The Effects of Humor in the Media: A Review of Experimental Research

1. Introduction

Media, both traditional and social, are pervasive in modern societies. Though early approaches promulgated a one-way flow of messaging from content producers to audiences, alternative views have cast receivers as active agents using media to meet their own goals (e.g., uses and gratification approach in Katz et al., 1974; mood management theory in Zillmann, 1988). In their review of media effects, Valkenburg et al. (2016) distinguish between media use and media effects. Media use involves the selection of media forms, while media effects refer to changes in cognitions, affect, attitudes, and behavior as a result of exposure to, or engagement with, media. Given the long history of research on “mass communication,” it is not surprising that a number of meta-analyses of media effects have been conducted, for example, risk-taking (Fischer et al., 2011), media violence (Ferguson & Kilburn, 2009), body image (Huang et al., 2020), and cultivation effects (Hermann et al., 2023). Clearly, a variety of topics fall under the media effects umbrella. However, one prominent function of media is entertainment (Zillmann & Bryant, 1994), often served through content that is humorous.

Humor has been “ever present in the mass media” (Kinsey, 1993/1994, p. 50) since its earliest days, a fact that has not gone unnoticed by humor researchers. Cantor (1977), for example, conducted a content analysis of humor on prime-time television and in cartoon magazines. Gruner (1976/1996) examined the role of humor as a tool for persuasion in mass communication, concluding that it might garner attention but was not effective in moving audiences. Additionally, humor in advertising has been the focus of voluminous research, including a number of meta-analyses (e.g., Kaur et al., 2022; Karpinska-Krakowiak et al., 2025). However, given the specific focus of advertising as a media genre, results of these endeavors may be less pertinent to a broader consideration of media effects.

On the other hand, reviews of the relationship between media effects and humor are sparse. The persuasive effects of satire have been explored meta-analytically by Dobmeier et al. (2023) and Burgers & Brugman (2022). A systematic review of social media humor in the COVID-19 pandemic was conducted by Alkaraki et al. (2024). Kaltenbacher and Drews (2020) provided a systematic narrative review of humor’s role in environmental communications. Becker (2020) reviewed research on political satire, offering various theoretical frameworks that might expand the understanding of satire’s effects. For the most part, these efforts are restricted to a specific type of humor or context. Thus, there appears to be a need for a broad-based review of humor and media effects research. The present scoping review seeks to address this gap.

1.1. Humor Theories and Media Effects

Combining insights from Becker’s (2020) review of political satire and Valkenburg et al.’s (2016) review of media effects, this section draws connections between media effects and the “big three” humor approaches (i.e. superiority, relief, and incongruity theories), offering what van der Wal et al. (2020) refer to as an “intertheoretical-perspective” (p. 480). Superiority theory involves “winner and losers” (Gruner, 1997). That is, humor involves social comparison processes whereby our laughter signals our feelings of elevation relative to the derogation of some target. In line with Becker, social identity theory (Tajfel & Turner, 1979), parasocial interaction (Horton & Wohl, 1956), and identification (Cohen, 2001) might all be relevant here. Media humor often involves stereotypical depictions deployed, for example, in situational comedies with stock characters or by comedians making jokes about public figures. These forms of disparagement humor often invoke group identities, reinforcing ingroup/outgroup dynamics (Martineau, 1972; Miczo, 2022). In line with prejudiced norm theory (PNT; Ford & Ferguson, 2004), those who are already prejudiced against some target may feel free(r) to act on their biases, post activation, if they are presented with a suitable opportunity. Additionally, affective disposition theory (ADT; Zillman & Cantor, 1976) argues that our feelings toward humor creators and humor targets affect our appreciation of humorous messages.

Relief (or release) theories focus upon laughter as a way to release tensions and inhibitions (Carrell, 2008; Morreall, 2009). Though early versions relied on a defunct hydraulic model of human emotions, the ideas persist today in the form of coping humor and humor as liberatory. Mood management theory (Zillmann, 1988) argues that we seek out opportunities to assuage negative moods or garner positive moods through media usage, even though the media selection process used to regulate mood operates without deliberate awareness. Mood management processes, then, are likely to drive the manner in which audience members seek out and enjoy entertainment (i.e., humorous) programming. Seeking out humorous programming should then be linked to broader coping efforts or desires to circumvent normative pressures.

Incongruity theory may be the most difficult to fit within the media effects framework. In part, this may be because incongruity is taken by many to be a defining feature of all humor. Suls’s (1972) model argued that humor involved a two-step process of comprehending an incongruity and then resolving it. More recently, the general theory of verbal humor (Attardo, 2001) has formalized this process with the notion of script oppositions (two scripts that are opposite in some sense) and the logical mechanism that makes sense of that oppositeness. Many forms of media humor involve a set-up and a punchline. Often, the set-up is a factual statement or a depiction of something “in reality.” The punchline presents the incongruity, requiring the receiver to “double back” and do the inferential work of making sense of the humor. Becker (2020) discusses hedonic and eudaimonic motivations for humor exposure and both may be applied to incongruity theory. For example, consider the following Conan O’Brien joke told in 2016, and reported by Young (2020): “Donald Trump has dropped to second place in a national poll. On the bright side, he’s still polling Number 1 among Germans of the 1930s.” Notice that the set-up contains a statement of fact, while the punchline presents a statement that is intended to be taken non-seriously. Hedonically, one could enjoy the joke because one is primed by the context of a late-night comedy program, by one’s liking for Conan O’Brien, the fact that others are laughing, that one is motivated to enhance one’s positive affect, or the joke is clearly targeting Trump supporters (presumably an outgroup to many O’Brien and/or late night comedy fans). The entire joke could be discounted, or the punchline might be discounted, even while the factual premise might have been “learned” (i.e., constitute new knowledge). Eudaimonically, if one seeks deeper meanings within the joke, one might see the implications that Trump supporters are old white people, with possible Nazi leanings (cf. Miczo, 2022; Young). Consistent with Valkenburg et al. (2016), these explanations require the positing of indirect and conditional effects.

The purpose of this section has been to make clear that there are many points of connection between media effects processes and the foci of humor theories. However, connection does not equal integration, though integration cannot proceed without reviews of the current state of research. The concern with media effects research is with the changes in audiences upon exposure to media messages (form, content, creator). From the perspective of humor research, the concern is with establishing how humor produces those changes. This means examining the extent to which humor theory is incorporated into research designs, how humor is operationalized, outcomes linked to humor, and the inclusion of conditional and indirect effects. Accordingly, the present review was guided by the following research questions:

RQ1: Is research on humor and media effects theory-driven?

RQ2: How has humor been operationalized?

RQ3: What post-exposure response measures have been used?

RQ4: What outcomes are examined?

RQ5: Does the research include conditional or indirect effects?

2. Methods

We conducted a scoping review (Arksey & O'malley, 2005) to synthesize the evidence on our topic. The searches for articles were conducted between September 2025 and January 2026.

2.1. Data Sources and Search Strategy

An initial limited search was conducted by one of the authors on the Communication and Mass Media Complete database, Google Scholar, PsychINFO as well as in the journals European Journal of Humor Research and Israeli Journal of Humor Research using the key words search string “humor” AND “media effects” with a publication date from 2008 to 2026. The initial search involves title and abstract screening. Subsequent full text screening and a secondary search were conducted by the two authors. The secondary search was done in the flagship journal of media effects, Media Psychology, using the key word “humor” as well as in the PsycARTICLES database using the search string “humor” AND “media effects”. Then, one of the authors did another targeted search in the references list of some existing records. Finally, the two authors discovered a few additional articles by the same first author when discussing excluding one of the ineligible ones. The initial screening returned 40 articles.

After each search, the two authors further screened all records for both abstract and full-text level. Disagreements were discussed and resolved between the two raters. As a result, thirty-four records met the inclusion criteria.

2.2. Inclusion and Exclusion Criteria

The eligibility criteria of the review can be found in Table 1. Articles were considered as eligible for inclusion if they examined the effects of humorous messages using experimental methods with the humorous content being the stimulus material in the experimental group. We excluded studies in advertising research due to the unique nature of advertising messages. We retrieved the following information for all eligible articles after initial screening: author(s), title, year, topic, study design, operationalization of humor, type of humor, source of humorous content, humor measure for manipulation check, outcome variables, and results. The data was coded in a Google Excel spreadsheet. Next, two authors independently evaluated half of the 40 articles to assess their eligibility for inclusion in this review.

2.3. Data Extraction and Study Variables

The authors iteratively developed and refined the codebook through discussions and revisions. An eligible sample of thirty-four studies were split in half and the two authors each independently coded their seventeen articles. Questions and borderline cases were addressed through discussions between the two authors.

2.3.1. Theoretical Guidance

Theoretical guidance was coded in four categories including lack of theoretical guidance, guided by humor theory, guided by theories in media psychology, social psychology, or other areas, or lastly, guided by a combination of humor theory and other theories.

2.3.2. Humor Types

The researchers conducted a literature search of humor types and derived eight humor categories, disparaging humor, nonsense humor (e.g., slapstick, pranks, and pun), political humor, satire, parody, self-disparaging humor, science humor, and sex humor. During the coding process, new categories emerged and were added to the codebook for this variable, which include irony, hostile humor, and a combination of two or more types of humor. If any article did not clarify humor types, it will be coded as a missing value.

2.3.3. Humor Source

We are interested in the sources of humorous stimuli used in the experiments. Therefore, humor source was operationalized to take into consideration both the creator of the content and the type of media platform hosting the content. It was coded into seven categories including researcher-developed content, researcher-adapted traditional media content, researcher-adapted social media content, traditional media content, user-generated social media content, general website content, and unclear (when it is not explicitly discussed in the article).

2.3.4. Manipulation Check of Humor Stimuli

Manipulation checks of experimental stimuli help increase the scientific rigor of research and help eliminate threats to internal validity. We are interested in whether a study conducted a manipulation check of humorous content before or after the experimental procedure. Accordingly, the manipulation check of humor stimuli was coded into a dichotomous measure of 1 = “Yes” or 0 = “No”.

2.3.5. Post-exposure Positive Emotion

The primary emotional response related to humor is amusement (Goldstein, 2021). Therefore, this review is interested in whether this emotion or any relevant positive emotion was measured in a given study after participants’ exposure to the humorous media content. This variable is coded based on presence and absence of a measure of positive emotion after humor message consumption.

2.3.6. Post-exposure Perceived Funniness

The cognitive response to humorous content is perceived funniness of such content (Goldstein, 2021). We are interested in whether a given study measured this perception after participants’ humorous content consumption. Thus, we coded the absence or presence of this measure in the studies being reviewed.

2.3.7. Types of Effects

To contextualize the review and examine different types of study outcomes, we also coded six different types of media effects including cognitive (e.g., elaboration), affective (including both self-report and behavioral observation/physiological measurement), attitudinal, knowledge, interpersonal (e.g., liking, trust) and behavioral (or proxy such as behavioral intention).

2.3.8. Indirect/Conditional Effects

We are interested in the mediators or moderators introduced in the studies. Therefore, we coded this variable with four categories: mediation, moderator, moderated mediation, and a combination of two or three of the previous categories.

2.4. Data Analysis

To answer the research questions, frequencies and other descriptive statistics such as mode and range were used.

3. Results

The first research question concerned the extent to which humor-media effects research was guided by theory. According to the results, 67.6% of studies were guided by a theoretical perspective. There was a slight tendency for media- and social psychology theories to be used (32.35%), rather than humor theory (20.59%). However, whereas there was more heterogeneity in the use of non-humor theories, humor theory usage was dominated by the affective disposition theory (5 articles), followed by prejudiced norm theory (2 articles).

RQ2 focused on the methodological operations for using humor within the experimental designs. Regarding the types of humor, satire (20.6%) and disparaging humor (20.6%) were the most common, with political humor (14.7%) following closely behind (See Table 2). The most common source of experimental stimuli was traditional media content (e.g., using clips from television programs) (44.1%) (See Table 3). Studies typically included a manipulation check to confirm that humor stimuli were perceived as higher in mirth or funniness compared to the non-humor conditions (61.8%).

To address RQ3, we examined whether each study measured emotional and cognitive responses to humorous media content after the content exposure. As a result, only five of the 34 studies (14.71%) assessed positive affective responses to humor. The majority of studies (67.65%) assessed perceived funniness of the humor. Only four of the studies (11.76%) measured both affective and cognitive responses to humor content. And, a total of 10 studies (29.41%) did not measure either type of responses.

Research question 4 focused on the outcomes included upon exposure to experimental manipulations. The most common outcomes were attitudinal (55.9%) and affective (52.9%). Cognitive, knowledge, and interpersonal outcomes were used infrequently.

The final research question concerned the exploration of conditional or indirect effects. The largest category was for the absence of such effects (41.2%). Where these effects were included, moderation was the most common (29.4%) with mediation and moderated mediation being equally likely to be used (at 11.8% each).

4. Discussion

The purpose of this review was to examine the use of humor in experimental media effects research. Based on a review of 34 published articles, we concluded that while two-thirds of the studies were guided by theory, only around 21% included a humor theory. Further, humor was often operationalized by using traditional media content, with the humor itself being disparaging humor or satire. Using some measure of humor as a manipulation check was common. However, only a few studies actually assessed positive emotion as a response to the humor content at post-exposure. A majority of studies assessed perceived funniness of humor content. Attitudinal and affective outcomes were used frequently, and moderation was included more frequently than mediation.

4.1. Humor and Media Effects

Two areas are heavily represented in the kind of research included here (i.e., experimental designs): politics and prejudice. These research foci align most closely with versions of the superiority theory, reflecting the hopes and fears of the media. On the one hand, the hope is that humor can promote engaged citizenship. Primarily relying on cognitive processes of learning and elaboration, exposure to the political comedy of hosts such as John Oliver or Samantha Bee can prompt viewers to engage with the issues selected as stimuli by the researchers. On the other hand, there is the fear that disparaging humor presented through media can reinforce stereotypes or problematic social hierarchies. Given that the political comedy most often used is from liberal hosts, and the disparaging humor is often linked to conservative ideologies, these may be two sides of the same coin. Such a conclusion aligns with Young’s (2020) argument that liberals enjoy wit and satire, while conservatives prefer humor that is more blunt and hurtful.

It is possible that a general model could be proposed drawing upon both PNT (Ford & Ferguson, 2004) and ADT (Zillmann, 1988). PNT argues that humor erects a conversational norm of levity, an atmosphere where we are all “just joking.” For those with pre-existent prejudices, this norm justifies subsequent behaviors expressive of that prejudice. ADT could be used to specify the conditions under which people are likely to seek out the situations where any particular kind of humor is likely to be displayed. In other words, people may be more likely to seek out situations for humor where they like the humorist and dislike the target(s). In those situations, three possible mechanisms may be operative: 1) people may seek to manage their mood by continued exposure to the humor (including their own participation, if possible, or sharing with co-viewers); 2) since much humor involves some sort of target, often depicted stereotypically, the display of humor may activate group identities; 3) rather than prejudice being “released,” viewers can engage with the content of humorous stimuli, whether this involves learning, cognitive elaboration, or sense-making. As a result of feeling good, or affirming an activated identity, or through sense-making, individuals might, if given the chance, act in predictable directions (i.e., humor influences behaviors). Such a model could account for disparagement humor’s effects, or political comedy viewing (insofar as hosts often have some target in mind) or the more prosocial behaviors studied by Wilson et al (2025).

One area that is under-represented here is the role of humor sought out for entertainment purposes. Lurking behind some studies may be a suspicion that humor cannot be mere frivolousness, part of the ornamentation of everyday life. Such a view collapses Mulkay’s (1988) distinction between the humorous mode and the serious mode. The serious mode is the world of Gricean maxims, whereby interlocutors bind themselves to a unified field of discourse. The humorous mode is one of ambiguity, contradiction, exaggeration, and distorted logics. According to this view, humor should not be treated as an instrument of the serious mode. On the other hand, many allow that a humorist may have a serious point to get across. Yet, it must be acknowledged that such a message is not simply a nugget of seriousness wrapped in a contain of humor that can be discarded and presented unproblematically to audiences. The medium may not be the message but it is part of the context of the message, a context that creates expectations among audience members. That is, just as people have expectations when they attend a lecture, or when a friend says “Can I tell you something?”, they have expectations for humorous contexts. The question then becomes: what sort of people seek out what sort of humor for what purposes of entertainment?

Based on these reflections, and guided by the results of this review, the following methodological recommendations are offered. First, researchers may consider more researcher-generated content in their experimental manipulations. Working with graphic designers and other media content producers, and utilizing current technological tools, experiments examining more general theoretical processes might be devised. Second, test different levels of humor or compare different types of humor. It was more common to compare a humorous stimulus against a non-humor manipulation. Given the subjective nature of what people find funny, such an approach risks masking nuanced differences. Third, measures of mirth/funniness/humor appreciation should be incorporated more frequently as an independent variable. Certainly, it is important to provide some estimate of the success of an experimental manipulation. When humor ratings are only used to confirm that a humorous stimuli is rated more humorous than a non-humorous stimuli at the group level, and those humor ratings are not utilized further, the very thing that is often presumed to cause an effect (i.e., humor) is left unspecified. Additionally, the emotional response to humor (e.g., amusement, joy) should be measured after content exposure not only to account for the variances in any outcome variable examined but also to identify humor-content-specific affective responses and their intensity. Finally, examining entertainment functions in relation to other outcomes (such as attitudes) might provide further specification to humor mechanisms. For example, humor comprehension is different from humor appreciation, and each might be related in a different way with attitudinal or affective outcomes (e.g., hedonic or eudaimonic motivations).

4.2. Limitations

There are several limitations that should be kept in mind regarding this review. First, there was some difficulty in identifying relevant articles to review. Changes in the media alter the meanings of “media effects.” Accordingly, our interpretation of the term restricting it to experimental designs that permitted quantifying the effect of a media message may seem over restrictive to some. Further, our decision to examine both traditional and social media may appear overly broad to others. Similarly, we used the term “humor” as a broad umbrella term without conducting extensive searches using terms such as “laughter,” “mirth,” or “funniness.” It is therefore possible that some studies were missed as a result.

Second, there were only a small number of articles that met our criteria for review. This meant that both humor and media effects theories that typically rely on surveys, such as uses and gratifications theory, were not represented. By the same token, active researchers who typically employ experimental designs in particular topic areas (e.g., Becker) were over-represented. Finally, there are limitations of the coding itself. Given the range of different topic areas, we decided not to attempt to summarize results. Rather, we were more interested in issues regarding the operationalization of humor and the outcomes associated with those operational decisions. Nevertheless, there were still difficulties in some cases with balancing the trade-offs between the number of categories and the coding of specific items.

5. Conclusions

The continued development of humor studies and media effects theories suggests a need for overviews to summarize, not just the results of research (as in meta-analysis), but the processes of research. Based on this review, a variety of topics are being covered, but political and disparagement humor are predominant. These more conflict-oriented approaches suggest a neglect of the more affiliative ways of using humor in the media. Similarly neglected is attention to basic humor processes, such as mechanisms of incongruities and their resolutions. Thus, we end with a call for greater integration of these two fruitful areas, with the development of theories of humorous media effects.