Printer Friendly

Sharing humour and laughter in autism and Down's syndrome.

Humour and laughter are fundamentally relational phenomena and can tell us a lot about the interpersonal, affective, socio-cognitive and cultural skills of young children. Despite their potential as developmental and diagnostic tools, however, we know little about their everyday occurrence in very early development and even less in atypical development. Humour and laughter in children with autism are of particular relevance because their difficulties centre specifically around relating to other people (Hobson, 1989), sharing cultural conventions (Lord, 1993; Loveland, 1991) and understanding others' emotional, attentional and intentional states (Baron-Cohen, 1989; Hobson, 1986a, b; Leslie, 1987; Mundy, 1995), and because the debate surrounding the nature of the primary deficit in autism is still unresolved. The present paper opens up a body of everyday phenomena for investigation in terms of interpersonal understanding and contributes to the debate about social understanding in autism with naturalistic data f rom children with autism and children with Down's syndrome.

Psychological theories of humour development in recent years have focused mainly on individual cognition, i.e. on the individual's processing of incongruity and its resolution, rather than on social, emotional or cultural processes. However, in older writings, humour and laughter are acknowledged as relational and social phenomena (Bergson, 1911; Freud, 1905; Johnson, 1978; Koestler, 1964; Lowenfield, 1935; Luquet, 1930; Wolfenstein, 1954/1978). A joke may be a joke when even one person laughs at it (Crawford, 1999), but one person's laughter is needed to make a joke funny. Whether humour and laughter could ever develop if they were never shared with another human being may be one of those questions that can only be indirectly answered. That humour exists between people rather than somehow in the joke itself, is also evident from the forms and topics of humour which do not travel well across cultures and contexts. Sharing humour may show a more complex acculturation than sharing tragedy, requiring subtle impl icit reference to socio-cultural rules rather than their explicit acknowledgement (Eco, 1986). Sharing laughter with others reveals both cultural and emotional attunement with them, namely a mutuality of interest in the topic of the laughter and/or an interest in the laughter of the others as an affective state in its own right. It is well known that the presence of others facilitates laughter to humorous events m adults and in children (Chapman, 1976; Glenn, 1989) and that laughter itself can be contagious (Provine, 1992). Although no clear evidence exists regarding the onset of contagious or socially facilitated laughter in infancy, it may begin from the middle of the first year (Reddy, personal observation).

The primary focus on cognitive processes underlying humour has led to a dearth of information about humour early in development. The cognitive skills for the humorous appreciation of incongruity (absurdity or the juxtaposition of different frames of reference) are believed to develop after 18 months (McGhee, 1979), although primitive precursors to humour can be seen earlier in laughter in response to tickling, peekaboo and chasing (Shultz, 1976). Potentially humorous behaviour in infants younger than 18 months has consequently been rarely explored, and study of the production and use of humour in actual engagement has been neglected in favour of studies of responses to controlled humorous 'stimuli'.

Early humorous exchanges in typical development

Humorous exchanges are central to social interactions with infants as young as 3 months of age (Ambrose, 1963; Darwin, 1872; Washburn, 1929), and during the first year, developments in infant laughter at humorous events reveal an increase in non-reflexive laughter, an increasing appreciation of culturally inappropriate acts, and a grasp of playful intentions in others. In the middle of the first year, infants laugh mainly at events involving physical or intense sensory contact--especially tactile stimulation such as tickling--but by 11 or 12 months of age, infants laugh more at distal events such as funny faces and sounds, or events involving socially inappropriate or incongruous acts such as mother waddling like a duck or sucking from the baby bottle (Sroufe & Wunsch, 1972). Laughter at increasingly more complex events may result from a tendency to respond most positively to challenges that are just at the edge of their capacities (Sroufe & Wunsch, 1972; Zigler, Levine, & Gould, 1966), and newly discovered patterns and norms are often violated and distorted in fun and exploration (Garvey, 1977; Groos, 1901/1976). Infant laughter at the violation of social expectancies (Sroufe & Wunsch, 1972) reveals the infant's grasp of the social conventions governing the use of objects and the performance of actions, and infant laughter at others' teasing and joking shows an understanding of playful versus serious intentions in others (Nakano & Kanaya, 1993; Trevarthen & Hubley, 1978).

From around 10 months of age, infants begin to be active humorists as well as appreciative respondents, laughing before rather than after the termination of humorous events (Sroufe & Wunsch, 1972), initiating key aspects of familiar playful sequences themselves (Bruner & Sherwood, 1976) and initiating novel games and humorous interactions as well as routine ones (Bates, Benigni, Bretherton, Camaioni, & Volterra, 1979; Reddy, 1991; Trevarthen & Hubley, 1978). Mischief begins at this age, specifically teasing and clowning, aimed at provoking laughter and other emotional reactions in others (Reddy, 1991). Clowning may involve the ritual violation of the sacred (Handelman, 1981), and the ability to accept intimacy (Lowenfield, 1935), but at its simplest level involves sensitivity to others' amusement and the repetition of, often exaggerated, acts that have previously elicited laughs. Teasing others requires some understanding of others' emotional attitudes, expectations and intentions (Dunn, 1988; Leekam, 1991; N akano & Kanaya, 1993; Reddy, 1991; Stern, 1985) and the provocative violation of social understandings, conventions and agreements (Alford, 1983; Miller, 1986; Pawluk, 1989; Reddy, 1991). Clowning and teasing require two corresponding 'skills' in both the comic and their 'audience': one, an interest in others' emotional reactions and a desire to elicit them (or let them be elicited); and two, the ability to perceive (and respond to) causal links between one's actions and others' reactions. Teasing, in addition, requires a shared knowledge of social meanings, conventions and agreements that may then be playfully violated. Both clowning and teasing may be precursors of joint attentional abilities (Bakeman & Adamson, 1984; Mundy, Sigman, & Kasari, 1993) as they involve the infant's active directing of others' attention to acts by the self before infants are able to direct attention to external targets (Reddy, Hay, Murray, & Trevarthen, 1997; Reddy, 1998, 1999).

Evidence of humour and laughter in autism

Evidence about humour in autism is sparse. Older writings and some anecdotal reports suggest that 'seeing the funny side of things' is a problem in autism, as is laughing in moderation or in balance with others (Asperger, 1952/1991; Lord, 1993). In typical development, humour is related to behaviours less evident in autism; playfulness, curiosity and disobedience (Barnett, 1991). Children with autism have deficits in 'showing-off' or drawing attention to themselves or to their acts (Dahlgren & Gillberg, 1989; Wetherby & Prutting, 1984), although it is unclear whether this extends to clowning. Although older autistic children are reported to tease, the nature of this teasing is unclear (Kanner, 1943; Lord, 1993), and while children with autism and Asperger's syndrome often get teased by peers, they are reported to have particular difficulty in responding to teasing with understanding, thus perhaps increasing the likelihood of being teased (Asperger, 1952/1991; Wing, 1991).

The very few recent studies on this topic suggest that children and adults with autism may not have a problem with humour per se, but only with humour involving complex cognitive skills. High functioning adults with autism produce a variety of forms of humour, including verbal and conceptual incongruities, multiple meanings and jocular acts of pretence with the functions of objects, and only show deficits in humour with high cognitive demands (Van Bourgondien & Mesibov, 1987). Parents of school-age children with autism do, however, report some problems in humour-related interactions, particularly in relation to clowning rather than to teasing (Briscall, 1995). In the only available exploration of humour and laughter in young school-age children with autism, St James and Tager-Flusberg (1994) suggest that the simpler forms of humorous interaction (which typically occur in the first year of infancy) may be unaffected. They observed parent-child interactions in the home for six children with autism and six child ren with Down's syndrome (DS) individually matched on language level with ages ranging from 3 to 7 years. Although there was a large variability within groups, they found more humour episodes overall in the DS group, but when total humour was a covariate, there was a significant difference between groups in only one category of humorous response--non-verbal incongruity (i.e. humorous inappropriate acts with objects)--with the children with DS producing (or responding to) relatively higher frequencies of such acts than the children with autism. In all other categories--tickling, familiar routines, silly/slapstick, funny sounds, teasing, verbal incongruity and riddles -- there were no significant group differences, although jokes and riddles were completely absent in the autism group. The authors suggest that it may be only the cognitively more complex forms of humour that are affected in autism (incongruity and above), since the developmentally simpler, more affective, forms of humour such as tickling, familia r routines and slapstick show no apparent deficit in autism.

However, we cannot yet conclude that children with autism are 'above' the 12-month level of typically developing infants in terms of humour and laughter, or that there are no simple affective and relational problems in humorous exchanges in autism. This study involved very small and heterogeneous groups of school-age children, and the findings may not generalize to a larger sample or to pre-school children. Only episodes where laughter actually occurred were analysed, and there may have been critical failures to laugh or unsuccessful attempts by others to elicit laughter. In more private contexts there may be other differences in the elicitors of laughter between the two groups of children that can be reported by the family. Furthermore, we do not yet know anything about children's interest in and responses to others' laughter or their own attempts to make others laugh, all of which are necessary before we can determine whether the affective aspects of humour and laughter are disrupted in children with autism .

Debates about the primary deficit in autism

Research on autism shows that deficits in young children with autism involve several aspects of social relations: affective engagement with others, the understanding of others' intentions, affective and attentional states and joint attention deficits. All of these skills are evident in the humorous interactions of typical infants before the end of their first year. However, there is considerable disagreement about the age of onset of the disorder and about the nature of the primary deficit. The mentality-specific conceptual deficit view (Baron-Cohen, 1989; Leslie, 1987; Perner, 1991) argues that the deficit appears at 18 months, following the failure of a critical cognitive representational mechanism. Affective disturbance views, however (Hobson, 1989; Trevarthen, Aitken, Papoudi, & Robarts, 1996), suggest an onset earlier in infancy, whose basis lies in disturbances in the ability to engage emotionally with others. Simple humorous exchanges involving emotional engagement may provide evidence to illuminate a part of the debate about autism. St James and Tager-Flusberg (1994), with their findings about the lack of group differences in simple humour, suggest support for a cognitive deficit view and conclude that the claim for an affective and interpersonal basis for the constellation of deficits in autism (Hobson, 1989) is weakened. However, even inappropriate acts with objects may require an affective and social basis. They need, for instance, an interest in what other people think of as appropriate actions (Williams, Costall, & Reddy, 1999; Williams, Reddy, & Costall 2001). More evidence is needed about other affective aspects of humour before we can resolve this issue.

Evidence about humour and laughter in Down's Syndrome

Exploration of the production of humour in children with DS is pertinent given anecdotal reports of their particular interest in others' emotional reactions and their own tendencies to clown around. In infants with DS, developmental changes in response to different laughter-eliciting events reportedly follow the same pattern as in typically developing infants, with physically intrusive events eliciting laughter earlier than distal events (Cicchetti & Sroufe, 1976). While the frequency of laughter is greater in DS than in 'other' learning disabilities (Norris, 1971), it may occur less than in typical development--infants with DS frequently smile where typical infants would laugh (Cicchetti & Sroufe, 1976). We do not know, however, whether they seek to ellcit laughter from others in the forms reported in typically developing infants and, if so, how this relates to their developmental age.

Comparing different aspects of humorous interactions in children with autism and children with DS with patterns reported in typical development may help us to identify different developmental paths in affective socio-cognitive development as well as to better understand specific affective and socio-cognitive problems in autism. The present study has two aims: (1) to provide a description of the contexts of humour and laughter and of responses to laughter in the everyday lives of children with autism and DS and (2) to examine the implications of these findings for current debates about autism. Of particular interest is the tentative finding in the literature that relating to others through humour in simple emotional ways may be unaffected in autism. Interactions that require affective attunement, shared cultural knowledge and fairly simple pre-symbolic and even pre-joint attentional skills are thus explored through interviews with parents and through videotaped observations of interaction, to see whether they are problematic in young children with autism compared with children with DS.



The participants in this study were 19 children with autism and 16 children with DS. Parents were sent letters and information about the study through a professional contact organization, inviting enquiries from those interested in participating. Further details were then given to those who enquired. Parents were assured of anonymity and of their right to withdraw from the study at any time. The parents of children with autism were contacted through professional diagnostic clinics and organizations associated with the National Autistic Society in the South of England, and through independent playgroups and nursery schools catering for children with special needs. All the children with autism displayed the triad of impairments (Wing & Gould, 1979) required for a diagnosis of autism, and all had received a primary diagnosis of childhood autism according to DSMIV or LCD 10, given by clinical psychologists or psychiatrists. One of the families came from a Middle-Eastern ethnic and cultural background. The remaini ng families were all white Caucasian and British. All families were of middle or low SES. The parents of the children with DS were contacted with the help of the Portsmouth Down's Syndrome Trust based at the University of Portsmouth and through independent playgroups and nursery schools. One of the families was of Afro-Caribbean origin, and the others were white Caucasian. All families were culturally British and were of middle or low SF5.

Matching and demographic details

Table 1 shows the characteristics of the children in the two groups. The mental scale of the Bayley Scales of Infant Development (BSID-Il) was used to match the groups. This test consists of items tapping non-verbal cognitive skills as well as verbal ability. A combined score gives a developmental age, but the score can be separated into a cognitive facet, giving non-verbal mental age in months, and a language facet, giving verbal mental age in months. The two groups were matched on the BSID measure of non-verbal mental age (which included items such as placing pegs in holes, matching colours and completing form boards) and were similar in overall developmental age. The groups were also similar on their daily living skills on the Vineland Adaptive Behaviour Scale. It was decided not to match the groups on language because communication problems are intrinsic to autism, and the behaviour being investigated in this study was likely to be part of any measure of early language and communication. However, correlat ions with language ability are explored and presented. As Table 1 shows, the groups were significantly different in their language abilities as measured by the BSID and by their production vocabulary (including signs) on the MacArthur Communicative Development Inventory. The production vocabulary of the MacArthur Inventory was also used for exploring correlations with other behaviours. The groups were not matched on chronological age (CA), and, although the age range was similar, the children with autism had a significantly higher CA, implying a lower developmental quotient. Children with DS may be more skilled in laughter and humorous interactions than children with other developmental disorders, suggesting a need for caution in interpretation of relative deficits in autism. However, the results are also compared with normative indications from other studies about typical development.

Design and procedure

A cross-sectional design was adopted with two to three visits (about 2 weeks apart) made to the home of each child. This study was part of a larger project including a longitudinal study of typically developing infants. The procedure during the visits consisted of a semi-structured parental interview, a parent questionnaire about language production (the MacArthur Communicative Inventory), psychometric tests (Bayley Scales of Infant Development, Vineland Adaptive Behaviour Scale), free play between parents and child and play-with-toys with and without the parents. The procedure was designed to be naturalistic and flexible to allow for difficulties in studying children with these disabilities. The free play and play-with-toys sessions were videotaped with a small hand-held camcorder, and the parental interview was audiotaped.


Parental reports about children's humour

Parents were interviewed on the first of two visits, given detailed instructions about the kinds of events to watch out for and the kinds of contextual details to include, and then given dictaphones for recording any incidents of relevance to the study that may occur in the gap between visits. Both in the parental interview and in their dictaphone recordings, parents were pressed to recall (or observe and report) specific incidents including contextual, expressive and background details rather than vague generalized responses to the questions. The interview was adapted from Bates et al. (1979) and used in a similar study of children's teasing based on parental reports (Reddy, 1991). Questions on humour were adapted from the experimental findings of Sroufe and Wunsch (1972). Key questions from the interview are shown in the following coding sections. The interview and dictaphone recordings were coded by the first author and a research assistant. The transcripts from 18 families (half from each group, represent ing 50% of the data) were double-coded, and satisfactory reliability measures obtained (K values ranged from .76 to 1.00, all significant at p <.001).

Coding parental reports

Types of events eliciting children's laughter

Parents were first asked an open-ended question: 'What sorts of things does X normally find funny or laugh at?' and were then questioned in detail about specific elicitors adapted from a list of the four main types of stimulus events and common examples identified by Sroufe and Wunsch (1972). A positive response to any one example within a 'type' was coded as affirmative for that type of event. Tactile events included tickling, rough and tumble play, blowing raspberries on the tummy and any other such physical contact; auditory events included funny sounds made by people and mechanical noises; visual events included funny faces and slapstick--usually on television; socially inappropriate acts included acts by others violating conventional codes of behaviour, such as putting a potty on the head or drinking from the child's beaker.(1)

Attempts to join in with others' laughter

Parents were asked whether their child ever laughed, or tried to join in with, others' laughter in his/her presence even when he/she did not understand what it was about. They were also asked whether their child had different types of laugh and whether he/she ever showed any 'artificial' or 'false' laughter. The contexts of occurrence of such laughter were discussed and coded.


Parents were asked whether their children ever attempted to deliberately re-elicit other people's laughter by repeating an act that had earlier--perhaps even accidentally--provoked laughter. Affirmative responses were coded as 'Yes' even if they were not supported by a clear example of a specific incident or it is not entirely clear that the child had indeed watched the other for a response before, during or after the repetition. This cautious coding was undertaken to run counter to the predictions in the present study. The number of incidents (a particular act repeatedly used to elicit laughter constituted one incident) reported per child was coded.

Teasing by the child

Three specific kinds of teasing were investigated (developed from Reddy, 1991) following initial open-ended questions about whether the child ever did anything to playfully tease others in any way; these were: teasing with offer and withdrawal of an object, teasing with provocative non-compliance, and teasing with disruption of routines or acts. Parents were given a number of examples of behaviour to aid memory and were prompted by the use of a number of overlapping terms such as 'joking', 'mucking about', 'being playful', 'being cheeky'. On each of the three types, teasing was coded as either 'Absent' (the child never teased in this way), present in a 'Game Only' (if the child did tease in this way but only within the context of a familiar game or in direct response to an initiation by the other) or 'Clearly Present' (if the child did tease not just within a game and not just in response). Reported incidents were probed for details about the history of the incident, including reactions from others (positive, negative or variable), frequency (never, rare, occasionally, regularly, frequently) and the child's emotional attitude during the teasing act (playful or sober/watchful).

Teasing by the parents

Parents were asked whether they ever playfully teased the child, and were given examples of the kinds of teasing that parents may sometimes do in order to help them remember. These were playful offer and withdrawal of object, making deliberate mistakes, blocking or obstructing the child's acts, and jumping out at the child. Children's responses to such teasing were described by the parents.


Clear responses were not always obtained from all parents in response to all questions, either due to problems with audibility or due to occasional conversational meandering. All group frequencies from the parental reports were thus counted as percentages of the number of clear responses available in each group rather than of the N in each group. Non-parametric statistical analyses were used for testing the significance of group differences and correlations.

Laughter by the children

There were no significant differences between the autism and DS groups in overall frequency of laughter reported by the parents. All of the children were reported to laugh at times, and most of the children were reported typically to laugh at least a few times per day. There were, however, some differences in what the children were reported to generally find funny or laugh at, as shown in Table 2.

All the children in both groups were reported to laugh at and enjoy tactile events, which include tickling, rough and tumble play and chasing, as well as more subtle forms of tactile contact such as being licked on the neck and being blown at. All but two parents (one from each group) spontaneously mentioned that the children loved being tickled. Similarly, there was no significant difference between groups in reported laughter to visual events; with the exception of a few of the children with autism, all children were reported to laugh at these, including slapstick and silly movements made by others or on the television. A greater proportion of the children with autism were reported not to laugh at auditory events (i.e. funny sounds made by people), but the difference was not significant. The greatest difference between groups was in laughter reported to socially inappropriate acts. None of the children with autism were reported to laugh at such acts (one parent was unsure, and six parents did not respond cl early), while 50% of the children with DS reportedly did so (three parents were unsure, and two did not respond clearly). In the DS group, laughter at socially inappropriate acts was related to various developmental test scores as shown in Table 1.

One of the direct probes yielded a large and surprising difference between groups. In response to the direct question 'Does he/she laugh at funny faces?', a significantly greater number of the parents in the autism group than in the DS group said 'No'. In the DS group, but not in the autism group, a 'Yes' response was significantly related to the developmental test scores. The largest group difference emerged from one of the spontaneous responses to the initial question 'What sorts of things does he/she laugh at?' All but one of the parents in the autism group spontaneously mentioned that the children laughed at strange or odd things, at odd times or at incomprehensible or inappropriate stimuli (responses were unclear or inaudible for two parents). In the DS group, only one parent mentioned anything close to this category, and this was laughter at 'horrible things' like being smacked or watching someone being hurt. This was unrelated to the developmental test scores in either group.

Children's laughter at/with others' laughter

There were significant group differences in responses to the question and related probes 'Does she/he laugh or try to join in when others are laughing, when she/he doesn't understand what it's about?' In the autism group, only five of the children (out of 18 whose parents were able to offer a clear reply) were reported to do so, whereas in the DS group, 12 out of 15 were reported to do so ([chi square](1) = 9.93, p < .003). Several parents in both groups reported (in response to a direct question) that their children sometimes put on an artificial or false laugh. The contexts for this kind of laughter were reported to be different in the two groups. In the autism group, artificial laughs were reported to occur mainly in non-social situations or in (sometimes echoic) imitation of others' laughter. In the DS group, it was reported to occur either in imitation or in order to elicit attention or laughter from others.

Trying to make others laugh--clowning

There were significant differences between groups in the numbers of children reported to attempt to re-elicit others' laughter by repeating acts that had previously elicited laughter. Even when doubtful instances of clowning were included, only five of the 19 children in the autism group were reported to have engaged in any clowning, whereas in the DS group, 13 of the 16 children were reported to do so (Table 3).

The reported incidents of clowning in the autism group were at a simpler level, largely involving repetition of imitated phrases such as 'You must be joking' or 'I'm tired' or involving the child tickling others, and the number of incidents reported per child was significantly lower (Mann-Whitney U(19,16) = 43.0,p < .0002). In the DS group, the incidents were varied, ranging from putting inappropriate objects such as pants on the head, imitating the mother doing her exercises behind her back to make the sibling laugh, pretending to tickle, to almost falling over. In the autism group, four of the 14 children who were reported to never engage in clowning were reported, none the less, to engage in some form of showing-off (i.e. repeating acts or phrases to get attention, either clever things in order to obtain praise, silly things in order to be noticed, or naughty things in order to provoke negative attention). The five children with autism who engaged in some clowning all showed some form of showing-off. In th e DS group, one of the three children who did not clown was reported to show-off, and all the 13 children who were reported to clown were also reported to show-off. This suggests that the failure to engage in clowning was not simply due to a global inability to link their own acts with others' reactions; rather, it could be a particular lack of interest in laughter. However, clowning was related to developmental age, to non-verbal mental age and to language age in both the autism and the DS groups, suggesting that a general developmental factor may underpin the occurrence of clowning.

Trying to provoke others--teasing

There were similar significant group differences in the frequency of teasing reported by parents, although teasing was more likely to be present in children with autism than was clowning. Eight of the 19 children in the autism group were reported to regularly (at least once a week) engage in at least one of the forms of teasing, compared with 13 of the 16 children in the DS group. In both groups, the most common form of teasing was teasing by disruption of acts or routines, and the least common was teasing with offer and withdrawal of objects; none of the children in the autism group spontaneously engaged in the latter (Table 4).

All the children who engaged in clear clowning also engaged in some form of teasing (with the exception of one child in the DS group). In the autism group, 11 of the 19 children engaged in no clowning or teasing, whereas two of the 16 children in the DS group engaged in neither. Teasing (sum of scores on the three types of teasing) was unrelated to clowning in either the autism or the DS group. In the DS group, teasing was strongly related to the developmental test scores, but in the autism group, the relation to developmental test scores was only moderate.

Teasing with disruption involved two sub-categories: deliberate mistake and obstruction or blocking. The latter was more common (but not significantly so) than the former in both groups. The number and variety of incidents were greater in the DS group. In the autism group, teasing with disruption involved the following incidents--fiddling with the music when someone is listening, blocking or turning the television off when someone is watching and increasing the behaviour if the reaction is negative, aggravating a younger sibling, deliberately misnaming parts of the body or misreading name cards, putting both legs into one trouser leg, calling for no reason and then laughing. In the DS group, teasing with disruption involved the following incidents--obstructively playing the piano while someone else is playing, switching the hoover off when mother is hoovering, playfully refusing to perform, offering to hug then moving away laughing, going to kiss then biting the nose playfully, refusing to give the other a cu p of tea in a game then giving it laughing, calling mother by name and laughing, taking shoes or clothes off when mother is in a hurry, urgent fake calling and then laughing, displacing mother from her chair, hiding things, deliberately refusing things then asking for them, waiting till other approaches then running, and many deliberate mistakes in naming and in familiar puzzles and tasks.

Teasing with non-compliance in the autism group involved the following incidents--throwing things or making a noise and increasing the speed of the activity when told not to, grinning cheekily when told off for making a noise, banging the door or turning a chair over while looking for a negative reaction and laughing when told not to, turning television off and watching for reaction, taking clothes off. In the DS group, the number and variety of incidents were again much greater and involved the following--deliberately blowing raspberries at mother at inappropriate times (e.g. in a shop), playing up as soon as mother is on the phone, putting finger in nose repeatedly and watchfully, hitting the baby for effect on mother, refusing to eat and laughing, touching forbidden things watchfully, and so on.

Teasing with offer withdrawal was always accompanied by positive affect in the child and followed by positive responses from the recipient, but the other two forms of teasing involved a variety of combinations of affect. More of the children in the DS group (seven of the 13 who were reported to tease) were reported to at least sometimes display watchful or sober affect while teasing than were the children in the autism group (one of the nine who were reported to tease), but this difference was not significant. In the autism group, teasing was occasionally reported to persist or increase in the face of distress from the recipient (e.g. a younger sibling). This was never reported in the DS group, although defiance or increase in the face of parental anger was common. Reported parental attitudes to children's teasing were no different between groups: most parents reported positive feelings about teasing; there were slightly more variable (i.e. sometimes negative and sometimes positive) attitudes reported in the DS group than in the autism group.

Teasing by the parents also differed significantly between the two groups ([chi square](2) = 10.82, p < .005). Of the 18 parents in the autism group who gave clear responses to this question, 15 reported that they either never teased because the child would not understand it or would throw a tantrum, or did so only through simple games such as jumping out from a hiding place or peekaboo. In the DS group, only four of 14 clear responses were in this category; 11 parents reported teasing such as mocking, deliberate mistakes such as reading the wrong story, stopping the child from getting out of the door, taking something as the child reaches for it and pretending to comply. More of the children with autism were reported to respond with blankness or uncomprehending distress at parental teasing than the children with DS, but the difference was not significant. Although parent teasing and child teasing were moderately related overall (p(33) = .48, p < .006), they were unrelated within the groups.


Videotaped observations

Coding observations of laughter

The video-taped free play and play-with-toys sessions were analysed by two of the authors and reliability measures obtained. The lengths of the sessions varied between children, partly reflecting the time it took to complete the tasks in the study, and partly the mood and cooperativeness of the children. All the interaction sessions recorded in each family were coded rather than selected segments of tapes, in order to avoid selective bias. Mean session lengths were similar between groups (autism group: M = 52.42 min, SD = 17.92; DS group: M = 47.09 min, SD = 9.00; t(32) = 1.07, n.s.). All coded behaviours were analysed relative to time or relative to the frequency of particular superordinate categories of behaviour. Coding began with the identification of a laugh either from the child or from any other person present in the room when the child was also present. Repeated laughs by a person within the same event were counted as one episode in order not to inflate the frequency of laughter; the presence of repea ted laughs was separately sub-coded. Reliability coding was done on data from six children (three from each group), with 10% of the video material (11% of the laughter episodes) by two of the authors. Blind coding of the interactions could not be achieved. The resulting Cohen's KS ranged from .70 to .92, and in all cases p < .001. The two lowest K values (.70 and .74) were for the coding of 'duration of laughter' by the child and by others.

Identifying laugh episodes

A laugh episode was identified by the occurrence of a laugh from any person present in the room (i.e. the child, parent, sibling or investigator) and consisted of one or more laughs by one or more people about a particular topic content. For example, the following description consisted of two laugh episodes: Mother (M) laughs at child (C)'s action with a puppet; C then laughs and carries on playing (episode 1). M then looks up at the investigator and makes a comment about the child; both M and investigator laugh (episode 2). The following incident would be counted as three laugh episodes: M tickles the child, the child laughs, M tickles again, the child laughs again, M tickles again, and the child laughs again (episode 1, with the child's laugh coded as extended over turns, see below); M stops the tickling, turns to investigator and speaks for a few seconds, then turns to the child and starts tickling again, and the child laughs again (episode 2), then M picks up a hand puppet to initiate a peekaboo action an d the child laughs (episode 3).

Duration of laugh (for child and for other)

Brief: when the laugh was merely a chuckle or snort and lasted no more than two 'HA's (the typical expirations in a laugh, e.g. 'ha ha') in length; Medium: when the laugh lasted between three and five 'HA's; Long: when the laugh was six 'HA's or more; Extended: when the laugh lasted over more than one 'turn' by the same person (e.g. a laugh by X followed by an act by the other followed by another laugh by X).

Laugh started by child

When the first laugh in the episode was by C, even if prompted by an act of the other person. When adult and child laughed together, the onset of the two laughs was never simultaneous (although such a situation is technically possible and did in fact happen with the adults), and it was always possible to identify the person who laughed first.

Child laugh direction

Solitary laugh: when the laugh was either not during interaction or was not directed to another person, e.g. C walks round the room, picks up a car, and laughing, walks on without looking at any person in the room. Content non-shared: when the laugh occurred within an interactive situation (i.e. with C sitting near or looking at the other person even from across the room) but the referent of the laughter was not the focus of either the joint interaction or the adult, but of the child alone, and the referent is not then brought into joint focus after the laugh. For example, M and C are sitting together on the floor with M trying to get C to play with a doll, C picks up a brick, laughs and plays with it. Or M and C are sitting together, M looks up and talks to investigator, C laughs looking towards another part of the room, does not look at M; turns to toys and picks one up. To an act by other: when the laugh was within an interactive situation and occurred in response to an act by the other person; e.g. M tick les C and C laughs; or M says to C 'You're a cheeky monkey, aren't you?' and C laughs. To external target: when the laugh was within an interactive situation and occurred in response to an external referent that was either already within joint attention or was brought into joint attention by the child's gaze to the other during the laugh; e.g. M and C sitting together on the floor both looking at a box that suddenly pops open, C laughs; or C walks across the room away from M, pulling a toy car with a string; the string comes off the car, C laughs looking at it and then looks up at M.

Child response to other's laughter

No attention: when the child did not react with any attentional response to the other's laughter; Look: when the child looked up at the person laughing without a smile or laugh; Smile: when the child laughed immediately following the other's laughter, even if without gaze to the person laughing; Laughs: when the child laughed immediately following other's laughter, even if without gaze to the person laughing; Unclear: which included vocalizations or looks, smiles or laughs, which were not clearly responses marking the other's laugh, but acts continuing from before the laugh.

Other's laughter direction

Laughter directed to the child: occurring either during direct interaction with the child, or when the other directs the laugh to the child; e.g. C playing with a toy animal, M makes a comment about the animal to C and then laughs; or C and M playing with sunglasses, C puts them on and M laughs at him. Laughter not directed to the child: when the other is interacting with another person in the room but not the child, and the laugh is not directed to the child; e.g. M and C on floor, M watching C playing with doll, C bangs doll's head on cooker, M looks up at investigator and laughs; no interaction or gaze between M and C before or after the laugh.

Laugh during game

Child laughter was coded as during tactile game if it occurred during a physical contact such as tickling, rough and tumble or the repetition of a simple act of physical contact; simple game including peekaboo, give and take game with an object, playing with a ball, repetitive blocking, putting on or taking off a mask; complex game: including either joint pretend play, complex ritual or novel games specific to the family; e.g. M and C playing game with doll where each pats the doll who is 'crying'; M laughs when C makes crying sound on behalf of the doll.


Laughter by the children: Frequencies, content and context

Table 5 shows that although frequencies of laughter varied between children, there were no group differences relative to time. Only one child from the autism group did not laugh at all during the filming sessions, supporting parental reports that laughter was common during the day in both groups. There was also no difference between groups in the rate per hour of laughter started by the child or of laughter started by the child in interactive situations.

Children in both groups displayed some amount of solitary laughter with no significant difference between groups. Within interactive situations, the most common reason for laughter in both groups was an act by the other; laughter with non-shared content was significantly higher in the autism group although also shown in the DS group; shared laughter at an external target was virtually absent in the autism group and significantly higher in the DS group. The relative frequency of solitary and non-shared laughter was significantly higher in the autism group than the DS group (U(16, 17) = 77.5, p < .04). The frequency of non-shared content in interactive situations or of solitary laughter or of the combined measure of solitary and non-shared laughter did not correlate with any developmental indices in either of the groups or overall; nor did they correlate with incidence of teasing or clowning. The lengths of laughter by children in the two groups did not differ significantly, nor did the relative incidence of ga me contexts for laughter.

Laughter by others

Table 6 shows that the frequency of laughter by others in the presence of the child was slightly, but not significantly, higher in the DS group than in the autism group. This was the case even when excluding laughter by the investigators. However, others' laughter in the presence of the child was significantly less frequently directed to the child in the autism group than in the DS group (the difference was in the same direction but missed significance when the laughter of the investigators was excluded from the analysis). The mean lengths of others' laughter did not differ significantly between the two groups.

Children's responses to laughter by others

As shown in Table 6, the most common reaction to others' laughter was no attention; this was the case in both groups but significantly more marked in the autism group.

When the other's direction of laughter was taken into account, this group difference was significant when the other's laughter was directed to the child but not when it was not directed to the child. Looking or smiling (even if without gaze) at the person laughing, were significantly more common in the DS group than in the autism group. This was the case for laughter directed to the child (p <.01) as well as for laughter not directed to the child (p <.02). The likelihood of the child laughing in response to others' laughter was very low in both groups and higher in the DS group, although missing significance and not different whether laughter was directed to the child or not.


The present study offers a description of naturally occurring humorous interactions in the everyday lives of children with autism and with DS using parental interviews as well as observations of interactions between children and parents. These descriptions show that there are several different aspects to humorous interactions not all of which co-occur in all children. These aspects include: the individual's own humour and laughter at different sensory; interpersonal and socio-cultural events, responding to others' humour and laughter with interest, attempts to join in or attempts to re-elicit it through clowning and teasing. Almost all the children with DS, like typically developing infants by the end of the first year (Reddy, 1991; Sroufe & Wunsch, 1972; Trevarthen & Hubley, 1978), showed or were reported to show all of these aspects in their interactions, and their presence was, in general, related to standard developmental indices. However, the children with autism showed or were reported to show only some of these aspects, and even when they did occur, some of them correlated differently with standard developmental indices from those in the DS group.

It was clear from the parental reports and from the videotaped observations that humour was regularly present in some form in the lives of all the children and there were no group differences in the frequency of laughter in daily life. Other studies have also found that, in contrast to the view that autism involves an absence of emotional reaction, parents do perceive emotional reactivity in children with autism, even though overall amounts of positive affect are reported or observed to be less than in control groups (Capps, Kasari, Yirmiya, & Sigman, 1993; St James & Tager-Flusberg, 1994).

Similar to the findings by St James and Tager-Flusberg (1994), there were no group differences in children's reported laughter to rough and tumble play, tickling by others or slapstick, but there was a significant group difference in laughter at socially inappropriate acts (non-verbal incongruities in St James and Tager-Flusberg's study). However, the difficulty with appreciating social incongruity may derive not from difficulties in symbolic or metarepresentational skills at 18 months of age, as suggested by St James and Tager-Flusberg (1994), but from an earlier difficulty (typically achieved by the end of the first year) in picking up the normal social conventions surrounding objects and actions, reflected in deficits in functional as well as pretend play (Williams, Costall, & Reddy, 1999; Williams, Reddy, & Costall, 2001). Typical infants do, in fact, appreciate such acts by 11-12 months of age (Sroufe & Wunsch, 1972), but even the children with autism who had higher developmental ages were not reported t o laugh at such acts.

Other simple events that provoke laughter also differentiated the groups. The children with autism unexpectedly were reported not to laugh at funny faces, and this was not related to any of the developmental abilities in this group, although it was significantly related to developmental age, non-verbal mental age and language production in the DS group. This reported response may be a result of the difficulty that children with autism have with faces (Hobson, 1986a, b), rather than of a difficulty with humour per se. Similarly, the children with autism did not respond to funny sounds with laughter as much as did the children with DS, and this may reflect a general tendency to either not react at all or react strangely to sounds (Dahlgren & Gillberg, 1989; Osterling & Dawson, 1994).

Although there were no group differences in frequencies of children's laughter overall or in frequencies of joint laughter in which the child laughed first, there were significant differences in the interpersonal relevance of the laughter. These difficulties were present at two levels. First, there were difficulties with triadic humorous exchanges, difficulties that might be predicted given the general difficulties in autism with triadic attentional acts (Mundy, 1995; Mundy et al., 1993). The laughter of the children with autism more often involved content that was not shared with others even when within a face-to-face interactive situation, and was less often related to a distal target involving triadic attention than was the laughter of children with Down's syndrome. That is, both in responding to other people's laughter at an external target, and in their own laughter, the children with autism showed difficulties: they did not respond normally to others' laughter and themselves often laughed at private and /or unshared content. Strong evidence of the atypicality and unshareability of the laughter of children with autism came from frequent reported episodes of inappropriate or inexplicable laughter in the children with autism but not in the children with DS, supporting previous reports of immoderate laughter in autism (Asperger, 1952/1991).

Second, there were also difficulties with the interpersonal relevance of laughter at a dyadic level. That is, the problem was also evident in humour-related behaviours involving no sharing of a topic. Although laughter is an expression of positive affect that is normally hard to ignore in others in close proximity, the children with autism showed a marked lack of interest in it, neither looking up nor smiling or laughing in response to others' laughter even when it was directed to them. As one parent of a child with autism eloquently put it, her son did not 'have the emotion of laughter' and so could not join in others' laughter normally or respond to it with interest whether or not it was directed to him. Such a lack of response to laughter itself is unlikely to be due to joint (or triadic) attentional difficulties or even to an inability to perceive the goal directedness of others' attentional behaviour (Baron-Cohen, 1993). It is more likely to be a result of difficulties in mutual attention and mutual emot ional sharing, which are also problematic in autism (Gomez, 1996; Hobson, 1986a, b; Hobson & Lee, 1998; Kasari, Sigman, Mundy & Yirmiya, 1990; Mundy, 1995; Reddy, 1998; Trevarthen et al., 1996).

The absence of interest in the laughter of others was also evident in the reported absence of attempts to re-elicit it through clowning and teasing, both of which may be precursors of joint attention (Bakeman & Adamson, 1984; Mundy, Sigman, & Kasari, 1990; Reddy, 1998). When clowning was reported, it tended to be limited to one or two specific routines, lending support to the suggestion that whereas social behaviours in typical development often emerge as part of routines, social behaviours in autism emerge out of routines (Lord, 1993). The relative absence of clowning in the autism group was more marked than the absence of teasing, supporting a similar previous finding with older children with autism (Briscall, 1995). Several of the children with autism were reported to show teasing or some form of showing-off, but not clowning, suggesting that in these children, in addition to a problem with triangulating attention, there is a more specific problem in relating to specific reactions. In this case, others' la ughter is not responded to in the way that matched children with DS or typical infants before the end of the first year respond (Reddy, 1991). They are either not interested in, or unable to engage with, others' laughter sufficiently to desire to evoke it. Alternatively, it could be the case that while clowning involves a sharing of an emotion, teasing does not necessarily do so. Teasing could involve a discrepancy of emotion between teaser and teased, e.g. enjoying someone's discomfort. The reported absence of clowning in autism, therefore, could be due not necessarily to a problem with sharing laughter, but to a problem with sharing emotion generally.

The reported presence of teasing but not clowning cannot easily be explained through cognitive-representational explanations: teasing is more, rather than less, likely to involve mental state understanding and metarepresentational skills than is clowning. The finding that the children with autism engage in some triadic reaction-seeking acts, but not others, supports the general finding in the literature that, although children with autism can demonstrate skills, they do not normally do so without an instrumental purpose or additional persuasion (Curcio, 1978); this discrepancy has been shown in relation to pretend play (Lewis & Boucher, 1988), perspective-taking (Baron-Cohen, 1991) and language production and communication (Abrahamsen & Mitchell, 1990).

Although children with autism show no difficulty in engaging in some forms of humour and laughter-related interactions, they clearly have difficulties in this domain compared with typically developing infants by the end of the first year or to mental-age matched children with DS. The difficulties relate to behaviour typically manifested before the development of symbolic or metarepresentational skills when the specific conceptual deficit theories of autism suggest the problem may begin (Leslie, 1987). The problems involve simple interpersonal and affective aspects of humour rather than only cognitively complex aspects, suggesting support for affective-relatedness theories of autism (Hobson, 1989). The difficulties appear to relate both to the appropriate perception of humorous affect in others and to the appropriate expression of the affect themselves, as is the case with other kinds of affect (Mundy et al., 1993).

Interestingly, group differences in patterns of others' laughter in the presence of the children mirrored the patterns shown by the children. Although the overall frequency of others' laughter did not differ between groups, there were significant group differences in whether or not others' laughter involved the child; in the autism group, others' laughter was significantly more often not directed to the child than in the DS group. Humour is a fundamentally relational activity, and any disturbances in the affective and communicative aspects of humour in children with autism can be seen in the behaviour of all interactants involved. Such relational effects have often been shown in other affective-communicative domains such as in the behaviour of strangers interacting with infants of depressed mothers (Field et al., 1988), joint attentional behaviours by caregivers directed to children with autism--such as pointing and making eye contact (Kasari, Sigman, Mundy, & Yirmiya, 1988) and the directive behaviours of pa rents with children with Down's syndrome.

Solitary laughter, or laughter whose content is not shared with others, clearly does sometimes occur. However, such laughter is generally perceived as unusual and some-how abnormal by those present. This may be the case not only for children, but also for adults when others present cannot imagine a reason for the laughter. The difficulties with sharing laughter and humour with others in young children with autism suggest that the implicit violations of cultural rules in humour (Eco, 1986) may require a degree of participation in culture that is too difficult for these children at this age (Baron-Cohen, 1993; Loveland, 1991; Tomasello, 1993). Not being able to perceive the same affordances for humour as do others and not being able to share or desire others' laughter can have negative implications for the development of relationships, of safe contexts for further development and for further cultural participation. In autism, this could lead to a serious secondary handicap by inhibiting the furthering of intima cy and destruction of fear and distance in a way that few other interactions than humour can achieve (Alford, 1983; Bakhtin, 1981; Lowenfield, 1935).

Interventions for children with autism that use humour and laughter clearly have a strong basis to work from inasmuch as all the children with autism do laugh and can respond appropriately to immediate sensory stimulation and to some humorous events. In particular, the high percentage of child laughter (with no difference between groups) in response to acts by others is a cause for optimism, reflecting parental skill at eliciting laughter and/or the children's interest in some of the things people do. However, laughter may usefully be used as a starting point for intervention only if there is a deeper understanding of the problems surrounding laughter in autism. The fundamentally affective and relational nature of the difficulties in sharing laughter needs to be appreciated and built into any intervention package before it can have a chance of working.

In conclusion, the present study suggests that humour and laughter are important aspects of early affective and socio-cognitive development. Young children with autism show difficulties in this domain compared with matched children with Down's syndrome and also compared with previous reports of infants within the first year (Reddy, 1991; Sroufe & Wunsch, 1972). These difficulties need to be explored further and could provisionally be added to the list of diagnostic criteria that are identified as pertinent to the autistic spectrum. They have the advantage of being prominent in their presence or absence in everyday life and are identifiable by parents with a little training. The findings suggest support for theories of early affective disturbances in autism. Like many other communicative behaviours, humour is evidently fundamentally relational both as phenomenon and as a problem.
Table 1

Characteristics of the children in the two groups

                                 M (SD) Min -- Max

Number (boys, girls)              19 (15, 4)
Chronological age                49.6 (9.00) 36 -- 65

BSID measures
Non-verbal mental age            24.1 (10.37) 8 -- 42
Developmental age                16.1 (4.76) 7 -- 26
Verbal mental age                12.86 (7.06) 0 -- 22

Vineland Daily Living Skills     18.89 (6.29) 13 -- 35

MacArthur production (including  11.32 (8.71) 0 -- 27

                                 Down's syndrome
                                 M (SD) Min -- Max      t

Number (boys, girls)             16 (9, 7)
Chronological age                41.3 (12.16) 18 -- 65  2.31 *

BSID measures
Non-verbal mental age            24.4 (9.33) 9 -- 42    .07
Developmental age                17.9 (5.17) 8 -- 25    1.05
Verbal mental age                23.00 (10.54) 6 -- 42  3.1 **

Vineland Daily Living Skills     20.06 (5.73) 10 -- 30  .57

MacArthur production (including  17.88 (5.34) 7 -- 26   2.73 *

* p<.05;

** p<.01;

*** p<.0001.
Table 2

Percentage of children in each group reported with certainty to laugh at
various events

                       Tactile    Visual    Auditory
Group                  events     events     events

Autism                  100% (c)   74% (d)    25% (d)
DS                      100% (c)   93% (d)    60% (d)
[chi square] (d.f.) p   n.s.       n.s.       n.s.

Group                  inappropriate acts   Funny faces

Autism                    0% (c)           16% (d)
DS                       50% (a)           73% (b)
[chi square] (d.f.) p     6.36 (1) <.02     9.21 (1) <.003

                       No apparent
Group                    reason

Autism                 88% (c)
DS                      6% (c)
[chi square] (d.f.) p  19.18 (1) <.001

Clear responses were not always available for these items; unclear
responses were excluded, and percentages of 'Yes' responses were
calculated on the number of clear responses for each item. Tactile
events referred to physical contact such as tickling and rough and
tumble; visual events referred to watching slapstick or any other visual
'stimuli' including funny faces; auditory events referred to funny
voices or sounds; socially inappropriate events referred to violations
of conventional codes such as putting a potty on the head; funny faces
referred to others making funny facial expressions; no apparent reason
referred to laughter that the parents could not understand the reason

Correlation with developmental test scores

                                             MacArthur (production
   Developmental age  Non-verbal mental age  vocabulary)

(a)  .77 ***          .69 **                 .63 *

(b)  .77 ***          .66 **                 .62 *

(c) Correlations could not be computed

(d) No significant correlations with any developmental test scores

* p < .05

** p < .01

*** p < .001.

Note. Spearman's correlations were conducted with Ns of between 13 and
Table 3

Number of children in each group reported to engage in clowning

Group                             Any reported clowning

Autism                                      5
DS                                         13
[chi square](1) = l0.49, p <.001

Group                             No clowning reported

Autism                                     14
DS                                          3
[chi square](1) = l0.49, p <.001

Any reported clowning: any instances reported by parents of repetition
of acts for re-eliciting laughter, including instances where the reason
for the act was in some doubt, or where it was unclear whether the child
actually watched the parent for a response.

No clowning reported: no clowning reported by parents at all.

Correlations with developmental test scores

                                                   MacArthur (production
         Developmental age  Non-verbal mental age   vocabulary)

Autism:  .74 ***            .55 *                  .60 **

DS:      .51 *              .56 *                  .56 **

* p <.05

** p <.01

*** p <.001.

Note. Spearman's correlations were conducted with Ns of between 15 and
Table 4

Number of children in each group reported to engage in three forms of

                                 Offer and withdrawal     Provocative
                                       of object          non-compliance

Group                           Yes       Game        No        Yes
Autism                           0          2         17          4
DS                               9          3          4         11
[chi square] (d.f.) p                17.12.(2),<.001

                             Provocative         Disruption of
                                                    acts or
                            non-compliance        or routines

Group                      Game      No            Yes
Autism                      3        12             6
DS                          2         3             13
[chi square] (d.f.) p  8.67(2),<.02

                        Disruption of acts or
                             or routines

Group                      Game      No
Autism                      1        12
DS                          0         3
[chi square] (d.f.) p  8.79(2),<.02

Yes: reported to tease in this way spontaneously (i.e. without immediate
initiation by other) and outside a regular game.Game: reported to tease
in this way, but only in response to an immediate request or initiation
by other, or within a regular game.

No: reported to never tease in this way.

Correlations between total teasing (sum of scores on the three types)
and developmental test scores

                                                   MacArthur (production
         Developmental age  Non-verbal mental age   vocabulary)

Austim:  .40                .54 *                   .47 *
DS:      .67 **             .56 *                   .75 ***

* p<.05;

** <p.01;

*** p<.001.
Note. Spearman's correlations were conducted with Ns of between 15 and
Table 5

Relative frequencies of laughter by children: mean (SD)

                             Solitary laughs
           Total laughs    (percentage of total
Group     (rate per hour)    laughs by child)

Autism      11.6 (12.8)         16% (17.9)
DS          12.1 (9.4)          11% (25)

                    Laughter in interactive situations
                    (percentages of laughs by child in
                         interactive situations)

                    Content            To act by   To external
Group            non-shared *            other       target *

Autism             35% (31)            62% (28.9)   3% (9)
DS                13% (19.5)           68% (27.8)  19% (23.9)

* p <.05

** p <.01

*** p <.001.

Note. Mann-Whitney U tests were conducted with group Ns of between 15
and 18.
Table 6

Frequencies of others' laughter and child responses: mean (SD)

                      Others' laughter                Child responses to
                                                        others' laughter
                              Percentage directed
Group     Laughs per hour         to child *        No attention ***

Autism       42 (21.4)             21% (16)              74% (16)
DS           49 (25.7)             36% (19)              47% (24)

           Child responses to others'
           Looking up or
Group      smiling ***       Laughing

Autism        13% (7)        1% (1.6)
DS            30% (14)       2% (3)

* p<.05;

** p<.01;

*** p<.001.

Note. Mann--Whitney U tests were conducted with group Ns of between 15
and 18.


We would like to thank all the parents and children who participated with such patience and enthusiasm in this study. Their support was vital for the research. The study was funded by Grant No. R000235481 from the Economic and Social Research Council.


Received 12 July 1999; revised version received 6 March 2001

(1.) The last two categories differ from Sroufe and Wunsch (1972); the latter classed socially inappropriate and absurd acts as visual stimuli, and classed visually presented events such as funny faces as social stimuli. In order to avoid the ambiguity of these categories, the present categorization involved three modality-specific categories (all of which are social at some level) and one category for socially inappropriate acts.

Abrahamsen, E., & Mitchell, J. (1990). Communication and sensorimotor functioning in children with autism. Journal of Autism and Developmental Disorders, 20, 75-85.

Alford, K. (1983). Privileged play: Joking relationships between parents and children. In F E. Manning (Ed.), The world of play (pp. 174-189). New York: Leisure Press.

Ambrose, A. (1963). The age of onset of ambivalence in early infancy: Indications from the study of laughing. Journal of Child Psychology and Psychiatry, 4(3-4), 167-181.

Asperger, H. (1952/1991). "Autistic psychopathy" in childhood. In U. Frith (Ed.), Autism and Asperger syndrome. Cambridge: Cambridge University Press.

Bakeman, R., & Adamson, L. (1984). Co-ordinating attention to people and objects in mother infant and peer-infant interaction. Child Development, 55, 1278-1289.

Bakhtin (1981). The dialogic Imagination (p. 23). Austin, TX: University of Texas Press.

Barnett, L. A. (1991). Characterising playfulness: Correlates with individual attributes and personality traits. Play and Culture, 4, 371-393.

Baron-Cohen, S. (1989). The autistic child's theory of mind: A case of specific developmental delay. Journal of Child Psychology and Psychiatry, 30, 285-298.

Baron-Cohen, 5. (1991). Precursors to a theory of mind: Understanding attention in others. In A. Whiten (Ed.), Natural theories of mind (pp. 233-251). Oxford: Blackwell.

Baron-Cohen, S. (1993). Are children with autism acultural? Behavioural and Brain Sciences, 16, 512-513.

Bates, E., Benigni, T., Bretherton, I., Camaioni, L, & Volterra, V. (1979). The emergence of symbols: Cognition and communication in infancy. New York: Academic Press.

Bergson, H. L. (1911). Laughter: An essay on the meaning of the comic. New York: Macmillan.

Briscall, L. (1995). A study of teasing, clowning, showing-off and humour in autistic and Asperger's syndrome children. Unpublished BSc dissertation, Department of Psychology, University of Portsmouth.

Bruner, J., & Sherwood, V. (1976). Early rule structure: the case of peekaboo. In J. Bruner, A. Jolly, & K. Sylva (Eds.), Play: Its role in evolution and development (pp. 277-285). Harmondsworth: Penguin.

Capps, L, Kasari, C., Yirmiya, N., & Sigman, M. (1993). Parental perception of emotional expressiveness in children with autism. Journal of Consulting and Clinical Psychology, 61, 475-484.

Chapman, A. J. (1976). Social aspects of humorous laughter. In A. J. Chapman & H. C. Foot (Eds.), Humor and laughter: Theory, research and applications (pp. 155-185). New Brunswick, NJ: Transaction.

Cicchetti, D., & Sroufe, L. A. (1976). The relationship between affective and cognitive development in Down's Syndrome infants. Child Development, 47, 920-929.

Crawford, M. (1999). To be perfectly frank. BBC2, 19 June 1999.

Curcio, F (1978). Sensorimotor functioning and communication in mute autistic children. Journal of Autism and Childhoocl Schizophrenia, 2, 1-8.

Dahlgren, 5. 0., & Gillberg, C. (1989). Symptoms in the first two years of life: A preliminary population study of infantile autism. European Archives of Psychiatry and Neurological Sciences, 238, 169-174.

Darwin, C. (1872/1998). The expression of the emotions in man and animals. London: Harper Collins.

Dunn, J. (1988). The beginnings of social understanding. Cambridge, MA: Harvard University Press.

Eco, U. (1986). The comic and the rule. In Travels in hyperreality. London: Pan Books.

Field, T. N., Healy, B., Goldstein, S., Perry, S., Bendell, D., Schanberg, S., Zimmerman, E. A., & Kuhn, C. (1988). Infants of depressed mothers show "depressed" behaviour even with non depressed adults. Child Development, 59, 1569-1579.

Freud, 5. (1905). Jokes and their relation to the unconscious. Harmondsworth: Penguin Books.

Garvey, C. (1977). Play. Cambridge, MA: Harvard University Press.

Glenn, P. (1989). Initiating shared laughter in multi-party conversations. Western Journal of Speech Communication, 53, 127-149.

Gomez, J. C. (1996). Second person intentional relations and the evolution of social understanding. Behavioural and Brain Sciences, 19(1), 129-130.

Groos, K. (1901/1976). The play of man: Teasing and love play. In J. Bruner, A. Jolly, & K. Sylva (Eds.), Play: Its role in evolution and development (pp. 68-83). Harmondsworth, UK: Penguin.

Handelman, D. (1981). The ritual clown: attributes and affinities. Anthropos, 76, 321-370.

Hobson, R. P. (1986a). The autistic child's appraisal of expressions of emotion. Journal of Child Psychology and Psychiatry, 27, 321-342.

Hobson, R. P. (1986b). The autistic child's appraisal of the expressions of emotion: A further study. Journal of Child Psychology and Psychiatry, 27, 671-680.

Hobson, R. P. (1989). Beyond cognition: A theory of autism. In G. Dawson (Ed.), Autism: Nature, diagnosis and treatment (pp. 22-48). New York: Guilford.

Hobson, R. P., & Lee, A. (1998). Hello and goodbye: A study of social engagement in autism. Journal of Autism and Developmental Disorders, 28(2), 117-126.

Johnson, R. (1978). Jokes, theories and anthropology. Semiotica, 22(3/4), 309-334.

Kanner, L. (1943). Autistic disturbances of affective contact. Nervous Child, 2, 217-250.

Kasari, C., Sigman, M., Mundy, P., & Yirmiya, N. (1988). Caregiver interactions with autistic children. Journal of Abnormal Child Psychology, 16(1), 45-56.

Kasari, C., Sigman, M., Mundy, P., & Yirmiya, N. (1990). Affective sharing in the control of joint attention interactions of normal, autistic and mentally retarded children. Journal of Autism and Developmental Disorders, 20, 87-100.

Koestler, A. (1964). The act of creation. New York: Dell.

Leekam, S. (1991). Jokes and lies: Children's understanding of intentional falsehood. In A. Whiten (Ed.) Natural theories of mind (pp. 159-174). Oxford: Blackwell.

Leslie, A. (1987). Pretense and representation: The origins of "theory of mind". Psychological Review, 94, 412-426.

Lewis, V., & Boucher, J. (1988). Spontaneous, instructed and elicited play in relatively able autistic children. British Journal of Developmental Psychology, 6(4), 325-339.

Lord, C. (1993). The complexity of social behaviour in autism. In S. Baron-Cohen, H. Tager-Flusberg, & D. Cohen (Eds.), Understanding other minds: Perspectives from autism (pp. 292-316). New York: Oxford University Press.

Loveland, K. (1991). Social affordances and interaction II: Autism and the affordances of the human environment. Ecological Psychology, 3, 99-120.

Lowenfield, M. (1935). Play in childhood. London: Victor Gollancz.

Luquet, G. H. (1930). Le rire dans les legendes oceaniennes. Journal de Psychologie, 27, 268-288.

McGhee, P. (1979). Humor: Its origin and development. San Francisco: W.H. Freeman.

McGhee, P. (1980). Development of the sense of humor in childhood: A longitudinal study. In P. E. McGhee & A. J. Chapman (Eds.), Children's humor (pp. 213-236). New York: Wiley.

Miller, P. (1986). Teasing as language socialisation and verbal play in a white working-class community. In B. B. Schieffelin & E. Ochs (Eds.), Language socialisation across cultures (pp. 199-212). Cambridge: Cambridge University Press.

Mundy, P. (1995). Joint attention and socio-emotional approach behaviour in children with autism. Development and Psychopathology, 7, 63-82.

Mundy, P., Sigman, M., & Kasari, C. (1990). A longitudinal study of joint attention and language development in autistic children. Journal of Autism and Developmental Disorders, 20, 119-128.

Mundy, P., Sigman, M., & Kasari, C. (1993). The theory of mind and joint attention deficits in autism. In S. Baron-Cohen, H. Tager-Flusberg, & D. Cohen (Eds.), Understanding other minds: Perspectives from autism (pp. 181-203). New York: Oxford University Press.

Nakano, S., & Kanaya, Y. (1993). The effects of mothers' teasing: Do Japanese infants read their mothers' play intentions in teasing? Early Development and Parenting, 2(1), 7-17.

Norris, D. (1971). Crying and laughing in imbeciles. Developmental Medicine and Child Neurology, 13(6), 756-761.

Osterling, J., & Dawson, G. (1994). Early recognition of children with autism: A study of first birthday home videotapes. Journal of Autism and Developmental Disorders, 24(3), 247-257.

Pawluk, C. (1989). The social construction of teasing. Journal for the Theory of Social Behaviour, 19(2), 145-167.

Perner, J. (1991). Understanding the representational mind. Cambridge, MA: MIT Press.

Provine, R. R. (1992). Contagious laughter: Laughter is a sufficient stimulus for laughs and smiles. Bulletin of the Psychonomic Society, 30(1), 1-4.

Reddy, V. (1991). Playing with others' expectations: teasing and mucking about in the first year. In A. Whiten (Ed.), Natural theories of mind. Oxford: Blackwell.

Reddy, V. (1998). Joining attention in infancy. Paper presented at the HREB Workshop on Joint Attention, University of Warwick, March 1998.

Reddy, V. (1999). Prelinguistic communication. In M. Barrett (Ed.), The development of language. Hove: Psychology Press.

Reddy, V., Hay, D., Murray, L., & Trevarthen, C. (1997). Communication in infancy: Mutual regulation of affect and attention. In G. Bremner, A. Slater, & G. Butterworth (Eds.), Infant development: Recent advances (pp. 247-273). Hove, UK: Psychology Press.

Shultz, T. R. (1976). A cognitive-developmental analysis of humor. In A. J. Chapman & H. C. Foot (Eds.), Humor and laughter: Theory, research and applications (pp. 11-36). New York: Wiley.

Sroufe, L. A., & Wunsch, J. P. (1972). The development of laughter in the first year of life. Child Development, 43, 1326-1344.

St James, P., & Tager-Flusberg, H. (1994). An observational study of humor in children with autism and Down's Syndrome. Journal of Autism and Developmental Disorder, 24(5), 603-617.

Stern, D. (1985). The interpersonal world of the infant. New York: Basic Books.

Stone, W., Lee, E. B., Ashford, L., Brissie, J., Hepburn, S. L., Coonrod, E. E., & Weiss, B. H. (1999). Can autism be diagnosed accurately in children under three years? Journal of Child Psychology and Psychiatry, 40(2), 219-226.

Tomasello, M. (1993). Cultural learning. Behavioural and Brain Sciences, 16, 495-552.

Trevarthen, C., Aitken, K., Papoudi, D., & Robarts, J. (1996). Children with autism: Diagnosis and interventions to meet their needs. London: Jessica Kingsley.

Trevarthen, C., & Hubley, P (1978). Secondary intersubjectivity: Confidence, confiding and acts of meaning in the first year. In A. Lock (Ed.), Action, gesture and symbol (pp. 183-229). London: Academic Press.

Van Bourgondien, M., & Mesibov, G. (1987). Humor in high-functioning autistic adults. Journal of Autism and Developmental Disorders, 17, 417-424.

Washburn, R. W. (1929). A study of the smiling and laughing of infants in the first year of life. Genetic Psychology Monographs, 6, 403-537.

Wetherby, A. M., & Prutting, C. A. (1984). Profiles of communicative and cognitive-social abilities in autistic children. Journal of Speech and Hearing Research, 27, 364-377.

Williams, E., Costall, A., & Reddy, V. (1999). Children with autism experience problems with both objects and people. Journal of Autism and Developmental Disorders, 29, 367-378.

Williams, E., Reddy, V., & Costall, A. (2001). A closer look at functional play in children with autism. Journal of Autism and Developmental Disorders, 31, 67-77.

Wing, L. (1991). The relation between Asperger's Syndrome and Kanner's autism. In U. Frith (Ed.), Autism and Asperger syndrome (pp. 93-121). Cambridge: Cambridge University Press.

Wing, L., & Gould, J. (1979). Severe impairments of social interaction and associated abnormalities in children: Epidemiology and classification. Journal of Autism and Developmental Disorders, 9(1), 11-29.

Wolfenstein, M. (1954/1978). Children's humour: A psychological analysis. London: The Free Press.

Zigler, B., Levine, J., & Gould, L (1966). Cognitive processes in the development of children's appreciation of humour. Child Development, 37, 507-518.

Vasudevi Reddy *

* Requests for reprints should be addressed to Vasudevi Reddy, Department of Psychology, University of Portsmouth, King Henry Bui1ding, King Henry 1st Street, Portsmouth PO1 2ER, UK (e-mail;
COPYRIGHT 2002 British Psychological Society
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2002 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Reddy, Vasudevi; Williams, Emma; Vaughan, Amy
Publication:British Journal of Psychology
Article Type:Statistical Data Included
Geographic Code:4EUUK
Date:May 1, 2002
Previous Article:Sex-typed preferences in three domains: do two-year-olds need cognitive variables?
Next Article:Misremembering Bartlett: a study in serial reproduction.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters