"A pirate goes nee-nor-nee-nor!" humor with siblings in middle childhood: A window to social understanding?

Humor Production in Childhood

According to McGhee (1979 , 2002 ), the development of children’s humor can be defined in stages in their emerging ability to recognize and produce incongruities. By the end of the first year of life, most infants intentionally produce humorous acts that are responsive to the behavior and emotional reactions of others, for example, clowning (e.g., producing odd movements, facial expressions, actions, and vocalizations) and teasing (e.g., pointing to an object and then refusing it) to elicit laughter from their caregivers ( Reddy, 2001 ; Reddy & Mireault, 2015 ). Such humorous actions are often discovered by accident; following an unintentional incongruity that made someone laugh, infants repeat the act deliberately to elicit the same reaction ( Mireault & Reddy, 2016 ).

From the second year, children begin to produce novel humorous acts, including incongruous language, such as chanting and creating nonsense words (e.g., “goo-joo-boo-joo”; Hoicka & Akhtar, 2012 ; Loizou, 2005 ). By middle childhood (around age 7), children produce more complex conceptual incongruities and word play, such as comedic hyperbole, puns, and riddles (e.g., “Whaddya call a test tube with a college degree? A graduated cylinder!”; Bergen, 2002 , 2012 ; Paine, Hashmi, et al., 2019 ; Varga, 2000 ). Evidence demonstrates that children’s repertoire of humor expands across development ( Bergen, 2002 ); incongruous actions (e.g., piling toy animals on a toy farm roof), physical play (e.g., silly posing, dancing, and facial expressions), sound play (e.g., shouting in a high-pitched voice, “Ew! I’ve been slimed!”), playful teasing and banter (e.g., “Ya dummy!”), and breaking taboos (e.g., “H, I, J, K, L, M, N, O… PEE? Get it? Pee!”) are all common forms of humor produced by 5-to-7-year-olds ( Paine, Howe, et al., 2019 ; Paine et al., 2021 ).

Connections Between Humor, Play, and Social Understanding

The ability to create humor is thought to be closely linked with children’s developing ability to understand the minds of others ( Bariaud, 1989 ; Bergen, 2002 ; Dunn, 1988 ; Leekam, 1991 ; Martucci, 2016 ). Children demonstrate an understanding of the social world through shared humor in their early interactions, long before they have developed a fully-fledged theory of mind ( Addyman & Addyman, 2013 ; Airenti, 2016 ; Mireault et al., 2011 , 2014 ; Reddy & Mireault, 2015 ; Sroufe & Wunsch, 1972 ). There is considerable evidence that awareness of others’ intentions, expectations, and beliefs emerges as early as 7 months of age ( Hamlin et al., 2007 ; Kovács et al., 2010 ). Clowning, teasing, and other “communicative games” in infancy and early childhood may require understanding of the minds of others (as argued by Mireault & Reddy, 2016 ), even if the incongruous acts themselves are not understood in a deep way ( Airenti, 2016 ; Bergen, 1998 ; Bergen, 2019 ). In line with social interactionist theories of development, humor during play may, therefore, act as a “melting pot,” providing a context for rehearsing existing social understanding skills and in turn, drive development and new competencies ( Gibson et al., 2020 ; Lillard et al., 2013 ).

Theorists have long hypothesized a close partnership between humor and play ( Bariaud, 1989 ; Bergen, 2012 ; Bergen, 2019 ; Wolfenstein, 1954 ). Both are enjoyable, internally motivated, and involve distortions of reality ( Bergen, 2012 ). Some researchers have identified the importance of humor in the personality construct of playfulness. Indeed, according to Barnett (2007) playfulness is defined as “… the predisposition to frame (or reframe) a situation in such a way as to provide oneself (and possibly others) with amusement, humor, and/or entertainment” (p. 955). Recent evidence demonstrates that 5-to-7-year-olds’ self-perceptions of their tendencies to tell jokes and funny stories, do silly things, and make others laugh contributes to their perceptions of their own playfulness ( Fink et al., 2020 ). In particular, childhood humor and pretend play may be underpinned by similar developmental processes, where both playful activities are argued to involve symbolic thinking ( McGhee, 1979 ; Rubin et al., 1983 ; Tower & Singer, 1980 ). Sharing humor requires the “performer” to understand and convey to their “audience” that their playful, nonliteral acts are of a “pretend,” and not an “incorrect” nature ( Bateson, 1955 ; Bergen, 2012 ; Leekam, 1991 ). But unlike some pretend play, children’s humor involves the creation of alternative or distorted realities through silliness or nonsense, with the goal of triggering amusement (i.e., the distinction between “joking pretend” and “serious pretend”; Bariaud, 1989 ; Wolfenstein, 1954 ).

Although evidence is mixed (for a review, see Lillard, 2013 ), many researchers have proposed that pretend play involves an understanding of one’s own and others’ mental states ( Weisberg, 2015 ). Leslie (1994 , 1987 ) proposed that in pretend play, a child mentally projects a nonliteral reality onto a situation, all the while knowing what the real situation is. For example, when a mental representation of a telephone is projected on a banana, a child simultaneously appreciates two contradictory models of reality ( Lillard, 1993 ). The same mental capacity, it is thought, is necessary for understanding when someone holds a belief that differs from one’s own (i.e., holds a false belief, Leslie, 1994 , 1987 ). For example, Harris (2000) proposed that pretend play involves social understanding skills, insofar that in pretend play, children simulate mental states they do not actually experience and states of the world that are not reality.

Sociological communications perspectives suggest that both shared pretend play and humor depend on children’s ability to understand that their partners share their intent to be playful ( Bariaud, 1989 ; Bergen, 2019 ; Semrud-Clikeman & Glass, 2010 ), as well as their ability to communicate ludic cues that “this is play” (i.e., smiling, laughter, sound effects, and exaggerated actions), which usually occur in the play frame ( Bateson, 1955 ; Bergen, 2019 ). In common with pretend play, producing humor successfully may depend on a child’s social understanding. Without a frame of mutual playful intent and complicity, incongruities might confuse or even threaten children’s play partners ( Semrud-Clikeman & Glass, 2010 ).

The theoretical and conceptual links between humor, pretend play, and social understanding are clear. However, most contemporary developmental research on humor has focused on describing children’s humor in play, rather than investigating associations between humor and other domains of child development ( Bergen, 2019 ). Is a child’s tendency to spontaneously produce humor associated with their tendency to engage in pretense? We examined this theoretical partnership by investigating children’s spontaneous humor production during social play with their use of objects in solo play with toys, distinguishing between (a) handling and setting up props; (b) transforming objects in expected ways (e.g., animating a horse to gallop); and (c) transforming objects in creative ways (e.g., animating a cow to fly; Howe et al., 2014 ). Insofar as transforming objects in pretend play requires symbolic, creative thinking ( Piaget, 1962 )—thought also to be required for humor production ( Bariaud, 1989 ; Wolfenstein, 1954 )—we were interested in associations between children’s tendency to produce humor and their creative object transformations during play.

In our previous work, we found that 5-year-old children’s spontaneous humor production was positively associated with their tendencies to talk about thoughts and knowledge (i.e., cognitive states) during play ( Paine et al., 2021 ). We now aim to expand knowledge about the association between children’s humor production during social play and

Sibling Interaction as a Context for Humor Production

Siblings share a long, co-constructed history ( Dunn, 2007 ; McHale et al., 2012 ) that makes their relationship a particularly important context for the development of social understanding ( Howe & Recchia, 2014 ; Howe et al., 2022 ) and humor ( Dunn, 1988 ; Paine, Howe, et al., 2019 ; Paine et al., 2021 ). In our previous work, we have observed that children’s humor produced with siblings often appears to be well-rehearsed, ritualistic, and based on the siblings’ intimate knowledge of, and experiences with, one another ( Paine, Howe, et al., 2019 ; Paine et al., 2021 ). Siblings who share day-to-day family life are highly familiar with one another, which can result in emotionally intense and uninhibited interactions ( Howe & Recchia, 2014 ). With siblings, children often engage in playful, mischievous behavior ( Paine, Howe, et al., 2019 ; Paine et al., 2021 ), such as taboo-themed humor and playful teasing (e.g., “Where’s your bum brain?”), as well as word and sound manipulations (e.g., chanting, “Chocolatina, chocolatina!”) and incongruous actions (e.g., playfully putting their foot on the sibling’s head). Given that sibling interactions are a rich context for humor production, we have now investigated children’s spontaneous production of humor during dress-up play with a younger sibling.

The Present Study: Aims and Hypotheses

A child’s developing ability to perceive, interpret, and understand the minds of others is a vital skill in being able to navigate the social world ( Carpendale & Lewis, 2015 ). Although the ability to conceive and express humor is likely to be associated with a sophisticated understanding of minds ( Bergen, 2002 ; Dunn, 1988 ; Leekam, 1991 ), few studies have tested this association ( Bergen, 2019 ; Paine et al., 2021 ). We sought to contribute to developmental theory that suggests childhood humor and pretend play may be underpinned by similar developmental processes ( Bergen, 2019 ) and that a child’s production of humor may reflect their social–cognitive development ( Bariaud, 1989 ; Bergen, 2002 ; Dunn, 1988 ; Leekam, 1991 ). Therefore, we drew on a corpus of observational data of interactions between siblings during play from a larger study of the development of first-born children ( Hay et al., 2021 ). As part of a home visit at age 7 years, we conducted detailed observational coding of the focal children’s humor production during dressing-up play with younger siblings. Focal children also completed a battery of social understanding tasks to assess emotion and second order belief understanding. This battery of tasks included a solo free play task with Playmobil toys, which was coded for children’s engagement in pretend play and tendency to talk about the cognitive states of others.

Our first overarching aim was to describe the humor that was spontaneously produced during play with a sibling. Previous evidence suggests that humor shared with siblings is influenced by sibling constellation factors (i.e., birth order and gender composition). In early childhood, first-born children tend to produce more types of humor with their siblings—including playful banter and teasing and action and object incongruities—than do second-born children of the same age ( Paine et al., 2021 ). In middle childhood, pairs of brothers produce more humor overall than pairs of sisters ( Paine, Howe, et al., 2019 ). Similarly, children’s social understanding skills are influenced by sibling constellation factors (e.g., age gap; Paine et al., 2018 ). The current analyses focus on the community sample of first-born 7-year-olds, which precluded analysis for the effects of birth order that would not be confounded with age. However, we did examine the 7-year-olds’ humor production in relation to the age gap between siblings and the gender of the second-born child. Given that closer-in-age children tend to have fewer developmental asymmetries and more mutual interests ( Hughes, 2011 ), we hypothesized that siblings with a smaller age gap would produce more humor. In line with previous work ( Paine, Howe, et al., 2019 ), we hypothesized that play between two brothers would produce more humor than other gender combinations.

The second overarching aim was to investigate associations between the focal children’s humor production during sibling play and other skills, in particular their engagement in pretend play and performance on social–cognitive tasks. Given the close theoretical partnership between humor and pretend play ( Bergen, 2019 ), we expected that children who play with objects in creative ways would also produce more humor. We also expected that humor production would be positively associated with social understanding skills, as indicated by children’s emotion understanding, tendency to talk about others’ cognitive states, and their understanding of second-order beliefs ( Bergen, 2019 ; Dunn, 1988 ; Leekam, 1991 ; Martucci, 2016 ; Paine et al., 2021 ). Given that any significant associations with social understanding could be explained by known covariates (i.e., child age, language ability, working memory, family demographics; Cole & Mitchell, 2000 ; Lecce et al., 2017 ; Paine et al., 2018 ), we controlled for these covariates.

Procedure

Families were visited in the home by research assistants for two two-hour sessions ( M age 6.91 years at session 1; 6.98 years at session 2). A questionnaire battery was provided during the visit to primary caregivers and, where possible, fathers. While the primary caregiver completed an interview with one trained research assistant, the child completed various cognitive, social, and emotional assessments in a quiet space with a second trained research assistant. At the end of the second session, children completed the dress-up task with younger siblings who were at home and consented to join the play. The child and caregiver then participated in a parent–child interaction task and some family games. When required, a third research assistant accompanied the research team to keep younger siblings from interfering with the child testing and parent–child interaction tasks. A remuneration of £20 was given to the caregiver and a book voucher of £10 to the child at the end of the home visits.

Children’s Receptive Vocabulary

Children’s vocabulary knowledge was assessed using the British Picture Vocabulary Scale (BPVS; Dunn et al., 1982 ). In this task, the experimenter spoke a word to the child, who was asked to point or say the number of the picture that corresponded to the word. Each child’s receptive vocabulary score was computed by age normalizing the data to produce a standardized score. All children in the subsample completed this assessment.

Children’s Working Memory

Children’s working memory was assessed using a task from the Amsterdam Neuropsychological Tasks (ANT), ( de Sonneville, 1999 ). The visuo-spatial sequencing (VSS) task measured children’s visuospatial working memory. Children were presented with a gray square containing nine circles symmetrically positioned in a 3 ×3 matrix on a computer screen. After a beep, a sequence of circles was pointed at by a computer animated hand, and after the sequence children took control of the mouse to replicate the sequence of circles. The test consisted of 24 trials and gradually increased in difficulty in the number of targets and complexity of the sequence. Working memory was assessed using the total number of correct targets in the correct order, with a total of 100 possible correct targets. Data were available for 105/110 (95.5%) families in the subsample.

Children’s Second-Order False Belief Understanding

Children were told a second-order false belief story to assess their higher-order understanding of belief ( Paine et al., 2018 ; Paine, Hashmi, et al., 2019 ). This age-appropriate task was adapted from the second-order belief paradigms used by Coull et al. (2006) and Perner and Wimmer (1985) . Children were told a story by the tester, enacted with Playmobil figures. In the story, the protagonist (Nick) shows his special teddy to the child and tucks the teddy inside the bed. The mother comes into the room and asks Nick to brush his teeth; they leave the room. In Nick’s absence, the sibling (Alex) removes the teddy from the duvet and hides the teddy in the cupboard. Unbeknownst to Alex, Nick returns and watches Alex hiding the teddy, before leaving the room again. When Nick comes back into the room he says, “I want my teddy.” Children had to answer the belief (“Where does Alex think Nick will look for the teddy?”), justification (“Why does Alex think Nick will look for the teddy in the [child’s answer]?”), and comprehension questions (a. “Does Nick know that the teddy is in the cupboard?”; b. “Does Alex know that Nick saw her hide the teddy?”; c. “Where will Nick look for the teddy?”) correctly to be classified as passing second-order false belief. Data were available for 108/110 (98.2%) children. Excellent reliability was established for passing this task (κ = 1.00, see Paine et al., 2018 ).