How Humorous Posts Influence Engagement With Political Posts on Social Media
The Role of Political Involvement
Abstract
Abstract. Individuals frequently process political posts on social media in the context of humorous, non-political posts, which research suggests may stimulate or dampen their engagement with the political posts depending on their political involvement. To clarify that claim, I conducted a 2 × 2 experiment (N = 286) in which individuals viewed political posts situated among either humorous or non-humorous posts, all presented as video-recorded posts on a social media newsfeed, in a condition of either low or high political involvement. Among the results, the humorous posts directly boosted general attention, the elaboration of political posts, and the acquisition of political knowledge in the low political involvement group and stimulated political participation in the high political involvement group. Further analyses revealed that, in the low involvement group, increased attention and the elaboration of the posts may have mediated humor’s effect on the acquisition of political knowledge. Meanwhile, its positive effect on participation in the high involvement group occurred independently from general attention and the elaboration of the posts. Altogether, the findings suggest that humorous social media contexts may benefit the acquisition of political knowledge and political participation.
Social media have become important sources of political information, especially for today’s younger generations (Heiss & Matthes, 2016). However, compared with pre-existing media contexts, social media present political information to be processed within an entirely new mode of reception. For one, individuals are exposed to political posts amid a myriad of other information on social media (Bode, 2016). Scrolling quickly through their newsfeeds and jumping from one post to the next, users have limited time to screen and evaluate individual posts on their newsfeeds. In that situation, contextual content may influence how they, as citizens, process political posts on their newsfeeds. A prominent part of that potentially influential content on social media is humor (Davis et al., 2018; Heiss & Matthes, 2021; Park et al., 2009; Thorson, 2014).
Despite that overlap in content, it remains unclear whether frequent exposure to political information on social media contextualized amid humorous, non-political content trivializes individuals’ engagement with politics (Postman, 1986; Wirth et al., 2010) or increases their attention to politics, especially if they rarely engage in politics in the first place (Baek & Wojcieszak, 2009; Baum, 2002). In fact, until now, research on the political effects of humorous, non-political posts on social media or how those effects may depend on individuals’ political involvement (Dimitrova & Matthes, 2018; Heiss & Matthes, 2021) has remained undeveloped. That gap in the literature needs to be filled, however, because social media do indeed provide “animated stimuli and a relaxing environment, in which political information mixes with updates about pets and babies” (Bode, 2016, pp. 28–29) such that the “lines between political and nonpolitical information become increasingly blurred” (Dimitrova & Matthes, 2018, p. 336).
Aiming to partly fill that gap, this article seeks to improve current understandings of how social media users process political posts when they appear amid humorous, non-political posts on their newsfeeds. In a study involving a forced exposure experimental design, I showed participants a simulated social media newsfeed containing video-recorded political posts situated amid either humorous or non-humorous posts. Added to that, I manipulated the participants’ political involvement with the political posts prior to exposure. Although the approach did not entail simulating an externally valid (i.e., interactive) social media environment, the results nevertheless provide initial evidence on social media users’ psychological capacity to process and engage with political posts when they appear amid unrelated humorous posts.
Political Effects of Humor
A large body of literature discusses the role of entertainment in citizens’ political engagement. Perhaps most prominently, Postman (1986) has argued that greater volumes of entertainment information increasingly inhibit citizens’ deeper engagement with political information. Extending that argument, Prior (2005) has furnished evidence suggesting that media environments affording users a high degree of choice may discourage users with a high relative preference for entertaining content from becoming exposed to and thus learning from political news. Finding similar evidence, Kim et al. (2013) observed that incidental news exposure online may positively predict political participation but only for individuals with a lower relative preference for entertaining content. In line with those findings, other authors have characterized social media as simply another high-choice environment in which less politically involved individuals’ attention to political issues is constantly distracted by more appealing humorous content and, as such, encourages slacktivism at best (Rothmund & Otto, 2016; Theocharis & Quintelier, 2016; van Aelst et al., 2017).
Against that trend, however, other authors have argued that incidental encounters with political content on social media can stimulate political exposure specifically among individuals less interested in politics (Valeriani & Vaccari, 2016; for a critical perspective, see Heiss & Matthes, 2019). One reason may be that less politically involved individuals often encounter political posts situated amid humorous ones, both of which demand cognitive resources. Research has shown that political humor can indeed activate attention and thus the acquisition of political knowledge and participation in politics (Baek & Wojcieszak, 2009; Baum, 2002; Bode & Becker, 2018). At the same time, other findings indicate that political humor’s positive effects may be limited. For example, Young (2008) found that humorous political messages, compared with non-humorous ones, may increase the use of cognitive resources for comprehending humor but also decrease their use for processing message-relevant content.
Taken together, findings on the effects of political humor remain mixed, while research on the political effects of non-political humor has rarely been conducted. However, the potential findings of such research could be especially relevant in the social media era, when political posts are frequently processed alongside humorous, non-political ones. In that new context, the distracting versus attention-enhancing function of humor should be closely re-examined.
Unrelated Humor on Social Media
In presenting humorous content, social media offer a different context from traditional media contexts because the humorous content among which political posts may be contextualized is usually not political in nature (Heiss & Matthes, 2021). On social media, such situations have become commonplace. Imagine scrolling down a newsfeed and encountering a funny video of a cat followed by a serious post about new tax law. Whereas the funny video is entirely unrelated to the news post, because nearly all posts are quite short, the video may not be processed entirely free from the influence of the news post. Although that phenomenon is underexplored, past studies conducted on traditional media contexts may help to clarify how politically unrelated humorous content on social media might affect the processing of political posts.
In one of the few existing studies on how humor unrelated to politics affects political outcomes, Matthes (2013) found that in political speeches, using such humor can distract individuals from engaging in political learning processes, while humor that is related to politics may increase the elaboration of the political content among individuals with high needs for humor. Even so, the contexts of political speeches, whose audiences tend to be highly attentive to political content, differ entirely from social media contexts. In another study on how funny advertisements preceding a political news show affected how individuals processed the news, Wirth et al. (2010) found that positive mood induced primarily negative effects, for the participants ultimately evaluated the news as being less important or serious.
In advertising, research has indicated that humor can stimulate individuals’ attention to information that they would not otherwise process. For example, evidence suggests that humor can stimulate attention even to products demanding low involvement (Chung & Zhao, 2003; Eisend, 2009) because humor rewards individuals with pleasant feelings and thus incentivizes them to pay close attention to the humorous components of messages (Eisend, 2009; Zillmann et al., 1980). Despite concern that topically unrelated humor may distract from the non-humorous parts of messages, research has also indicated that such humor may exert weaker but nevertheless positive effects on persuasion, though evidence on recall and recognition remains too weak to recommend any conclusions (Eisend, 2009).
All told, although research on unrelated humor’s effects has focused on traditional media contexts, especially ones presenting commercial advertising, social media constitute a new context in which political and non-political content often occur separately from but in juxtaposition with each other. In such cases, the humorous content is not only unrelated to the political content but also distinct in storyline and topic. Thus, the extent to which scholars can learn and draw conclusions about the effects of humorous posts on social media from the literature remains limited.
The Role of Political Involvement
In this article, I am specifically interested in how political involvement may moderate the effects of humorous, non-political posts on engagement with political news posts in social media contexts. Political involvement, defined as a high motivation to engage with political issues that prompt attention to, deeper knowledge of, and stronger behavioral engagement with politics (Huddy & Khatib, 2007), is triggered by well-known social factors, including education, family background, and/or the wider social network. At the same time, gaps in political involvement may be explained by the media choices made by individual citizens (Norris, 2001; Prior, 2010). For example, individuals who begin using social media with lower levels of political involvement may prefer non-political sources on social media and, in turn, curate their newsfeeds to meet their non-political preferences (Knoll et al., 2018).
However, citizens are not the sole curators of their social media newsfeeds. In fact, network acquaintances, strategic actors (e.g., journalists and politicians), and algorithmic choices also co-design personal newsfeeds (Thorson & Wells, 2016). For that reason, users may thus incidentally encounter political information that they did not seek in the first place (Valeriani & Vaccari, 2016). As a consequence, just as they may more often be exposed to political information in their newsfeeds, ones with low political involvement may also be exposed to such information from time to time (Heiss & Matthes, 2021). Even so, the degree to which they process incidental encounters with political content on social media may heavily depend on their political involvement (cf. Knoll et al., 2018). In fact, if users are actively involved in politics, then they are likely to appraise encounters with political content as being highly relevant. As such, they may not only screen the content but also develop certain goals for information processing; they may read the post, click on the link for further information, and thus engage in more thorough learning processes (Heiss & Matthes, 2021; Karnowski et al., 2017).
Based on that reasoning, initial involvement in politics may determine how individuals process and engage with political posts on social media. However, other reasoning suggests that the often humorous posts preceding and following political posts in the newsfeed may influence how users process the political ones (Bode, 2016).
Effects of Humor on Political Outcomes
Interested in identifying humorous content’s effects on general attention, the elaboration of political content, the acquisition of political knowledge, and political participation depending on the individual’s political involvement, I measured political involvement as intended political participation – that is, the self-reported likelihood of performing a political act. Because those four outcome variables are all interlinked, however, one could assume that they are affected sequentially. To begin, evidence strongly suggests that humor’s presence is associated with individuals’ attentiveness and thus their processing capacity. Thus, I conceptualized attention as to whether individuals pay attention to posts on their newsfeeds in general (i.e., both political and non-political posts). Such a general level of attention may increase the likelihood of the in-depth elaboration of political posts, defined as whether individuals generate their own thoughts about the political posts encountered (Petty & Cacioppo, 1986). Of course, the effect’s extent may depend on how much of the increased attention is attributed to the processing of the political instead of the humorous posts (Keib et al., 2018; Tam & Ho, 2005).
The elaboration of political posts, by contrast, has been identified as an important prerequisite for gaining political knowledge and may affect political participation (Eveland, 2001; Shah et al., 2007). For example, individuals may increase their knowledge only if they process and thus store encountered content in their memories. Furthermore, they may develop participatory intentions only when they engage with the content, for only then may they identify a problem and develop a desire to change, or prevent the change of, the current political state (Knoll et al., 2018; Kruglanski et al., 2015). Even though those sequential theoretical effects may make sense theoretically, I could not test them in my study, because I manipulated humor and involvement only. Because I can make only causal claims concerning the direct effects of humor and involvement, additional experimental studies are needed to test the causal relationships between attention, elaboration, participation, and knowledge.
Hypotheses
The effect of humorous social media posts on attention to content, the elaboration of content, and knowledge about and participation with such content may depend on individuals’ prior political involvement. For one, I assumed that individuals with low political involvement do not engage with political posts unless their attention is stimulated by an external driver (e.g., humor). Thus, even humorous, non-political posts in social media newsfeeds can positively affect users’ level of cognitive activation, because people need to activate cognitive resources in order to comprehend humorous messages and are thus rewarded with positive feelings (Heiss & Matthes, 2021; Matthes, 2013; Suls, 1983). That dynamic may trigger learning processes specifically among individuals with low levels of political involvement, who are generally not attentive in humor-free environments. As a result of their cognitive activation, they may become attentive and experience passive learning processes, through which they may inadvertently absorb political content (Krugman & Hartley, 1970; Zukin & Snyder, 1984). They may also become more likely to intentionally elaborate upon political content (Baek & Wojcieszak, 2009; Baum, 2002; Bode & Becker, 2018). Therefore, individuals with low political involvement, upon being exposed to humor, may be more likely to store political content in their memories and retrieve it later. On top of that, processing political content more actively may induce stronger behavioral effects, because such processing can create new participatory goals via new knowledge structures or at least prime for realizing existing goals (Higgins, 1996; Knoll et al., 2018).
Following that logic, I assumed that if individuals lack prior political involvement, then the humorous posts may activate their cognitive resources to process the political content and thus stimulate knowledge acquisition and behavioral effects. Thus, my first hypothesis was:
Non-political humor increases (a) general attention, (b) the elaboration of political posts, (c) the acquisition of political knowledge, and (d) intended political participation among individuals in the low political involvement condition.
Drawing on goal systems theory and the limited capacity model, I additionally assumed that humorous posts would negatively affect individuals with high political involvement, who generally have strong initial political processing motivations and may thus pay attention to political content without needing external drivers (e.g., humor). In fact, additional, politically unrelated humorous messages may even distract them from their initial processing goals (Knoll et al., 2018; Matthes, 2013). According to goals systems theory, individuals form goals based on personal needs and desires, and in a given situation, some goals are prioritized above others (Kruglanski et al., 2015). Thus, if individuals have already formed strong processing goals when it comes to political information, then the in situ activation of a competing goal – in the present case, experiencing pleasure from humor – may inhibit their actions toward realizing the initial goal of reading a political post (Shah et al., 2002), because “a message may require more resources than the message recipient has available to allocate to the task” (Lang, 2000, p. 51). In short, if message recipients switch their goal from encoding political content to encoding the politically unrelated humorous parts of the message, then they have less capacity available to retrieve and store the political content (Lang, 2000).
Following that theoretical reasoning, individuals may spend their limited capacity on processing the content of a newsfeed by encoding and storing more information from either the political post or the unrelated humorous posts. As a consequence, politically involved individuals may become distracted and at least partly replace their initial political processing goal with a competing non-political processing goal (i.e., consuming humor). In that context, “the activation of one goal automatically leads to the inhibition (i.e., lower accessibility) of another, competing goal” (Shah et al., 2002). Thus, I assumed that among highly politically involved individuals, humor may hinder the in-depth processing of political content, dampen the acquisition of political knowledge, and lower the intention to participate politically. Thus, my second hypothesis was:
Non-political humor is unrelated to attention but decreases (a) the elaboration of political posts, (b) the acquisition of political knowledge, and (c) intended political participation in the high involvement condition.
By contrast, I did not pose a hypothesis about the effect of humor on attention in the high involvement condition. After all, attention may already be high in that group due to the political content, and the additional humorous content may shift some of the attention to the humorous aspects, but may not increase overall attention to the newsfeed.
Method
I conducted a 2 × 2 experiment using a sample of university students (N = 286; Mage = 22.70, SD = 3.32), 80.42% of whom had high school degrees, 19.58% of whom had college degrees, and 67.48% of whom were women. Using a sample of students was appropriate for my study because students tend to have a common concept of humor, which was important for successful manipulation in the study (Darke et al., 1998). However, it also limits the generalizability of my findings, as addressed in the limitations section at the end of the article. The study was part of a research course at a large university in Austria, such that students contributed to creating the stimulus material and recruiting independent participants. All respondents were randomly assigned to one of four experimental conditions, and data were collected from June 5 to June 10, 2018.
Stimulus Material
I manipulated involvement with the political issue presented in the political post as a proxy for political involvement. To that end, participants were either assigned to read a news article about a new law set to increase tuition at universities (i.e., high involvement condition) or a news article about math scores on recent national high school examinations (i.e., low involvement condition). Following that approach, the high involvement group was already familiar with the political issue before exposure to the posts on the newsfeed. Following Petty and Cacioppo (1986), I also manipulated the personal consequences associated with the issue. That is, if participants believe that an issue may personally affect them, then they “become more motivated to process the issue-relevant arguments presented” (Petty & Cacioppo, 1986, p. 146). Both news articles, presented as online news from a non-identifiable source, were the same length and appeared in the same layout across the conditions (see Appendix). I also calculated readability statistics in terms of grade-level using the quanteda package in R (Benoit et al., 2018), which uses a German adaption of the SMOG grading (McLaughlin, 1969). The grade levels of the texts – 8.25 in the high involvement condition and 6.16 in the low involvement condition – were deemed acceptable, considering that the participants were university students.
To manipulate the context (humor vs. no humor), I exposed participants to three political posts embedded in 10 non-political posts from a Facebook newsfeed, all captured in a short video such that they appeared successively in 10-second intervals. In the humorous condition, participants saw humorous, non-political posts, whereas, in the control group, their counterparts saw posts that were neither political nor humorous. In both groups, the non-political posts were paired in terms of structure and content but varied in terms of the presence or non-presence of humorous cues. For example, a non-political post in the non-humorous condition included a horse standing in a field, whereas the funny version of that post in the humorous condition included a meme with a horse taking a selfie (see Appendix, Figure A3). The source of the political posts, their content, and the time of their appearance in the video were constant across both groups.
By source, one political post came from the local student union, whereas the other two came from the Austrian Public Broadcaster, the ORF. The former, in which the local student union emphasizes its opposition to the tuition fees, guarantees support for affected students, and provides information about a petition against the new law. By contrast, the first post from the ORF explained opposition to the new law because it will not solve existing problems in teaching and research and because it neglects expert opinions. By further contrast, the second post from the ORF stated that experts expected high psychological pressure on students, especially among students from lower-income families.
I used the forced exposure technique to guarantee internal validity and ensure that each participant saw each post for the same amount of time (for similar approaches, see Heiss & Matthes, 2016; Kruikemeier et al., 2016). Nevertheless, the approach did not afford a real-life experience on social media, in which users can switch from one post to the next or even engage with posts by liking and/or sharing content (de Vreese & Neijens, 2016).
Measures
Unless stated otherwise, all items were measured on a 7-point Likert scale, ranging from 1 = disagree to 7 = agree.
Dependent Variables
Attention (α = .83, M = 4.64, SD = 1.57) was measured with three items asking participants whether they agreed that they (a) had paid attention to the content in the newsfeed, (b) had concentrated while following the posts in the newsfeed, and (c) had thought about other things while following the newsfeed (reversed). To measure cognitive elaboration (α = .87, M = 4.59, SD = 1.52), I asked participants whether they agreed that they (a) had intensively thought about the content of the political posts, (b) had focused on the facts in the political posts, (c) had critically reflected upon the content of the political posts while reading, and (d) had not really thought about the content of the political posts (reversed; Eveland, 2001). Next, intended political participation (1 = unlikely, 7 = very likely; α = .81, M = 2.99, SD = 1.29) was measured by asking participants how likely they were to participate in political activities related to the issue of the political posts, including by (a) signing a petition, (b) liking or sharing the political posts that they had seen, (c) writing a short comment about the issue, (d) attending a related political event, (e) participating in a related demonstration, and (f) discussing the issue with friends (e.g., Kim et al., 2013). Last, to measure the acquisition of political knowledge (M = 3.28, SD = 1.65), I asked participants six multiple-choice questions about the three political posts, and in each section, they could select one of four choices, including “I don’t know.” I added the sum of the correct responses, which resulted in an additive index from 0 to 6.
Control Variables
Political interest (1 = not at all interested, 7 = very interested; α = .92, M = 4.87, SD = 1.50) was measured with three items asking participants how interested they were in (a) political issues, (b) political news, and (c) politically relevant social developments. By comparison, social media use (1 = never, 7 = often; M = 5.97, SD = 1.47) was measured with a single item asking participants how frequently they use social media (e.g., Twitter, Facebook, and Instagram). Next, entertainment user motivation (α = .84, M = 5.17, SD = 1.31) was assessed by asking participants whether they used media (a) to be entertained, (b) to watch entertaining pictures or videos, (c) to pass time, and (d) to find entertainment. Last, political user motivation (α = .91, M = 3.99, SD = 1.52) was measured with four questions asking participants whether they used media (a) to access political information, (b) to familiarize themselves with different perspectives about politics, (c) to follow political news, and (d) to follow current political events (Park & Lee, 2014; Quan-Haase & Young, 2010).
Treatment Check Variables
Perceived funniness (α = .94, M = 3.95, SD = 1.82) was measured with four items asking participants whether they agreed that the posts in the newsfeed were (a) funny, (b) entertaining, (c) tedious (reversed), and (d) boring (reversed). Issue-specific political involvement (α = .88, M = 5.40, SD = 1.489) was measured with four items asking whether participants agreed that (a) the issue was personally important to them, (b) the developments around that issue were personally important to them, (c) they were personally interested in the results of the discussion on the issue, and (d) the issue did not have any significance to them personally (reversed).
Results
Randomization and Treatment Checks
To perform randomization and treatment checks for the variables, I ran simple regression models and switched reference groups to test differences between the four experimental groups. I used linear regression with ordinary least squares for continuous outcome variables and logistic binomial regression for binary outcome variables (i.e., gender and education). The participants were equally distributed across the four groups: the low involvement/no humor group (n = 73), the low involvement/humor group (n = 77), the high involvement/no humor group (n = 65), and the high involvement/humor group (n = 71). I conducted randomization checks by comparing the means of groups or the proportions of control variables measured prior to stimulus exposure across the groups. Although I did not detect any significant differences between groups for age, gender, education, social media use, social media users’ political motivations, or their entertainment motivations, I did find that political interest was distributed somewhat unevenly across the groups. More precisely, participants in the low involvement/no humor condition scored lower on political interest than participants in the high involvement/humor group (unstandardized mean difference: b = −0.69, p < .01) and the high involvement/no humor group (b = −0.65, p < .05). No significant differences emerged between the two low involvement groups or between the two high involvement groups. Political interest was used as a control variable in all models testing treatment effects.
I tested the humor manipulation by asking participants about the perceived funniness of the posts on the newsfeed. Participants in the humor condition scored significantly higher on the pooled measure of those items (i.e., mean scale) than their counterparts in the non-humor group (unstandardized mean difference: b = 1.39, p < .001). Meanwhile, participants in the high involvement condition scored higher on issue-specific involvement (i.e., mean scale) than ones in the low involvement group (b = 0.69, p < .001). Those effects remained highly significant across conditions when I controlled for political interest.
Hypotheses Testing
To test my hypotheses, I used ordinary least squares regressions while controlling for demographic characteristics, political interest, social media use, and political and entertainment motivations in media use (Darlington & Hayes, 2017). All control variables were measured prior to stimulus exposure, and in all models, the low involvement/non-humorous condition served as the reference group. Table A1 in the Appendix provides an overview of the intercorrelations among the dependent variables. I also performed a confirmatory factor analysis of the theoretically linked dependent variables (Rosseel, 2012), but excluded acquisition of political knowledge, which is a formative variable (e.g., Bollen & Diamantopoulos, 2015). The analysis revealed support for a three-factor solution with an acceptable model fit (RMSEA = 0.08; SRMR = 0.06; χ2(62) = 184.55, p < .001; CFI = 0.93). All factor loadings appear in Table A2 in the Appendix.
In the next step, I tested the effects of humor manipulation on general attention, the elaboration of the political content, acquisition of political knowledge, and intended political participation. H1 assumed that in the low political involvement group, humor would increase all four variables. The results of the regression models, shown in Table 1 and visualized in Figure 1 (i.e., with predicted mean values calculated from Table 1 and covariates set to mean values; Fox, 2003), indicated support for H1a, H1b, and H1c. In the low involvement group, participants in the humor versus the non-humor group scored significantly higher for general attention (b = 0.67, p < .01), the elaboration of political posts (b = 0.45, p < .05), and the acquisition of political knowledge (b = 0.75, p < .01). However, no effect on the intention for political participation emerged, meaning that H1d was rejected.
H2 assumed that in the high political involvement group, humor would decrease (a) the elaboration of political content, (b) the acquisition of political knowledge, and (c) intended political participation. However, within the condition, the humor group did not significantly differ from the non-humor group in the elaboration of political content (b = 0.01, SE = 0.24, p = .96) or acquisition of political knowledge (b = −0.07, SE = 0.27, p = .81). As expected, humor also did not affect attention (b = 0.17, SE = 0.26, p = .52). By contrast, in the high involvement group, humor did significantly increase political participation (b = 0.41, SE = 0.19, p = .04). However, the effect was positive and thus points in the opposite direction than I expected. Thus, I found no support for H2.
Additional Analyses
I also examined whether the identified effects of humor on knowledge and participation were mediated by general attention and the elaboration of political posts – that is, the two variables indicating the depth of cognitive processing that may thus precede learning and behavioral effects. I did so by adding attention and elaboration as independent variables to the two models predicting knowledge and participation scores in Table 1 (see Table A3 in the Appendix for full results).
Among the results, general attention (b = 0.28, SE = 0.07, p < .001) and the elaboration of the political content (b = 0.17, SE = 0.08, p = .03) were both significant predictors of the acquisition of political knowledge. Including the two variables in the knowledge model modified humor’s effect in the low involvement group, and the original mean difference of 0.75 points weakened and lost significance (b = 0.48, SE = 0.25, p = .05). Both variables played a similar role in diminishing the treatment effect. To be precise, with only general attention in the model, humor’s effect was 0.50 (b = 0.50, SE = 0.25, p = .05), and with only elaboration, humor’s effect was 0.58 (b = 0.58, SE = 0.25, p = .02). Such results may indicate the mediating role of general attention and the elaboration of the political content in predicting the acquisition of political knowledge.
In predicting political participation, only the coefficient of elaboration achieved statistical significance (b = 0.28, SE = 0.06, p < .001). However, including the two variables did not substantially change humor’s significant coefficient in the high involvement group, thereby indicating that the two variables did not play a mediating role in the group.
Discussion
The results of the study shed new light on how humorous, non-political posts affect social media users’ engagement with political posts in situations of low and high political involvement. The results, indicating positive but no negative effects of humorous posts, suggest that such posts can increase general attention to social media newsfeeds, the elaboration of political content encountered there, and the acquisition of political knowledge among less politically involved individuals. Additional analysis also indicated that both general attention and the elaboration of the political content were significant independent predictors of knowledge that partly explained the treatment effect.
The reason for those positive effects on knowledge may be that among less politically involved individuals, humor can activate cognitive resources, which are consequently also used to process political content. For that dynamic, I identified two possible routes. The first route is a more or less unconscious process, in which individuals absorb more humor-unrelated information inadvertently via passive learning (Krugman & Hartley, 1970; Zukin & Snyder, 1984). By definition, passive learning occurs in environments that reduce the resistance to learning and in which users become more accepting of the information encountered (Bode, 2016). A humor-laden context may facilitate such an environment. In the second route, individuals engage in the intentional elaboration of political content. In that process, initially less politically involved individuals activate cognitive resources to process not only the humorous content but also the political content in the newsfeed.
Among other results, humor did not affect political participation in the low involvement group, possibly because political participation is a goal-oriented behavior (Kruglanski et al., 2015). Thus, if individuals are uninvolved with a political issue and lack initial political goals, then mere humor-induced processing may not be strong enough to induce behavioral intentions (Knoll et al., 2018). In other words, some level of systematic cognitive engagement with the political content is needed to form goals and develop behavioral intentions to those ends. Along those lines, only the elaboration of the political content, not attention to the newsfeed per se, significantly related to political participation. Given humor’s positive effect on the elaboration of political content in the low involvement group, positive, indirect, across-time effects of humor on political participation via the increased elaboration of political content remain possible.
Second, and against my expectation, some evidence suggested that humor may also positively affect more politically involved individuals. Although I found no evidence of effects on general attention, the elaboration of political content or the acquisition of political knowledge, more politically involved individuals did report higher levels of intended political participation when they encountered the political posts in a humorous environment. One explanation may be that exposure to humor elicits positive feelings, which may consequently increase willingness to engage in more effortful behavior (Gardner, 1985). However, inducing such effects requires a certain level of political involvement and, in turn, cognitive engagement, which may explain why no direct positive effect surfaced for humor among the politically less involved.
All in all, I found positive outcomes for both less and more politically involved individuals. First, among the less involved, the humorous context may have boosted the acquisition of political knowledge, mostly due to their increased general attention and a higher likelihood for the more in-depth elaboration of political content. While I did not find a direct effect of humor on political participation among the less involved, indirect effects via the elaboration of political content remain possible and need to be further explored. Second, among the highly involved, a humorous environment may have directly stimulated political participation. Because encounters with humor may increase positive feelings, when those feelings are combined with high political involvement, individuals may report a greater willingness to engage in effortful behavior. However, that second effect needs to be further investigated and specifically tested in media environments marked by free choice. In such environments, positive feelings may induce individuals to selectively expose themselves to mood-congruent entertainment content, thus leading to more pronounced distracting effects.
Limitations
The study involved a few notable limitations. First, to keep internal validity high and allow robust causal conclusions, I employed a forced exposure design, in which participants could not select or interact with the content encountered. Thus, the results allow only inferences about how new situations for receiving messages, in which multiple posts are quickly processed in sequence, may influence the processing of political posts. In the future, researchers need to replicate my findings by using more interactive experimental designs and observational studies (e.g., panel studies or mobile experience sampling). Second, I used self-reported measures of general attention, the elaboration of political content, and intended political participation. Such measures may, however, over-or underestimate actual behavior (Junco, 2013). Thus, researchers also need to address that possibility by tracing physical reactions (e.g., eye tracking) and by observing actual behavior. Third, the findings preclude any causal conclusions about the interrelationships between general attention, the elaboration of political content, the acquisition of political knowledge, and intended participation. Those interrelationships need to be tested in a separate experimental setting. Fourth, because my sample consisted of college students, the effects found the need to be replicated for other social groups and by using samples that are more representative of the general population. Likewise, the stimulus material was aligned to the specific target group, meaning that I could not portray a large variety of political or humorous content. However, the effects of humorous posts may depend on not only whether they include emotional cues or are presented with strong visual appeals but also whether the accompanying political posts contain soft or hard news. Last, I tested the effects of humorous posts because they appear frequently on social media – for example, as funny memes and videos. However, other non-political content, including non-humorous entertainment musical and video content, may also elicit attention and affect political information processing. Researchers should address all of those issues to broaden current understandings of humorous as well as non-humorous posts.
Conclusion
Those limitations notwithstanding, the findings contribute to theoretical understandings of how politically unrelated humor affects the processing of political posts on social media. Above all, humorous posts may trigger active and passive learning processes among the politically less involved and stimulate behavioral intentions among their highly involved counterparts. Thus, concerns that a mix of humorous and more serious political posts on social media may inhibit political learning and dampen political participation may be unfounded. The findings also inform research on news exposure, which has largely neglected the role of context effects on social media. For example, incidental political exposure may be more likely to boost political learning if less politically involved individuals consume them under the condition of elevated attention, even if that attention is triggered by something else in the newsfeed. Those context effects are key characteristics of today’s social media environments and need to be considered to better understand the political consequences of changing information environments.
Raffael Heiss (PhD, University of Vienna) is a lecturer and postdoctoral researcher at the Management Center Innsbruck. His research interests include digital media, political communication, and civic engagement.
References
(2009). Don’t expect too much! Learning from late-night comedy and knowledge item difficulty. Communication Research, 36(6), 783–809. https://doi.org/10.1177/0093650209346805
(2002). Sex, lies, and war: How soft news brings foreign policy to the inattentive public. American Political Science Review, 96(1), 91–109. https://doi.org/10.1017/S0003055402004252
(2018). quanteda: An R package for the quantitative analysis of textual data. Journal of Open Source Software, 3(30), Article 774. https://doi.org/10.21105/joss.00774
(2016). Political news in the news feed: Learning politics from social media. Mass Communication and Society, 19(1), 24–48. https://doi.org/10.1080/15205436.2015.1045149
(2018). Go fix it: Comedy as an agent of political activation. Social Science Quarterly, 99(5), 1572–1584. https://doi.org/10.1111/ssqu.12521
(2015). In defense of causal-formative indicators: A minority report. Psychological Methods, 22(3), 581–596. https://doi.org/10.1037/met0000056
(2003). Humour effect on memory and attitude: Moderating role of product involvement. International Journal of Advertising, 22(1), 117–144. https://doi.org/10.1080/02650487.2003.11072842
(1998). Accuracy motivation, consensus information, and the law of large numbers: Effects on attitude judgment in the absence of argumentation. Personality and Social Psychology Bulletin, 24(11), 1205–1215. https://doi.org/10.1177/01461672982411007
(2017). Regression analysis and linear models: Concepts, applications, and implementation. Methodology in the social sciences. The Guilford Press.
(2018). Seriously funny: The political work of humor on social media. New Media & Society, 20(10), 3898–3916. https://doi.org/10.1177/1461444818762602
(2016). Measuring media exposure in a changing communications environment. Communication Methods and Measures, 10(2–3), 69–80. https://doi.org/10.1080/19312458.2016.1150441
(2018). Social media in political campaigning around the world: Theoretical and methodological challenges. Journalism & Mass Communication Quarterly, 95(2), 333–342. https://doi.org/10.1177/1077699018770437
(2009). A meta-analysis of humor in advertising. Journal of the Academy of Marketing Science, 37(2), 191–203. https://doi.org/10.1007/s11747-008-0096-y
(2001). The cognitive mediation model of learning from the news. Communication Research, 28(5), 571–601. https://doi.org/10.1177/009365001028005001
(2003). Effect displays in R for generalised linear models. Journal of Statistical Software, 8(15), 1–27. https://doi.org/10.18637/jss.v008.i15
(1985). Mood states and consumer behavior: A critical review. Journal of Consumer Research, 12(3), 281–300. https://doi.org/10.1086/208516
(2016). Mobilizing for some: The effects of politicians’ participatory Facebook posts on young people’s political efficacy. Journal of Media Psychology, 28(3), 123–135. https://doi.org/10.1027/1864-1105/a000199
(2019). Does incidental exposure on social media equalize or reinforce participatory gaps? Evidence from a panel study. New Media & Society, 21(11–12), 2463–2482. https://doi.org/10.1177/1461444819850755
(2021). Funny cats and politics: Do humorous context posts impede or foster the elaboration of news posts on social media? Communication Research, 48(1), 100–124. https://doi.org/10.1177/0093650219826006
(1996).
Knowledge activation: Accessibility, applicability, and salience . In E. T. HigginsA. W. KruglanskiEds., Social psychology: Handbook of basic principles (pp. 133–168). Guilford Press.(2007). American Patriotism, national identity, and political involvement. American Journal of Political Science, 51(1), 63–77. https://doi.org/10.1111/j.1540-5907.2007.00237.x
(2013). Comparing actual and self-reported measures of Facebook use. Computers in Human Behavior, 29(3), 626–631. https://doi.org/10.1016/j.chb.2012.11.007
(2017). From incidental news exposure to news engagement. How perceptions of the news post and news usage patterns influence engagement with news articles encountered on Facebook. Computers in Human Behavior, 76, 42–50. https://doi.org/10.1016/j.chb.2017.06.041
(2018). Picture this: The influence of emotionally valenced images on attention, selection, and sharing of social media news. Media Psychology, 21(2), 202–221. https://doi.org/10.1080/15213269.2017.1378108
(2013). Stumbling upon news on the Internet: Effects of incidental news exposure and relative entertainment use on political engagement. Computers in Human Behavior, 29(6), 2607–2614. https://doi.org/10.1016/j.chb.2013.06.005
(2020). The social media political participation model: A goal systems theory perspective. Convergence, 26(1), 135–156. https://doi.org/10.1177/1354856517750366
(2015). The rocky road from attitudes to behaviors: Charting the goal systemic course of actions. Psychological Review, 122(4), 598–620. https://doi.org/10.1037/a0039541
(1970). Passive learning from television. Public Opinion Quarterly, 34(2), 18–190. https://doi.org/10.1086/267788
(2016). The effect of website interactivity on political involvement. Journal of Media Psychology, 28(3), 136–147. https://doi.org/10.1027/1864-1105/a000200
(2000). The limited capacity model of mediated message processing. Journal of Communication, 50(1), 46–70. https://doi.org/10.1111/j.1460-2466.2000.tb02833.x
(2013). Elaboration or distraction? Knowledge acquisition from thematically related and unrelated humor in political speeches. International Journal of Public Opinion Research, 25(3), 291–302. https://doi.org/10.1093/ijpor/edt005
(1969). SMOG grading – a new readability formula. Journal of Reading, 12(8), 639–646.
(2001). Digital divide: Civic engagement, information poverty, and the Internet worldwide. Communication, society and politics. Cambridge University Press. https://doi.org/10.1017/CBO9781139164887
(2009). Being immersed in social networking environment: Facebook groups, uses and gratifications, and social outcomes. Cyberpsychology, Behavior and Social Networking, 12(6), 729–733. https://doi.org/10.1089/cpb.2009.0003
(2014). College students’ motivations for Facebook use and psychological outcomes. Journal of Broadcasting & Electronic Media, 58(4), 601–620. https://doi.org/10.1080/08838151.2014.966355
(1986). The elaboration likelihood model of persuasion. Advances in Experimental Social Psychology, 19, 123–205. https://doi.org/10.1016/S0065-2601(08)60214-2
(1986). Amusing ourselves to death: Public discourse in the age of show business. Viking Penguin.
Elisabeth Sifton books (2005). News vs. entertainment: How increasing media choice widens gaps in political knowledge and turnout. American Journal of Political Science, 49(3), 577–592. https://doi.org/10.1111/j.1540-5907.2005.00143.x
(2010). Post-broadcast democracy: How media choice increases inequality in political involvement and polarizes elections. Cambridge studies in public opinion and political psychology. Cambridge University Press.
(2010). Uses and gratifications of social media: A comparison of facebook and instant messaging. Bulletin of Science, Technology & Society, 30(5), 350–361. https://doi.org/10.1177/0270467610380009
(2012). lavaan: An R package for structural equation modeling. Journal of Statistical Software, 48(2), 1–36. https://doi.org/10.18637/jss.v048.i02
(2016). The changing role of media use in political participation. Journal of Media Psychology, 28(3), 97–99. https://doi.org/10.1027/1864-1105/a000204
(2007). Campaign ads, online messaging, and participation: Extending the communication mediation model. Journal of Communication, 57(4), 676–703. https://doi.org/10.1111/j.1460-2466.2007.00363.x
(2002). Forgetting all else: On the antecedents and consequences of goal shielding. Journal of Personality and Social Psychology, 83(6), 1261–1280. https://doi.org/10.1037//0022-3514.83.6.1261
(1983).
Cognitive Processes in humor appreciation . In P. E. McGheeJ. H. GoldsteinEds., Handbook of humor research: Volume 1: Basic issues (Vol. 56, pp. 39–57). Springer. https://doi.org/10.1007/978-1-4612-5572-7_3(2005). Web personalization as a persuasion strategy: An elaboration likelihood model perspective. Information Systems Research, 16(3), 271–291. https://doi.org/10.1287/isre.1050.0058
(2016). Stimulating citizenship or expanding entertainment? The effect of Facebook on adolescent participation. New Media & Society, 18(5), 817–836. https://doi.org/10.1177/1461444814549006
(2014). Facing an uncertain reception: Young citizens and political interaction on Facebook. Information, Communication & Society, 17(2), 203–216. https://doi.org/10.1080/1369118X.2013.862563
(2016). Curated flows: A framework for mapping media exposure in the digital age. Communication Theory, 26(3), 309–328. https://doi.org/10.1111/comt.12087
(2016). Accidental exposure to politics on social media as online participation equalizer in Germany, Italy, and the United Kingdom. New Media & Society, 18(9), 1857–1874. https://doi.org/10.1177/1461444815616223
(2017). Political communication in a high-choice media environment: A challenge for democracy? Annals of the International Communication Association, 41(1), 3–27. https://doi.org/10.1080/23808985.2017.1288551
(2010). Trivializing the news? Affective context effects of commercials on the perception of television news. Mass Communication and Society, 13(2), 139–156. https://doi.org/10.1080/15205430903128924
(2008). The Privileged role of the late-night joke: exploring humor’s role in disrupting argument scrutiny. Media Psychology, 11(1), 119–142. https://doi.org/10.1080/15213260701837073
(1980). Acquisition of information from educational television programs as a function of differently paced humorous inserts. Journal of Educational Psychology, 72(2), 170–180. https://doi.org/10.1037/0022-0663.72.2.170
(1984). Passive learning: When the media environment is the message. Public Opinion Quarterly, 48(3), 629–639. https://doi.org/10.1086/268864
Appendix
Stimulus Material and Additional Analysis
