research humor

January 26, 2016

A Scientific Theory of Humor

The “entropy” of nonsense words is linked to their funniness, research finds

By Cindi May &

Given that humor is such a powerful tool for social success, it’s not surprising that scientists have sought to determine the perfect formula for funny.  

©iStock

George W. Bush was not known for his cunning intellect, but he did have a good sense of humor.  In a commencement address at Southern Methodist University, he famously told the graduates, “For those of you graduating with high honors and distinctions, I say well done.  And as I like to tell the “C” students, you too can be president.”  Like Bush, many of us use humor to diffuse difficult situations, mask nervousness, soften criticism, and cope with failure.  Humor also serves the role of locksmith in both platonic and romantic social interactions, as it helps us break the ice, gain social acceptance, and initiate romantic overtures.  Both men and women tend to seek mates who have a good sense of humor, and we perceive funny people as smarter, more attractive, and more personable.

Given that humor is such a powerful tool for social success, it’s not surprising that scientists have sought to determine the perfect formula for funny.  Although there are many competing theories (and no definitive answers) about how humor functions, new research by Chris Westbury, Cyrus Shaoul, Gail Moroschan, and Michael Ranscar suggests that at least one key ingredient can be found in a 200 year-old theory proposed by philosopher Arthur Schopenhauer .

In a nutshell, Schopenhauer suggested that humor derives from an incongruous outcome of an event for which there is a very specific expectation.  It is the violation of the specific expectation that creates humor.  Consider this pun: “ When the clock is hungry it goes back four seconds .”  The notion of a clock eating is incongruous with our knowledge of the world, but that alone is insufficient to create humor.  The statement, “ When the clock is hungry it eats a cheeseburger ,” is also incongruous, but “ eating a cheeseburger ” does not violate any

On supporting science journalism

If you're enjoying this article, consider supporting our award-winning journalism by subscribing . By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.

specific expectation about a clock and so the statement is far less amusing.  It is the ending, “ goes back four seconds ,” that elicits a humorous response (albeit an extremely mild one), and it does so because of our understanding of the dual meanings of the words “four” and “seconds,” and our expectation about which of those meanings apply to a clock.

In three experiments, Westbury and colleagues tested the idea that greater incongruity between expectations and outcomes produces a stronger feeling of humor.  They did so by examining the humor in non-words, which are strings of letters (e.g., digifin) that form a pronounceable but meaningless unit.  Non-words offer a unique advantage in the analysis of the role of expectation-violation in humor, as they are relatively devoid of meaning and thus allow a more pure assessment of influence of incongruency on funniness. 

In their first study, Westbury et al. assessed whether there is any consistency in the funniness ratings of non-words.  Although specific instances of humor are not always considered universally funny (consider for example popular skits from Saturday Night Live , which tickle many people but also offend others), something is objectively humorous only if there is some consensus about its hilarity. Thus Westbury and colleagues asked nearly 1000 students to rate a total of almost 6000 randomly-generated non-word strings (e.g., artorts ) on funniness.  The results indicated that these random non-words had reliably consistent humor ratings.  If one participant found a given non-word funny, it was likely that others found that same item funny as well, and vice versa.

Westbury et al. next tried to understand what made certain non-words funny (and others not so much).  In two additional studies, they directly examined funniness ratings for non-words that varied with respect to entropy, that is, the extent to which the combination of letter strings was incongruous or unexpected.  To understand incongruity in non-words, it is important to know that some letters are more likely than others in the English language (e.g., “E” is more frequent than “Q”), and furthermore that some letter combinations are more likely than others.  Thus the entropy of a non-word is essentially a measure of the summed probabilities of the individual letters in each string.  Non-words with unusual letters and/or combinations have low entropy and offer more surprise.  In line with Schopenhauer’s theory, Westbury et al. predicted that items with low entropy would receive the highest humor ratings, as these items were most likely to violate expectations about letters and words.

In one study, participants saw two non-word strings (e.g., quarban, mestead ) that appeared simultaneously on a computer screen, and on each trial had to select the non-word that they perceived to be more humorous.   Each participant made judgments for 50 pairs.  In another study, participants saw non-word strings that appeared one at a time on a computer screen, and had to rate the humor of each item on a scale from “least humorous” to “most humorous.”  Participants each rated the humor of 100 non-words.  The findings from both of these studies supported the hypothesis that non-word strings with low entropy are perceived as more humorous.  Strings with low entropy (e.g., himumma ) were reliably chosen as more humorous than paired strings with higher entropy (e.g., tessina) , and strings with lower entropy were judged to be funnier than strings with higher entropy. When we expect one thing, even something as simple as letter combinations, and that expectation is violated, we chuckle. 

It is important to note, however, that at this point we cannot pinpoint low entropy as the definitive source of humor. While these studies demonstrate that expectation violation increases perceived humor, only one type of entropy (i.e., the probability associated with letter strings) was studied here, and with more complex stimuli other types of expectation violation may contribute to amusement.  Even for non-words, many other layers of expectation violation are possible (e.g., how many double letters, such as “zz,” are included in a string, how unusual is the string’s phonology). Indeed, although Westbury et al. intentionally used non-word stimuli because non-words are fairly meaningless, they still found that a handful of the non-word items that were rated most humorous were not necessarily those with lowest entropy, bur rather those that were similar to or contained parts of dirty words (e.g., whong, nip, poo ).  Of course one could argue that this finding demonstrates a different kind of expectation violation, as taboo words are arguably unexpected in a serious scientific study, and so are likely to be perceived as funny in that context. 

Unfortunately, understanding that outcomes that violate expectations tend to be perceived as funny doesn’t necessarily make it easier to say or write something humorous.  If creating humor involved a simple scientific calculation, more of us nerdy researchers would be out of the classroom and into the late night comedy circuit (or perhaps we too could be president).  Instead, we’ll likely go back to the lab and tweak our non-word generators. Himumma!

Cindi May is a professor of psychology at the College of Charleston. She explores avenues for improving cognitive function and outcomes in college students, older adults and individuals who are neurodiverse.

International Society for Humor Studies Journal

The Society's journal, HUMOR , provides an interdisciplinary forum for the publication of high-quality articles on humor as an important and universal human faculty. Contributions include theoretical papers, original research reports, scholarly debates, and book reviews. The journal is currently published by DeGruyter, and all submissions are peer reviewed. Since 1988, HUMOR has published over 1000 articles and book reviews.Below are links to sample articles and book reviews with free online access on the DeGruyter website.

Comedy Bootcamp: Stand-up Comedy as Humor Training for Military Population Authors: Andrew Olah, Janelle Junkin, Thomas Ford, and Sam Pressler (2022: HUMOR 35.4)

Satire without Borders: The Age-Moderated Effect of One-Sided versus Two-Sided Satire on Hedonic Experiences and Patriotism Authors: Mark Boukes and Heather LaMarre (2023: HUMOR : 36.1)

The Humor Transaction Schema: A Conceptual Framework for Researching the Nature and Effects of Humor Authors: Jessica Milner Davis and Jennifer Hofmann (2023: HUMOR : 36.2)

Laughing to Keep from Dying: African American Satire in the Twenty-First Century Author: Danielle Fuentes Morgan Reviewers: Danielle Bobker and Catherine Sauvé Dowers (2022: HUMOR 35.4)

Political Humor in a Changing Media Landscape: A New Generation of Research Editors: Jody C. Baumgartner and Amy B. Becker Reviewer: Sara Polak (2022: HUMOR 35.4)

That’s not funny: how the right makes comedy work for them Editors: Jody C. Baumgartner and Amy B. Becker Reviewer: Sara Polak (2022: HUMOR 35.4)

Sexual Humour in Africa: Gender, Jokes, and Societal Change Author: Ignatius Chukwimah Reviewer: Tessa Dowling (2023: HUMOR 36.1)

Current Editor-in-Chief Christian F. Hempelmann Literature & Languages Department Texas A & M University, Commerce humorjournal-at-gmail.com

Former Editors-in-Chiefs Salvatore Attardo Literature & Languages Department Texas A & M University, Commerce Thomas E. Ford Psychology Department Western Carolina University Giselinde Kuipers Sociology Department University of Amsterdam Lawrence Mintz American Studies Department University of Maryland Victor Raskin Linguistics Department Purdue University

ISHS Executive Secretary Martin D. Lampert Psychology Department Holy Names University humorstudies-at-outlook.com

Editorial Board Members Salvatore Attardo Texas A & M University, Commerce Nancy Bell, Washington State University Delia Chiaro University of Bologna Wladislaw Chlopicki Jagiellonian University Jessica Milner Davis University of Sydney Thomas E.Ford Western Carolina University Gil Greengross Aberystwyth University Giselinde Kuipers University of Amsterdam Liisi, Laineste, Estonian Literary Museum Sharon Lockyer Brunel University Moira Marsh Indiana University John Morreall College of William & Mary Alleen Nilsen Arizona State University Don Nilsen Arizona State University Elliott Oring Cal State University, Los Angeles Rene Proyer Martin Luther Univ. Halle-Wittenberg Victor Raskin Purdue University Willibald Ruch Zürich University Limor Shifman Hebrew Univ. of Jerusalem Villy Tsakona National and Kapodistrian Univ. of Athens

• Architecture and Design
• Asian and Pacific Studies
• Business and Economics
• Classical and Ancient Near Eastern Studies
• Computer Sciences
• Cultural Studies
• Engineering
• General Interest
• Geosciences
• Industrial Chemistry
• Islamic and Middle Eastern Studies
• Jewish Studies
• Library and Information Science, Book Studies
• Life Sciences
• Linguistics and Semiotics
• Literary Studies
• Materials Sciences
• Mathematics
• Social Sciences
• Sports and Recreation
• Theology and Religion
• Publish your article
• The role of authors
• Promoting your article
• Abstracting & indexing
• Publishing Ethics
• Why publish with De Gruyter
• How to publish with De Gruyter
• Our book series
• Our subject areas
• Your digital product at De Gruyter
• Contribute to our reference works
• Product information
• Tools & resources
• Product Information
• Promotional Materials
• Orders and Inquiries
• FAQ for Library Suppliers and Book Sellers
• Repository Policy
• Free access policy
• Open Access agreements
• Database portals
• For Authors
• Customer service
• People + Culture
• Journal Management
• How to join us
• Working at De Gruyter
• Mission & Vision
• De Gruyter Foundation
• De Gruyter Ebound
• Our Responsibility
• Partner publishers

Your purchase has been completed. Your documents are now available to view.

International Journal of Humor Research

• Online ISSN: 1613-3722
• Print ISSN: 0933-1719
• Type: Journal
• Language: English
• Publisher: De Gruyter Mouton
• First published: January 1, 1988
• Publication Frequency: 4 Issues per Year
• Audience: Researchers, students and practitioners with an interest in the scholarly study of humor and related phenomena

The European Journal of Humour Research

Current Issue

The EJHR is an open-access, academic journal published by Cracow Tertium Society for the Promotion of Language Studies and endorsed by The International Society for Humor Studies (ISHS) . The EJHR publishes full research articles, shorter commentaries, which discuss ground-breaking or controversial areas, research notes, which provide details on the research project rationale, methodology and outcomes, as well as book reviews. The journal has a special focus on supporting PhD students and early career researchers by providing them with a forum within which to disseminate their work alongside established scholars and practitioners.

The EJHR welcomes submissions that combine research and relevant applications as well as empirical studies detailing their usefulness to the study of humour. All contributions received (apart from book reviews) undergo a double-blind, peer-review process. In addition to established scholars within humor research, we invite those as yet unfamiliar with (or wary of) humor research to enter the discussion, especially based on less known or less covered material. The elaboration of joint methodological frameworks is strongly encouraged. For further details or inquiries you may contact the Editors.

No charges are applied either for submitting, reviewing or processing articles for publication.       

The journal is now listed in important international indexing bases including Scopus and Scimago ranking :

This publication is supported by the CEES and ELM Scholarly Press.

Announcements

Interesting scillit data on the ejhr journal.

Browse and enjoy some statistics on our journal in the decade from 2014-2024.

EJHR has joined the Reviewer Credit system

We are pleased to announce that the European Journal of Humour Research  is now registered with ReviewerCredits, the system for supporting and appreciating journal reviewers for their work. 

Most read and cited papers 2013-2024

Here is the updated list of papers published in EJHR in the years 2013-2024  that received most views. Below is also the list of most cited papers . Congratulations to the Authors.

Commentary articles

EDITORIAL article

Editorial: humor and laughter, playfulness and cheerfulness: upsides and downsides to a life of lightness.

• 1 Department of Psychology, University of Zurich, Zurich, Switzerland
• 2 Department of Psychology, University of Sunderland, Sunderland, United Kingdom
• 3 Department of Psychology, Martin-Luther-University Halle-Wittenberg, Saxony-Anhalt, Germany
• 4 Department of Psychology, National Taiwan Normal University, Taipei, Taiwan

Editorial on the Research Topic Humor and Laughter, Playfulness and Cheerfulness: Upsides and Downsides to a Life of Lightness

Introduction

This research topic brings together the four research areas of humor, laughter, playfulness, and cheerfulness. There are partial overlaps among these phenomena. Humor may lead to laughter but not all laughter is related to humor. Playfulness is considered the basis of humor (a play with ideas), but not all play is humorous. Cheerfulness is considered the temperamental basis of good humor, a disposition for laughter and for keeping humor in face of adversity but it mostly overlaps with the socio-affective component of humor. Laughter was considered a play signal and to indicate the annulment of seriousness, but there is play without laughter and laughter outside of play. Cheerfulness might facilitate play and cheerful state might be raised due to play but again the conceptual overlap is only partial. They all contribute to levity in life and their apparent similarity suggests studying them together to map out the territory; i.e., to see where they overlap and what is specific. While these traits and behaviors have the potential to contribute to a good life, there is the danger of overlooking their non-virtuous facets; that is, laughter may not only be expressing amusement but scorn directed at people, humor may be benevolent but there is also sarcasm, and playfulness may elicit positive emotions but also risk prone behaviors. While this research topic solicited articles to these four domains without the aim to connect them, a few articles did and it is expected that growing together will be one outcome of this compilation of articles.

Currently, these fields are studied mostly in isolation. A literature search (using the psychology database of Web of Science Core Collection from 1900, 06.08.2018) yielded that humor is clearly leading in terms of number of publications ( n = 3,006), followed by laughter ( n = 1,412), playful(ness) ( n = 629), and cheerful(ness) ( n = 204). As a comparison, antonyms were studied as well, and yielded higher numbers, such as for crying ( n = 1640), serious-mindedness (or seriousness) ( n = 892), and sadness ( n = 3,654). The latter indicates that sadness is 18 times more frequently researched than cheerfulness.

Next, the frequency of articles combining terms was investigated. Combinations of humor and one of the other key terms are rather infrequent with the exception of “humor and laughter” ( n = 454), suggesting that about 10% of all articles on humor also refer to laughter. Humor and playfulness ( n = 59) and humor and cheerfulness ( n = 53) represent only 2% of all articles on humor, and these numbers are still much higher than any combination among the other three. This clearly shows that work is needed integrating these areas to examine how the concepts overlap both regarding their defining substance but also in predicting third variables. It should be mentioned that in a pioneering publication preceding the renaissance of empirical humor research three of the keywords were considered together. Toronto-based English psychologist ( Berlyne, 1969 ) gave an account of laughter, humor, and play in a chapter in a handbook of social psychology. The compilation of research in the four fields is aimed at deepening our understanding of these concepts and stimulating research combining them.

Overview of Studies

There are 32 manuscripts in this research topic. Not surprisingly, most articles are on various aspects of humor, followed by laughter (including dispositions to ridicule and being laughed at), playfulness and cheerfulness. To highlight some prevalent issues beforehand: Individual studies relate to introducing new concepts, or new scales or working on existing ones ( Aykan and Nalçaci ; Bruntsch and Ruch; Heintz et al.; Hofmann et al.; Hofmann et al.; Ruch and Heintz; Ruch et al.; Ruch et al. ). Furthermore, substantial attempts are made to develop and evaluate trainings and interventions ( Auerbach; Linge-Dahl et al.; Tagalidou et al.; Wellenzohn et al. ). There is also a significant number of cross-cultural comparisons ( Heintz et al.; Pang and Proyer; Tosun et al. ) and systematic literature reviews ( Chadwick and Platt; Linge-Dahl et al. ). What research questions were posed and what have we learned in the different fields?

Humor and Humor-Related Traits

Seven contributions relate to humor. Two are systematic reviews summarizing the use of humor in their related fields. Chadwick and Platt 's paper draws upon the 32 existing articles on humor with regards to intellectual disability, which they found grouped into eight emergent themes. The paper showed humor to be of importance in social interactions, not only for people with intellectual disabilities but those who support them and highlighted both the positive and the negative role of humor for both groups. However, the authors suggest that future studies should aim for more empirical rigor when investigating this important, yet complex construct. As Heintz et al. highlighted, the terminology of a dichotomized thinking of positive and negative humor may be a too simplistic approach, especially when thinking about fostering positive relationships. For example, employing carers with a propensity for benevolent humor may help forge more than a work relationship, but a friendship.

In the study of humor assessment and interventions in palliative care, Linge-Dahl et al. reviewed 13 papers. The review found that although the papers were difficult to compare, it was clear that humor is an appropriate and useful resource in palliative care of terminally ill patients (in different settings, such as hospices or oncology wards). Given this review accounts for the last 20 years, the authors note that research is still exceptionally limited, although humor interventions showed promising results on many well-being outcomes.

Humor as a quality that can be trained and developed evidently has potential not only to increase well-being in the terminally ill but also to reducing stress, depressiveness, and anxiety in a population of sub-clinical individuals ( Tagalidou et al. ). This pilot intervention demonstrated encouraging evidence that a humor training can have a stable, long-lasting impact on increasing positive affective states and reducing levels of stress, depressiveness and anxiety. This study also reported a relatively low attrition rate, which would suggest that participants were enjoying themselves, whilst having an overall positive impact on their mental health.

Wellenzohn et al. studied who benefits from online humor-based positive psychology interventions. In Study 1, personality traits were tested and it was the extraverts that benefitted more from the three funny things intervention than introverts did. Remembering emotional events allows reliving the emotion and the extraverts' tendency to positive emotions (i.e., the amusement due to the funny events during the day) apparently contributed to increasing their level of happiness and to lowering their depressive symptoms. In Study 2, no moderating effects were found for sense of humor on the effectiveness of the five humor-based interventions tested. Interestingly, however, changes in sense of humor from pretest to the 1-month follow-up predicted later changes in happiness and depressive symptoms. Thus, increases in sense of humor during and after the intervention are associated with the interventions' effectiveness.

Instruments that measure aspects of humor were investigated in five studies. Heintz et al. investigate responses to the BenCor in 25 samples from 22 countries. The BenCor measures humor aiming at the good and may be seen as a character (as different from personality or temperament) approach to humor. Benevolent humor treats human weaknesses and wrongdoings benevolently, while corrective humor aims at correcting and bettering them. The 12 items exhibited sufficient psychometric qualities in most of the samples. Metric measurement invariance was supported across the 25 samples, and scalar invariance was supported across age and across gender. This study supported the suitability of the 12 marker items of benevolent and corrective humor in different countries, enabling cross-cultural research and eventually applications of humor aiming at the good. Importantly, benevolent and corrective humor were clearly established as two positively related, yet distinct dimensions of virtue-related humor.

Ruch and Heintz study the construct and criterion validity of the HSQ ( Martin et al., 2003 ), which assesses humor styles. They argue that each item entails construct-relevant content (i.e., humor) but also (unwanted) variance produced by the item context. The 32 items were experimentally manipulated to strip off the context or to substitute the humor content by non-humorous alternatives (i.e., only assessing context). Study 1 shows that humor is not the primary source of the variance in three of the HSQ scales with the self-defeating humor style being primarily determined by the context. Study 2 shows that also the relationships of the HSQ with personality were reduced and those with subjective well-being vanished when the non-humorous contexts in the HSQ items were controlled for. For self-defeating, removing the context rendered the results to a positive rather than a negative view of the humor in this humor style. The results suggest that the items of humor instruments warrant careful examination.

Ruch et al. enlarge the list of styles of humor by adding fun, benevolent humor, non-sense, wit, irony, satire, sarcasm, and cynicism and by providing first evidence for the reliability and validity of a set of 48 marker items for their assessment, the Comic Style Markers (CSM). Exploratory and confirmatory factor analyses showed that the eight styles could be distinguished in English- and German-speaking samples, and studying self- and other-reports supported both convergent and discriminant validity. Studies also showed that the scales tapped differentially into personality, intelligence, and character strengths; for example, wit correlated with verbal intelligence, fun with indicators of vitality and extraversion, and while benevolent humor was related to strengths of the heart, the styles related to mock/ridicule (i.e., sarcasm, cynicism, but also irony) correlated negatively with character strengths. The results suggest that more styles may be distinguished than was done hitherto, which is also confirmed by Heintz and Ruch (2019) .

Two more studies examine irony in more detail and distinguish between two forms. Bruntsch and Ruch investigate irony in ironic criticisms (i.e., mock positive evaluation of negative circumstances) and ironic praise (i.e., mock negative evaluation of positive circumstances). They introduce the TOVIDA (Test of Verbal Irony Detection Aptitude) containing 26 scenario-based items for the detection of ironic criticism vs. ironic praise. Initial validation is provided by exploring personality and ability correlates of the two TOVIDA scales. Relatedly, Milanowicz et al. study mocking compliments and ironic praise from an interactional gender perspective. The ability to create irony is assessed and related to state and trait anxiety. Male responses were consistently more ironic but both genders used more irony in response to male ironic criticism than to female ironic praise. Anxiety predicted irony comprehension and willingness to use irony. The results enrich the discussion within the framework of linguistic intergroup bias and natural selection strategies.

Also Aykan and Nalçaci introduce a new instrument (ToM-HCAT) for assessment of ToM (i.e., theory of mind) by humor comprehension and appreciation suitable for healthy adult populations. This performance test consisting of cartoons measures perceived funniness, reaction time to perceived funniness decision, and meaning inference. While a first validation is presented (individuals high and low in the Autism Spectrum Quotient differ in the meaning-inference scores of the subscale with the ToM cartoons) it awaits further validation to support the claim it is useful to detect variations in ToM ability in the healthy adult population.

While Heintz et al. study country differences in measured humor traits, Tosun et al. explore lay conceptions of an ideal sense of humor in three countries, namely Iran, United States, and Turkey. As in prior US studies they find that the embodiment of an ideal sense of humor is predominantly a male figure. Country and gender had an impact on relative number of specific humor characteristics. For example, Americans mentioned hostility/sarcasm and caring more often than participants from the other countries. Further work is needed to replicate the observed group differences and to identify their sources.

Canestrari et al. use the Theory of the Pleasures of the Mind to study the enjoyment derived from both humor and insight problem solving as they share similar cognitive mechanisms. The results show that finding the solution to a problem is associated with a positive evaluation, and curiosity, virtuosity and violation of expectations are the most frequent explanations. Understanding a joke is accompanied by the joy of verification and a feeling of surprise. However, the choice for the most enjoyable cartoons related to other factors, such as recognizing a violation of expectations and experiencing a diminishment in the cleverness attributed to the characters in the cartoon.

Mendiburo-Seguel et al. investigate the effects of political humor on an individual's trust toward politics and politicians. They conducted two experiments, in which participants were exposed to political disparagement humor to non-humorous political information, or to non-political humor. Study 1 showed that an exposure to political disparagement humor and non-humorous political contents negatively affects trust in politicians immediately after the exposure. Study 2, in which semidaily messages were sent to the participants, did not yield significant effects.

The study by Wagner nicely demonstrates how close upside and downside of humor are together by showing that class clown behavior was positively related to different indicators of social status and peer-rated popular-leadership behavior, but also to aggressive-disruptive behaviors and negatively to prosocial behaviors. Thus, humor is involved in making a student popular but it may also be used in destructive ways. The study also demonstrates that it is important to distinguish among different dimensions of class clown behavior, as they yielded different results.

Laughter and Dispositions to Ridicule and Being Laughed at

Laughter is both a social signal and an expression of emotion with several behavioral and physiological components (e.g., respiratory, acoustic, facial, postural, hormonal). There are different motivations for laughter (with laughing with and laughing at being a minimal distinction made by many) and there are individual differences to be considered regarding both the laughing person and the one perceiving the laughter. Laughter is studied among the healthy but also within psychopathology. Clearly, the section of this research topic devoted to laughter and laughter-related dispositions received a variety of submissions.

Ritter and Sauter investigated whether listeners can identify in- and out-group members from laughter. They showed that listeners were unable to accurately identify group identity from laughter and the exposure to a group did not affect the classification performance. In conclusion, group membership cannot be inferred from the way people laugh.

Curran et al. test the notion that laughter is an ambiguous signal, which is only interpreted correctly in the context it occurs. They provide supportive data from two experiments in which participants judged the genuineness of audio–video recordings of social interactions containing laughter (either original or replacement laughter). When replacement laughter was matched for intensity, genuineness judgments were similar to judgments of the original recordings. When replacement laughter was not matched for intensity, genuineness judgments were generally significantly lower.

Stewart et al. used the 2016 US presidential debates to study laughter together with other responses of audience, such as applause, cheering, laughter, and even booing. In three interconnected studies the impact of the norm-violating audience behavior on those watching or listening was studied. Applause–cheering significantly enhanced liking of the speaking candidate, whereas laughter did not, and party identity mediated the response to applause–cheering, but not for laughter. Thus, in such settings, cheering may be more socially contagious and laughter more stereotypic and likely to be mimicked.

The study by Auerbach confirms that it is important to distinguish between Duchenne Displays as an indicator of joy and non-Duchenne displays. Only the former go along with a variety of indicators of positive experience during a visit of hospital clowns in a rehabilitation center. Thus, also in such interventions it pays off to invest into the fine-grained assessment of facial expressions; i.e., to use the Facial Action Coding System to code the patients' affective responses. Only the Duchenne displays are affected by trait cheerfulness and they can serve as an indicator that hospital clown interventions are beneficial for patients.

The study of laughter also includes the dispositions to laughter—more precisely individual differences in qualities relating to laughing at and being laughed at. They are still the new kid on the block of variables related to humor and laughter with a research tradition of about 10 years. Gelotophobia (i.e., the fear of being laughed at) represents one form of humorlessness and gelotophobes see humor and laughter as weapons directed at them not as a basis for a pleasant experience to be shared with others. Together with gelotophilia (i.e., the joy of being laughed at) and katagelasticism (i.e., the joy of laughing at others) gelotophobia forms the dispositions to being laughed at and ridicule.

Two of the articles in the present collection of articles relate to their assessment. Ruch et al. utilize a picture completion task to derive a more unobtrusive semi-projective test of gelotophobia. This alternative instrument for the assessment of gelotophobia turns out to yield comparable results to the standard assessment. Hofmann et al. fulfill the need for an ultra short instrument for the assessment of these three dispositions and extends research into the workplace. They propose (and confirm in a nationally representative sample of employees) that if friendly teasing and laughter of co-workers, superiors, or customers are misperceived as malicious, one may feel less satisfied with work and life and experience more work stress. Conversely, gelotophilia went along with positive evaluations of one's life and work, and katagelasticism was negatively related to work satisfaction and positively related to work stress. Torres-Marín et al. provide evidence that gelotophobia is related to a potential bias in gaze discrimination in two experiments. Interestingly, the nature of the emotion did not play the expected role raising the question what elements are necessary for smiling faces to elicit the effect among gelotophobes.

Renner and Manthey investigate humor creation abilities in their study of self-presentation styles and dispositions to ridicule and being laughed at. They derive scores for quantitative (e.g., number of punch lines) and qualitative (e.g., wittiness of the punch lines and wittiness of the person as evaluated by three independent raters) aspects of humor creation abilities. Results show that both gelotophilia and histrionic self-presentation are supported by fluency and quality of humor creation abilities.

Three manuscripts examine gelotophobia in circumscribed groups. Kohlmann et al. investigated the associations between the experience of weight-related teasing and mockery with overweight, self-perceptions of weight, and gelotophobia in youth. Deviations from normal weight were related to experiencing teasing, which in turn was related to the fear of being laughed at. The four studies suggest that research on well-being of youth with weight problems would benefit from studying weight-related teasing and mockery in connection with gelotophobia. Tsai et al. study the relation between the dispositions toward ridicule and being laughed at, personality, and presence of autism spectrum disorder (ASD) in high school students. As in prior studies, the ASD group was found to have a higher level of gelotophobia and the present study reveals that they also have lower levels of gelotophilia and katagelasticism. However, extraversion fully accounted for the observed lower gelotophobia scores among the ASD sample, and partly for the differences found for gelotophilia. Brück et al. investigated the prevalence of gelotophobia among Borderline Personality Disorder patients. They showed an extraordinarily high level of the fear of being laughed at (i.e., 87%) compared to other clinical and non-clinical reference groups.

Playfulness

The section on playfulness consists of five contributions of which two have a qualitative approach, while the others are quantitative in nature. Two contributions focus on play (the behavior associated with trait playfulness) and playfulness in school and the others employ adult samples. With 1,235 Tweets reaching an upper bound of 3,945,511 followers (March 25th, 2019) 1 , Barnett's article attracted much attention on social media. Her analyses show that teachers react differently—more negatively—toward playfulness expressed by boys than by girls (kindergarten-aged children followed up across 3 years). In contrast, playfulness in girls did not seem to be a concern for the teachers. The methodology employed and the study of gender differences provides a valuable update on earlier literature. Overall, the emerging question is how teachers, schools and societies in general may benefit from playfulness in the classroom.

Pinchover 's pilot study examines the interplay of playfulness in teachers and their students. Taking the limitations of this initial study into account, this may indicate that teacher behavior impacts children's playfulness. Given that there is initial evidence for a contribution of playfulness to academic achievement and more robust data on a beneficial use for stress coping, some functions of playfulness may be helpful for students in their learning experience and development.

The idea that a playful state of mind contributes to innovativeness and creativity has received much interest in the literature (for overviews see Proyer et al., 2019 ) and, for example, it has been argued that “[…] a child who experiences truly “playful play” learns cognitive and behavioral processes that enhance his creative potential” ( Bishop and Chace, 1971 ; p. 321). Heimann and Roepstorff introduce microphenomenological interviews as a method for research in playfulness. In this initial study, they found that autonomy and self-expression were of particular importance for achieving a playful state of mind.

Proyer et al. test associations of playfulness with self-reported health, activity, and physical fitness. Self- and peer-ratings (i.e., ratings by knowledgeable others; Study 1) and a series of behavioral tests (Study 2) to assess playfulness were collected. Overall, playfulness is linked to some facets of physical functioning. Future research will have to clarify the pathways and moderators of these associations (e.g., causality or indirect ways of impacting greater physical activity).

Finally, Pang and Proyer present first data on a comparison of playfulness scores in samples from two regions in the P.R. China and a sample from German-speaking countries—using measures from both, the East and the West. The article provides details on cultural differences and linguistic challenges in the translation of the term playfulness. Overall, the findings indicate that differences are smaller than expected, but that the differentiation between private and public situations impacts how people in the two regions enjoy expressing their playfulness. This study narrows a gap in the literature by providing initial data on cross-cultural differences (see also Barnett, 2017 ) and highlights that larger scale cross-cultural comparisons are encouraged.

These five studies support the notion that playfulness has an impact on various domains of life, but also that more research will be needed for a better understanding of its role across different age groups.

Cheerfulness

Cheerfulness has a tradition in psychological research for more than 100 years (e.g., Morgan et al., 1919 ). Trait cheerfulness, seriousness, and bad mood have been proposed to form the temperamental basis of humor. Bypassing the vague folk concept of the “sense of humor” they were expected to predict humor-related thoughts, feelings, and actions. Washburn in her early studies claimed that a person in the attitude of cheerfulness is incapable of a depressing thought, and meanwhile there is ample evidence that trait cheerful individuals maintain being in a cheerful state (i.e., keep humor) in the face of adversity. The contributions of the present collection of articles are diverse. First, a training of humor yielded outcomes for cheerfulness, seriousness, and bad mood) in the desired direction with medium to large effect sizes ( Tagalidou et al. ). Different to a recent study ( Ruch et al., 2018 ) the state version was utilized. Congruent with the assumption that cheerfulness predicts smiling and laughter, Auerbach shows that trait cheerful patients showed more genuine smiling and laughter during a hospital clown intervention than low trait cheerful individual do. Hofmann et al. present an adaptation of the instrument measuring state and trait cheerfulness using samples from the USA and the UK to providing the basis for studies with English-speaking participants. Next to the long version with 106 items, they provide the standard short form with 60 items and deliver initial validation data. López-Benítez et al. investigate a cognitive mechanism associated with trait cheerfulness. Utilizing a task-switching paradigm they find that while trait cheerfulness does not influence switching costs it modulates preparation and repetition effects. Studies like this are needed to further illuminate the processes associated with the traits be it cheerfulness, playfulness, or humor. Bruntsch and Ruch find trait cheerfulness and low bad mood facilitating the detection of ironic praise.

Conclusions

The individual contributions show how humor, laughter, playfulness, and cheerfulness are related and yet heterogeneous. Each field would profit from starting to talk to each other, see overlaps in scope, finding common structure, common language, and work on theories connecting these fields. Combining the domains in the prediction of important criteria might be important too. The topics studies in this research topic (plus others) may be understood as nodes in a larger net and the interrelations need to be better explored.

It is positive to see that integrative models within the domains are now developed. This indeed needs to be the prime goal, namely to work on a solid structure within the four fields. It took research of personality and intelligence more than half a century to arrive at models that are shared by many. Also in these fields we once had “schools” that did believe into one model and defended it a lifetime. Later generations of researchers then found that the competing models were incomplete variants and do fit into a more general, often hierarchical model. We recommend concerted efforts to solve those basic questions, perhaps by compiling special issues on pertinent topics.

Author Contributions

All authors listed have made a substantial, direct and intellectual contribution to the work, and approved it for publication.

Conflict of Interest Statement

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Acknowledgments

We would like to thank all the authors who agreed to participate in this Topic with their original contributions, and to all the reviewers who promoted the quality of research and manuscripts with their comments. Furthermore, special remarks go to Frontiers staff and Professors Marcel Zentner and Anat Bardi for the opportunity they gave to us.

1. ^ https://frontiers.altmetric.com/details/33125117/twitter

Barnett, L. A. (2017). The inculcation of adult playfulness: from west to east. Int. J. Play 6, 255–271. doi: 10.1080/21594937.2017.1383010

CrossRef Full Text | Google Scholar

Berlyne, D. E. (1969). “Laughter, humor, and play,” in Handbook of Social Psychology, Vol. 3 , eds G. Lindzey and E. Aronson (Reading, MA: Addison-Wesley), 795–852.

Bishop, D. W., and Chace, C. A. (1971). Parental conceptual systems, home play environments, and potential creativity in children. J. Exp. Child Psychol. 12, 318–338. doi: 10.1016/0022-0965(71)90028-2

Heintz, S., and Ruch, W. (2019). From four to nine styles: an update on individual differences in humor. Pers. Individ. Diff. 141, 7–12. doi: 10.1016/j.paid.2018.12.008

Martin, R. A., Puhlik-Doris, P., Larsen, G., Gray, J., and Weir, K. (2003). Individual differences in uses of humor and their relation to psychological well-being: development of the humor styles questionnaire. J. Res. Pers. 37, 48–75. doi: 10.1016/S0092-6566(02)00534-2

Morgan, E., Mull, H. K., and Washburn, M. F. (1919). An attempt to test the moods or temperament of cheerfulness and depression by directed recall of emotionally toned experiences. Am. J. Psychol. 30, 302–304.

Google Scholar

Proyer, R. T., Tandler, N., and Brauer, K. (2019). “Playfulness and creativity: a selective review,” in Creativity and Humor , eds S. R. Luria, J. Baer, and J. C. Kaufman (San Diego, CA: Academic Press), 43–60. doi: 10.1016/B978-0-12-813802-1.00002-8

Ruch, W., Hofmann, J., Rusch, S., and Stolz, H. (2018). Training the sense of humor with the 7 Humor Habits Program and satisfaction with life. Humor Int. J. Humor Res . 31, 287–309. doi: 10.1515/humor-2017-0099

Keywords: humor, playfulness, laughter, cheerfulness, gelotophobia, wit

Citation: Ruch W, Platt T, Proyer RT and Chen H-C (2019) Editorial: Humor and Laughter, Playfulness and Cheerfulness: Upsides and Downsides to a Life of Lightness. Front. Psychol . 10:730. doi: 10.3389/fpsyg.2019.00730

Received: 22 February 2019; Accepted: 15 March 2019; Published: 09 April 2019.

Edited and reviewed by: Nadin Beckmann , Durham University, United Kingdom

Copyright © 2019 Ruch, Platt, Proyer and Chen. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY) . The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

*Correspondence: Willibald Ruch, [email protected]

Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.

Cover Story

The science of humor is no laughing matter.

• Communication
• Interpersonal Communication
• Nonverbal Communication
• Psychological Science
• Social Psychology

In 1957, the BBC aired a short documentary about a mild winter leading to a bumper Swiss spaghetti crop in the town of Ticino. In a dry, distinguished tone, BBC broadcaster Richard Dimbleby narrates how even in the last few weeks of March, the spaghetti farmers worry about a late frost, which might not destroy the pasta crop but could damage the flavor and hurt prices. The narration accompanies film footage of a rural family harvesting long spaghetti noodles from trees and laying them out to dry “in the warm Alpine sun.”

Naturally, the hundreds of people who called the BBC asking where they could get their own spaghetti bushes hadn’t noticed the air date of the news clip: April 1st. The prank was so successful that even some BBC staff were taken in, leading to some criticism about using a serious news show for an April Fool’s Day joke.

Why April 1st became a holiday devoted to pranks and laughter remains a mystery, although some historians trace it back to the Roman holiday of Hilaria. Humans start developing a sense of humor as early as 6 weeks old, when babies begin to laugh and smile in response to stimuli. Laughter is universal across human cultures and even exists in some form in rats, chimps, and bonobos. Like other human emotions and expressions, laughter and humor provide psychological scientists with rich resources for studying human psychology, ranging from the developmental underpinnings of language to the neuroscience of social perception.

The Hidden Language of Laughter

Theories focusing on the evolution of laughter point to it as an important adaptation for social communication. Studies have shown that people are more likely to laugh in response to a video clip with canned laughter than to one without a laugh track, and that people are 30 times more likely to laugh in the presence of others than alone.

“The necessary stimulus for laughter is not a joke, but another person,” writes laughter expert and APS Fellow Robert R. Provine, professor emeritus at University of Maryland, Baltimore County, in an article in Current Directions in Psychological Science .

Just look at the canned laughter in TV sitcoms as an example: The laugh track has been a standard part of comedy almost from the birth of television. CBS sound engineer Charley Douglass hated dealing with the inappropriate laughter of live audiences, so in 1950 he started recording his own “laugh tracks.” These early laugh tracks were intended to help people sitting at home feel like they were in a more social situation, such as sitting at a crowded theater. Douglass even recorded varying types of laughter, including big laughs and small chuckles, as well as different mixtures of laughter from men, women, and children.

In doing so, Douglass picked up on one of the qualities of laughter that is now interesting researchers: A simple “ha ha ha” communicates an incredible amount of socially relevant information.

For example, a massive international study conducted in 2016 found that across the globe, people are able to pick up on the same subtle social cues from laughter. Samples of laughter were collected from pairs of English-speaking college students — some friends and some strangers — recorded in a lab at the University of California, Santa Cruz. An integrative team made up of more than 30 psychological scientists, anthropologists, and biologists then played audio snippets of this laughter to 966 listeners from 24 diverse societies spanning six continents, from indigenous tribes in New Guinea to urban working-class people in large cities in India and Europe. Participants then were asked whether they thought the two people laughing were friends or strangers.

On average, the results were remarkably consistent across all 24 cultures: People’s guesses about the relationship between the laughers were correct approximately 60% of the time.

Researchers also have found that different types of laughter can serve as codes to complex human social hierarchies. Across the course of two experiments, a team of psychological scientists led by Christopher Oveis of University of California, San Diego, found that high-status individuals had different laughs than low-status individuals, and that strangers’ judgments of an individual’s social status were influenced by the dominant or submissive quality of the person’s laughter.

“Laughing in the presence of others indicates the interaction is safe,” the researchers explain. “While the norms of most social groups prevent direct, unambiguous acts of aggression and dominance, the use of laughter may free individuals to display dominance because laughter renders the act less serious.”

In the first study, the researchers wanted to know whether high- and low-status individuals actually do laugh differently.

To test this, 48 male college students were randomly assigned to groups of four, with each group composed of two low-status members (“pledges” who had just joined a fraternity a month earlier) and two high-status members (older students who had been active in the fraternity for at least 2 years).

Laughter was recorded on video as the group members engaged in a teasing task. Each member of the group took a turn in the hot seat, receiving light teasing from his peers. The teasers came up with a nickname based on randomly generated sets of initials (e.g., L. I. became “Loser Idiot”) and then told joking stories about why they chose the nickname.

One team of coders (naïve to the study hypotheses) identified all of the instances of laughter in the video, and a second team of coders (also blind to the study hypotheses) watched the video and rated how submissive or dominant each laugh sounded using a scale of −3 (definitely submissive) to 3 (definitely dominant). Laughs receiving average ratings of 2 or higher were classified as dominant, whereas laughs receiving average ratings of −2 or lower were classified as submissive.

A third team of coders, also blind to the hypotheses, coded the audio of each laugh on specific sound characteristics — loudness, pitch, pitch range, pitch modulation, airiness, and burst speed — that are associated with disinhibited behavior.

“If dominant laughs are more disinhibited than submissive laughs, as we hypothesize, they should exhibit greater vocal intensity, more pitch range and modulation, and greater burst speed,” Oveis and colleagues explain.

The analysis revealed that, as predicted, high-status fraternity brothers produced more dominant laughs and fewer submissive laughs relative to the low-status pledges. Dominant laughter was higher in pitch, louder, and more variable in tone than submissive laughter. In this regard, dominant laughter appears to share some of the features researchers have identified in genuine (compared with fake) laughter: greater irregularities in pitch and loudness and faster bursts of sound.

Previous research published in  Psychological Science demonstrated that holding a position of power can influence the acoustic cues of our speech. The voices of individuals primed with high-power roles tended to increase in pitch and were, at the same time, more monotone. Listeners who had no knowledge of the experiment were able to pick up on vocal cues signaling status: They correctly rated individuals in the high-power group as being more powerful with a surprising degree of accuracy — about 72% of the time.

Findings from the fraternity-brother experiment also showed that low-status individuals were more likely to change their laughter based on their position of power; that is, the pledges produced more dominant laughs when they were in the “powerful” role of teasers. High-status individuals, on the other hand, maintained a consistent pattern of dominant laughter throughout the teasing game regardless of whether they were doing the teasing or being teased themselves.

In another study, the research team tested out whether naïve observers could detect an individual’s social status based just on their laughter, and whether the type of laugh (dominant or submissive) could influence judgements of social status.

A group of 51 college students was randomly assigned to listen to a set of 20 of the laughs recorded from the fraternity brothers. Each participant listened to an equal number of dominant and submissive laughs from both high- and low-status individuals. Participants then estimated the social status of the laugher using a series of 9-point ratings scales. And indeed, laughers producing dominant laughs were perceived to be significantly higher in status than laughers producing submissive laughs.

“This was particularly true for low-status individuals, who were rated as significantly higher in status when displaying a dominant versus submissive laugh,” Oveis and colleagues note. “Thus, by strategically displaying more dominant laughter when the context allows, low-status individuals may achieve higher status in the eyes of others.”

However, regardless of whether raters heard a dominant or a submissive laugh from a high-status individual, they rated that person as being relatively high in status.

It’s unclear whether this was because high-status laughs include characteristics that were not measured in the current study or whether high-status fraternity brothers just didn’t have very convincing low-status laughs while being teased.

When it comes to comedy, it’s often a thin line between love and hate. What qualities make something funny (or not) is a question that philosophers have been attempting to answer for thousands of years. But a pair of psychological scientists have come up with a theory that explains why we might laugh at a dark joke about murder as well as a silly pun or play on words.

Psychological scientists Peter McGraw (University of Colorado, Boulder) and Caleb Warren (University of Arizona) propose that negativity is an intrinsic part of humor — without violating a norm or rule of some kind, a joke just isn’t funny. But violations can’t stray too far; otherwise, they become unappealing or even disgusting and upsetting. According to the researchers’ Benign Violation Theory, a violation is humorous when it breaks a rule or norm but is benign.

McGraw and Warren’s Humor Research Lab (HuRL) has conducted several studies examining the exact criteria that cause us to perceive a comedic situation as benign or not. Along with the severity of the norm violation, a sense of psychological distance from the violation — by space, time, relationships, or imagination — is a key ingredient for turning an unpleasant situation into a humorous one, they posit.

For example, in a study published in Psychological Science , the researchers looked at the effect of psychological distance in terms of time. Inspired by the classic Mark Twain quote, “Humor is tragedy plus time,” the research team investigated how the passage of time can influence one’s perception of an event as funny or painful.

“If distance increases the humor in severe violations (i.e., tragedies), but decreases the humor in mild violations (i.e., mishaps), then autobiographical events that get funnier over time should feature more severe violations than those that get less funny over time,” the researchers write.

One study found that the events from people’s lives that became funnier over time were more severe events (like a car accident), while events that lost their comedic effect over time were seen as minor violations (like stubbing a toe).

Another study examined distance by manipulating whether an image was seen as hypothetical or real. A group of 67 students was asked to rate the humor of images from a website. Those in the  close  condition were told they would be looking at real photos that “have not been altered using image design software”; participants in the  distant  condition were told they would be viewing “fake pictures” that “have been altered using image design software.”

One picture portrayed a severe abnormality: a Cronenbergian image of a man sticking a finger up through his nose out of his eye socket. The other portrayed a mild abnormality — a man with large icicles hanging from his frozen beard. Using a 6-point scale, participants rated how funny they thought the photos were.

The students rated the more disturbing image of the empty eye socket as more humorous when they were told it was fake, and they reported the less disturbing frozen-beard image as more humorous when they thought it was real.

“These findings suggest that there’s a real sweet spot in comedy — you have to get the right mix between how bad something is and how distant it is in order for it to be seen as a benign violation,” McGraw said.

The Energizing Effect of Humor

Having trouble finishing a project on deadline? Well, put down that Red Bull and head over to YouTube. No joke — watching funny cat videos at work may not be such a bad thing after all. A study conducted by Australian National University management professors David Cheng and Lu Wang suggests that exposure to humorous stimuli may actually help people persevere in completing tedious tasks.

Across two studies, Cheng and Wang found that people who watched a funny video clip before a task spent approximately twice as long on a tiresome task compared with people who watched neutral or positive (but not funny) videos.

Prior research has found that humor can help facilitate recovery from stressful situations, even prolonging people’s tolerance for physical pain. In the business world, many successful organizations such as Zappos, Virgin, and Google deliberately build play areas into their workspaces and organize fun events to ameliorate the stressful nature of work, boost morale, and increase productivity.

Indeed, in a 2007 article published in  Current Directions in Psychological Science , APS William James Fellow Roy F. Baumeister (Florida State University), APS Fellow Kathleen D. Vohs (University of Minnesota), and APS Fellow Dianne M. Tice (Florida State University) point to humor as a factor that can moderate or counteract the effects of mental depletion.

In line with this idea, Cheng and Wang hypothesized that humor may provide a respite from tedious situations in the workplace. This “mental break” might not only prevent work-related depletion, but also might facilitate the replenishment of mental resources, ultimately allowing people to persist longer on difficult tasks.

To test this theory, for their first study the researchers recruited 74 students studying in a business class to come into the lab, ostensibly for an experiment on perception. First, the students performed a mentally depleting task in which they had to cross out every instance of the letter “e” contained in two pages of text. The students then were randomly assigned to watch a video clip eliciting either humor, contentment, or neutral emotions.

For the humorous video, students watched a clip of the BBC comedy “Mr. Bean.” In the contentment condition, participants watched a scene with dolphins swimming in the ocean. The students in the neutral condition were treated to an 8-minute video about the management profession designed for students studying business. Immediately after watching the videos, participants reported their responses to a list of 16 discrete emotions (e.g., amusement, anger, disgust) using a 7-point scale.

Then the students completed a persistence task in which they played what amounted to an unwinnable game. The students were asked to guess the potential performance of employees based on provided profiles and were told that making 10 correct assessments in a row would lead to a win. However, the computer software was programmed such that it was nearly impossible to achieve 10 consecutive correct answers. Participants were allowed to quit the task at any time.

Students who watched the humorous “Mr. Bean” video clip ended up spending significantly more time working on the task, making twice as many predictions as the other two groups.

Cheng and Wang then replicated these results in a second study, during which they had participants complete long multiplication questions by hand. Again, participants who watched the humorous video spent significantly more time working on the task and completed more questions correctly than did those who did not watch the funny video.

“Although humor has been found to help relieve stress and facilitate social relationships, the traditional view of task performance implies that individuals must concentrate all their effort on their endeavors and should avoid things such as humor that may distract them from the accomplishment of task goals,” Cheng and Wang conclude. “We suggest that humor is not only enjoyable but more importantly, energizing.”

Kathleen D. Vohs will speak at the 2017 APS Annual Convention, May 25–28, in Boston, Massachusetts.

Baumeister, R. F., Vohs, K. D., & Tice, D. M. (2007). The strength model of self-control.  Current Directions in Psychological Science, 16 , 351–355. doi:10.1111/j.1467-8721.2007.00534.x

Bryant, G. A., Fessler, D. M. T., Fusaroli, R., Clint, E., Aarøe, L., Apicella, C. L., … Zhou, Y. (2016). Detecting affiliation in colaughter across 24 societies.  Proceedings of the National Academy of Sciences ,  113 , 4682–4687. doi:10.1073/pnas.1524993113

Cheng, D., & Wang, L. (2015). Examining the energizing effects of humor: The influence of humor on persistence behavior.  Journal of Business and Psychology, 30 , 759–772. doi:10.1007/s10869-014-9396-z

Ko, S. J., Sadler, M. S., & Galinsky, A. D. (2014). The sound of power: Conveying and detecting hierarchical rank through voice.  Psychological Science, 26 , 3–14. doi:10.1177/0956797614553009

McGraw, A. P., & Warren, C. (2010). Benign violations: Making immoral behavior funny.  Psychological Science ,  21 , 1141–1149. doi:10.1177/0956797610376073

McGraw, A. P., Warren, C., Williams, L. E., & Leonard, B. (2012). Too close for comfort, or too far to care? Finding humor in distant tragedies and close mishaps.  Psychological Science , 23 , 1215–1223. doi:10.1177/0956797612443831

Oveis, C., Spectre, A., Smith, P. K., Liu, M. Y., & Keltner, D. (2016). Laughter conveys status.  Journal of Experimental Social Psychology ,  65 , 109–115. doi:10.1016/j.jesp.2016.04.005

Provine, R. R. (2004). Laughing, tickling, and the evolution of speech and self.  Current Directions in Psychological Science ,  13 , 215–218. doi:10.1111/j.0963-7214.2004.00311.x

Chemical Communication is Nothing to Sniff At

Research is uncovering just how much our noses know about our social environments.

Communicating Psychological Science: The Lifelong Consequences of Early Language Skills

“When families are informed about the importance of conversational interaction and are provided training, they become active communicators and directly contribute to reducing the word gap (Leung et al., 2020).”

Student Notebook: Beginning Your Collaborative Writing Journey 

Nidhi Sinha explores the benefits of collaborative writing for graduate students: “The more people you involve in your research, the better experience, productivity, and research satisfaction you will receive in the long run.”

Privacy Overview

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals
• My Account Login
• Explore content
• About the journal
• Publish with us
• Sign up for alerts
• Open access
• Published: 03 November 2023

Humor appreciation can be predicted with machine learning techniques

• Hannes Rosenbusch   ORCID: orcid.org/0000-0002-4983-3615 1 , 2 &
• Thomas Visser 1  

Scientific Reports volume  13 , Article number:  19035 ( 2023 ) Cite this article

851 Accesses

Metrics details

• Scientific data

Humor research is supposed to predict whether something is funny. According to its theories and observations, amusement should be predictable based on a wide variety of variables. We test the practical value of humor appreciation research in terms of prediction accuracy. We find that machine learning methods (boosted decision trees) can indeed predict humor appreciation with an accuracy close to its theoretical ceiling. However, individual demographic and psychological variables, while replicating previous statistical findings, offer only negligible gains in accuracy. Successful predictions require previous ratings by the same rater, unless highly specific interactions between rater and joke content can be assessed. We discuss implications for humor research, and offer advice for practitioners designing content recommendations engines or entertainment platforms, as well as other research fields aiming to review their practical usefulness.

Similar content being viewed by others

Predicting and improving complex beer flavor through machine learning

Michiel Schreurs, Supinya Piampongsant, … Kevin J. Verstrepen

TacticAI: an AI assistant for football tactics

Zhe Wang, Petar Veličković, … Karl Tuyls

Song lyrics have become simpler and more repetitive over the last five decades

Emilia Parada-Cabaleiro, Maximilian Mayerl, … Eva Zangerle

Introduction

What makes things funny? For millennia, philosophers and empirical scientists have pursued this question. Consequentially, we now have many theories, often backed up by experimental evidence, that highlight who labels what funny under which conditions. Despite these academic insights, the world outside the lab is certainly not freed from bad jokes, and likely never will be. What can humor research do to optimize humor exposure, for instance, on content platforms? To optimally guide practitioners, we set out to assess the usefulness of amusement predictors forwarded in academic research.

Specifically, we will quantify the empirical accuracy with which humor research can predict amusement by combining prominent psychological constructs with cutting-edge prediction models from the field of machine learning 1 . If decades of empirical research are enabling accurate predictions of funniness, research could focus on optimizing real-life applications. Conversely, if we are not able to predict humor appreciation accurately, we need to focus on constructing theories and predictors with higher predictive value. Therefore, the current work also ranks relevant constructs from most to least predictive.

In the current work, we focus on predicting perceived funniness, a component of humor appreciation. The term amusement is the emotional “state of experiencing something funny” 2 , 3 . Other components of humor appreciation like perceived offensiveness or boringness are not investigated here 4 , 5 .

Predictors of amusement

Factors that influence whether someone will be amused by a humor stimulus can be divided into different categories. One of these categories is the nature of the joke, including its content and structural qualities 4 , 6 . Early theories suggested that jokes with certain contents are funnier than others because they tie in more closely with our supposed reasons for laughing. For instance, Freud suggested a cathartic function of jokes and laughter, as a valve for releasing tension from societally suppressed drives towards sex and aggression 7 . Thus, a hypothesis was formed (and sometimes supported 8 ) that sexual or aggressive humor stimuli are funnier than others. In fact, even earlier theorizing suggests that all instances of humor consist of a satisfactory domination of others 9 , 10 . Other theories posit that people are more likely to laugh, when a joke elicits a strong but manageable cognitive violation (for a distinction with surprisingness/incongruity, see 11 ) or when a joke has a strong elaboration potential (i.e., many funny implications 12 ). To zoom out, such theories state that some jokes are simply funnier than others; they imply that to predict amusement, one should evaluate how good the respective joke is.

There is parallel stream of empirical evidence, sometimes coming from the same groups of researchers, showing that characteristics of the audience members are also predictive of humor appreciation. In short, some people will laugh at most jokes, whereas others remain hard to amuse 13 . One of the most differential traits when it comes to humor appreciation is a person’s dispositional cheerfulness, the temperamental basis for humor, often used synonymously with the fragmented term ‘sense of humor’ 14 . Much like a person’s mood (e.g., 15 ), cheerfulness has both a dispositional baseline as well as time-varying components that have been observed to be aligned with people’s appreciation of humor stimuli 13 , 16 .

Other prominent personality constructs that have been linked to a person’s general level of humor appreciation include the Big Five personality traits. Specifically, extraverted 17 , 18 , 19 , 20 , 21 , conscientious 17 , 20 , open-minded 17 , 20 , and emotionally stable people 17 , 19 , 20 , 22 have all shown heightened levels of humor appreciation, albeit not consistently. These positive relationships are likely restricted to personality- coherent humor stimuli (e.g., neurotic people specifically disliking absurd humor 6 , 17 ). Similarly, sensation-seekers appreciate nonsensical humor, whereas their counterparts prefer humor with a clear uncertainty-resolving punchline 23 , 24 . The proposed explanation of the authors highlights that nonsense humor involves stronger cognitive stimulation, thereby dividing sensation-seekers and avoiders. Comparable to humor appreciation, humor production is divided across personality dimensions. For instance, extraverts produce more affiliative humor, whereas agreeable people tend to produce non-aggressive humor 25 , 26 .

Next to mood and personality, people’s capabilities can predict humor appreciation. Specifically, people with a higher competence for producing funny jokes 20 rated humorous cartoons comparatively low, potentially due to a heightened desire for dominance (humor production) as opposed to submission/affiliation (humor appreciation). Thus, a broad string of research focuses on the attributes of the audience when predicting amusement, rather than merely scoring the content of the joke.

Note that some of the listed findings go beyond simple person-effects and advance to the interaction between person characteristics and the characteristics of the joke (e.g., an aggressive person enjoying an aggressive joke). Similarly, work focusing on stimuli differences often acknowledges the notion of interactivity. For instance, Freud himself gave examples of people differing in their habitual suppression of urges which in turn affected their sense of humor 7 . Proposed advantages of specific joke contents, like sex and violence, were merely due to the observation that the suppression of these urges were common, thus rendering such joke contents more effective on average. In short, no theory on humor appreciation claims that it is exclusively the joke, or exclusively the nature of the listener that determines amusement. The question is merely, to which degree the two sources and their interaction contribute.

In a variance-decomposition study, ICC values ascribed higher importance to raters than to stimuli when it came to funniness ratings, and even more variance by pooling variance for specific rater X stimulus pairs 27 . This suggested that the totality of interaction effects harbors the highest predictive power, but it remains to be determined which specific effects account for this finding, and how accurate resulting prediction models are, especially when applied outside the training sample. The importance of person-stimulus interactions is also highlighted in some humor appreciation theories like the Benign Violation Theory which posits that amusement requires a simultaneous categorization of jokes as violating and benign 11 , 28 , both judgments being influenced by the rater’s specific relationship to the joke content 29 . In the current set of studies, we review the practical usefulness of person-level predictors, stimulus-level predictors, and their interactions, when building cutting-edge machine learning systems trained to predict people’s appreciation of humor content.

Predictive accuracy as practical relevance

Traditionally, psychological sciences focus on offering coherent explanations for human behavior. Usually, this is achieved by accumulating and mapping empirical observations to causal theories. Many such theories imply patterns of statistical associations between two or more relevant measurements. While theories often state that one measure should ‘predict’ the other (i.e., a non-zero statistical association), there is usually no pre-defined cut-off for how accurate this prediction should be to support the theory.

For example, various findings suggest that bad quality of sleep is associated with depression and suicide 30 . A theorist might wonder about the mechanism and directionality of the association. Conversely, a practitioner might wonder whether it is now feasible to build the researchers’ statistical model into their fitness tracker app to enable an “early-warning” feature for poor sleepers 31 . If the model can accurately predict depression risk for new participants, one could even consider a direct alert going out to local health care professionals, similar to car crash detection in mobile phones. However, most analyses in psychological research do not inform practitioners about the predictive accuracy that their theory affords 32 . Integrating machine learning models and, more importantly, out-of-sample accuracy assessments would therefore constitute a clear step from theoretical to practical usefulness 33 .

Regarding humor, most machine learning systems are focused on humor detection , meaning that models automatically infer which texts or images were meant to convey humor 34 , 35 , 36 , 37 . Interestingly, state-of-the-art performance in humor detection is achieved by models that are designed based on psychological humor theories 38 . While these research efforts do not focus on humor appreciation , they highlight that computational and psychological research can mutually reinforce each other when analyzing humor stimuli.

Some research also exists on machine predictions of humor appreciation. Specifically, models are trained to predict which of two humor stimuli will be rated as funnier, when aggregating ratings of multiple participants 39 . The efforts are often only partially successful meaning models make many mistakes in ranking the funniness of humor stimuli 40 , 41 .

In this work, we change the level of analysis to pairings of individual raters with individual stimuli (i.e., no aggregation). Ruch reviewed the stark effect of the aggregation choice on later results (e.g., meaning and magnitude of correlations between funniness ratings and auxiliary variables 42 ). Generally, aggregation boosts reliability and thus our focus on single experiences of humor by single individuals makes prediction more difficult. However, a compensatory advantage is that individual-level characteristics of both, the stimulus and the rater, can be used for the prediction of amusement.

The benefit of reviewing the predictive power of humor research is that practitioners can use the new insights to build content recommendation websites, revise their comedy routine, or forecast the success rate of their clinical humor treatments. Generally, it gives insight into the practical usefulness of the many variables that appear in recent humor theories.

We assess whether prominent constructs from humor research can accurately predict humor appreciation and which specific constructs offer the largest gains in empirical accuracy. All participants provided informed consent. The study procedure in accordance with the APA ethics code as well as the Declaration of Helsinki, and was approved by the Faculty Ethics Review Board of the first author’s research institution under leadership of Lourens Waldorp.

In an initial survey, participants filled out a questionnaire assessing constructs relating to personality, morality, ability, humor attitudes, and demographic data. Two days later, they received an invitation to a second survey which assessed their current mood as well as humor appreciation for 10 stimuli. It was further asked how well participants understood the English texts of the humor stimuli. All participants passed an attention check asking for the sum of eight and two. The data from the initial study was merged with the second survey using the participants’ unique user ID on the platform.

When trying to estimate the predictive value of psychological research it is of central importance to select promising predictor variables. All constructs and variables outlined below were selected based on their prominence in the literature (cf., sources above, and per measure below). Some noteworthy omissions are discussed in the limitation section. We relied on published short versions of psychological scales. All stimuli and newly generated questions about people’s attitude towards the content of the humor stimuli are included in the Supplementary materials .

Various meta-analyses and reviews of the relationship between the Big Five personality traits and humor have been conducted 25 , 26 . The Big Five Inventory 10 (BFI-10) is a ten-item version of The Big Five Inventory (BFI) developed by Rammstedt and John 43 and measures agreeableness, extraversion, openness to experience, conscientiousness, and neuroticism. Answers are given on a 5-point scale (“Disagree strongly” to “Agree strongly”). The scale authors report an average six-to-eight-week retest-reliability of 0.75.

Cheerfulness

In humor appreciation, state cheerfulness is one of the central constructs as it forms part of the temperamental basis for humor 44 . The State-Trait Cheerfulness Inventory State Version–Short Form (STCI-S18 45 ) is a short version of the State-Trait Cheerfulness Inventory State Version (STCI 16 ). It is an 18-item survey that measures cheerfulness on three dimensions: cheerfulness, seriousness, and bad mood. Because mood is already measured separately, only the 6 cheerfulness items were used here. The answers to these items were given on a 4-point scale (“Strongly disagree” to “Strongly agree”). The authors report a four-to-five-week retest-reliability of 0.85. The Cronbach's alpha coefficient in the current study was 0.77.

Moral identity

In humor research, a negative effect of morality on humor appreciation was observed by various groups 46 , 47 , 48 for at least some forms of humor. The five-item internalization dimension of the Moral Identity Scale 49 was developed to measure moral beliefs. The participant is asked to imagine a person that is “caring, compassionate, fair, friendly, generous, helpful, hardworking, honest, kind”. After this, five identification items are answered on a 5-point scale (“Strongly disagree” to “Strongly agree”). The original authors report a retest-reliability of 0.49 over a four-to-six-week time period, arguing that the construct varies over time. The Cronbach's alpha coefficient in the current study was 0.7.

Sensation seeking

While the relationship between dispositional sensation-seeking and humor appreciation is sometimes positive and sometimes negative (depending on the nature of the humor stimuli) it remains a potent predictor of humor appreciation across cultures 23 . The Brief Sensation Seeking Scale (BSSS 50 ) is an eight-item survey created to measure the components of sensation seeking. The answers to the BSSS are given on a 5-point scale (“Strongly disagree” to “Strongly agree”). The original authors of the scale did not report a retest-reliability but a Dutch version achieved a two-week retest reliability of 0.93 51 . The Cronbach's alpha coefficient in the current study was 0.81.

Humor production

Moran and colleagues 20 found that people that were better at producing humor, were less appreciative of humor stimuli than others (see 19 for a set of opposing results). The Multidimensional sense of humor scale (MSHS 52 ) assesses respondents’ self-rated skill in being funny. The relevant subscale encompasses four items. Answers are given on a 5-point scale (“Strongly disagree” to “Strongly agree”). The original authors did not report a retest-reliability, but a Finnish version showed a three-year retest reliability for the subscale of 0.75 53 . The Cronbach's alpha coefficient in the current study was 0.75.

It is an ongoing question which mood facets influence humor appreciation (the most 13 , 15 , 54 ). The International Positive and Negative Affect Schedule Short Form (I-PANAS-SF 55 ) is a short version of the Positive and Negative Affect Schedule (PANAS 56 ). It is a 10-item measurement in which people rate how they generally feel (e.g., “upset”, “nervous”) on a 5-point scale (“never” to “always”). The authors report an eight-week retest-reliability of 0.85.

Stimulus-specific attitudes

Broad traits, as listed above, are prominent in humor research, but narrow constructs can be highly predictive as well if they fit the context well. In the case of humor appreciation this pertains specifically to people’s attitudes towards specific joke contents 57 . Knowing whether someone is anti-Trump will likely be informative for their amusement from anti-Trump jokes, albeit less informative for broader humor categories. Here, we assess stimulus-specific attitudes with newly developed questions about the specific humor stimuli displayed in the survey Examples of these are “I like pictures of cute animals” or “I am very familiar with Game of Thrones”. The respective jokes included a humorous picture of a dog and a meme featuring a scene from Game of Thrones. Attitude items were rated on a 5-point scale from (“Strongly disagree” to “Strongly agree”).

Humor appreciation

The prediction target, humor appreciation, was measured for ten different humor stimuli with two items per stimulus. These stimuli were selected from a previously published corpus of 105 diverse humor stimuli 27 . The selected humor stimuli were categorized by the original authors as falling into five categories (there were more in the corpus): affiliative, self-defeating, aggressive, self-enhancing, and sexual 58 , 59 . For the current study, one textual and one image stimulus were selected from each category.

Note that the assignment of the stimuli into one of the five categories is not objective. The original authors write “Note that most stimuli fell into multiple categories […] because everyday humor usually combines different dimensions” ( 27 , p. 1390). Table 1 shows an example stimulus for each of the five categories. Note, for instance, that jokes which weren’t assigned to the category ‘aggressive’ still vary in their aggressiveness.

Here, humor appreciation was measured with two questions: “How funny do you find this {text/image}?” (1 = “not funny at all” to 7 = “very funny” 27 , 48 , 57 , and “Final evaluation:” with answer options “Good joke” and “Bad joke”). We selected a continuous and a binary measure to have complementary accuracy metrics available during the out-of-sample accuracy analysis.

A German panel provider offered participation through their mobile application where anonymous users can generate and distribute their own opinion polls, as well as answer polls of others. The app includes gamified rewards such as levels and coins that can be gathered by generating and participating in surveys. Once a certain number of coins is reached, participants can donate 10 Euro to protect the rainforest or get 10 Euro transferred to their PayPal account. However, the main driver for participation is entertainment rather than monetary rewards, as well as the opportunity for anyone to conduct representative polls outside one’s own social bubble 60 . Participants from Germany (58.87%), the UK (24.79%), and the USA (16.33%) entered the study for a total of 5473 responders (50.23% male, 49.77% female). Ages ranged from 18 to 84 (M age  = 32.68, SD age  = 10.76). All participants indicated in a previous assessment that they can read and speak English and passed an English attention check question.

Analysis plan

The ultimate goal of the current work is to assess how accurately one can predict humor appreciation on unseen cases, and which predictors are the most useful in that regard. Thus, the focal pieces of the current study are prediction models relating humor appreciation to a range of predictors from psychological research. Given the tabular nature of the data, we utilized boosted decision trees (XGboost 61 ) as the prediction model (ensemble). The XGboost algorithm is currently considered the state-of-the-art technique for predicting target variables based on tabular data 62 . Tabular data refers to the data commonly found in social science investigations with multiple observations scored on qualitatively distinct variables which can be organized in a table-like format. Other collections of data include images, audio, or text, for which other approaches, most notably neural networks, trump the performance of tree-based algorithms like XGboost (for a review of both approaches see 63 ).

Much like in OLS regression models, the XGboost algorithm relates predictor variables to an outcome variable, and its inner parameters are optimized for prediction accuracy. However, rather than fitting a linear combination of predictors, XGboost is based on decision trees that funnel observations down different “branches” based on logical statements (e.g., “Age > 30” goes left, other datapoints go right). These splits are optimized to create end nodes (i.e., “leaves”) on which data points have similar scores on the outcome variable. The XGboost algorithm then iteratively introduces new trees which predict the residuals of previous trees rather than the raw target scores. Thereby, the predictions of the summed tree outputs approximate the target score while actively combating mispredictions. The added complexity of the algorithm usually results in substantial improvements of prediction accuracy compared to OLS regression. For details and implementation guidance, see González and colleagues 64 .

The entire pipeline of data transformation, model tuning, and final analysis was conducted separately for the continuous measure of humor appreciation (metrics: R 2 , RMSE) and the binary measure (metrics: AUC, accuracy). Data was transformed into a long format with a single response per row (54,730 rows). One response per participant was randomly selected to form part of the test set which was only accessed once during final evaluation (5473 rows). We used tidymodels and its associated packages 65 in R 66 for both tuning and evaluating the models. We used tenfold cross-validation to assess the performance of 42 hyperparameter settings and selected the best performing setting to fit the model on the full training data. We compute variable importance and final performance on the test set. Variable importance scores were generated by permutating the values for individual predictor variables and quantifying associated model performance drops in the test set. The generation of all results was done in the same way but separately for three sets of predictors: (1) all psychological constructs (including demographic info), (2) average ratings of respectively the current joke and of the current participant, and (3) the combination of both of these predictor sets. Responses in the test set were excluded when computing rating aggregates. All data and scripts are in the Supplementary materials . We compare model accuracies by inspecting 95% bootstrap confidence intervals. Formal tests for comparing the predictor sets were not necessary given the width and separation of intervals (see below).

Distributions of humor appreciation for the binary measure (overall: 56.96% “good joke”) and the continuous measure (overall: M  = 4.34, SD  = 1.92) for each of the 10 stimuli are depicted in Fig.  1 .

Top left: Proportion of ‘Good joke’ ratings for each stimulus including 95% confidence intervals (binary scale). Top right: Average funniness ratings for each stimulus including 95% confidence intervals (7-point scale). Bottom left: Each participant made ten binary ‘good vs bad joke’ decisions averaged into a personal proportion score (0–1). The histogram shows these proportion scores for all participants. Bottom right: Each participant provided ten continuous funniness rating averaged into a personal rating score (1–7). The histogram shows these average rating scores for all participants.

Hyperparameter settings for the XGboost algorithms were selected based on cross-validation performance and can be reproduced with the supplementary scripts. Table 2 shows all accuracy metrics of each model when predicting humor appreciation.

The Supplementary material includes scripts and results for an alternative predictor setup, where individual survey items are used as predictors instead of the underlying psychological scales . The performance is virtually the same as shown here.

Variable importance scores (see Fig.  2 ) were quantified as average performance drops on the test set when randomly permutating predictor variables (100 iterations).

Drops in prediction performance when a given predictor variable is scrambled.

It is apparent that the model rests primarily on how other people rated the current joke, and the current rater’s appreciation of other jokes. Scrambling any of the other psychological constructs and variables leads to negligible deviations in model performance. If one assumes that participants responded to all the jokes equally, were not indicating their authentic appreciation of the stimuli, their responses could artificially inflate the predictive power of past ratings. However, after removing this group of 249 people (4.5%), and retraining the models, predictions were still primarily based on past responses.

Figure  3 shows the zero-order associations between individual psychological variables from psychological research and the separate funniness rating for all 10 stimuli.

Statistical associations (Pearson correlations and Cohen’s D’s) between predictors and continuous humor appreciation scores. The content of the individual humor stimuli (boxes 1–10) can be reviewed in the Supplementary materials .

The accuracy with which humor appreciation could be forecasted was around 70% and R 2  = 0.36. While this is substantially better than chance, predictions for individual people are thus far from reliable even when considering a wide variety of predictors from psychological research. While we calculated lower accuracy limits by establishing guessing baselines, we could not quantify a higher limit that accounts for random (i.e., non-predictable) noise in the humor appreciation scores. Study 2 quantifies this limit as the retest reliability of the predicted variable. This will clarify whether the remaining errors are due to weak predictors or unreliable measurements of humor appreciation.

A positive take-away for practitioners is that simple aggregates of a user’s past behavior on the platform (i.e., here operationalized as funniness ratings) were almost sufficient to achieve the maximal model performance with only small incremental gains coming from the assessment of psychological constructs. Before concluding that much psychological work has little incremental value for the prediction of humor appreciation there is an important caveat to consider; many psychological constructs were previously highlighted to matter because they interact with the nature of the material when steering humor appreciation. While the nomenclature differs, various research groups have found person-specific differentiation of stimuli into aggressive vs affiliative 58 , violating vs non-violating 28 , or adaptive vs maladaptive humor 67 . The resulting humor appreciation depends on a person’s unique appraisal of these attributes. In Study 1, we did not actively encode such rater X stimulus interaction terms warranting an extended analysis for predictor importance in Study 2.

Bi-variate analyses in Study 1 replicated much previous work. For instance, cheerfulness and mood were consistent predictors of higher humor appreciation 68 . Further, we also observe lower appreciation scores for people scoring high on morality 48 . A more surprising observation was that people finding the English language in the materials generally harder to understand, rated jokes funnier. This is counterintuitive as understanding (or “getting”) the joke is usually a strong predictor for finding it funny 69 . We assume that the current combination of sample (mostly native/fluent English speakers) and materials (no challenging vocabulary) led to virtually everyone understanding the jokes without any problems (median = 4; 5-point scale). Thus, the variation in self-indicated understanding could be mostly due to other factors like self-ascribed intellect. This surprising finding is followed up on in Study 2.

Study 2: Pre-registered follow-up

In Study 2, we aim to address three questions remaining after Study 1:

Can we predict humor appreciation more accurately by including rater X stimulus interactions?

What is the reliability of the humor appreciation scores (i.e., the maximally achievable prediction accuracy)?

Do we replicate the findings of Study 1 with new stimuli (including the finding that lower understanding of the stimuli is associated with higher stimulus appreciation)?

To that end, we conduct a pre-registered replication of Study 1, including different participants and humor stimuli. Further, we extend the measures to include interactions between rater and stimulus characteristics, Lastly, we estimate the reliability of humor appreciation scores through a test–retest procedure.

The procedure, assessed constructs (Cronbach’s alpha deviations from first study < 0.02), and analysis plan stayed the same as in Study 1 with four exceptions:

We used 10 new humor stimuli from the corpus of jokes from Rosenbusch and colleagues 27 . All stimuli are available in the Supplementary materials . Again, we picked one textual and one image joke from each of the pre-annotated categories of self-defeating, affiliative, aggressive, self-enhancing and sexual humor to ensure stimulus diversity 58 , 59 . Table 3 shows the text stimuli.

The general item “How hard was it for you to understand the English language/vocabulary in the jokes?” was replaced with 10 stimulus-specific items: “I understood this joke” (1 = strongly disagree, 5 = strongly agree). This was implemented to shed light on the observation from Study 1 that generally poorer language understanding was associated with more humor appreciation overall.

Two Likert-items per humor stimulus were added for each stimulus (“The joke seems aggressive” and “The joke seems friendly”; 1 = strongly disagree, 5 = strongly agree). These will be used to quantify the interaction between the nature of specific stimuli and the unique perception of each rater.

In a second wave, participants were invited to re-evaluate the same humor stimuli to allow for an estimation of re-test reliability of the humor appreciation measures. Notice, that we quantify the reliability of individual responses, as this aggregation-free reliability score poses the theoretical ceiling for the prediction models.

Participants were collected through the same online survey platform as in study 1. Participants from Germany (67.48%), the UK (24.54%), and the USA (8%) entered the study for a total of 5502 responders (59.47% female, 40.53% male). Ages ranged from 16 to 99 (M age  = 29.58, SD age  = 10.06). All participants indicated in a previous assessment that they can read and speak English and passed an English attention check question. None of the participants from study 1 were admitted to study 2. In a smaller, second wave, the stimuli ratings were re-provided by 407 participants (52% female, 48% male). Ages ranged from 16 to 75 (M age  = 35.17, SD age  = 11.86). Of the re-invited participants, 90% provided their answer between four and seven days after the first wave.

Distributions of binary humor appreciation scores (overall: 60.3% “good joke”) and continuous scores (overall: M  = 3.32, SD  = 1.21) for each of the 10 stimuli are depicted in Fig.  4 .

Hyperparameter settings for the XGboost algorithms were selected based on cross-validation performance and can be reproduced with the supplementary scripts. Table 4 and Fig.  5 show all accuracy metrics of each model when predicting humor appreciation on new data.

Out-of-sample prediction accuracies achieved with the three different predictor sets. The dashed line highlights for each accuracy metric which score would be achieved by simply always guessing the mean/majority class. The width of the points is wider than their respective 95% confidence intervals. The dotted line represents the retest-reliability of the outcome variable.

When taking the retest-reliability of the humor appreciation as the theoretical ceiling of prediction accuracy, one can see that the prediction of binary appreciation scores (top panels) virtually reaches the maximal limit. For continuous appreciation scores, the machine prediction even surpasses this threshold. This is likely related to the low stability of continuous appreciation scores over time ( r  = 0.49). Humor appreciation varies from situation to situation, making a retest-reliability score a very strict measure of error-freeness. The issue of construct stability vs. measurement error is discussed below.

Variable importance scores (see Fig.  6 ) were quantified as average performance drops on the test set when randomly permutating predictor variables (100 iterations).

It is apparent that past appreciation by the participant (of other stimuli) and other participants’ ratings of the current stimulus remain potent predictors. However, the predictors of perceived friendliness, understanding, and aggressiveness—all rater X stimuli measures—score high as well. This explains why the accuracy of the scale-based model (including these variables) now surpasses the performance of the purely “past ratings”-based model.

Figure  7 shows the zero-order associations between individual variables from psychological research and the separate funniness rating for all 10 stimuli.

Statistical associations (Pearson correlations and Cohen’s D’s) between predictors and continuous humor appreciation scores. The content of the individual stimuli (boxes 1–10) can be reviewed in the Supplementary materials .

In Study 2, where understanding of jokes was measured separately for each stimulus, the association between understanding and appreciating a joke is positive. This addresses the previously paradoxical finding that more (general) understanding predicts less (general) appreciation (cf., Study 1).

General discussion

Imagine you just thought of a fantastic joke. There are 100 contacts in your phone, but you are usure who to tell the joke to. Unbeknownst to you, 60 of them would laugh and 40 would not. Thus, if you told the joke to a randomly selected person, you would have a 60:40 chance of success. Applying much of the psychological research on humor appreciation from the last decades would boost this chance to about 76:24, a noticeable improvement but far from certainty. An obvious catch is that one needs to closely analyze specific characteristics of your contacts to achieve this improvement. Specifically, our results highlight the predictive value of past amusement proclivity, above all other constructs. The strongest “laugher” among your contacts is likely a safe bet. The next best class of predictors relate to the interaction between the specific joke content and the specific receiver characteristics including whether the person is likely to understand the joke or consider it overly malicious.

One reason for imperfect model accuracies appears to be the difficulty of measuring , rather than predicting, individual instances of amusement reliably. People’s average reaction to humor stimuli can be measured with higher reliability but predicting these scores would disallow the usage of stimuli-level predictor variables that were of interest here (for measurement of humor appreciation, see 70 ; for response aggregation, see 42 ). Without reliable amusement responses, one cannot obtain flawless predictions for individual reactions. Situational fluctuations of amusement are well-known in the humor literature and integrating state constructs like a user’s current rating tendency pushed the models to (and sometimes beyond) the outcome’s temporal stability. To which degree the remaining errors are due to suboptimal model performance vs. poor measurement in the validation data remains uncertain and is methodologically challenging to address 71 .

The high predictive value of rating aggregates is encouraging for practitioners aiming to build humor applications and tools. For instance, recommendation engines that gradually collect user behaviors appear more promising than systems relying on tests of user personality. Further, many collaborative and content-based recommendation engines are designed to slowly capture optimal combinations of user and content characteristics. Based on the current research, integrating user tendencies with user-by-content interaction features is optimal to achieve a high recommendation performance. This is in line with findings achieved in a variance-decomposition approach 27 . Note that most humor appreciation theories acknowledge such interactivity in some way. For instance, the Benign Violation Theory forwards that person-specific perceptions of stimuli’s aggressiveness and benignity lie at the core of amusement 28 .

The funniness ratings used in the current work are similar to explicit content ratings requested by many online entertainment platforms. Importantly, many platforms also use implicit measures to assess user preferences (e.g., searches, view time). Measures of laughter or smiling, the closest behavioral companions to amusement, are usually not available to content providers. These behaviors are also rare and difficult to interpret in solitary settings 72 . Due to its social functions, laughter might come with an altered pattern of predictors (e.g., a higher predictive value of social personality traits) compared to the current study.

In order to obtain explicit preferences, entertainment platforms sometimes ask their consumers to describe what kind of content they would like to see. Our investigations show that this method works on occasion. Participant’s explicit attitudes (e.g., whether they like cheesy pick-up lines, or jokes about people’s appearance) were predictive of their later humor appreciation. However, these topic-specific questions did not perform consistently and would require effortful questionnaire work by users who want to be entertained. Similarly, correlations between personality/demographics and humor appreciation from the humor literature were largely replicated, but offer little incremental value in terms of prediction power. For instance, gender, age, and the Big 5, do show face-valid relationships (e.g., agreeable people disliking some aggressive jokes), however these relationships were generally weak and inconsequential for model performance. Thus, humor research on these variables continues to be insightful, but more so for theory development than practical applications.

We replicated the finding that people’s rating aggregates were more predictive than the stimuli’s rating aggregates 27 . However, the current study relativizes the previous authors’ advice to prioritize rater characteristics when predicting humor appreciation. Specifically, if we accepted the predictive value of prominent psychological constructs as relevant, one would be well advised to consider the success rate of the joke (among other raters) as well, because it outperformed all personality variables. Thus, we would extend the cited recommendation into “focus on audience tendencies, specifically their past ratings, while not discarding average stimulus performance ”.

Limitations and future directions

A general limitation that can be addressed gradually is the inclusion of additional predictor variables. Past research points towards certain interaction effects that were not included here but nonetheless steer humor appreciation: people’s dispositional preference for certain joke structures . An example is conservatism which steers people’s preference for a full resolution of incongruity at the end of jokes (as opposed to nonsense jokes 70 ). Similarly, preferences for certain joke formats (e.g., text vs. video) can be tied to rater attributes (e.g., the ability to read) and can, depending on the stimuli and rater sample, become informative.

Additionally, we assume that additional sources of predictors, such as attributes of the joke teller or the relationship between joke teller and receiver can offer incremental gains. In past research, social cues and interpersonal perception affected the appreciation of the joke substantially (e.g., a halo effect of attractiveness 73 ; gender dynamics between teller and listener 74 ). To make this point clear, if you, the reader, feel like you could outperform the current models in predicting which of your contacts would laugh at your joke, you might rely on social cues that were not part of the current study. For instance, you are aware of the status dynamics at play when you speak to your personal contacts, and these affect humor responses significantly 75 . It is challenging to include such predictors in large-scale studies as the source of the joke has to be varied alongside the stimuli and rater characteristics. Here, we concentrated on performance humor, meaning jokes and images that can be implemented across social situations (e.g., content produced for online consumption). We did not include spontaneous humor that is afforded by specific social interactions, thus foregoing the predictive power of interpersonal predictor variables but also reducing sample size requirements, which were already substantial due to the highly-parametrized models we used. We assume that large web-scraping jobs, potentially including video recordings of social exchanges, alongside detailed annotations of the humor instances, could function to include such social and situational variables as predictors of humor appreciation.

We want to highlight the many analyses that can be conducted with the newly generated datasets. Close to 11.000 participants from multiple countries rated the humor stimuli. The dataset includes annotations of many demographic and psychological variables, allowing to re-test many statistical associations proposed in the literature. For some analyses, like the correlation between aggressiveness and funniness ratings, it might be reasonable to reformat and aggregate responses across either stimuli or subjects 42 . Data quality appears to be satisfactory (based on face-valid inter-item and inter-construct correlation, as well as successful attention checks, and sensitivity analyses), however it remains worthwhile to consider in which way measurement noise could have influenced our results. For instance, being able to measure predictors with higher reliability (e.g., moral internalization is assumed to be temporally unstable 49 ) can increase their predictive power. People aiming to improve the performance of the prediction algorithms presented here, can extend the supplementary scripts through, for instance, advanced feature engineering.

Humor appreciation is predictable. Variables like a person’s amusement propensity, joke popularity, and specific interactions between humor and rater characteristics allowed for prediction accuracies close to the suspected maximum. Main effects of broad psychological constructs were not useful to improve predictions. Humor research can be useful by forwarding the optimal range of predictors to content platforms. Field studies on such platforms would constitute a natural follow-up to the current work. Other research fields can apply a similar, prediction-focused approach to identify which of their studies and theories provide practical value.

Data availability

The datasets generated and analyzed during the current study are available in the OSF repository: https://shorturl.at/kmDGW .

Hofman, J. M. et al. Integrating explanation and prediction in computational social science. Nature 595 (7866), 181–188 (2021).

Article   ADS   CAS   PubMed   Google Scholar  

Hofmann, J., Platt, T., Ruch, W., Niewiadomski, R. & Urbain, J. The influence of a virtual companion on amusement when watching funny films. Motiv. Emot. 39 (3), 434–447 (2015).

Article   Google Scholar  

Ruch, W. Amusement. In The Oxford Companion to the Affective Sciences (eds Sander, D. & Scherer, K.) 27–28 (Oxford University Press, 2009).

Google Scholar  

Heintz, S. Separating content and structure in humor appreciation. J. Individ. Differ. 41 (1), 37–44 (2020).

Ruch, W. & Rath, S. The nature of humor appreciation: Toward an integration of perception of stimulus properties and affective experience. Humor Int. J. Humor Res. 6 (4), 363–384 (1993).

Ruch, W. & Hehl, F.-J. A two-mode model of humor appreciation: Its relation to aesthetic appreciation and simplicity-complexity of personality. In The Sense of Humor: Explorations of a Personality Characteristic 2nd edn (ed. Ruch, W.) 109–142 (Mouton de Gruyter, 2007).

Chapter   Google Scholar  

Freud, S. Der Witz und Seine Beziehung zum Unbewußten (Jokes and its Relationship to the Unconscious) (Deuticke, 1905).

McCauley, C., Woods, K., Coolidge, C. & Kulick, W. More aggressive cartoons are funnier. J. Pers. Soc. Psychol. 44 (4), 817 (1983).

Article   CAS   PubMed   Google Scholar  

Gruner, C. R. The Game of Humor: A Comprehensive Theory of Why We Laugh (Routledge, 2017).

Book   Google Scholar  

Hobbes, T. Leviathan (Hackett, 1994) ( Original work published 1668 ).

Warren, C. & McGraw, A. P. Differentiating what is humorous from what is not. J. Pers. Soc. Psychol. 110 (3), 407 (2016).

Article   PubMed   Google Scholar  

Wyer, R. S. & Collins, J. E. A theory of humor elicitation. Psychol. Rev. 99 (4), 663–688 (1992).

Ruch, W., Köhler, G. & Van Thriel, C. Assessing the “humorous temperament”: Construction of the facet and standard trait forms of the state-trait-cheerfulness-inventory—STCI. Humor Int. J. Humor Res. 9 , 303–339 (1996).

Ruch, W. & Carrell, A. Trait cheerfulness and the sense of humour. Pers. Individ. Differ. 24 (4), 551–558 (1998).

Wicker, F. W., Thorelli, I. M., Barron, W. L. & Willis, A. C. Studies of mood and humor appreciation. Motiv. Emot. 5 (1), 47–59 (1981).

Ruch, W., Köhler, G. & Van Thriel, C. To be in good or bad humour: Construction of the state form of the state-trait-cheerfulness-inventory—STCI. Pers. Individ. Differ. 22 (4), 477–491 (1997).

Galloway, G. & Chirico, D. Personality and humor appreciation: Evidence of an association between trait neuroticism and preferences for structural features of humor. Humor Int. J. Humor Res. 21 (2), 1–10 (2008).

Ku, L. C., Chang, Y. T. & Chen, H. C. How do extraverts process jokes? An event-related potential study on humor processing. Brain Cogn. 141 , 105553 (2020).

Köhler, G. & Ruch, W. Sources of variance in current sense of humor inventories: How much substance, how much method variance?. Humor Int. J. Humor Res. 9 (3/4), 363–397 (1996).

Moran, J. M., Rain, M., Page-Gould, E. & Mar, R. A. Do I amuse you? Asymmetric predictors for humor appreciation and humor production. J. Res. Pers. 49 , 8–13 (2014).

Ruch, W. & Deckers, L. Do extraverts, like to laugh?: An analysis of the situational humor response questionnaire (SHRQ). Eur. J. Pers. 7 (4), 211–220 (1993).

Deaner, S. L. & McConatha, J. T. The relation of humor to depression and personality. Psychol. Rep. 72 (3), 755–763 (1993).

Carretero-Dios, H. & Ruch, W. Humor appreciation and sensation seeking: Invariance of findings across culture and assessment instrument?. Humor 23 (4), 427–445 (2010).

Forabosco, G. & Ruch, W. Sensation seeking, social attitudes and humor appreciation in Italy. Pers. Individ. Differ. 16 (4), 515–528 (1994).

Mendiburo-Seguel, A., Páez, D. & Martínez-Sánchez, F. Humor styles and personality: A meta-analysis of the relation between humor styles and the Big Five personality traits. Scand. J. Psychol. 56 (3), 335–340 (2015).

Plessen, C. Y. et al. Humor styles and personality: A systematic review and meta-analysis on the relations between humor styles and the Big Five personality traits. Pers. Individ. Differ. 154 , 109676 (2020).

Rosenbusch, H., Evans, A. M. & Zeelenberg, M. the relative importance of joke and audience characteristics in eliciting amusement. Psychol. Sci. 33 (9), 1386–1394 (2022).

McGraw, A. P. & Warren, C. Benign violations: Making immoral behavior funny. Psychol. Sci. 21 (8), 1141–1149 (2010).

McGraw, A. P., Warren, C., Williams, L. E. & Leonard, B. Too close for comfort, or too far to care? Finding humor in distant tragedies and close mishaps. Psychol. Sci. 23 (10), 1215–1223 (2012).

Alvaro, P. K., Roberts, R. M. & Harris, J. K. A systematic review assessing bidirectionality between sleep disturbances, anxiety, and depression. Sleep 36 (7), 1059–1068 (2013).

Article   PubMed   PubMed Central   Google Scholar  

Fukuda, S., Matsuda, Y., Tani, Y., Arakawa, Y., & Yasumoto, K. Predicting depression and anxiety mood by wrist-worn sleep sensor. In 2020 IEEE International Conference on Pervasive Computing and Communications Workshops (PerCom Workshops) , 1–6. (IEEE, 2020).

Yarkoni, T. & Westfall, J. Choosing prediction over explanation in psychology: Lessons from machine learning. Perspect. Psychol. Sci. 12 (6), 1100–1122 (2017).

Rosenbusch, H., Soldner, F., Evans, A. M. & Zeelenberg, M. Supervised machine learning methods in psychology: A practical introduction with annotated R code. Soc. Pers. Psychol. Compass 15 (2), e12579 (2021).

Bertero, D. & Fung, P. Deep learning of audio and language features for humor prediction. in Proceedings of the Tenth International Conference on Language Resources and Evaluation (LREC'16) , 496–501 (2016).

Kamal, A. & Abulaish, M. Self-deprecating humor detection: A machine learning approach. in International Conference of the Pacific Association for Computational Linguistics , 483–494. (Springer, 2019).

Mihalcea, R. & Strapparava, C. Making computers laugh: Investigations in automatic humor recognition. in Proceedings of Human Language Technology Conference and Conference on Empirical Methods in Natural Language Processing , 531–538 (2005).

Yang, Z., Hooshmand, S. & Hirschberg, J. CHoRaL: Collecting humor reaction labels from millions of social media users. in Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing , 4429–4435 (2021).

Annamoradnejad, I. & Zoghi, G. Colbert: Using bert sentence embedding for humor detection. arXiv:2004.12765 (2020).

Shahaf, D., Horvitz, E. & Mankoff, R. Inside jokes: Identifying humorous cartoon captions. in Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining , 1065–1074 (2015).

Hessel, J. et al . Do Androids Laugh at Electric Sheep? Humor" Understanding" Benchmarks from The New Yorker Caption Contest . arXiv:2209.06293 (2022).

Winters, T. Computers learning humor is no joke. Harvard Data Sci. Rev. 3 (2), 1–10 (2021).

Ruch, W. Will the real relationship between facial expression and affective experience please stand up: The case of exhilaration. Cogn. Emot. 9 (1), 33–58 (1995).

Rammstedt, B. & John, O. P. Measuring personality in one minute or less: A 10-item short version of the Big Five Inventory in English and German. J. Res. Pers. 41 (1), 203–212 (2007).

Lau, C., Chiesi, F. & Saklofske, D. H. The heart of humor: A network analysis of the temperamental basis of humor and humor personality traits. Pers. Individ. Differ. 185 , 111253 (2022).

Lau, C., Chiesi, F., Hofmann, J., Saklofske, D. H. & Ruch, W. Development and linguistic cue analysis of the state-trait cheerfulness inventory: Short form. J. Pers. Assess. 103 (4), 547–557 (2021).

Chang, C. How morality judgments influence humor perceptions of prankvertising. Int. J. Advertis. 40 (2), 246–271 (2021).

Erzi, S. Dark Triad and schadenfreude: Mediating role of moral disengagement and relational aggression. Pers. Individ. Differ. 157 , 109827 (2020).

Yam, K. C. et al. Why so serious? A laboratory and field investigation of the link between morality and humor. J. Pers. Soc. Psychol. 117 (4), 758 (2019).

Aquino, K. & Reed, A. II. The self-importance of moral identity. J. Pers. Soc. Psychol. 83 (6), 1423 (2002).

Hoyle, R. H., Stephenson, M. T., Palmgreen, P., Lorch, E. P. & Donohew, R. L. Reliability and validity of a brief measure of sensation seeking. Pers. Individ. Differ. 32 (3), 401–414 (2002).

van Dongen, J. D., de Groot, M., Rassin, E., Hoyle, R. H. & Franken, I. H. Sensation seeking and its relationship with psychopathic traits, impulsivity and aggression: A validation of the Dutch Brief Sensation Seeking Scale (BSSS). Psychiatry Psychol. Law 29 (1), 20–32 (2022).

Thorson, J. A. & Powell, F. C. Measurement of sense of humor. Psychol. Rep. 69 (2), 691–702 (1991).

Kerkkånen, P., Kuiper, N. A. & Martin, R. A. Sense of humor, physical health, and well-being at work: A three-year longitudinal study of Finnish police officers. Humor Int. J. Humor Res. 17 , 1–10 (2004).

Carretero-Dios, H., Delgado-Rico, E., López-Benítez, R. & Acosta, A. Differential effects of affective arousal and valence on humor appreciation in female university students. Humor 36 (2), 225–243 (2023).

Thompson, E. R. Development and validation of an internationally reliable short-form of the positive and negative affect schedule (PANAS). J. Cross-cult. Psychol. 38 (2), 227–242 (2007).

Watson, D., Clark, L. A. & Tellegen, A. Development and validation of brief measures of positive and negative affect: The PANAS scales. J. Pers. Soc. Psychol. 54 (6), 1063 (1988).

Burmeister, J. M. & Carels, R. A. Weight-related humor in the media: Appreciation, distaste, and anti-fat attitudes. Stigma Health 1 (1), 92–107 (2015).

Martin, R. A., Puhlik-Doris, P., Larsen, G., Gray, J. & Weir, K. Individual differences in uses of humor and their relation to psychological well-being: Development of the Humor styles questionnaire. J. Res. Pers. 37 (1), 48–75 (2003).

Ruch, W. Assessment of appreciation of humor: Studies with the 3 WD humor test. In Advances in Personality Assessment, Vol. 9 (eds Spielberger, C. D. & Butcher, J. N.) 27–75 (Lawrence Erlbaum Associates, 1992).

Appinio. https://www.appinio.com/ . Accessed 21 Jun 2023.

Chen, T. & Guestrin, C. Xgboost: A scalable tree boosting system. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining , 785–794 (2016).

Borisov, V. et al. Deep neural networks and tabular data: A survey. IEEE Trans. Neural Netw. Learn. Syst. 1 , 1–21 (2022).

McElfresh, D. et al . When Do Neural Nets Outperform Boosted Trees on Tabular Data? arXiv:2305.02997 . (2023).

González, S., García, S., Del Ser, J., Rokach, L. & Herrera, F. A practical tutorial on bagging and boosting based ensembles for machine learning: Algorithms, software tools, performance study, practical perspectives and opportunities. Inf. Fusion 64 , 205–237 (2020).

Kuhn, M. & Wickham, H. Tidymodels: A Collection of Packages for Modeling and Machine Learning Using Tidyverse Principles. https://www.tidymodels.org . (2020).

R Core Team. R: A Language and Environment for Statistical Computing . (R Foundation for Statistical Computing, 2022). https://www.R-project.org/ .

Guenter, H., Schreurs, B., Van Emmerik, I. H., Gijsbers, W. & Van Iterson, A. How adaptive and maladaptive humor influence well-being at work: A diary study. Humor 26 (4), 573–594 (2013).

Ruch, W. State and trait cheerfulness and the induction of exhilaration. Eur. Psychol. 2 (4), 328–341 (1997).

Dunbar, R. I. M., Launay, J. & Curry, O. The complexity of jokes is limited by cognitive constraints on mentalizing. Hum. Nat. 27 , 130–140 (2016).

Article   CAS   PubMed   PubMed Central   Google Scholar  

Ruch, W. Assessment of appreciation of Humor: Studies with the 3 WD Humor test. Adv. Pers. Assess. 9 , 27 (2013).

Dejonckheere, E. et al. Assessing the reliability of single-item momentary affective measurements in experience sampling. Psychol. Assess. 34 , 1138–1154 (2022).

Provine, R. R. Laughing, tickling, and the evolution of speech and self. Curr. Direct. Psychol. Sci. 13 (6), 215–218 (2004).

Cowan, M. L. & Little, A. C. The effects of relationship context and modality on ratings of funniness. Pers. Individ. Differ. 54 (4), 496–500 (2013).

McLachlan, A. The relationship between familiarity, gender, disagreement, and status and bouts of solitary and joint laughter. Curr. Psychol. 1 , 1–15 (2022).

Oveis, C., Spectre, A., Smith, P. K., Liu, M. Y. & Keltner, D. Laughter conveys status. J. Exp. Soc. Psychol. 65 , 109–115 (2016).

Download references

Author information

Authors and affiliations.

Department of Psychological Methods, University of Amsterdam, Amsterdam, The Netherlands

Hannes Rosenbusch & Thomas Visser

Appinio GmbH, Hamburg, Germany

Hannes Rosenbusch

You can also search for this author in PubMed   Google Scholar

Contributions

All authors conceptualized and ran the studies. All authors conducted the statistical analyses. All authors prepared data visualizations. All authors wrote and edited the manuscript. All authors contributed to the revision.

Corresponding author

Correspondence to Hannes Rosenbusch .

Ethics declarations

Competing interests.

The authors declare no competing interests.

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Supplementary information., rights and permissions.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Cite this article.

Rosenbusch, H., Visser, T. Humor appreciation can be predicted with machine learning techniques. Sci Rep 13 , 19035 (2023). https://doi.org/10.1038/s41598-023-45935-1

Download citation

Received : 21 June 2023

Accepted : 25 October 2023

Published : 03 November 2023

DOI : https://doi.org/10.1038/s41598-023-45935-1

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

By submitting a comment you agree to abide by our Terms and Community Guidelines . If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

The Psychology Behind Humor

Are jokes just a bit of harmless fun.

Posted March 29, 2024 | Reviewed by Tyler Woods

• Jokes are ever-present, but especially so on the first of April.
• For a joke to be seen as funny, it should violate a norm.
• Violating norms can be risky and backfire on the joke teller.

As April Fools’ Day is just around the corner, when better to look at the psychology of jokes? Over the years, April Fools’ Day has become larger than life, especially with social media spreading jokes at a rapid rate. Whilst it may be seen as just a bit of fun, there is more to jokes and humour than may be apparent, begging the question: are the jokes on the 1 st of April a good or a bad idea?

Every year, companies, individuals, papers, and newsreaders present jokes on the 1 st of April that may be entertaining, a bit confusing, and at times causing an unexpected upheaval. One announcement made in 1949, on a radio station in New Zealand, that a mile-wide swarm of wasps was headed for Auckland, was perhaps taken a bit too seriously. The news came with helpful advice on how to prepare for the incoming insects and it resulted in many tucking their trousers into their socks as well as setting honey traps outside their doors. Other examples of April Fool’s Day jokes include a claim that Stephen King would run for governor of Maine and that Heinz was launching a chocolate-flavoured mayonnaise.

Sometimes, though, April Fool’s Day jokes are simply not viewed as funny. For example, a joke by Hooters ended up in litigation as one of their servers thought she had won a new Toyota car. Instead, upon collection, she was given a Yoda doll.

What makes something funny?

This is a million-dollar question and one that is not easily answered. First, it is worth noting that people have different opinions about what is humorous. It is influenced by the context, the culture, and subjective perspectives , meaning that it is not consistent across populations. However, there are aspects of jokes that may determine, more broadly, if they are funny.

If a joke is violating a norm or rule of some kind, it is more likely to be perceived as funny. Such violations need to be benign, as they can otherwise be perceived to be unpleasant or even revolting. What determines whether it is benign is linked to how bad something is and how far removed we are from it. In terms of personal experience, serious events may be considered funnier over time, while stubbing a toe, which would be considered a minor violation, would lose its comedic effect over time.

Interestingly, people have been found to be more likely to laugh when they are near others compared to when they are alone. Children, in particular, have been found to laugh much more when they are with others, suggesting that laughter may be a social signal rather than a direct response to humour.

Not just for fun

Jokes and laughter may seem frivolous but can, in fact, serve social functions like managing strong emotions such as embarrassment and showing affiliation . Humour can also be used to cope in work environments. For example, studies have found that there is a meaningful function of humour in healthcare settings . It can have a beneficial effect on patient-carer interaction as it strengthens the connection between the two and eases stressful symptoms linked to end-of-life care among palliative care nursing professionals.

Humour can also function as a facilitator in insight problem-solving . The rationale for this relates to attentional processes, in that humour relieves stress . By doing so, it dilutes the degree of attention being devoted to the problem and, in turn, stimulates the problem solver's “peripheral focus,” destabilizing perceptual and thought patterns, thus helping people to change their perspective to restructure the problem.

Humour can be risky

Because jokes can be misunderstood in cross-cultural settings, it has been suggested that they should be used cautiously , especially if used in political or diplomatic settings. Couple this with the fact that humour requires a violation and it can be a risky tactic, especially as perceptions of appropriateness are subjective .

Risky jokes may also be seen differently by people over time. This is something research has not really explored. For example, in 2021, David Letterman faced backlash for an interview he had conducted with Lindsay Lohan in 2013. At the time, the audience had laughed at his repeated questions about her going to rehab. However, eight years later, his comments were seen as offensive. Possibly, this means that risky jokes made by companies or individuals for April Fool’s Day can cause the teller to lose status, raising some questions about whether it is good PR to engage in joke telling, even if it is for April Fool’s.

The fact that humour can generate valuable results when integrated into daily life means that telling jokes and making people smile or laugh can have a positive impact, though it is important to ensure that any violations made are not significant enough to damage your future reputation.

Cathrine Jansson-Boyd, Ph.D., is a Consumer Psychologist based at Anglia Ruskin University, Cambridge, UK.

• Find a Therapist
• Find a Treatment Center
• Find a Support Group
• International
• New Zealand
• South Africa
• Switzerland
• Asperger's
• Bipolar Disorder
• Chronic Pain
• Eating Disorders
• Passive Aggression
• Personality
• Goal Setting
• Positive Psychology
• Stopping Smoking
• Low Sexual Desire
• Relationships
• Child Development
• Therapy Center NEW
• Diagnosis Dictionary
• Types of Therapy

Understanding what emotional intelligence looks like and the steps needed to improve it could light a path to a more emotionally adept world.

• Coronavirus Disease 2019
• Affective Forecasting
• Neuroscience

101 Lab Jokes

In the world of science , where curiosity meets discovery, there’s a special place where laughter finds its way into the serious realm of research and experimentation. Welcome to the world of lab jokes, where scientists and researchers let their hair down and share a good laugh over the peculiarities of their work .

From biology to chemistry , from atoms to genes, these witty one-liners and puns lighten the atmosphere and remind us that even in the pursuit of knowledge, humor has its place. So, put on your lab coats and safety goggles as we delve into a collection of lab jokes that will tickle your scientific funny bone .

Top 101 Lab Jokes:

• Why don’t biologists trust atoms? Because they make up everything!
• Why did the microbiologist refuse to play cards with the leprosy bacteria? He heard they like to play with a full-deck of genes!
• What did one cell say to his sister cell when she stepped on his foot ? Mitosis!
• Why do chemists like high pressure? It’s the only time they can be under the weather and still be in their element.
• Why was the biology book so full of itself? Because it had all the “organ”ization!
• How does a biologist make a hormone ? They don’t pay it.
• Why do chemists call helium, curium, and barium the healing elements? Because if you can’t helium or curium, you barium!
• Why was the geneticist good at repairing his car ? He had the genes for it.
• Why did the chemist go broke? Because he had no solutions.
• What did the biologist wear to impress their date ? Designer genes.
• Why don’t biologists have mid-life crises? They’re too busy with their cell-f reflection.
• Why did the physicist break up with the biologist? There was no chemistry.
• What do you call a tooth in a glass of water ? A one molar solution.
• Why did the biology student get an A? Because he had all the organs down to a T.
• Why are chemists so good at solving problems? They always have a solution.
• Why did the lab tech go to therapy ? He had separation anxiety .
• What did the DNA say to the other DNA? Do these genes make me look fat?
• What did the biologist use to fix his jeans? A gene splicer.
• What do chemists call a clown who inhaled helium? A noble gas .
• Why did the biologist go to the party alone? He had no body to go with.
• What’s a chemist’s favorite type of tree ? Chemistry.
• What do you call a microbiologist that’s gone bad? A germ-inator.
• Why was the microbiologist always calm? He always kept his cell-f control.
• Why was the chemist’s report seven pages long ? He had a lot of elements to cover.
• How did the biology student know he was going to do well on his test? He felt it in his bones.
• What did the enzyme say to the substrate? I’ve got a crush on you.
• Why was the biologist never lonely ? He always had his buds around.
• What’s a chemist’s favorite type of dog ? A lab-rador.
• Why did the DNA go to the party? Because he wanted to express himself.
• How does a biochemist spice up their life ? They get into the proteins.
• What did the biologist say when they found the missing link? “I’ve been looking for you gene-etically!”
• What does a chemist put on their bed ? A periodic blanket.
• Why don’t physicists trust particles? Because they can’t measure their position and momentum at the same time.
• What did the scientist say when they found two isotopes of helium? HeHe.
• Why was the DNA sequence so embarrassed? Because it was caught in the middle of replication.
• What did the biologist say to the physicist? “Let’s combine our genes and make a theory of everything.”
• Why do biologists hate statistics? Too many mean averages.
• How can you spot a chemist in the restroom? They wash their hands before and after using the toilet.
• What did the scientist say when he discovered a new species of bacteria? “I’ve got culture!”
• Why are lab safety rules so strict? Because broken glass and chemicals mix like acid and base.
• Why did the biologist become a gardener ? They wanted to branch out.
• What do you call two atoms of helium laughing together? HeHe.
• How do you know you’re speaking to an extroverted physicist? They look at your shoes when they’re talking to you.
• Why do biologists go to the bar? To get a round of shots under the microscope.
• How do you cut a sea in half? With a sea-saw.
• Why don’t we tell secrets in labs? Because the walls have ears and the test tubes have eyes !
• What did the chemist say after a failed experiment? “We’ll just have to go back to the drawing Bunsen.”
• Why do biologists make bad comedians ? They have too many inside jokes.
• What do you do with a sick chemist? If you can’t helium and you can’t curium, you might as well barium!
• How do you know a physicist has been in your house? Your kids have lecture notes about their toys.
• Why did the DNA go to therapy? It had too many issues to unwind.
• Why don’t chemists make good chefs? They’re always over-reacting!
• How do we know that atoms are generous? They always give away their electrons.
• Why do biologists study fish ? They want to get to know their cells better.
• What did the geneticist say after a successful experiment? “I’ve got the code!”
• Why was the biologist always late? He took his time in the gene pool .
• What do you call a biologist with a large brain ? An organ donor.
• What’s the first rule of the Chemistry Club? Never mix acid with base.
• Why do chemists like nitrates so much? They’re cheaper than day rates.
• What do you call a biologist who studies their own body? Self-centered.
• Why did the biologist go on a diet ? He wanted to reduce his mass.
• Why do biologists go to school ? To get a degree in Celsius!
• Why do chemists love coffee ? It’s all about the brew-tal reaction.
• Why are chemists great at solving problems? They have all the solutions.
• Why did the chemist sit on a cold bench? He wanted to test his reaction.
• What’s the most relaxing thing for a biologist? Listening to some cell-o music .
• What’s a physicist’s favorite food ? Fission chips.
• Why did the lab mouse join a dating app? He heard there was a lot of chemistry.
• Why do chemists make the best DJs ? They always drop the base.
• Why did the biologist buy an extra microscope? He needed to double-check his work.
• How do we know the moon isn’t made of cheese ? The mice haven’t left yet!
• Why did the atom break up with his girlfriend ? He couldn’t bond with her anymore.
• What did the RNA say to the DNA? “Stop being so negative!”
• Why are chemists great for solving problems? They always have a solution!
• Why was the computer cold in the lab? It left its Windows open!
• Why was the physics book heavy? It had a lot of mass.
• What do you call a nervous chemist? A shaky solution.
• What’s a biologist’s favorite dessert ? Berry DNA-lato.
• Why did the chemist start doing yoga ? He wanted to achieve a state of equilibrium.
• Why was the microscope always unhappy? It had a tough time focusing.
• What’s a biologist’s favorite gym equipment? The gene machine.
• What do you call a swimming electron ? A Michael Phelpsicle.
• Why was the photon arrested? For speeding .
• What’s a biologist’s favorite band ? The Cell Out Boy .
• How does a physicist exercise ? By doing quantum leaps.
• Why did the DNA take antidepressants? It was feeling coiled up.
• Why was the chemist sad ? He lost an electron.
• What’s the first rule in a lab? Don’t trust atoms, they make up everything!
• What does a microbiologist use to make a call? A cell phone .
• What did the biologist say when her colleague found a new species? “Cell-a-brate good times, come on!”
• Why do chemists always work in a team? Because they have good chemistry.
• How do scientists freshen their breath? With experi-mints.
• What did the virus say to the bacterium? “Stop copying me!”
• Why do biologists hate math ? Because they can’t find the square root of a tree.
• What did one charged atom say to the other? “I’ve got my ion you.”
• Why was the protein always the life of the party? Because it knows how to unwind.
• Why do biologists wear glasses? To improve their cell-vision.
• What do you call a laughing test tube? A cracking good time.
• Why did the lab tech feel safe in the lab? Because he was surrounded by glasses.
• What’s a physicist’s favorite dance move? The Quantum Jump.
• Why was the bio lab always so noisy? Because all the cells kept ringing!

From clever wordplay to hilarious situational humor, these lab jokes have offered a lighthearted glimpse into the world of scientists, biologists, chemists, and physicists. They showcase the ability of these dedicated professionals to find humor in their daily endeavors, even when faced with complex concepts and intricate experiments.

In the midst of their pursuit of understanding the mysteries of life and the universe, scientists embrace the power of laughter, reminding us that humor can be found in the most unexpected places.

So, the next time you find yourself in a laboratory or discussing scientific theories, remember to share a lab joke and bring a smile to the faces of those around you. After all, a little laughter can make the scientific journey all the more enjoyable.

Similar Posts

108 Common Sense Jokes

Laughter is contagious, and one of the best ways to tickle our funny bones is through clever and witty jokes. Common sense jokes, with their puns and clever wordplay, bring a smile to our faces and make us appreciate the simple yet hilarious aspects of life. These jokes often play on familiar concepts and everyday…

101 Red Jokes

Red jokes, often centered around the color red and various red objects, offer a delightful blend of humor and wordplay. From witty puns to clever observations, these jokes are sure to bring a smile to anyone’s face. The color red has a unique ability to capture attention and evoke emotions, and these jokes capitalize on…

150 Weather Jokes

Weather jokes, like the ever-changing nature of the elements they humorously depict, bring a delightful gust of laughter into our lives. From raindrops cracking jokes to clouds attending school and hurricanes attempting detective work, these witty quips playfully personify the forces of nature that shape our daily lives. With puns, wordplay, and clever observations, weather…

100 Hip Jokes

Get ready for a lighthearted journey filled with laughter and clever wordplay as we delve into the world of hip jokes. These punny and witty quips will have you grooving with joy as we explore the comical side of hips and their playful interactions with other body parts. So, let’s limber up our funny bones…

101 PTO Jokes

In the fast-paced world of work, everyone looks forward to a much-needed break. Yes, we’re talking about the beloved phenomenon called PTO, or Paid Time Off. It’s that magical time when employees get to escape the daily grind, recharge their batteries, and have a little fun. But have you ever wondered how everyday objects and…

150 Band Jokes

Music is not only a universal language but also a great source of joy and entertainment. And within the realm of music, bands have their own unique charm. They bring together talented musicians, each playing their own instruments, to create harmonious melodies that captivate our hearts and souls. But behind the scenes, in the midst…

Leave a Reply Cancel reply

Your email address will not be published. Required fields are marked *

Save my name, email, and website in this browser for the next time I comment.

Javascript not detected. Javascript required for this site to function. Please enable it in your browser settings and refresh this page.

Personalize Your Experience

Log in or create an account for a personalized experience based on your selected interests.

Already have an account? Log In

Free standard shipping is valid on orders of $45 or more (after promotions and discounts are applied, regular shipping rates do not qualify as part of the $45 or more) shipped to US addresses only. Not valid on previous purchases or when combined with any other promotional offers.

Register for an enhanced, personalized experience.

Receive free access to exclusive content, a personalized homepage based on your interests, and a weekly newsletter with topics of your choice.

Home / Healthy Aging / The health benefits of humor

The health benefits of humor

Please login to bookmark.

Years ago, I went to what I thought would be a typical yoga class — stretching and meditative silence in a darkened room. To my great surprise, I actually showed up for a laughter yoga course. The class involved deep breathing, making loud “ha” noises with a hand on the belly or chest, and walking around to other class members, laughing while making eye contact.

As an introvert, this type of class was pretty much my worst nightmare. Still, I went through the motions of each exercise and as I slowly loosened up, I realized I actually felt pretty good. The stress of the day had receded. I felt present. My body finally relaxed. Though the laughter was forced at first, I eventually found myself giggling right along with everyone else.

Looking back, my yoga instructor was clearly tapping into the therapeutic potential of laughter — which can have a range of potential benefits for your mental, social and physical health.

According to Mayo Clinic expert and oncologist Edward T. Creagan, M.D., staying healthy isn’t just about diet and exercise. It’s about finding moments of humor, joy and human connectedness — even if they’re not always quite what you expect.

Finding joy is especially important today, as many people are navigating heightened burnout, health concerns and social isolation in the wake of the pandemic. Below, Dr. Creagan explains how laughter can impact — and improve — your well-being.

The social and mental health benefits of laughter

Though most people know the cliche “laughter is the best medicine,” there’s actually scientific evidence to back up the benefits. Humor can actually be a powerful — if often overlooked — strategy to support mental health.

Laughing can alleviate stress by decreasing stress hormones like epinephrine and cortisol, says Dr. Creagan. These hormones play a role in mental health conditions like depression and anxiety, as well as insomnia.

“When we laugh, when we’re funny, when we’re engaged with funny people who make us feel better, there is a decrease in cortisol and there’s an increase in the endorphins,” says Dr. Creagan.

During times of connectedness and joy — like when you laugh with a friend or watch a funny video online — Dr. Creagan says the body is flooded with feel-good hormones like dopamine, oxytocin and endorphins.

In fact, some studies suggest that laughter can help to not just alleviate but actually reverse the body’s stress response.

“These are the love and chocolate hormones,” says Dr. Creagan, referring to the fact that these hormones can show up when you fall in love or eat chocolate. “Nothing is as effective (at relieving stress) as running a marathon, holding your cat or your dog, or having an explosion of laughter,” says Dr. Creagan.

In addition, recent studies have shown that laughter and introducing humor into mental health settings like therapy may reduce the symptoms of depression and anxiety, boost both self-esteem and sense of humor, improve social skills and communication, and even help reduce the intensity of mental health conditions. Meanwhile, there is evidence to suggest that laughter can enhance memory, creative thinking, friendliness, vigor and even hopefulness.

But Dr. Creagan says humor isn’t just about feeling good — it can help us cope with the hard parts of life as well.

“If a patient can have a moment of levity in the face of crisis, I think it helps them better cope and better deal with the uncertainties of their problems,” he says.

This has certainly been the case for Roberta Gold, president of the Association for Applied and Therapeutic Humor , an advocacy and research organization. Like many people in her family, Gold has a genetic disease that can impact her day-to-day life.

“I use humor a lot with the medical issues we’ve had in our family,” says Gold. “It makes us more resilient and able to deal with it.”

Perhaps unsurprisingly, research suggests that people who are more resilient to life challenges tend to be happier and healthier.

Laughter, longevity and physical health

Laughter also may help support your physical health.

“A hearty laugh may decrease blood pressure, help regulate heart rhythm and just provide an element of joy, that elusive factor in well-being,” says Dr. Creagan.

Additionally, laughter may have a positive impact on nearly your entire body — it may benefit your immune system and respiratory function, aid muscle relaxation, assist with pain relief, and stimulate circulation.

Laughter also may have a role in the process of healthy aging. In a group of over 14,000 older adults, researchers found that people who laughed less were more likely to develop a functional disability. Meanwhile, another small study reported that laughter therapy reduced insomnia and improved overall well-being in the elderly.

Outlook and social connectedness also play a role in aging. Evidence suggests that optimists — or those who look to the sunny side of life — live longer than those who tend to focus on the bad. In one study, highly optimistic people had a 29% lower risk of early death and were significantly less likely to die of a heart attack or stroke. Though scientists aren’t sure exactly why optimism impacts longevity , it’s clear that outlook, social connectedness, community, and sense of purpose can all lend to a longer, healthier life. For many people, laughter plays a role in all of these factors.

Simply put, laughter is often an important component of not just how long someone lives, but how well they live.

How to incorporate more humor in your life

Perhaps the biggest benefit of laughter is that it’s so easy to enjoy — and incorporate into your life.

To get started, consider where you could bring levity into your routine. Gold, for example, keeps toys like a stuffed animal or stress ball in high-pressure environments like her office or car. Technology, too, can help. You may listen to your favorite comedian during your lunch break or watch a funny movie before bed — as long as you’re laughing, you can reap the health benefits. But Dr. Creagan says that human interactions are still the best opportunity for laughter, like a face-to-face conversation with an old friend or loved one.

However, it’s also important to remember that humor is social, regional and contextual, as well as connective. And if not used with care, even well-meaning jokes can harm others and your relationships.

“Healthy humor is inclusive,” says Gold. “It includes everybody and lightens up somebody’s day instead of putting them down.”

In other words, humor should never be at the expense of a single person, says Dr. Creagan, or at the expense of a culture or group of people, like a certain race, gender or sexual orientation. When in doubt, Dr. Creagan recommends leading with curiosity.

“Ask another person about themselves, but a guaranteed link is to ask that person to speak about the pets and their lives. No one can talk about their pet without smiling,” he says.

Relevant reading

Mayo Clinic on Incontinence

This easy-to-understand book is a source of hope and assurance for all individuals who experience bladder and bowel control problems. Its message is that incontinence isn’t something people need to live with; most incontinence can be improved or cured. Mayo Clinic on Incontinence is a helpful resource to successfully manage…

Mayo Clinic Guide to Arthritis, Second Edition

If you have joint pain, you know it can lead to frustrating limitations in daily life. In fact, arthritis is the most common cause of disability in the United States. This complex group of joint diseases, which includes osteoarthritis and rheumatoid arthritis among others, affects at least 54 million Americans…

Discover more Healthy Aging content from articles, podcasts, to videos.

You May Also Enjoy

Privacy Policy

We've made some updates to our Privacy Policy. Please take a moment to review.

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List
• Behav Sci (Basel)

A Study on Instructional Humor: How Much Humor Is Used in Presentations?

Vera paola shoda.

1 Center for Computational Social Science (CCSS), Research Institute for Economics and Business Administration (RIEB), Kobe University, 2-1 Rokkodai, Nada, Kobe 657-8501, Japan

2 Degree Programs in Systems and Information Engineering, Graduate School of Science and Technology, University of Tsukuba, 1-1-1 Tennodai, Tsukuba 305-8573, Japan

Toshimasa Yamanaka

3 Graduate School of Comprehensive Human Sciences, University of Tsukuba, Tsukuba 305-8577, Japan; pj.ca.abukust.ustujieg@mayt

Associated Data

The data presented in this study are openly available on Kaggle at https://kaggle.com/vpshoda/instructional-humor-research (accessed on 23 November 2021).

Humor is applied in pedagogy to create a positive learning environment. Recent research focuses on the theories, effects, individual differences, and qualitative aspects of humor for instruction. However, there is a lack of studies focusing on quantitative features. Therefore, this research explored the quantitative characteristics of instructional humor in a naturalistic setting and applied techniques from natural language processing (NLP). This paper describes the results of two studies. The first study focused on instructional humor frequency and the placement of humor, while the linguistic features of instructional humor and non-instructional humor were compared in the second study. Two corpora were used in this research: TED Talks and user-submitted jokes from “stupidstuff.org” The results found that educators used humor 12.92 times for popular talks, while less popular talks only had 3.92 times. Humor is also more commonly placed during the first parts of the talk and lessens toward the end. There were also significant differences between the linguistic features of instructional and non-instructional humor in terms of readability scores and sentiment. These results provide a substantial update on quantitative instructional humor research and help educators understand how to use humor in the classroom in terms of quantitative and linguistic features.

1. Introduction

Educators face the challenge of creating a learning environment that is advantageous to student learning. Especially with the rapid changes and advances in society, new methodologies and tools for education are highly valuable. Humor is an essential tool used by educators to improve the learning environment of their students. The use of humor is a prevalent communication behavior in pedagogical settings. In 1983, Robinson argued that “What is learned with laughter is learned well.” [ 1 ] (p. 121). Likewise, much research provides evidence that humor positively affects student learning [ 2 , 3 , 4 , 5 , 6 , 7 , 8 , 9 , 10 , 11 , 12 , 13 , 14 ]. Therefore, research in instructional humor is significant to educators.

1.1. Definition and Theories of Humor

Researchers define humor in a variety of ways. According to Scheel [ 15 ], superiority, incongruity, and arousal relief are the most popular theories in humor research. Superiority theory, which has been prevalent since the time of Plato and Aristotle, explains that laughter is an effect of a feeling of superiority due to the depreciation of other people [ 16 ]. Incongruity theory argues that something is perceived as humorous when there is a contradiction or unexpected outcome [ 17 ]. In the arousal theory, Berlyne [ 18 ] said that “Humorous situations always contain factors that can be expected to raise arousal and other factors that can be expected to lower arousal or else keep it within moderate bounds.” [ 18 ] (p. 861). Another interesting theory in humor research is the anxiety theory, which states that laughter results from tension release [ 19 ].

1.2. Humor and Laughter

Laughter is more ancient than humor or speech [ 20 ] and is considered a social stimulus [ 21 ]. Similarly, laughter is also defined as a component of a universal language of basic emotions, which all people have in common and recognize [ 22 , 23 ]. According to Willman, “Laughter occurs when a total situation causes surprise, shock, or alarm, and at the same time induces an antagonistic attitude of playfulness or indifference.” [ 24 ] (p. 70). Several works in humor recognition have used laughter to indicate an instance of humor [ 25 , 26 ]. However, literature is divided on the relationship between humor and laughter. Some researchers argue that laughter is a physical manifestation of humor, while others believe the opposite. Laughter is a phonetic activity, while humor is treated as a cognitive concept [ 27 ]. On the contrary, humor is a psychological state characterized by the likelihood to laugh wherein contradictory ideas are held simultaneously [ 11 , 28 ]. Other research supporting the humor and laughter relationship has proved that laughter occurs in all the theories of humor, whether they be the superiority theory, incongruity theory, or the relief theory [ 29 ]. While we recognize that not all instances of laughter are effects of humor and not all humor can elicit laughter, in this paper, which follows research supporting the relationship between humor and laughter, we consider laughter as an indication for humor usage in the learning environment.

1.3. Theories of Instructional Humor

Since humor has several functions aside from being a tool used in education [ 30 ], in this paper, we will refer to humor used in an educational context as instructional humor. In the academic context, the usage of humor in the classroom has proved to have positive effects such as generating attention and arousing curiosity [ 31 ]. Prominent theories in instructional humor include the instructional humor processing theory (IHPT), which explains that the students need to perceive and solve the paradox in humor to ease their learning [ 32 ]. Another related research links the cognitive load theory [ 33 ] to humor application in STEM education [ 34 ]. According to Hu et al. [ 34 ], humor in STEM education should be integrated into the intrinsic cognitive load to be effective. The studies in instructional humor can be divided further into quantitative, qualitative, or individual differences; effects; and theories [ 30 ]. IHPT also states that, for instructional humor to have positive results, students must be motivated and comprehend the lesson content [ 13 , 32 , 35 ]. Application of the IHPT at different levels of education and learning environments has similar effects and conditions. For instance, at the higher-education level, students’ cognitive learning is predicted by the instructor’s related humor [ 36 ]. In online social networks, instructors’ humorous posts improved student engagement and instructor credibility [ 37 , 38 ]. Application of IHPT was also observed in a study in fifth- to tenth-grade students wherein teacher humor had to be related to the course to affect the learning experience positively [ 39 ]. Likewise, for adult learning environments in which students come from different nationalities and cultures, it is said that humor also increases the cultural competence and metalinguistic awareness of the students [ 40 ]. Regarding gender research, past research findings show that humor used by male instructors was more effective than that of female instructors [ 41 ].

1.4. Quantitative Instructional Humor

While research in instructional humor is abundant, there is a significant lack of works focusing on the quantitative aspects of humor [ 30 ]. Examples of research in quantitative instructional humor focus on instructional humor frequencies. For instance, Bryant et al. [ 42 ] found that professors utilized humor once every 15 min during a 50 min class session. In contrast, in similar research, Javidi and Long [ 43 ] found an average of 4.05 humorous messages. Likewise, Downs et al. [ 44 ] found that acclaimed professors used an average of 7.44 instances of humor per class. Other research includes that by Gorham and Christophel [ 13 ], who found only 1.37 humor attempts per class.

On the other hand, more recent research on instructional humor frequency focuses on self-reported measures. Moreover, frequency counts were obtained via survey forms or ratings rather than through actual counts of humor instances [ 45 , 46 , 47 , 48 ]. However, using self-reported measures yields lower rates of instructional humor [ 30 ]. There is a great need for studies that update our understanding of the field. While those studies provide extensive results, the learning environment and educational settings have changed drastically, mainly with the applications of technology such as e-learning [ 30 ]. Therefore, we conducted experiments on the instructional humor frequency in modern educational settings using a naturalistic approach. In addition, we also compared the linguistic features of instructional humor and non-instructional humor to provide educators with insights on how instructional humor differs from humor used for non-educational purposes.

1.5. Education Technology (EdTech) and E-Learning

Education technology (EdTech) is a growing field wherein information communication technology (ICT) supports learning and instruction [ 49 ]. Some examples of EdTech are e-learning [ 50 ], mobile learning [ 51 ], gamification [ 52 ], virtual reality and augmented reality platforms [ 53 ], and virtual meeting platforms [ 54 ]. One of the most discussed educational technology applications is e-learning, which uses the Internet to access educational materials outside traditional classrooms [ 52 ]. Some of the popular instructional content providers in e-learning include online course providers (Udacity [ 55 ], Course Hero [ 56 ], and Coursera [ 57 ]), language-learning website and mobile apps (Duolingo [ 58 ] and Rosetta Stone [ 59 ]), game-based learning platforms (Kahoot! [ 60 ] and Quizlet [ 61 ]), and online conference lecture streaming sites (TED Talks [ 62 ] and Talks at Google [ 63 ]). These providers heavily innovate the education field. Past literature revealed that these EdTech platforms impacted the learning environment with positive effects such as increased student performance [ 64 ], the lessened workload of teachers [ 65 ], and heightened student and teacher engagement [ 66 ]. E-learning gained popularity among educators and students for several reasons, such as functionality and affordability [ 67 ], social interaction [ 67 ], and collaborative learning [ 68 ].

1.6. Natural Language Processing (NLP)

The literature describes natural language processing (NLP) as using computational techniques to analyze texts at single or multiple levels of linguistic analysis, which aims to output human-like language processing in various tasks [ 69 ]. NLP is built on mathematical and linguistic foundations. NLP heavily draws from elementary probability theory and essential information theory in mathematics and parts of speech, morphology, semantics, and pragmatics in the linguistics field [ 70 ]. NLP techniques are applied in humanities, natural sciences, and social sciences research. NLP can help researchers in text data analysis by performing tasks such as assessing subjectivity, linguistic features, and classification. Examples of NLP techniques include sentiment analysis, which can classify text as negative, neutral, or positive [ 71 ]. Other methods include named-entity recognition (NER), wherein important nouns and pronouns are identified in a text [ 72 ], and sentence segmentation, which splits a large chunk of text into sentences [ 73 ].

1.7. Corpora: TED Talks and User-Submitted Jokes

Previous studies in quantitative instructional humor used corpora from lectures conducted in offline traditional classroom settings [ 42 , 43 , 44 ]. However, since we are interested in assessing instructional humor frequency in non-traditional classroom settings, this paper presents and discusses research findings from two studies using two corpora—TED Talks [ 62 ] and user-submitted jokes from stupidstuff.org ( https://stupidstuff.org (accessed on 1 March 2021)) [ 74 ].

TED Talks are recordings of presentations done at TED conferences and related TED events. Talks vary in length, with most being about 20 min in length. Presenters in these talks come from various fields, and discussions range from science, design, technology, entertainment, and business, to global issues. The TED Talks corpus has been used in previous research on several topics such as resources for best practices in teaching [ 75 ], commenting behavior [ 76 ], academic listening exercises [ 77 , 78 ], and speech recognition [ 79 ]. In humor research, TED Talks are more used for building automatic humor recognition algorithms [ 25 , 26 ]. Although research on TED Talks is abundant, few have used it to describe the quantitative features of instructional humor. We chose the TED Talks corpus since the presentations follow a naturalistic setting wherein there were no restrictions on the humor usage of presenters, which is recommended by previous works [ 30 ].

Next, we also used user-submitted jokes from stupidstuff.org to represent non-instructional humor. We define non-instructional humor as sentences containing humor that was not initially intended for use in the classroom. Users can submit and publish their jokes on the website, and other users can also view and rate these jokes. The data on studpidstuff.org consist of jokes and their respective user ratings [ 74 ]. Research on humor using user-submitted jokes is also commonly used in building humor recognition systems and humor generation algorithms [ 80 , 81 ]. Thus, it is acceptable to use user-submitted jokes to represent non-instructional humor.

This paper presents and discusses research findings from two studies using these two corpora. In the first study, we observe the instructional humor frequency and the placement of humor using audience laughter in the TED Talks transcripts as a marker for humor instance. The second study compared the linguistic features of instructional humor and non-instructional humor. Natural language processing (NLP) techniques were applied in both studies, and all data analyses were conducted in the Python programming language.

Previous research has focused on self-reported methods [ 45 , 46 , 47 , 48 ] and offline settings [ 13 , 42 , 43 , 44 ] for calculating instructional humor frequency. This research is unique as we use corpora from online media (TED Talks and user-submitted jokes) and conducted the analysis using NLP techniques to better understand the context in a digital age. Past works have also only looked at traditional classroom settings. However, in this paper, we discuss findings in non-traditional classroom settings to account for the changing learning environment brought about by technological advances. These findings are relevant as they provide new insights into humor usage in online settings and update the literature on instructional humor frequency research. Our results also benefit teachers in incorporating humor in their lectures and engineers involved in NLP projects for humor recognition.

2. Materials and Methods

We conducted two studies in this research using computational linguistics and natural language processing (NLP) techniques. The field of computational linguistics is interdisciplinary and focuses on understanding both written and spoken language from a computational perspective [ 82 ]. Similarly, NLP is used to do text analysis using computerized approaches [ 83 ]. The research in this study was conducted using the programming language Python. For the first study, we focused on instructional humor frequency as found in presentations, while the second study compares the linguistic features of instructional and non-instructional humor. For this study, we set the alpha level for all statistical tests conducted in this research to 0.05.

2.1. Study 1: Instructional Humor Frequency in TED Talks

We used TED Talks as our corpus to examine the instructional humor frequency. Our study limited the talks to those conducted in the English language. We scraped to get the fifty most popular talks and the fifty least popular talks. To ensure that the least popular talks were not affected by the upload count date, we only considered talks that had been published at least one year earlier on the TED website. Likewise, only talks with a duration of 15 to 20 min were included in the dataset to make comparisons possible. We divided our corpus into the most popular and least popular talk datasets since previous works found significant differences in the instructional humor frequency usage based on the educator’s teaching experience and popularity among their students and peers [ 84 ]. Afterward, we extracted the transcript of the talks and created a command-separated values (CSV) file. We used the Pandas library [ 85 ] in Python to conduct the analysis.

While previous works had different methods to measure humor rates and locate humorous messages [ 13 , 42 , 43 , 44 , 45 , 46 , 47 , 48 ], in our study, we decided to locate the humor instances using the special markup "Laughter" found in the TED Talks transcripts. This markup occurs whenever the audience laughs during the presentations. We then examined these laughter occurrences using statistical tests. Furthermore, the placement of humor in the presentation timeline was analyzed. The results are explained in Section 3 .

2.2. Study 2: Comparison of the Linguistic Features of Instructional and Non-Instructional Humor

In this experiment, we used two corpora: TED Talks and user-submitted jokes from stupidstuff.org. For the TED Talks, we extracted the transcripts of 2000 talks. After scraping the website, we created two CSV files for each corpus. We split the talks into sentences using the Stanza module (formerly the Stanford Core NLP) [ 86 ], then labeled sentences containing or immediately followed by the special markup "Laughter" in Python to get the sentences containing humor from the TED Talks transcripts. After data cleaning and processing, we were able to get 8906 humorous sentences from the TED Talks dataset. For stupidstuff.org, the dataset contained transcripts of 3200 user-submitted jokes. Finally, using NLP techniques and descriptive and inferential statistics, we looked at several linguistic features to compare the humorous sentences from the two corpora.

2.2.1. Word Frequency, Bigrams, and Trigrams

Previous works identified that humor has a variety of functions, both positive and negative [ 87 ]. Therefore, it is vital to see whether there is a difference between the choice of words in instructional and non-instructional humor. To get the word frequency or the most frequent words appearing in our corpora and the most popular bigrams and trigrams of humorous sentences, we utilized Python’s open-source NLTK (Natural Language Toolkit) library [ 88 ].

2.2.2. POS (Part of Speech)

Works in computational humor research found that humorous messages use personal nouns and proper nouns, such as when referring to human-related scenarios [ 89 , 90 , 91 ]. To see whether the humorous sentences in the instructional and non-instructional humor dataset follow this theory, we used an open-source library, TextBlob [ 92 ], in Python for POS tagging.

2.2.3. Readability Score

We need to look at the readability score of the humorous sentences since the instructional humor processing theory (IHPT) emphasized that humor needs to be understandable and should not distract from the instructional message [ 32 ]. Likewise, the cognitive load theory (CLT) that focused on humor integrated into science, technology, engineering, and mathematics (STEM) education states that, if humor is not integrated into the lesson content, it will increase the students’ cognitive load and lower learning [ 33 , 34 ].

We used the Flesch reading ease and the Gunning Fog Index to calculate the readability of the humorous sentences. The Flesch reading ease scores range from 0 to 100, with 0 being extremely difficult and 100 being very easy to read [ 93 ]. On the other hand, the Gunning Fog Index rates text from 6 to 17, and each of these scores has an equivalent educational level that determines the text’s difficulty [ 94 ]. For example, a text with a Gunning Fog Index of 6 can be read by sixth-grade students, while a score of 17 can be read by college graduates [ 94 ]. We used the open-source library TextStat [ 95 ] and descriptive and inferential statistics in Python to conduct a readability score analysis.

2.2.4. Sentiment Analysis

Research on the types of humor explains that the sentiment of the humorous messages can negatively or positively affect the listener [ 20 , 96 ]. For instance, humorous sentences having a negative feeling can lower students’ learning performance [ 20 , 96 ]. Thus, it is essential to conduct sentiment analysis on the instructional and non-instructional humorous sentences. We used the open-source library TextBlob [ 92 ] in Python to compute the polarity or sentiment of our dataset’s humorous sentences.

3.1. Study 1: Instructional Humor Frequency

In this study, we first looked at how humor was placed throughout the presentation timeline. Then, we looked at the speakers’ frequency of humor usage and compared the results for popular and unpopular talks using the TED Talks dataset.

3.1.1. Placement of Humor in the Presentation Timeline

The placement of humor in the presentation timeline was calculated using the sentence position when audience laughter occurred and its frequency (see Figure 1 ). We observed a maximum of six occurrences and four occurrences for the same sentence position in the presentation timeline for the most popular and least popular talks, respectively. Although popular talks had more observed laughter instances than less popular talks, the placement of humor in the presentation timeline seemed similar for both datasets. As shown in Figure 1 , humor was more commonly observed during the first part of the talk and gradually lessened toward the end.

The frequency of audience laughter in the top 50 least popular ( left ) and top 50 most popular ( right ) TED Talks. The x -axis shows the presentation timeline or the n th position of the sentence when audience laughter occurred during the presentation. Frequency counts for each sentence’s n th position are shown in the y -axis.

3.1.2. Humor Frequency in Popular and Unpopular TED Talks

Popular talks incorporated humor an average of 12.92 times per 15 to 20 min, while unpopular talks only used humor for an average of 3.92 times. A Welch’s t-test on the humor frequency usage of popular and unpopular talks revealed that the difference was statistically significant between the two datasets ( p < 0.001).

Instructional humor frequency usage in popular talks tended to vary more (M = 12.62, SD = 12.65) than in unpopular talks (M = 3.92, SD = 5.23). For instance, we found that the highest humor frequency for popular talks was 69 times while the lowest was 0 times. Out of 50 talks, 2 had zero humor usage for popular talks. On the other hand, 13 talks showed no humor usage for unpopular talks. Table 1 describes the statistics for the humor frequency in popular and unpopular talks.

Summary of statistics for the humor frequency of TED Talks.

3.2. Study 2: Linguistic Features of Instructional and Non-Instructional Humor

The results from NLP techniques such as calculation of word frequencies, n -grams, POS tagging, readability scores, and sentiment analysis applied to TED Talks and stupidstuff.org corpora are described in this section.

3.2.1. Word Frequencies, Bigrams, and Trigrams

The top 10 most frequently used words for the TED Talks and stupidstuff.org datasets are described in Figure 2 . The most common word for humorous sentences in TED Talks was “like”, and it was “said” for studpidstuff.org. The two datasets share some similarities in word frequencies. For example, the words “one”, “said”, “say”, and “get” both appeared in the top 10 most frequently occurring words for both datasets.

The 10 most frequently occurring words for humorous sentences in TED Talks ( left ) and user-submitted jokes from stupidstuff.org ( right ).

Next, we looked at the most frequently used bigrams or two-word combinations. In Figure 3 , we see that “I’m, going” and “one, day” were the most common bigrams for TED Talks and Stupidstuff.org datasets, respectively. Notably, the bigram “don’t, know” appeared in both datasets.

The 10 most frequently occurring bigrams for humorous sentences in TED Talks ( left ) and user-submitted jokes from stupidstuff.org ( right ).

Finally, the research looked at the most common trigrams (see Figure 4 ). “New, York, City” was the most frequently used trigram in the TED Talks dataset, and it was “take, change, light” for the stupidstuff.org dataset. We found no similarities between the two datasets regarding the most frequently used trigrams.

The 10 most frequently occurring trigrams for humorous sentences in TED Talks ( left ) and user-submitted jokes from stupidstuff.org ( right ).

3.2.2. Part of Speech (POS)

We observed a high usage of possessive ending (POS) and proper nouns in singular form (NNP) for both datasets. This result supports previous works that state that humorous messages tend to use possessive forms and proper nouns [ 88 , 89 , 90 ]. Figure 5 below contains information on the two datasets’ top 10 most frequently used POS. Strikingly, the top 10 commonly used POS were similar for the two datasets. For example, the POS verbs in gerund or present participle form and verbs in the past tense form were frequently observed.

The 10 most frequently occurring POS for humorous sentences in TED Talks ( left ) and jokes from stupidstuff.org ( right ). Legend on POS tag definitions: POS (possessive ending), NNP (proper noun, singular), VBG (verb, gerund/present participle), VBD (verb, past tense), NN (noun, singular), IN (preposition/subordinating conjunction), RB (adverb), CD (cardinal digit), JJ (adjective), NNS (proper noun, plural).

3.2.3. Readability Score

Using the Flesch reading ease score, results showed that humorous sentences for both datasets tended to have scores from 60 to 100, with the peak at 80 and an average of 72–74. The results mean that the humorous sentences range from reasonably difficult to very easy to read. Humorous sentences from the TED Talks dataset (M = 73.89, SD = 17.32) were slightly easier to read and had minor variance compared to those of user-submitted jokes from stupidstuff.org (M = 72.45, SD = 18.88). Table 2 summarizes the results of the analysis. Welch’s t-test on the readability scores using the Flesch reading ease shows that they were statistically significant ( p < 0.001).

Summary of statistics for the readability scores using the Flesch reading ease score.

The number of samples used in this analysis removed outliers or talks that had readability scores out of the range of the scores determined by the Flesch reading ease score (see Figure 6 ). This result is favorable since it is recommended that instructional humor should be easy to understand and not increase the students’ cognitive load [ 32 , 33 , 34 ].

Histogram of readability scores for humorous sentences in TED Talks ( left ) and jokes from stupidstuff.org ( right ) using the Flesch reading ease method.

We also looked at the Gunning Fog Index for both datasets to further assess the readability of the humorous sentences. We removed outliers or talks that had readability scores out of range of the scores determined by the Gunning Fog Index in this analysis. Table 3 describes the results of the statistical tests. A Welch’s t-test on the readability scores using the Gunning Fog Index revealed that the difference was statistically significant ( p < 0.001).

Summary of statistics for readability scores using the Gunning Fog Index.

Figure 7 describes the detailed results of the Gunning Fog Index assessment. The stupidstuff.org dataset (M = 10.19, SD = 2.82) tended to have slightly more variation in scores than the TED Talks dataset (M = 10.55, SD = 2.65). Nevertheless, both datasets returned scores with an average of 10, meaning lower grade levels can comprehend the sentences containing humor.

Histogram of readability scores for humorous sentences in TED Talks ( left ) and jokes from stupidstuff.org ( right ) using the Gunning Fog Index method. The score ranges from 6 (can be understood by sixth-grade students) to 17 (comprehensible to college graduate students).

3.2.4. Sentiment Analysis

Figure 8 shows the histogram of the polarity of humorous sentences for the two datasets. We observed that humorous sentences for both datasets returned a neutral sentiment with an average of 0.07 and 0.06 polarity for TED Talks and stupidstuff.org, respectively. Negative scores imply a negative emotion, while positive scores indicate positive feelings, and near-zero scores usually express a neutral sentiment.

Histogram of sentiment for humorous sentences in TED Talks ( left ) and jokes from stupidstuff.org ( right ).

A Welch’s t -test on the sentiment scores of the two datasets revealed that the difference was statistically significant ( p = 0.01). Table 4 summarizes the results of the statistical tests on the sentiment scores. The observations from the sentiment analysis of sentences containing humor in TED Talks (M = 0.07, SD = 0.27) and user-submitted jokes from stupidstuff.org (M = 0.06, SD = 0.21) showed that the two were very similar.

Summary of statistics for sentiment analysis.

4. Discussion

This study aimed to investigate the quantitative and linguistic features of instructional humor in the EdTech learning environment and provide an update to quantitative instructional humor research by using techniques in NLP and statistical tests on presentations from TED Talks and user-submitted jokes on stupidstuff.org.

The results from our investigation on the frequency of instructional humor in TED Talks showed that educators’ incorporation of humor in their lectures had increased significantly (M = 12.62) as compared to the results from previous research dating back to three decades ago [ 24 , 25 , 26 ]. While earlier research findings in traditional classroom settings found humor usage once every 15 min [ 42 ], our research findings revealed humor usage once every 1.58 min. These findings show that humor usage by educators is significantly more frequent in non-traditional classroom settings. Learning environments in TED Talks incorporate more humor than traditional classroom settings do. The popularity of TED Talks among online viewers also follows past research that higher frequencies of humor applications receive higher satisfaction and popularity ratings from students [ 84 ]. Therefore, whether in traditional or non-traditional classroom settings, high frequencies of humor are associated with high popularity among audiences. Although, in this study, we were not able to identify what might cause the high humor frequency, this increase in humor usage in EdTech learning environments can be another point of interest for researchers to study.

On another note, we observed that the humor usage of more popular educators (M = 12.62) was significantly higher than that of less popular educators (M = 3.92). This observation is in line with previous research, wherein there is a correlation between the amount of usage of humor and the popularity, credibility, and experience of teachers [ 83 ]. Likewise, our results show that the humor used in TED Talks followed humor’s positive effects as stated in the instructional humor processing theory (IHPT) [ 32 ] since audiences express their positive experience through frequent laughter. This study confirms that despite changes in the learning environment such as those brought about by technology, the theory that more experienced, popular, and credible teachers tend to use humor more in their lectures [ 30 ] still holds.

Another important factor is that we looked at the placement of humor in the presentation timeline. This study is novel as far as we know, as previous studies only focused on the frequency of humor usage and not where humor was placed during the lectures. The results showed that humor is more commonly set during the start or first parts of the presentation rather than in the middle or end. These findings support the findings of previous works in which humor was used to seek attention [ 31 ]. Likewise, these results can also be linked to prior theories in instructional humor, wherein placing humor at the start of the lectures is more beneficial to student learning. To give a specific example, as suggested by Sweller in the cognitive load theory (CLT), humor should not add additional cognitive load for it to have positive effects [ 33 ]. Since the start of the lectures has the lowest cognitive load, it is understandable that the educators prefer early placement of humor in their classes. Similarly, as stated in the IHPT, since students’ cognitive load is lesser, students can quickly solve the contradiction of the humor, hence, increasing the positive effects of humor in their academic performance [ 32 ].

On the other hand, our comparison between the linguistic features of instructional humor and non-instructional humor provided several intriguing results. First, in terms of word frequencies, there seemed to be no difference between the word usage of humor used for education and that used for other purposes. Our findings support past studies wherein some words were more commonly used in humorous contexts, such as when referencing human-related scenarios [ 89 ]. In our results, words corresponding to human-centered scenarios such as “I’m”, “people”, and “man” are common in non-instructional and instructional humor cases. Similarly, previous studies suggest humorous sentences use more possessive pronouns and nouns [ 88 , 89 , 90 ]. The results from POS tagging are also in line with past literature since POS (possessive ending) and NNP (proper noun, singular) were the most frequently occurring POS for instructional and non-instructional humorous sentences. Our research further supports describing humorous sentences as human-centric and focusing on personal opinions, as observed in previous studies [ 88 , 89 , 90 ]. These results are also beneficial to the field of using NLP for building systems for humor recognition for machines since we can devise algorithms that take the linguistic components of sentences as features to recognize the humor in sentences automatically. Furthermore, the methodology of past works using words and pronouns as features for automatic humor recognition in machines [ 25 , 26 , 80 , 81 ] is also supported and validated through our research results.

When we looked at the readability scores of instructional and non-instructional humor sentences, we observed a slight difference in scores when using different methods for calculating the readability. Using the Flesch–Kincaid reading ease method [ 93 ], instructional humor had a higher mean, making it easier to understand. However, when using the Gunning Fog Index [ 94 ], non-instructional humor was much easier to comprehend. In this research, we only used two methods for calculating the readability score. Therefore, other researchers might apply other readability scoring methods to obtain comparable results.

Lastly, we expected to see more negative sentiment for non-instructional humor and more positive emotion for instructional humor in terms of sentiment analysis. However, both returned scores leaned more toward a neutral view. We expected a more positive sentiment for instructional humor since previous research suggests that humor used for education should contain positivity rather than negative feelings to create a positive learning environment [ 30 ]. Likewise, we expected to see more negative sentiment in non-instructional humor since past humor research studies showed that humor uses negative words, adult slang, and swear words [ 89 ]. Perhaps, the methodology we used for calculating the sentiment of the humorous sentences might not have been accurate enough, leading to neutrality. Since humor contains incongruity and ambiguity [ 17 , 32 ], the algorithm we used might not have detected the sentiment correctly. In future research, better algorithms and methods for sentiment analysis are recommended for getting more accurate results.

4.1. Implications of the Current Study

The current study provides teachers with insight into incorporating humor in their lectures in terms of quantitative features. The results also give a positive light to engineers in humor recognition research as even humor for different purposes tends to have the same linguistic features. The findings also update instructional humor frequency research, where the prominent studies date from 30 years ago with insights on the case of instructional humor frequency in non-traditional classroom settings.

4.2. Limitations and Future Research

The current study has several limitations. First, although we were able to find the results to be statistically significant, the number of samples we used to compute the instructional humor frequency and placement of humor in the presentation timeline was limited to a small sample size for both popular (N = 50) and unpopular talks (N = 50). Second, the method for calculating humor rates was highly dependent on the transcriptions of the presentations using the special markup “Laughter” since we did not conduct any test to validate whether “Laughter” instances accounted for humor usage. For instance, the audience’s laughter might have been caused by other factors and not by the actual humor usage of the presenter, but in this study, we assumed that the presenter incorporated humor when the audience in the talk laughed. Third, the presentation timeline in this study was expressed with the sentence position rather than the actual time. The reason for this was that the presentations had varying lengths (15 to 20 min), and we thought that occurrences would be better described by sentence position rather than time to help teachers plan their lectures using the sentence position or word count rather than time. In our future research, we will take time as the x -axis for the presentation timeline and study longer presentations (more than 20 min) and see whether there are any changes in the results. Fourth, the algorithms in this study for studying the linguistic features such as word frequencies, POS tagging, readability scores, and sentiment analysis of instructional and non-instructional humor were limited to those we used. In future research, we can try incorporating different algorithms for each analysis and see if there is a difference in the results. Fifth, we were only able to assess the effects of humor frequency regarding the popularity of the TED Talks and not determine the audience’s comprehension. In future studies, it would be interesting to see if high frequencies of humor in TED Talks positively or negatively affect the audiences’ learning performance. Sixth, we used non-traditional classroom settings in this study to observe instructional humor frequencies since we wanted to see how humor is used in the digital context. However, further research using offline lectures can also be of potential interest for more direct comparisons on how humor frequencies changed in traditional classroom settings. Lastly, we only used Welch’s t-test for statistical tests to confirm whether our dataset’s difference was significant. We can use more statistical tests to improve our findings in future research.

5. Conclusions

Overall, this study updates instructional humor frequency research by assessing humor usage in non-traditional classroom settings. We observed that educators’ use of humor increased significantly compared to research results 30 years ago. Humor usage in non-traditional classroom settings such as TED Talks is more frequent. On the other hand, this research finding supports previous theories stating that the more experienced, credible, and popular educators are, the higher their usage of humor. Whether in learning environments supported by technology or not, humor positively affects the presentations’ popularity. In addition, results showed that humor is more commonly placed at the start of presentations. This study also showed comparisons on the linguistic features between humor used for educational purposes and those used for other purposes. Whether for instructional or non-instructional purposes, humorous sentences follow similar linguistic features. Our findings are beneficial to teachers in incorporating humor in their lectures and to engineers involved in NLP projects for humor recognition.

Author Contributions

Conceptualization, V.P.S. and T.Y.; methodology, V.P.S.; software, V.P.S.; validation, V.P.S.; formal analysis, V.P.S.; investigation, V.P.S.; resources, V.P.S.; data curation, V.P.S.; writing—original draft preparation, V.P.S.; writing—review and editing, V.P.S.; visualization, V.P.S.; supervision, T.Y.; project administration, V.P.S. All authors have read and agreed to the published version of the manuscript.

This research received no external funding.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Data availability statement, conflicts of interest.

The authors declare no conflict of interest.

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.