels of openness to experience are more likely to appreciate ideas contrary to their own opinions. Future research should further explore this assumption in other contexts.
Thirdly, this study compares two different measures of emotional response to political satire. The findings reveal that expressions of happiness (i.e., the autonomic measure) are only moderately related to self-reported humor appreciation. Moreover, the explorative analysis shows that self-reported humor appreciation is related to the viewer’s attitudes toward the satirized subject. Meanwhile, expression of happiness does not relate to the viewer’s attitudes toward the satirized subject. These findings suggest that self-reported humor appreciation and expressions of happiness may theoretically differ. It seems that the viewers’ unconscious bodily reactions are different from their conscious evaluations of their
emotions. The negative relationship between the humor appreciation and attitudes toward the MeToo movement suggests that the viewers’ evaluations of their emotions are consistent with their cognitive evaluation of the satirized object. It is possible that self-report
measurements of emotion do involve cognitive judgements, especially when one’s attitude is important in the context. Future researches in political humor processing should distinguish these two constructs.
Moreover, this research demonstrates the possibility of reevaluating the current humor processing mechanism (Boukes et al., 2015; Nabi et al., 2007). Instead of considering the self-reported humor appreciation (i.e., perceived funniness) as the beginning point in the current models, there might be a more accurate and concise alternative with the two following parallel routes. Facial expression is adopted as a construct to indicate one’s emotional response, while simultaneously another cognitive indicator (i.e., processing depth or absorption) is included. Future study can adopt Facereader to further investigate this assumption. It is also suggested that studies in other fields (e.g., health messages, news messages, or political speeches) introduce facial expressions as a distinct construct of emotional response, especially when social desirability or other factors would influence the viewers’ evaluations of their emotions.
Also, though there is no manipulation of exposure to political satire, the results of this research might also indicate that the persuasive effect of political satire is minimal. Attitude toward the satirized object is highly predictable by political preference. It seems that
exposure to political satire has a marginal effect on attitude change, which is consistent with current literature (Boukes et al., 2015; Holbert, 2013; Nabi et al., 2007)
In sum, this study pioneers in its introduction of facial expressions to the study of political satire. As an emotional response construct, facial expressions can be adopted in contexts where self-report measures are insufficient, since these two constructs may theoretically differ. It also further demonstrates the possibility to reevaluate humor
processing mechanism. Though the statistical results are not significant, this study showed the possibility of confirming the disposition theory of humor, and that openness to experience
can affect the relationship between political preferences and emotional responses to humorous messages.
What’s more, political turbulence sometimes makes people unwilling to share their opinions. For example, in 2016, Trump won the US presidential election, which was
unexpected based on polling results. It was speculated that voters were unwilling to express their true positions, since supporting Trump seemed to be “politically incorrect.” This research further demonstrates that facial expressions are easily collected and analyzed without strict laboratory settings, which can be adopted into more research fields or other political practices in the future.
References
Baumgartner, J., & Morris, J. S. (2006). The Daily Show effect: Candidate evaluations, efficacy, and American youth. American Politics Research, 34, 341–367.
Baym, G. (2005). The Daily Show: Discursive integration and the reinvention of political journalism. Political Communication, 22, 259–276
Becker, A. B. (2014). Humiliate my enemies or mock my friends? Applying disposition theory of humor to the study of political parody appreciation and attitudes toward candidates. Human Communication Research, 40(2), 137–160. doi:10.1111/hcre.12022 Bore, I. L. K., & Reid, G. (2014). Laughing in the face of climate change? Satire as a device
for engaging audiences in public debate. Science Communication, 36(4), 454–478. doi:10.1177/1075547014534076
Boukes, M., Boomgaarden, H. G., Moorman, M., & de Vreese, C. H. (2015). At odds: Laughing and thinking? The appreciation, processing, and persuasiveness of political satire. Journal of Communication, 65(5), 721–744. doi:10.1111/jcom.12173
Brandt, M. J., Chambers, J. R., Crawford, J. T., Wetherell, G., & Reyna, C. (2015). Bounded openness: The effect of openness to experience on intolerance is moderated by target group conventionality. Journal of Personality and Social Psychology, 109(3), 549–568. Colletta, L. (2009). Political satire and postmodern irony in the age of Stephen Colbert and
Jon Stewart. Journal of Popular Culture, 42(5), 856–874. doi:10.1111/j.1540-5931.2009.00711.x
Costa, P. T., & McCrae, R. R. (1992). Revised NEO Personality Inventory (NEO PI-R) and NEO Five-Factor Inventory (NEO-FFI) Professional Manual. Psychological
Assessment Resources Inc., Odessa.
Haigh, M. M., & Heresco, A. (2010). Late-night Iraq: Monologue joke content and tone from 2003 to 2007. Mass Communication and Society, 13(2), 157–173. doi:10.1080/
15205430903014884.
Hayes, A. F. (2012). PROCESS: A versatile computational tool for observed variable mediation, moderation, and conditional process modeling.
Higgie, R. (2017). Public engagement, propaganda, or both? Attitudes toward politicians on political satire and comedy programs. International Journal of Communication, 11, 930–948.
Hmielowski, J. D., Holbert, R. L., & Lee, J. (2011). Predicting the consumption of political TV satire: Affinity for political humor, the Daily Show, and The Colbert Report. Communication Monographs, 78(1), 96–114.
Holbert, R. L., Hmielowski, J., Jain, P., Lather, J., & Morey, A. (2011). Adding nuance to the study of political humor effects: Experimental research on juvenalian satire versus horatian satire. American Behavioral Scientist, 55(3), 187–211.
doi:10.1177/0002764210392156
Holbert, R. L., & Young, D. G. (2013). Exploring relations between political entertainment media and traditional political communication information outlets: A research agenda. In E. Scharrer (Ed.), The International Encyclopedia of Media Studies (pp. 484–504). Malden, MA: Blackwell
Lamarre, H. L., Landreville, K. D., & Beam, M. A. (2009). To see what you want to see in The Colbert Report. The International Journal of Press/Politics, 14(2), 212–231. doi:10.1177/1940161208330904
Lewinski, P., & Butler, C. (2015). Automatic facial coding: Validation of basic emotions and facs aus recognition in noldus facereader.
Lewinski, P., Fransen, M. L., & Tan, E. S. H. (2014). Predicting advertising effectiveness by facial expressions in response to amusing persuasive stimuli. Journal of Neuroscience, Psychology, and Economics, 7(1), 1–14. doi:10.1037/npe0000012
Matthes, J., & Rauchfleisch, A. (2013). The Swiss “Tina Fey effect”: The content of
late-night political humor and the negative effects of political parody on the evaluation of politicians. Communication Quarterly, 61(5), 596–614.
McCrae, R. R., & Costa, P. T., Jr. (1997). Conceptions and correlates of openness to
experience. In R. Hogan, J. A. Johnson, & S. R. Briggs (Eds.), Handbook of Personality Psychology (pp. 825–847). San Diego, CA: Academic Press.
doi:10.1016/B978-012134645-4/ 50032-9
Meyer, J. C. (2000). Humour as a double-edged sword: Four functions of humour in communication. Communication Theory, 10(August), 310–331.
doi:10.1111/j.1468-2885.2000.tb00194.x
Nabi, R. L., Moyer-Gusé, E., & Byrne, S. (2007). All joking aside: A serious investigation into the persuasive effect of funny social issue messages. Communication Monographs, 74(1), 29–54. doi:10.1080/03637750701196896
Onraet, E., Hiel, A. V. A. N., Roets, A., & Cornelis, I. (2011). The closed mind: ‘Experience’ and ‘cognition’ aspects of openness to experience and need for closure as psychological bases for right-wing attitudes, European Journal of Personality, 197(March 2010), 184– 197.
Peifer, J. T. (2018). Liking the (funny) messenger: The Influence of news parody exposure, mirth, and predispositions on media trust. Media Psychology, 1–29.
doi:10.1080/15213269.2017.1421470
Peifer, J. T., & Holbert, R. L. (2016). Appreciation of pro-attitudinal versus
counter-attitudinal political humor: A cognitive consistency approach to the study of political entertainment. Communication Quarterly, 64(1), 16–35.
Perry, R., & Sibley, C. G. (2013). Seize and freeze: Openness to Experience shapes judgments of societal threat. Journal of Research in Personality, 47(6), 677–686. doi:10.1016/j.jrp.2013.06.006
Poels, K., & Dewitte, S. (2006). How to capture the heart? Reviewing 20 years of emotion measurement in advertising. Journal of Advertising Research, 46, 18–37.
doi:10.2501/S0021849906060041
Polk, J., Young, D. G., & Holbert, R. L. (2009). Humor complexity and political influence: An elaboration likelihood approach to the effects of humor type in The Daily Show with Jon Stewart. Atlantic Journal of Communication, 17(4), 202–219.
doi:10.1080/15456870903210055
Speck, S. P. (1991). The humorous message taxonomy: A framework for the study of humorous ads. Current Issues and Research in Advertising, 13(1–2), 1–44. doi:10.1080/01633392.1991.10504957
Warner, B. R., Hawthorne, H. J., & Hawthorne, J. (2015). A dual-processing approach to the effects of viewing political comedy. Humor, 28(4), 541–558.
doi:10.1515/humor-2015-0099
Webster, D. M., & Kruglanski, A. W. (1994). Personality processes and individual
differences in need for cognitive closure, Journal of personality and social psychology, 67(6), 1049–1062.
Weinberger, M. G., & Gulas, C. S. (1992). The impact of humor in advertising: A review. Journal of Advertising, 21, 35-59.
Zillmann, D., & Cantor, J. R. (1976). A disposition theory of humour and mirth. In A. J. Chapman & H. C. Foot (Eds.), Humor and Laughter: Theory, Research and
